se one Hi eave gh Sarre Me meee aa : : cherageset sete Sree r of etngee nee syria ‘ : hag ete rae Pope as? Sr pene be RES ae ero teat arate r Perey oie im oy wwe ry te fest : a went x a 210 SA sets Scottish Birds THE JOURNAL OF THE SOC Vol 24 No 1 June 2003 Mandarin Ducks and the potential consequences for Goldeneye Scottish Chough Census 2002 Scotland’s endemic subspeciess Eurasian Reed Warblers in Scotland Scottish Birds - The Journal of the SOC Editor: Dr S R D da Prato Assisted by: Dr I Bainbridge, Professor D Jenkins, Dr M Marquiss, Dr J B Nelson and R Swann. Business Editor: Admin Officer, The SOC, Harbour Point, Newhailes Road, Musselburgh, EH21 6SJ. Scottish Birds, the official journal of the SOC, publishes original material relating to ornithology in Scotland. Papers and notes should be sent to The Editor, Scottish Birds, Harbour Point, Newhailes Road, Musselburgh, EH21 6SJ. Two issues of Scottish Birds are published each year, in June and December. Scottish Birds is issued free to members of the SOC, who also receive the quarterly newsletter Scottish Bird News, the annual Scottish Bird Report and the annual Raptor Round Up. Published by: The Scottish Ornithologists’ Club, Harbour Point, Newhailes Road, Musselburgh EH21 6SJ. Design by: Pica Design, 51 Charlton Crescent, Aboyne, Aberdeenshire AB34 SGN. Printed by: Meigle Colour Printers Ltd, Block 11, Tweedbank Industrial Estate, Galashiels TD1 3RS Scottish Birds (2003) 24: 1-10 Mandarin Ducks and the potential consequences for Goldeneye i Mandarin Ducks in northern Scotland and the potential consequences for breeding Goldeneye P COSGROVE During 2002, an exceptional number of non native Mandarin Duck sightings were reported in and around Strathspey, including at least 2 breeding attempts. Recent records from the north of Scotland were collated and suggest that Mandarins are increasing in number and have begun to colonise sites close to and inside the core British Goldeneye breeding area. The pattern and origin of these records is investigated and suggests that Mandarin are now migrating or dispersing more regularly into the north of Scotland during the spring. The potential consequences for breeding Goldeneyes in northern Scotland of the hole nesting Mandarin Ducks becoming established and competing for nest sites are considered and recommendations made. Introduction The Mandarin Aix galericulata is a colourful non native duck from eastern Asia that was first introduced to Britain in the 19th century. Since then the species has been popular in wildfowl collections and has bred ferally (from both escapes and deliberate releases) for many years before being admitted to the official British and Irish list in 1971 (BOU 1971). The Mandarin nests chiefly in tree holes up to 10m above the ground, but also readily takes to elevated nest boxes (Owen et al 1986). In Britain, Mandarins compete for nesting sites with such hole nesting species such as Jackdaws Corvus monedula, the introduced Grey Squirrel Sciurus carolinensis (Lever, in Wernham et al 2002) and apparently Goosanders Mergus merganser and Tawny Owls Strix aluco (Anderson and Petty 1996). The distribution of Mandarins in the British Isles is primarily in central and southern England, with 3 Scottish populations on the River Tay outside Perth (c100 birds, but possibly declining recently) and on the Eye Water in Berwickshire (c50 birds) in 1993 (Lever, Gibbons et al 1993). Petty and Anderson (1994) recorded the first breeding records from Argyll between 1991 and 1993 in nest boxes erected for Tawny Owls around Loch Eck. The Argyll population has continued to expand with 25 ducklings hatching during 1992-1994. In 1995, 5 pairs laid 42 eggs from which 17 ducklings hatched and 4 pairs in 1996 laid 55 eggs and hatched 47 ducklings (Anderson and Petty 1996). Most records appear to refer to incidental sightings, or counts from specific well recorded water bodies. For example, the 1998 Scottish Bird Report (Murray 2000) refers to 2—3 birds (pairs?) breeding in Perth, 50 Glenbranter, Argyll and 18 breeding pairs at Loch Eck. In their native range, Mandarin Ducks are both migratory and dispersive. Some authors (eg Lever, in Wernham et al 2002) contend that Mandarins in Great Britain have lost their migratory instinct, although they do acknowledge a tendency for some seasonal dispersal and vagrancy. This has happened with other introduced/released wildfowl, eg Ruddy Duck Oxyura jamaicensis, Barnacle Goose Branta leucopsis and Canada Goose Branta canadensis (apart from the development of a moult migration) and is presumably a response to living somewhere which permits year round occupancy (Malcolm Ogilvie pers comm). Ringing recoveries of Mandarin have originated from several parts of the species’ British range, but most have arisen from ringing in the southern English stronghold between September and March. There are also some remarkable and swift long distance movements to and from the near continent (Lever, in Wernham et al 2002), where general movements tend to be longitudinal, rather than latitudinal. At an international level, more research into the movements and _ dispersal of European Mandarins might shed light on the alleged loss of instinct to migrate. Comparisons between the 2 BTO Breeding Atlases (Lever, in Gibbons et al 1993) suggest that there was a marked extension in the Mandarin Duck’s range and a 459% increase in numbers, but this level of expansion has been disputed by others such as Davies (1988) who believe that Mandarins were always more numerous than originally suspected due to the species’ shy and secretive nature leading to under recording. In 1988, the British Mandarin population was estimated to be c7,000 individuals using extrapo- lation from gathered evidence of distribution, counts and trapping (Davies 1988) and this figure was inexplicable translated into 3,500 pairs by Lever, in Gibbons et al (1993). Davies (1988) makes no statement regarding the time of year that the British extrapolated population is just under 7,000 individuals and to convert this to 3,500 pairs is not acceptable. Furthermore, these unsubstan- tiated figures have been used to argue that the British Mandarin population is of conservation significance because it is believed to be an important component of the total world population of 25,180 pairs (Lever, in Wernham et al 2002). Unfortunately, it not possible to ascertain the conservation significance of the British Mandarin population (whatever it is) because the global figure of 25,180 pairs quoted by Lever is SB 24(1) incorrect. The most up to date information on the global population size of Mandarin Ducks is of c65,000 pairs (Delany and Scott 2002) (Table 1). The principal reason for the decline of the Mandarin in its native range has been habitat loss due to widespread deforestation. However, recent figures from the eastern Palearctic suggest that the species has staged a recovery in its native area, where it is now classified as of least concern (Birdlife International 2000). Despite their gaudy colouration, Mandarins are surprisingly elusive and probably the most poorly monitored of all inland ducks species by WeBS counts (Cranswick et al 1999). Consequently, there are no systematic or reliable counts of this species over significant areas of Scotland. Furthermore, some observers do not even bother to count or submit their sightings of Mandarins because of the bird’s non native status. This lack of reliable data makes it difficult to determine Mandarin population trends in Scotland. Mandarin in Badenoch, Strathspey, Strathnairn and NE Scotland No Mandarins were recorded in Badenoch and Strathspey until 1985. Dennis (1995) lists 3 records of 4 birds since 1985 including a male at Loch Garten on 23-25 October 1985, a pair at Loch Vaa on 6 May 1990, then a single male there between 7-13 May 1990, and one at Newtonmore in April 1992. Mandarins have also been recorded in the spring in recent years on the Allt Mor, near Nethy Bridge (Roy Turnbull pers comm), where birds were released on an ornamental pond. However, during spring 2002 it became clear that an exceptional number of Mandarin sightings had been made in Strathspey. These sightings (Table 2 and Figure 1) suggest that at least 2 well travelled pairs of birds were prospecting for nest sites in Strathspey. Most of these sightings occurred on Scottish Birds (2003) wetland and water habitats used by breeding Goldeneyes. For example, Mandarin records are from near the main stem of the River Spey, lochs on the main Strath and small but relatively high sections of small tributaries (<10m wide). Nearby in the Highlands, escaped and released Mandarins, from local collections, have occurred in Strathnairn and Cannich since the 1970’s and 1980s (Roy Dennis pers comm). During the last decade, Mandarin have repeatedly bred at several localities in Strathnairn, the nearest of which is only 25km away from Strathspey. Regular monitoring by Ray and Val Collier has shown that at least 7 pairs bred in 2000, with variable numbers in other years (Table 3). A post breeding flock of 18 birds together in September 1999 suggests that a significant population of Mandarins has now become established in Strathnairn. It is perhaps surprising that Mandarin Ducks have been deliberately released because the species is listed Mandarin Ducks and the potential consequences for Goldeneye 2) on Schedule 9 (Part 1) of the Wildlife and Countryside Act 1981. Section 14 of the Act states: 14(1) Subject to the provisions of the Part, if any person releases or allows to escape into the wild any animal which (a) is of a kind which is not ordinarily resident in and is not a regular visitor to Great Britain in a wild state; or (b) is included in Part 1 Schedule 9, he shall be guilty of an offence. It is not clear where all the Mandarins in Badenoch, Strathspey and Strathnairn have come from. Some records can probably be attributed to local wildfowl collections, such as one on the Allt Mor, near Nethy Bridge, but not all records relate to local escapes or releases. A Dutch ringed Mandarin was recovered in Strathspey on 19 April 1998. It was hatched in Rotterdam Zoo in June 1991 and escaped in 1994, before making its own way to Scotland (Pete Moore pers comm). Figure 1 Locations of Mandarins in Strathspey in 2002. (Map courtesy of SNH) Other rivers i); River Spey Lochs A Settlements m Mandarin sightings Scale 1:100000 4GRANTOWN- GF , Allt Mor vA -Dorback Burn In 1989 a Mandarin nest was found in a nest box, 2km up a wooded tributary of the River Dee near Ballater in Aberdeenshire (Mick Marquiss pers comm). On 18 April a female was incubating 3 eggs, which had increased to 11 eggs on | May. The box was visited on 31 May and all the eggs had hatched and the young had successfully left the box. No brood was seen on the river despite thorough searching. The provenance of the birds was assumed to be a collection at Ballater, which had free flying birds. More recent records from across the other side of the Cairngorms in NE Scotland indicate that a few birds are now recorded almost annually, with peak sightings in the March to June period (Table 4). Most recently, a pair of Mandarins bred at Glen Tanar, Deeside in 2002 and are still present there at the time of writing (May 2003). The increase in records suggests that Mandarin are now migrating or dispersing more regularly into the north of Scotland during the spring. This is supported by sightings from Orkney, Shetland, Faeroe and Iceland which have increased in recent years. There are 7 records from Orkney (although apparently none since 1998) and apart from a pair in May 1979, all other sightings were of males, with 5 records between 6 April and 10 June, and one of a bird present from 7 September to 28 October (Booth 1998). Shetland has 6 records (Pennington ef a/ in prep) comprising 8 Goldeneye by Dan Powell SB 24(1) individuals (both males and females), all but one in the spring. All records are post 1985, apart from one male on Foula in June 1942. There have also been at least 3 records in Iceland and 2 in the Faeroe Islands since 1984, all of these have also been in April and May (Pennington et al in prep). The importance of Badenoch and Strathspey for Goldeneyes In 1961, the RSPB started a major nest box scheme throughout the Highlands and a pair of Goldeneyes successfully nested in one of these boxes at Loch an Eilein in 1970. This pair nested again in 1971 and 1972; increased to 3 pairs in 1973 and then annually increased to 41 in 1981, 54 in 1984, 85 in 1989, 95 in 1990 and at least 130 pairs in 1995, nearly all of which were in nest boxes (Dennis 1995). In 2002, a concerted effort was made to monitor the Scottish population and this fieldwork indicated that 58 clutches were incubated and 33 were laid but not incubated in 223 nest boxes checked (Goldeneye Study Group 2002 newsletter). Unpublished studies have revealed that as many as two thirds of nests contained eggs laid by more than one female, making a firm population estimate problematical (Begg 2002). However, tentatively assuming that two thirds of the 91 clutches laid involved more than one female, the breeding population in Scotland could involve as many as 150 egg laying females (Carl Mitchell pers comm). Goldeneyes readily find nest boxes, even those away from water, and once an individual starts to nest it tends to return to the same box, or nearby in later years. The provision of nest boxes certainly helped the birds to colonise Scotland, where no or few suitable natural nest holes previously existed and the provision of nest boxes is key to their continued presence as a British breeding species. The birds have Scottish Birds (2003) remained remarkably faithful to the lochs and rivers of Badenoch and Strathspey with the vast majority of British breeders in the area. In 1984, 2 pairs bred north of the main area and up to 4 pairs have continued to nest, on 3 localities, away from the main area (Dennis; in Gibbons et al 1993). In 2002, only 2 nests were discovered outwith Badenoch and Strathspey, including one in Deeside (Goldeneye Study Group 2002 newsletter). The potential impact of Mandarin Ducks on breeding Goldeneyes in Badenoch and Strathspey The principal reasons for the successful establishment of Mandarin Ducks in Britain are firstly, the existence of a vacant ecological niche for a hole nesting duck which feeds on aquatic invertebrates in the spring and summer; secondly, the abundance of suitable habitats and thirdly, the founder stock was a particularly vigorous one derived directly from the wild (Lever; in Gibbons et al 1993). In Badenoch and Strathspey the niche for a hole _ nesting insectivorous duck is not vacant. This is the first time in Britain (and possibly Europe?) that breeding Mandarins have come across another hole nesting, insectivorous duck species in the wild. Is it possible to predict the outcome of this encounter? In the eastern Palearctic both species’ natural ranges meet, but there appears to be little, if any, published information on competitive interactions. The impact of the introduced North American Ruddy Duck on the conservation of the threatened White-headed Duck Oxyura leucocephala has become a cause celebre within the non native species debate (Walton 2002) and there are several lessons that can be learned from it. Firstly, when comparing and contrasting any potential threat from the non native Mandarin to the native Goldeneye it is important to have Mandarin Ducks and the potential consequences for Goldeneye 5 good data and information on which to assess risk and make informed decisions. The Goldeneye breeds across the northern coniferous forests of Europe and the former USSR. Unlike the White-headed Duck, the Goldeneye is not a threatened species and across its range it is apparently doing well (Cramp and Perrins 1998). The British breeding population of Goldeneyes is not threatened by hybridisation with Mandarin Ducks and food is not known to be limiting. The Mandarin is a bird of temperate broadleaved forests, primarily in the lowlands, but occupies valleys and forested uplands to 1500m (Cramp and _ = Perrins 1998). Consequently, it would probably do quite well in the mixed broadleaved and conifer forests of northern Scotland. Although both species feed on aquatic invertebrates, they do so in different niches with Mandarins largely surface feeders whilst Goldeneyes are mainly benthic feeders, reducing potential competition. Suitable nest sites are apparently limiting for Goldeneyes, with populations still expanding in the core area given suitable nest boxes. However, not all boxes are used every year, providing many ‘empty’ boxes in the core Goldeneye breeding areas. Why some boxes are used by Goldeneye and not others is still somewhat of a mystery. In his 1996 review of Strathspey nest box selection, Landridge found that Goldeneyes were unaffected by habitat, tree species and height up the tree. However, the closer a box was to the water the more successful it was and boxes facing NW were slightly more successful than other directions. The primary cause of nest failure in Scottish Goldeneye has been attributed to Pine Marten Martes martes predation (Langridge 1996) and this probably remains the single most important limiting factor on Goldeneye breeding success. Pine Marten predation may also significantly affect Mandarin breeding success, as it has already been the cause of at least one nest failure at Farr in 1996 (Table 2), and could slow down or even halt successful colonisation of suitable nest sites in the core Goldeneye breeding areas. The timing of both species breeding is identical according to BWP, with mid April the earliest date and the main period being late April and early May. Aggressive interactions between Mandarins and breeding Goldeneyes have already taken place at least once in Strathspey (Table 1), incidentally clkm from water. There appears to be no published information on Mandarin nest box selection, as there is with Goldeneye, but it would appear that both species tend to have similar nest site preferences. So, given these circumstances, any potential impact of Mandarins on breeding Goldeneyes is likely to focus primarily on nest site competition. Ironically, the presence of the Goldeneye as a regular British breeding species is due to the provision of artificial nest boxes in and around the Badenoch and Strathspey area. Today’s woodlands in northern Scotland do not have many suitable natural tree holes because of the historical loss of native woodlands and possibly a lack of large woodpeckers, such as Black Woodpecker Dryocopus martius, creating suitable holes. Four thousand years ago, pine dominated native mixed forests covered around 1.5 million ha of the Scottish Highlands. The remaining fragments now represent about 1% of this former range (Biodiversity Steering Group 1995). The provision of nest sites for Goldeneyes in northern Scotland has inadvertently provided Mandarins with suitable nest sites. Although it would appear that relatively small numbers of Mandarins are presently breeding in and adjacent to the core Goldeneye areas, due to their secretive nature during the breeding season, the Mandarins recorded might be an underes- SB 24(1) timate of true numbers involved. Pair formation and courtship occurs mainly in poor light at early morning or evening and on dark days or shade (Cramp and Perrins 1998); a common occurrence in northern Scotland. Thus, the data on Mandarins presented in this paper is likely to be an absolute minimum estimate of numbers present. In Argyll, the origin of the founding pair at Loch Eck could not be determined (Anderson and Petty 1996), but this single pair increased quickly to 18 pairs within 5 years (Murray 2000) and seems primed for further expansion. With perhaps as many as 7000 individuals in the wild in Great Britain, Mandarins have been able to rapidly establish themselves in suitable habitats. Of particular concern in northern Scotland is the issue of a non native species getting a foothold in the area and becoming a significant problem in the future. There is a legal and moral respon- sibility towards conserving our native species and this means acting prior to a problem becoming established. By removing the natural biogeographic barriers between species we should expect to see some problems for native wildlife (Walton 2002). Not all introduced non native species will have a detectable effect on native wildlife, but given that 2 species with very similar, and limiting, nest sites have come together for the first time in the wild in Britain, we cannot predict what will happen next. Recommendations As Mandarin Ducks become established in Badenoch, Strathspey and Strathnairn there are 4 options that could be considered: (1) Ask the keepers of tame Mandarins to pinion ducklings and wing clip their adult birds annually or stop keeping them in such sensitive areas; (2) The relative ease of trapping Mandarin at nest boxes, means either a lethal or non lethal control programme could quickly be developed in core Goldeneye areas, with Scottish Birds (2003) wild Mandarins either killed humanely or removed under licence, wing clipped and perhaps given to responsible wildfowl collections. A less effective variation of this would be to prick eggs or dip them in paraffin, which would keep the female incubating until it was too late to relay; (3) Increase the number of nest boxes available in an attempt to reduce nest box competition between Goldeneyes and Mandarins, segregated if possible according to the respective optimal habitat; and (4) Nothing, leave both Mandarin Ducks and Goldeneyes to their own devices and monitor the outcome. Whilst it is important to promote the understanding of the problems caused by non native species through awareness raising, this should not preclude appropriate proactive management from taking place before a problem becomes significant. Both Mandarins and Goldeneyes should be monitored regularly in Badenoch, Strathspey and Strathnairn along with adjacent areas, in Aberdeenshire, Morayshire and Highland. If Mandarins are found to threaten native breeding Goldeneyes through interspecific competition for nest sites, then suitable measures should be undertaken immediately to protect Great Britain’s only population of breeding Goldeneyes. Acknowledgements This paper would not have been possible without the help of a large number of people. I am grateful to all the observers who submitted records and made comments about Mandarin Ducks and Goldeneye in Scotland: Stuart Benn, Roy Dennis, Keith Duncan, Ian Francis, Ian Hastie, Alastair and Ann MacLennan, Mick Marquiss, Carl Mitchell, Ray Nowicki, Steve Petty, Jonny Pott, Kevin and Carol Shaw, Richard Thaxton and Roy Turnbull. I would particularly like to thank Ray and Val Collier for allowing me to use their breeding Mandarin Mandarin Ducks and the potential consequences for Goldeneye 7. data. Finally, I would also like to thank Malcolm Ogilvie for his very helpful comments and additional sources of information. References Anderson D I K and Petty S J 1996. Population growth and breeding of Mandarins Aix galericulata in Cowal, Argyll. 1996 Argyll Bird Report, pp 82-84. Begg T 2002. Aspects of the breeding biology of the Goldeneye Bucephala clangula in Scotland. Unpublished MSc _ Thesis, University of Glasgow. Birdlife International 2000. Threatened Birds of the World. Lynx Edicions, Barcelona, and Birdlife International, Cambridge. Biodiversity Steering Group 1995. Biodiversity: The UK Steering Group Report. Vol 2 Action Plans, HMSO. Booth C and J 1998. Status and checklist of The Vertebrate Fauna of Orkney. The Orcadian, Kirkwall. BOU Records Committee 1971. Sixth Report. Ibis 113: 420-423. Cramp S and Perrins C M (eds) 1998. The Birds of the Western Palearctic: Concise edition. Oxford University Press. Cranswick P, Pollitt M, Musgrove A and Hughes B 1999. The Wetland Bird Survey 1997-98 Wildfowl and Wader Counts. BTO, WWT, RSPB and JNCC. Davies A K 1988. The distribution and status of Mandarin Duck Aix galericulata in Britain. Bird Study 35: 203-208. Delany S and Scott D 2002. Waterbird Population Estimates (3rd edition). Wetlands International. Dennis R 1993. Goldeneye in The New Atlas of Breeding birds of Britain and _ Ireland: 1998-1991 (ed Gibbons DW, Reid JB and Chapman RA). T & AD Poyser, Calton. Dennis R 1995. The Birds of Badenoch and Strathspey. Colin Baxter Photography Ltd. Goldeneye Study Group 2002. Breeding season newsletter. RSPB Insh Marshes. 8 SB 24(1) Langridge A 1996. The status of the breeding population of Goldeneye Bucephala clangula in Scotland from 1974-1995. Unpublished RSPB Report. Lever C 1993. Mandarin in The New Atlas of Breeding birds of Britain and _ Ireland: 1998-1991 (ed Gibbons DW, Reid JB and Chapman RA). T & AD Poyser, Calton. Lever C 2002. Mandarin Duck in The Migration Atlas: Movements of the Birds of Britain and Treland. (ed Wernham C, Toms M, Marchant J, Clark J, Siriwardena G and Baillie S) Miyabayashi and Mundkur (1999) Callaghan and Green (1993) Miyabayashi and Mundkur (1999) Yuan-Hsun, University of Science and Technology, Taiwan, in litt. BTO, Poyser. © Murray R (ed). The 1998 Scottish Bird Report. = SOC, Edinburgh. a North-East Scotland Bird Report. 1996, 1997, 1998, 1999, 2000, 2001. North-East Scotland e «2 Bird Club. z = ££ 2 2 . : a) s) 5) rO xe) Owen M, Atkinson-Willes G L and Salmon D G & v v s s , ae - A a Dn n 1986. Wildfowl in Great Britain. 2nd Edition. Cambridge University Press. ns Pennington M G, Osborn K, Harvey P V, 3 zloo Oo S Riddington R, Heubeck M, Ellis P M and = o/s = ae! Okill JD The Birds of Shetland. in prep. ae & ww FSF & Petty S J and Anderson D I K 1994. First breeding records of the Mandarin in Argyll. » Sp Scottish Birds 17: 164-165. 3 ee Stone B H, Sears J, Cranswick P A, Gibbons D W, § a = Rehfisch M M, Aebischer N J and Reid JB -& || 2 1997. Population estimates of birds in Britain ~ BE | P 2 we é a) fo) foe} c tse] and the United Kingdom. British Birds 90: § || & ¢ = 5 s 5 1-22. ieee] O @ 8S & Thom V M 1986. Birds in Scotland. T & AD 8 a P = oe) S oyser. 5 o 2 Walton P 2002. The ruddy duck and the white- ms vy, = headed duck: the case for ruddy duck 6§ Z. 8 population control in the UK. In, Alien species: 3 | 4 £ friends or foes? The Glasgow Naturalist. Vol: & 5 q s Q : Zs > 23 Supplement 2001. pp 91-98. = 2 g = a sS < n So oO = s 5 = a) Peter. Cosgrove, 14 The Square, Grantown on © es a 2 Z S j = ons n iS) Spey, Morayshire, PH26 3HG FA 2 E 2 e : E = Ye DB bY ee Pa) Revised manuscript accepted March 2003 Se a iS a Z S A Scottish Birds (2003) Mandarin Ducks and the potential consequences for Goldeneye 9 Table 2 Mandarin records from Strathspey in 2002 Date 27/03/02 April 2002 April 2002 10/04/02 12/04/02 12/04/02 13/04/02 14/04/02 15/04/02 18/04/02 19/04/02 24/04/02 25/04/02 27/04/02 28/04/02 01/05/02 09/05/02 10/05/02 11/05/02 13—14/05/02 15/05/02 17/05/02 22/05/02 30/05/02 Location and activity Pair on Loch Garten!. Pair inspecting natural hole in riparian tree, Lettoch, Dorback Burn’. Up to 3m and If together on the Allt Mor, Nethy Bridge’. 2m, If flying around house, Allt Mor, Nethy Bridge’. Pair of birds sitting on the Osprey nest and the Osprey CCTV camera Loch Garten, Nethy Bridge’. 2m, If inspecting rabbit burrow and Goldeneye nest box, Allt Mor, Nethy Bridge’. 2m, 1f on Allt Mor, Nethy Bridge’. 1f on Allt Mor, Nethy Bridge”. 2m, 2f on Allt Mor, Nethy Bridge’. 1f on Dorback Burn, Clachaig?. Pair of birds on Loch Garten, Nethy Bridge®. 2m, 1f on Allt Mor, Nethy Bridge’. 2m, 1f on Allt Mor, Nethy Bridge’. 2m, If on Allt Mor, 1f flew out of Goldeneye nest box in the evening, 13 Mandarin eggs in box, Nethy Bridge’. 2m on Allt Mor, Nethy Bridge’. 2m, If on Allt Mor, Nethy Bridge’. Pair of birds attacked and attempted to evict a sitting female Goldeneye from its nest box, Balliefurth, Nethy Bridge for over an hour. The Goldeneye nest subsequently failed (not predated)’. Pair swimming close together on the Duaick burn, Nethy Bridge’. 2m, 1f on Allt Mor, only 4 Mandarin eggs left in nest box (eggs cold), Nethy Bridge’. Pair of birds on the River Nethy, Nethy Bridge’. 2 (cold) eggs in nest box, Allt Mor, Nethy Bridge’. 2m, If flying down Allt Mor, Nethy Bridge’. 1m flying up Allt Mor, Nethy Bridge. 1m flying up Allt Mor, Nethy Bridge’. Observers: Roy Dennis, pers comm!, Keith Duncan, pers comm, Roy Turnbull, pers comm’, Richard Thaxton, pers comm‘, Peter Cosgrove pers obs°, Jonny Pott, pers comm®, Alastair and Ann MacLennan pers comm’, Kevin and Carol Shaw pers comm’. 10 SB 24(1) Table 3 Strathnairn Mandarin records 1993-2002 Date Location and activity 1993 Successfully bred in a natural hole, Blar-Buidhe, Farr. 1994 Two pairs bred, Achvaneran, Farr!. 1995 Two pairs bred, Achvaneran, Farr!. 1996 Three pairs bred, at least one pair hatched, Achvaneran, Farr. One nest was lost when the female was taken on eggs by a Pine Marten!. 1997 Four pairs bred, Achvaneran, Farr'. 1998 Four pairs bred, Achvaneran, Farr'. 1999 Three pairs bred, Achvaneran, Farr. In September 1999, 18 Mandarin were seen on the pond at Achvaneran, Farr!. 2000 Five pairs bred, Achvaneran, Farr. Maximum 8 males and 5 females in April!. 2000 Pair of birds nested in a garden nest box, Littlemill, Daviot. Success unknown!. 2000 Pair successfully bred on a lochan, Littlemill, Daviot!. 2001 Two pairs bred, Achvaneran, Farr!. 2001 Pair of birds nested in a garden nest box, Littlemill, Daviot. Success unknown!. 2001 Pair successfully bred on a lochan, Littlemill, Daviot!. 2002 Pair of birds nested in a garden nest box, Littlemill, Daviot. Success unknown!. 2002 Two pairs bred, Achvaneran, Farr'. Early March-early April 2002. Pair prospecting at a pond, Inverarnie but did not nest’. Observers: Ray and Val Collier', Stuart Benn. Table 4 NE Scotland Mandarin records 1996—2003 Date Location 7-18 May 1996 1 f, Stoneywood Mill! 29 Sept—2 October 1996 1 m, Kinloch! 19 October 1996 1 m, Banchory! 3 May 1998 2 m, Dinnet? 3-18 May 1999 2 m, River Dee, Inchgarth* 9 February 2000 1 f, River Don, Inverurie* 25 March 2000 1 f, River Urie, Inverurie* 18 April 2000 1 pair, River Urie, Inverurie* 8 June 2000 1 pair, River Urie, Inverurie* 28 June4 July 2001 1 eclipse m, Marywell? 10 April 2002 1 pair, Lossiemouth® 19 May 2002 2 m, River Carron, Stonehaven’ 23 May 2002 1 pair, Meikle Loch® 30 October to at least 18 November 2002 1 m, River Dee, Inchgarth® Spring 2002 1 pair bred, Glen Tanar? Spring 2003 1 pair present, Glen Tanar? Reference/observer: 1996 NE Scotland Bird Report', 1998 NE Scotland Bird Report”, 1999 NE Scotland Bird Report*, 2000 NE Scotland Bird Report*, 2001 NE Scotland Bird Report?, Jonny Pott®, Tan Hastie’, Various observers’, Ian Francis’. Scottish Birds (2003) 24: 11-17 Scottish Chough Census 2002 i The distribution and status of the Red-billed Chough in Scotland in 2002 S K FINNEY'* & D C JARDINE? A survey of Red-billed Choughs in Scotland in spring 2002 estimated the total population size to be 256-257 birds; 83 pairs were at nest sites and of these 66 were confirmed as breeding. This was an increase of 26% in the number of pairs since the last compre- hensive census in 1998. Breeding was restricted to the Hebridean islands of Islay, Colonsay and Oronsay with the exception of a single breeding pair on the mainland. Productivity was relatively low with a mean of 1.7 chicks fledged per pair. Introduction In northwest Europe, the Red-billed Chough Pyrrhocorax pyrrhocorax has a highly localised and fragmented distribution. Red-billed Choughs are predominantly found in areas with a relatively mild, wet climate along with low intensity agricultural systems (Monaghan et al 1989, Bignal et al 1997). Pastoral farming provides an abundance of soil and dung invertebrates on which the birds feed and the mild climate ensures prey availability throughout the winter months. In Britain, the range of the Red-billed Chough has contracted over the last 200 years and the breeding population is now restricted to western coasts and islands (Gibbons et al 1997). The reasons for the population decline are not clear but may be related to persecution, changes in climate and the intensification of farming practices (Bullock et al 1983, Warnes 1983). The Scottish population of Red-billed Choughs is the most northerly in Europe. The Inner Hebridean islands of Islay and Colonsay have been the birds major stronghold in Scotland for many years (Thom 1986, Monaghan et al 1989). Between 1986 and 1998, when the last Scottish census was carried out, over 90% of the Red-billed Chough population was found on these islands. Over the same period, the Red-billed Chough population in Scotland showed a dramatic decline from 105 pairs in 1986 to 66 pairs in 1998. The Red-billed Chough is now listed as one of the UK’s Birds of Conservation Concern (Gibbons et al 1996, Gregory et al 2002). As a result of the species rarity, population decline and restricted distribution, Red-billed Choughs are given full legal protection under the Wildlife and Countryside Act and the European Directive on the Conservation of Wild Birds. Under the EC Directive, member states are responsible for conserving both the species and its habitat. Regular censuses are a critical part of this respon- sibility in order to monitor the status of the population and to assess the impact of land use changes. Unlike the last census in Scotland in 1998 (Cook et al 2001), in 2002 Red-billed Choughs were also counted in other parts of their range in the British Isles. The aim of this survey in Scotland was primarily to establish the population size and distribution of Red-billed Choughs and secondarily to measure breeding success. Methods Survey area The census was carried out between 6 April and 15 June 2002. Surveys were undertaken throughout the known breeding range of the Red-billed Chough in Scotland. On the islands of Islay and Colonsay, where most of the Scottish population are known to breed, nearly 12 COLONSAY SB 24(1) Oronsay , Elsewhere a visited at least twice VW visited once | | not included in survey Figure 1 Map showing survey coverage on Islay, Jura and Colonsay during the 2002 Red-billed Chough census. Regions of Islay mentioned in the text are also shown. all the coastline and all suitable inland sites were checked (Figure |). On Jura, all historical sites on the west coast and most of the east coast were covered. In addition, inland areas of Jura where Red-billed Choughs have been recorded visiting in the past were also checked (Figure 1). On Mull, Mull of Kintyre and Galloway only histor- ically known nest sites were checked. Many other areas of Scotland were covered by other workers as part of the National Peregrine Census (March—June 2002). Survey methodology All potential nest sites were first visited between 6 April and 7 May, under licence from Scottish Natural Heritage. Coastal sections were surveyed by slowly walking along the cliff tops, constantly looking and listening for Red-billed Choughs. At least one hour was spent near each known or suitable nest site. An observation period of one hour was chosen as adult Red- billed Choughs normally visit the nest every 40-50 minutes during both incubation and chick rearing. For extensive sections of cliff with many potentially suitable nest sites but where birds had not previously been recorded breeding, observations, of at least one hour, were made from vantage points approximately 1 km apart. At least one hour was also spent at all known or suitable inland sites: buildings and crags. All areas where Red-billed Choughs were found, or where Red-billed Choughs had been recorded breeding historically, were checked for a second Scottish Birds (2003) time between 8 May and 15 June. Thus, historical sites on Mull and on the mainland were surveyed twice; survey coverage on Islay, Jura and Colonsay varied and is shown in Figure 1. No additional nest sites were found during the second survey period indicating that all breeding Red-billed Choughs were successfully located during the first visit. Any birds located during the survey were observed to establish their breeding status. Solitary birds or pairs that were found feeding or flying were followed to determine whether they returned to a nest site. Many birds were ringed with unique colour ring combinations, which were used to distinguish between close nesting pairs. If no birds were seen after one hour, and if access was possible, potential nest sites, ie buildings, caves and gullies, were searched thoroughly for evidence of breeding activity. Assessment of breeding status followed the criteria of Monaghan et al (1989) and was categorised as follows: Confirmed: Sites where nests with eggs or young were observed, where adult birds were observed incubating, feeding young, leaving the nest with egg shell or a faecal sac, where nestlings were heard or where dependent young were observed with parents. Pairs were included in this category even if no young fledged. Probable: Sites where Red-billed Choughs were present and were suspected of breeding but where this was not confirmed, largely because of inaccessibility of the site. Pair present: Sites where Red-billed Choughs were known to visit but where nest building was never completed or laying was not believed to have occurred. Scottish Chough Census 2002 vist Where a breeding attempt was confirmed, additional visits were made during June and July to establish breeding success and number of fledged young. Nests which were known to fail were included in assessments of overall breeding SUCCESS. Numbers of non breeding birds seen in flocks were also recorded to allow an estimate of the total population size to be made. The standard Red-billed Chough census methodology involves 2 visits to potential nest sites as described above. On Islay, however, time and resources were available for additional visits to be made. These were carried out both at the end of the season to establish breeding success and, in some cases, during the main survey period. This offered the opportunity to assess the effectiveness of the standard census methodology at correctly classifying the breeding status of Red-billed Choughs. This was done by comparing the results obtained based on the standard methodology to those obtained when the additional information collected during the breeding season was taken into account. Results Population size Eighty three pairs of Red-billed Choughs were recorded in Scotland in 2002. Of these, 66 were confirmed breeders, 5 probably bred but breeding was not confirmed and a further 12 pairs were active at suitable nest sites but there was no evidence that laying occurred (Table 1). All breeding Red-billed Choughs were on the islands of Islay, Colonsay and Oronsay with the exception of a single breeding pair in Galloway. On Islay, the maximum number of non breeding birds recorded at any one time was 61. These birds were seen at feeding sites on 24 April. On 14 Colonsay, 29-30 non breeding birds were seen regularly on the island. There were no non breeding flocks on Jura, Mull or on the mainland. This gives a minimum Scottish Red- billed Chough population size (breeders and non breeders, but excluding young from 2002) of 256-257 birds of which 74% were on Islay and 25% on Colonsay. Breeding success Breeding success was calculated using data from the 48 confirmed breeding pairs on Islay for which brood size at fledging was known. A mean of 1.69 (+ 0.19 se) chicks fledged per pair. Breeding success did not differ significantly between Red-billed Choughs nesting on natural sites (1.78 chicks/pair) and those nesting in buildings (1.60 chicks/pair; t-test, df=46, t=0.64, n.s.). The mean number of young produced from successful nests was 2.38 (+ 0.16 se, n=34). It was not possible to obtain a value for mean breeding success of Red-billed Choughs on Colonsay as infrequent visits prevented breeding failures from being confirmed; the data were therefore biased towards successful pairs. However, at least 10 of the 13 confirmed breeding pairs successfully fledged young (mean for successful nests = 3.0 chicks/pair, + 0.26 se). The pair breeding on the mainland fledged 2 young. 120 100 D S go =, 60 (aD) Ke E 40 20 0 1980 1985 1990 1995 2000 2005 SB 24(1) Nest site type and location Red-billed Choughs used a variety of nest sites including ledges in caves, crevices and both used and abandoned buildings. The proportion of pairs nesting in buildings, ie derelict cottages, farm outbuildings and barns, has increased substantially on Islay over the last 15-20 years with 52% (33/64) of pairs nesting in buildings compared to 29% (25/86) in 1986. On Colonsay, almost all Red-billed Choughs nested in natural sites. The greatest number of breeding Red-billed Choughs on Islay was found on the Rhinns and on the Oa (Table 1), with only 14 of the 64 pairs being found elsewhere on the island. This distri- bution is consistent with that found during previous Red-billed Chough — surveys (Monaghan et al 1989, Bignal et al 1992, Cook et al 2001; chi-squared test, c2=3.67, df=6, n.s.). Effectiveness of basic census methodology Out of a total of 64 Red-billed Chough pairs on Islay, 6 (9%) were reclassified based on the information obtained from the additional nest visits (Table 1). Four pairs were classified as “pairs present” or probable breeders based on the standard methodology alone but breeding was confirmed following a third visit, when fledglings were observed leaving the nest. Two pairs, both nesting in buildings, were classified as probable breeders based on the standard Figure 2 Changes in the number of pairs of Red-billed Choughs recorded at nest sites on Colonsay & Oronsay (squares, dotted line), Islay (open circles, dashed line) and in Scotland (closed circles, solid line) between 1982 and 2002 (data from Monaghan et al 1989, Bignal et al 1992, Cook et al 2001 and current study). Scottish Birds (2003) Scottish Chough Census 2002 V3) Table 1 The number of nest sites at which pairs of Red-billed Choughs were recorded in Scotland during 2002. Numbers in parentheses show classification based on the standard census methodology, the final figures take into account information collected during additional nest visits. Area Confirmed Islay 32 (48) Rhinns 36 (33) Oa 7 Elsewhere 9 (8) Colonsay (& Oronsay) 13 Jura 0 Mull 0 Kintyre 0 Galloway l Total Scotland 66 (62) methodology but were finally classified as “pairs present” based on additional checks of the nest site during the season, which showed that no eggs had been laid. Despite these discrepancies, this comparison of results between the standard methodology and that including additional information showed no difference in the estimated number of pairs (Table 1). Discussion Eighty three pairs of Red-billed Choughs were recorded in Scotland in 2002. This was an increase of 26% since the last comprehensive census in 1998. The increase in the number of Red-billed Chough pairs on Islay was the first recorded on the island since 1986, after which the population declined by over 50% (Fig 2). In contrast, the number of Red-billed Choughs recorded breeding on Colonsay increased consis- tently between 1986 and 2002 (Bignal et al 1992, Cook et al 2001). However, there is evidence of an interchange of birds between the 2 islands and, as such, the increase in the number of breeding pairs on Colonsay may have involved the recruitment of young birds from Islay. Probable Pair Present Total 4 (8) 8 64 0 (3) 4 40 2 1 10 DiS) 5 14 l 4 18 0 0 0 0 0 0 0 0 0 0 0 | 5 (9) 12 83 On Islay, breeding success was 1.7 chicks fledged per pair. This was low compared to that recorded in previous years (Bignal et al 1987, Cook et al 2001). The reasons for the low productivity are unknown but the exceptionally wet weather experienced during the 2002 breeding season may have played a role to some extent. The size of the non breeding flock on Islay (minimum 61 birds) was the largest recorded since 1992. Furthermore, the 29-30 non breeding birds on Colonsay was the largest non breeding flock recorded on the island for many decades (see Bignal et al 1992 and Cook et al 2001 for previous recent records). Despite the increase in the population size of Red-billed Choughs in Scotland over the last 4 years, the range of the birds has contracted. Having lost the pairs on Mull and Jura the entire Scottish breeding population is now restricted to the islands of Islay, Colonsay and Oronsay with the exception of a single breeding pair in Galloway. Historical records suggest that the distribution of Red-billed Choughs in Scotland used to be much more widespread with birds regularly breeding in several areas on the mainland but that both the size and range of the 16 breeding population has been declining for at least 200 years (Warnes 1983). However, between 1982 and 1998, when detailed population monitoring was first carried out, some long abandoned breeding areas (Galloway and Mull) were recolonised, although stable breeding populations were never established. Previous work has suggested that joining a flock in the autumn is crucial for the first winter survival of young birds, which interact with other individuals within the flock (Bignal et al 1997). The recent recolonisation of sites on Mull and on the mainland may have occurred through the chance arrival of single pairs, as the dispersal distances of juvenile birds are normally very short (Bignal et al 1997). It therefore seems unlikely that breeding by several pairs will take place in these areas for some time. The results obtained following the standard census methodology were slightly different to those obtained when information collected during additional nest visits was taken into account. The greatest difference arose when pairs that were classified as “pairs present” or probable breeders based on the standard methodology alone, had breeding confirmed following a third visit, when fledglings were observed leaving the nest. Despite these changes, the results obtained using the 2 methods were very similar. Therefore, the use of information from additional nest visits did not affect the overall conclusions drawn from this study. However, the Islay Red-billed Chough population has been intensively studied and monitored and the methods used in this census are therefore as consistent as possible with those used during previous Red-billed Chough surveys in Scotland (Monaghan et al 1989, Bignal et al 1992, Cook et al 2001). Although there will inevitably be some variation between surveys in the classifi- cation of breeding status based on the interpre- tation of a birds behaviour, the results do still provide a reliable measure of population change. SB 24(1) In conclusion, although there has been a recent increase in the size of the Red-billed Chough population in Scotland, Red-billed Choughs remain of conservation concern as a result of their small population size and very restricted distri- bution (Gregory et al 2002). With the majority of birds concentrated in such a small area the population is vulnerable to threats such as changes in land use or the alteration of farming practices as well as changes in weather conditions (Reid et al in press). Research that is currently being undertaken, which aims to identify the factors that most strongly influence population growth rate and the stage of the life cycle that is most affected, should help focus future conser- vation effort in the most appropriate areas. Acknowledgements We would like to thank the landowners and farmers who kindly gave permission to survey their land. Thanks also to John Adair, John Armitage, Eric Bignal, Roger Broad, Ian and Margaret Brooke, Fiona Harmer, James How, Angus Keys, Ronnie McIndoe, Clive McKay, Mike Peacock, Fiona Rout and Guy Thomas for help in the field and to Pat Monaghan and Eric Bignal for comments on an earlier draft of this manuscript. The Scottish Chough Study Group provided information on the location of historical nest sites. The census was run as part of the Statutory Conservation Agencies/RSPB Annual Breeding Birds Scheme (SCARABBS) and was funded jointly by the Countryside Council for Wales and the RSPB. Nest visits were licensed by Scottish Natural Heritage. References Bignal E, Bignal S & McCracken D 1997. The social life of the chough. British Wildlife 8: 373-383. Bignal E, Bignal S, Moore P, Clarke J & Clarke P 1992. The second international chough census in Scotland, 1992. Unpublished report to the RSPB. Scottish Birds (2003) Bignal E, Monaghan P, Benn S, Bignal S, Still E & Thompson P M 1987. Breeding success and post-fledging survival in the chough Pyrrhocorax pyrrhocorax. Bird Study 34: 39-42. Bullock I D, Drewett D R & Mickleburgh S P 1983. The chough in Britain and Ireland. British Birds 76: 377-401. Cook A S, Grant M C, McKay C R & Peacock M A 2001. Status, distribution and breeding success of the Red-billed Chough in Scotland in 1998. Scottish Birds 22: 82-91. Gibbons D W, Avery M I, Baillie S R, Gregory R D, Kirby J, Porter R F, Tucker GM & Williams G 1996. Bird species of conser- vation in the United Kingdom, Channel Islands and Isle of Man: revising the Red Data List. RSPB Conservation Review 10: 7-18. Gibbons D W, Reid J B & Chapman R A 1997. The new atlas of breeding birds in Britain and Treland, 1988-1991. T&AD Poyser, London. Gregory R D, Wilkinson N I, Noble D G, Robinson J A, Brown A F, Hughes J, Proctor D A, Gibbons D W and Galbraith C A 2002. The population status of birds in the United Kingdom, Channel Islands and Isle of Man: an analysis of conservation concern 2002-2007. British Birds 95: 410-450. Monaghan P, Bignal E, Bignal S, Easterbee N & McKay C R 1989. The distribution and status of the chough in Scotland in 1986. Scottish Birds 15: 114-118. Reid J M, Bignal E, Bignal S$, McCraken D & Monaghan P. Environmental variability, life history covariation and cohort effects in the red billed chough. Journal of Animal Ecology — in press. Thom V M 1986. Birds in Scotland. T&AD Poyser, Calton. Warnes J M 1983. The status of the chough in Scotland. Scottish Birds 12: 238-246. Scottish Chough Census 2002 17 *SK Finney, RSPB, Dunedin House, 25 Ravelston Terrace, Edinburgh, EH4 3TP Tel: 0131 311 6500 Email: suki.finney @rspb.org.uk DC Jardine, 49 Bellfield Road, North Kessock, Inverness, IVI 3XX *To whom correspondence should be addressed. Revised manuscript accepted January 2003 18 Scotland’s endemic subspecies SB 24(1) R Y MCGOWAN, D L CLUGSTON & R W FORRESTER ON BEHALF OF THE SCOTTISH BIRDS RECORDS COMMITTEE. ILLUSTRATED BY C ROSE Scottish Crossbill Loxia scotica is the only species currently considered to be endemic to Scotland. Scotland does however possess a further 9 endemic subspecies, that appear on the Scottish List (Clugston et al 2001). These are Rock Ptarmigan Winter Wren (Shetland Wren) Winter Wren (Fair Isle Wren) Winter Wren (Hebridean Wren) Winter Wren (St Kilda Wren) Song Thrush (Hebridean Song Thrush) Crested Tit (Scottish Crested Tit) Common Starling Common Linnet (Shetland Starling) (Scottish Linnet) Many of these races are little known and have been poorly studied. From the early years of the 20th century until the 1950s many bird books treated Scottish birds at the subspecies level, but this practice largely disappeared by the 1960s. As a result most of Scotland’s endemic races have to a great extent been ignored during the second half of the twentieth century. In recent years, however, there has been a development in the interest of subspecies amongst many birding enthusiasts, as noted by Rogers et al (2002). This has largely been driven by the availability of diagnostic information and the expectation of possible future elevation to species status. Whilst most Scottish species are now well known, we lack detailed knowledge about identification, population and even distribution of many of our endemic races. The purpose of this paper is to provide an overview of what is currently known and to highlight gaps in our knowledge. We also hope to stimulate further study of our endemic birds and to encourage attempts at identification of the various races in the field. (Scottish Ptarmigan) Lagopus mutus millaisi Troglodytes troglodytes zetlandicus Troglodytes troglodytes fridariensis Troglodytes troglodytes hebridensis Troglodytes troglodytes hirtensis Turdus philomelos hebridensis Parus cristatus scoticus Sturnus vulgaris zetlandicus Carduelis cannabina autochthona The “Hebridean” race of Hedge Accentor Prunella modularis hebridium would be included in the list of endemic races, except that it is currently considered that birds in Ireland belong to this race, so it is not a true Scottish endemic. This race was first described by R Meinertzhagen in 1934 (Meinertzhagen 1934), with the type locality South Uist, Outer Hebrides. As far as it is now known its distri- bution within Scotland covers the Hebrides, Ayrshire, Renfrewshire, Dunbartonshire the Clyde Islands and Argyll. Two million pairs of Hedge Accentor breed in Britain (Stone et al 1997). From the species distribution within Britain (occupied 10 kilometre squares) and its judged density, it would be reasonable to estimate the Scottish population of hebridium to be in the region of 75,000—125,000 pairs. Although the Fair Isle Wren, with less than 50 breeding pairs, is the rarest of our endemic races, both Song Thrush and Linnet are UK Red listed species, as a result of rapid declines (over 50%) in population during the last 25 years. There are io Scotland’s endemic subspecies 18-35 Scottish Birds (2003) 24 (ISI) YINOg ‘aTepstoqysoT) 76:96 DO 11" (amysyyieg ‘yoouury Your) 9/ *L9 DO 11Ng (1A8rVy ‘opepdeuy]) ZZ :€L OO 11" (ast) WING) CC:OL Sq] (amrysyseur’y ‘[TeyuowwIOYL) 76:78 S!4] (amysyreue’y “yoouunuutey) ¢9:1$ OO 11g (amysyed ‘ATYIOPd) T:€L OO 11" (SITYS-SSOUIOAUT ‘SNYSINUTSTYIOY) 99:€9 DO 11"F (puepioying “yoousog *xt[eaq) plEe:€8 S47 (amysmoruoy ‘AoTwIed) PE :L9 YNV (omysmoiuoy ‘Aopureq) LEE:¢7 VSOUNIT (amYysmorUoY “Wesped) 6rL :08 S29] (amysmorguoy ‘Aopured) OSL ‘08 874] ({[ASry ‘opepdeuy YON) Z6€ *L9 YNV (SI) YINOS) YE-9L Sq] (OITYSMOIJUDY “SUTIBIPA, UDIMON) OLE:TT VSOMNT (puepioying “yooulod) r1:79 00g 11" (UBLIY ‘JOOJIOIeMAOeT) 8S:79 DO 11" (asIE) YINog ‘aTepstoqysoT) ¢¢:9¢ DO 11" (SI, YINog “YBInqIeq) 71:89 DO 11" (emysyyog “‘yoouuey yoorury) *812d1aT Jorg 114§ (AouyIO ‘MydIO) 9:7p DO" 11" (SIA) YNog ‘e[NApIy eYPNY) 61 E:77 VOUT (AouysO ‘AUeysny) TOT ‘pl 12M f28 aig (omTYssuTINg ‘ormopsieg) 0L:6$ OO 11" PULTIONS 0} POJOLNSaI SI SULIOJ dso} JO UOTNGMISIC x OL6l ‘PAG lalzuayoou SNjIIUaOYIS D2UAaquUey Le] ‘AaourlD impppiva vinyssdd vynyssdg €ES6I ‘AdouR[D sydnasip vauunvyf syanpavy pl] UsseyZoulay] wnsouosuag SiujsolAv]f siyJUuvdy (=Syanpsvd) OP6] ‘A2our[D MosiADy S140]YI S1LO]YID (=SYanpsvdd) LES] ‘UOSLLIRY VI1NO0IS Sqajaod DIPSUls J €C6| ‘UOLODH puke pooMjezeH Sisuauopajyd Sniuvpuvn]s snjn4sivyH) Cp6 ‘Aoour[D snjooiid sao snavg Ip6I ‘Agoura sounjamopoysXupjyo snjppnvd sojoyjisay OS6I ‘Aeoue[D isunpsol stunumuod piajx¢ OS6I ‘Ad0URID 40JUaraId SNLOAIISIA SNPANT QC6I ‘Aoour[D avulsayjvo sojamojpiyd snpAn], QC6] ‘Aoour[D Nsunyaol DjINAIW SNPANI, OS6I ‘AdouRID “asajui ayjuvuao ayjuDUua— PEG] ‘UaseYyZOUIO| avsasay] DIVNbAO] VjOIIXDS 661 ‘AgouR[D vpiydosadsay vajaqna DjOIIxXDS C6] ‘AvourlD vjlsodiajul Siupjnpou vyjaunid Tb] ‘Aoour[D snuvisadsay vyjajouids snyjuy OL6L ‘PIG luaspyZjauiawu DyJajoulds snyjuy Lp6l ‘Aoour[D suasiaaip sisuadiv Dpnv]y OS6I ‘AsouR[D Ypimyosuiazy snqunjod pqunjoD IZ6[ ‘UsseYyzyoutoyy soj1ydoaso piavoidy sypianid 66] ‘Aoour[D iuosiuapy vynoyo1y snlAposDYD 676] ‘UULUINON Sypjuap19z0 snsajp.jso sndojowavy 66] ‘AgouryD syo1sysaa sndosojyo vjnuyjoyH ,sUUNG pody «GOUT[[Ng UOUTWO,) ]jodpey uouru0d OMT 4 YOuTJUsdIH uvadoing «4ouryey) y Ae UBISeING «ILL [ROD 4 ILL poptey-3u0T JEOIMOTY AA, UOUTWIOD, Yysmy,L, Sst xYsny], Su0g «PIIQYSe[ A UOUUO Teayeoy A WIOyVON 4 JBYIOU0IS [,Jeyouty y1OWUIIIV I8poH alidig Joven alidig Joie «eT AAS UuOddsIg POOAA UOWUIODZ IQAO[d USpjOH uvsedoing IOAO]d posuny Joyojeo10jsAOQ ueIseIN| ,UdYIOO WOUTWOD Ayyeso] ad 4} puv uoye}Id Saqiiosep YjIA ‘soueU dTQUIIIS pue YsI[suyq yNO| 4q pajdasan sasuo] ou pup jouysip Kyuaioiffns aq 0} pasapisuod JOU MOU yng (SJWapUua YSIIOIS SV palfisspj) ag pjnom YIIYM fo JsouL) puvjjoIg Worf pagisasap Saizadsqng { aQBI, 20 no data available to indicate whether the populations of the races endemic to Scotland (Hebridean Song Thrush and Scottish Linnet) have declined at the same rate or exceeded the rate of decline of the species as a whole. The Scottish List is based upon the taxa accepted by the British Ornithologists’ Union Records Committee (BOURC) as they appear on the British List (British Ornithologists’ Union 1992). A number of other forms have been described from Scotland, though these are not now considered to be taxonomically distinct (Vaurie 1959; 1965) and their names are accordingly synonymised with current British races. These are listed in Table 1. Sixteen (possibly 17) of these taxa had breeding distri- butions limited to Scotland when originally described and are marked with an asterisk. Scottish Endemic Subspecies Rock Ptarmigan (Scottish Ptarmigan) Lagopus mutus millaisi Hartert 1923. British Birds 17: 106. (Ben Wyvis) Millaisi is the only race found in Britain. The entire British population is now confined to Scotland, having become extinct in the English Lake District during the 19th century. Restricted to the higher mountains with the main population in the Grampians and North West Highlands. It is scarce on the Caithness hills. Small numbers are resident on the islands of Arran, Mull, Skye and Outer Hebrides. With few exceptions, such as Applecross, the habitat in the west is poorer than in the Cairngorms and east Mounth (A Watson pers comm). The most recently published population estimates of a minimum of 10,000 pairs (Gibbons et al 1993; Stone et al 1997; Mead 2000) appear all to have been based on Ratcliffe’s (1990) data. Gibbons et al (1993) do draw attention to the fact that few data from counts exist. A Watson (pers comm) SB 24(1) confirms that no accurate estimate is possible because counts have been so few in number, though he believes it unlikely that there would be less than 10,000 birds in spring and seldom more than 100,000 in autumn. Identification Summer and autumn dorsal plumage rather greyer and less barred, lacking the white markings of the nominate race. In comparison, nominate mutus is browner and blacker, more barred and vermiculated. In winter, most individuals of millaisi retain a few dark feathers. Wing length of male millaisi significantly shorter than nominate mutus (Cramp and Simmons 1979). A general impression from skins in the NMS is that Scottish birds have narrower and less deep bills than Scandinavian ones. Winter Wren (Shetland Wren) Troglodytes troglodytes zetlandicus Hartert 1910. Die Vogel der palaark- tischen fauna page 777. (Shetland Islands) Bannerman (1954) describes this race as ‘A resident confined to the Shetland Islands (except Fair Isle)’. More recently, Pennington et al (in press) draw attention to a few records of this subspecies from Fair Isle on 21 October 1985 and 17 May 1986. There is some evidence from FIBO reports that these records may be a regular, though perhaps less than annual, phenomenon (Collinson 2002). A further extra limital record of a corpse from Newburgh, Aberdeenshire on 13 January 1968 was considered by Kenneth Williamson to be a definite zetlandicus (Macmillan 1969, Pennington et al in press), though Collinson alludes to a suggestion that this may have been a large migrant from Scandinavia. Unfortunately, none of these birds had been ringed. The current breeding population is estimated as 3,000-6,000 pairs (Pennington et al in press). Scottish Birds (2003) Winter Wren (Fair Isle Wren) Troglodytes troglodytes fridariensis Williamson 1951. Ibis 93:599. (Fair Isle, Shetland) The resident breeding population is confined to Fair Isle, an island lying between Orkney and Shetland, and 25 miles from each. They rarely nest away from the sea cliffs. There are no records of this race away from Fair Isle. A population estimate by Williamson (1965) suggested 40-50 pairs whereas Thom (1986) indicates that the population ‘...fluctuates between 10 and 40 pairs and was estimated to include only 10 singing males in 1980.’ Pennington et al (in press) currently refer to a population that has fluctuated between 10 and 50 singing males during the last 30 years, with 29 territories recorded in 2000. Winter Wren (Hebridean Wren) Troglodytes troglodytes hebridensis Meinertzhagen 1924. Scottish Naturalist page 135. (Butt of Lewis, Outer Hebrides) Bannerman (1954) describes this race as ‘A resident in the Outer Hebrides, Lewis, Harris, North and South Uist and probably Skye’. Earlier and subsequent writers add virtually nothing to our understanding of its complete distribution, nor any real idea about the size of the population. Population estimates are surprisingly broad, ranging from 1,000—10,000 pairs (B Rabbitts pers comm, A Stevenson pers comm), with the upper figure possibly attainable in good years. Winter Wren (St Kilda Wren) Troglodytes troglodytes hirtensis Seebohm 1884. Zoologist page 333. (St Kilda) Scotland’s endemic subspecies 75) This species is confined to the archipelago of St Kilda, lying almost 40 miles west of the Outer Hebrides. Baxter & Rintoul (1953) indicate a population range of 50—100 pairs with 68 pairs in 1931, though later estimates by others generally suggest a higher figure. Williamson (1958) calculates a total population of 233 pairs from a study in 1957. Specific figures for each island were 116 pairs on Hirta, 25 on Dun, 45 on Soay, 45 on Boreray and 2 on Stac an Armin. A later census on Hirta only, between May 1961 and September 1962 (Waters 1964) gives a population of 92 pairs, an apparent decrease on the 1957 figure. More recently Murray (2002) writes that ‘there is no evidence of gross changes in numbers and the 1957 estimate of not less than 230 pairs for the islands is probably still valid’. Field identification of Scotland’s Winter Wren subspecies (Plate 2) As noted by Vaurie (1959) geographical variation within the Winter Wren is substantial with between 26 and 30 races now usually being recognised in the Palearctic. Wing length, size and length of bill, overall plumage colouration and extent of barring are the features that show the most noticeable difference. The degree of colour saturation is believed to correlate with prevailing humidity. The following descriptions compare these subspecies with the mainland Scottish wren T t indigenus. The latter is the predominant subspecies throughout the British Isles, though nominate troglodytes occurs towards the south. In comparison with nominate troglodytes, indigenus is darker and richer brown on the upperparts. For a fuller discussion on the geographical variation of wren taxa in the Palearctic see Cramp (1988) and Collinson (2002). As wren populations are generally sedentary, identification of individuals will not usually be a problem; eg most wrens observed on Fair Isle 22 will be fridariensis. Instances of ‘vagrancy’ amongst these endemics, however, have been recorded eg nominate troglodytes in St Kilda (Murray 2002) and possible zetlandicus in Aberdeenshire (King 1969; Collinson 2002). However, for birds outwith their normal ranges, subspecific differences may not always permit field identification of individuals. Whilst expertise in the recognition of vocalisations of particular populations may also be of some assistance with identification, local variation in song dialects would make this difficult. Shetland Wren: Darker and more rufous brown than the mainland form with underside heavily barred, barring generally extending from abdomen to breast, bill longer and stouter, stronger legs. Fair Isle Wren: In comparison with zetlandicus, generally paler; wing, bill and tarsus length intermediate between mainland form and zetlandicus; heavier than mainland form. Hebridean Wren: Darker and more heavily barred than the mainland form; size similar. From zetlandicus tends to be buffier, with less heavy and extensive barring; bill slightly shorter and weaker. St Kilda Wren: Heavily barred, bill strong and long, distinctly paler and greyer undersides than all of the above subspecies. From the mainland form by larger size, heavier barring, generally greyer and less rufous. Song Thrush (Hebridean Song Thrush) Turdus philomelos hebridensis Clarke 1913. Scottish Naturalist page 53. (Barra, Outer Hebrides) The BOU (1971) states ‘confined to the Outer Hebrides and the Isle of Skye; populations in the Inner Hebrides other parts of western Scotland and Kerry are intermediate between this and the race T p clarkei’, the latter being the race which breeds SB 24(1) throughout Britain. Clements (2000) shows the distribution to be Outer Hebrides and Isle of Skye. Thom (1986) makes no mention of this race at all. Due to the lack of census work current population estimates for the Outer Hebrides are understandably vague, with A Stevenson (pers comm) suggesting 750—2,000 pairs, whereas B Rabbitts (pers comm) prefers the slightly lower range of 500 —1,000 pairs. No survey data are available from Skye, although Andrew Currie (pers comm) considers it unlikely that the breeding population exceeds 200 pairs. The total population of hebridensis cannot therefore be estimated more accurately than the broad approximation of 500—2,200 pairs. In Britain, breeding Song Thrushes are largely sedentary with only c25% of breeding birds recovered in winter moving more than 20km (Wernham et al 2002). Birds specifically from the Outer Hebrides and Skye are generally believed to be similarly sedentary, apart from a proportion which move away from their breeding grounds between July and February (Witherby et al 1938b). A number of these birds are quite reasonably assumed to overwinter in Ireland (Cramp 1988). However, apart from Ruttledge’s (1966) reference to a bird ‘showing the character- istics’ of hebridensis which was caught in County Tipperary on 1 January 1963, there is, to date, little in the way of empirical evidence to support this. Some of the Hebridean birds are replaced in winter by continental (= nominate philomelos) birds (Witherby et al 1938b; Baxter & Rintoul 1953) and one was trapped at Uig, Lewis on 23rd October 2001 (A Stevenson pers comm). Examination of winter collected skins from the Outer Hebrides in the NMS (this study) indicates that some mainland Scottish (= clarkei) birds also move to these western areas. A small number of clarkei were claimed to breed at Stornoway (Witherby et al 1938b), but A Stevenson (pers comm) says that this is no longer the case, with all Scottish Birds (2003) Stornoway breeders now being hebridensis, as are all Song Thrushes currently breeding in the Outer Hebrides. Baxter & Rintoul (1953) refer to autumn and spring records of hebridensis from Ayrshire, Pladda (off Arran), and Wigtownshire. The Ayrshire record relates to a specimen collected from a ‘small flock’ by Richmond Paton on 6 October 1922 and deposited in the Dick Institute, Kilmarnock (Baxter & Rintoul 1953; Paton & Pike 1929). Validation of this record should therefore be possible if the specimen still exists. The Pladda records may allude to birds (ex coll H Whistler in NMS) obtained at Pladda lighthouse in April 1937 and October 1938. In fact, these skins are perfectly good clarkei and not hebridensis. Specimen based and/or ringing evidence for hebridensis movements is virtually nonexistent. It is certainly insufficient to substantiate N F Ticehurst’s theory, as quoted by Bannerman (1954), by which Redwings Turdus iliacus coming south by the west coast route “pick up” a proportion of hebridensis. A bird ringed in its first year on 2 August 1983 near Inverness was found dead on Lewis on 16 May 1984. The race was not recorded nor is it known if the specimen was preserved. It could either be a hebridensis returning to Lewis to breed or a wandering clarkei or philomelos still to return to its breeding area. As mentioned above, birds from the Inner Hebrides and parts of western Scotland are described as intermediate between hebridensis and clarkei (Witherby et al 1938b; Vaurie 1959; BOU 1971) or, at least, generally darker in plumage (Bannerman 1954). Whilst researching this paper, an examination of the NMS skin collection revealed several skins from Islay (all collected in June 1926) that show a mixture of hebridensis and clarkei characters. Three adults exhibit hebridensis features of dark spotting, Scotland's endemic subspecies 23 greyer rump and flanks, richer buff underwing and more boldly marked face, though they also have lighter rufous brown crown and mantle, these latter features being more typical of clarkei or philomelos. A fourth adult in the same series closely resembled typical hebridensis even in crown and mantle colouration. A juvenile bird collected at the same time is heavily spotted, though its immaturity makes assessment of its affinities more difficult. A further specimen showing similar intermediate characters is a bird collected in April 1911 at Barns Ness Lighthouse, Dunbar. This individual is heavily spotted with darkish buff underwing and hint of grey on the rump, though it does have a richer rufous crown and mantle. It is clear that we are still largely ignorant of the status, distribution and movements of Hebridean Song Thrush even 90 years after its first description. Identification (Plate 3) Six features are useful in differentiating hebridensis from clarkei and nominate philomelos. Ventrally these are (a) bold, almost black spots, (b) flanks smokey, greyish brown rather than buffish, (c) underwing coverts rich, rufous buff, darker than other races. On the dorsal surface, (d) the forehead, crown and mantle is dark brown with a less rufous tone than clarkei, and (e) the rump and upper tail coverts have a grey wash. Finally, (f) the dark markings of the face and throat are generally bolder. Regarding (d), it is worth emphasising that the forehead and crown colour is noticeably darker in hebridensis when directly compared with the other 2 races. As some clarkei and nominate philomelos may superficially resemble hebridensis eg with heavy dark spotting, it is important that as many characters as possible are used to verify the diagnosis. 24 Most Hebridean Song Thrushes should be separable in the field given good views. Bold spotting alone should not be relied on. Greyish wash on rump and upper tail coverts, smokey brown flanks and richer rufous buff underwing, are important identification features. Crested Tit (Scottish Crested Tit) Parus cristatus scoticus (Prazak, 1897). Journal fur Ornithologie 45: 347. (Scotland) The entire British population is confined to Highland pine forests of Scotland and belongs to the race scoticus. It appears likely that they became isolated from the widespread Continental birds along with the Caledonian pine forest after the last glaciation. Scoticus 1s largely restricted to the pinewoods of Strathspey, the Beauly catchment, parts of the Great Glen, east Ross and the coasts of Nairn, Moray and Banffshire. It is absent as a breeding species from Deeside, despite the presence of apparently suitable habitat and occasional records. Studies have shown that 35% of the population occurs in ancient native pinewoods, which comprises only 4.7% of the woodland, with the rest nesting in pine plantations (Summers et al 1999). The Scottish subspecies is by far the most restricted in range of any of the European populations. Estimates of population size have previously varied from Nethersole-Thompson’s 300-400 pairs in the 1960s (Thom 1986) to c900 pairs in Cook’s 1980 study (Cook 1982). This latter figure is quoted in The New Atlas of Breeding Birds (Gibbons et al 1993) and also used by Stone et al (1997) and Mead (2000). In the most recent quantitative assessment Summers et al (1999) calculate a winter population range of 5,600 — 7,900 birds over three winters in the early 1990s and estimate 2,400 territories holding pairs or unmated birds. SB 24(1) The population estimates obtained in this study are greater than earlier ones, because of the different methods used and small sample sizes of densities in earlier works. There is no evidence of any real population increase. Generally 6 races of Crested Tit including scoticus are recognised with all populations being considered extremely sedentary (Cramp & Perrins 1993). Of cl10 English records, only 2 have been identified to race, one nominate cristatus (Whitby, Yorkshire, 1872) and one mitratus (Isle of Wight, pre 1844) (Bannerman 1953b). There is an interesting report of a Crested Tit attending a garden feeder in Tobermory, Mull from 19 October 2002 to at least 9 November 2002. Although the race was undetermined, this presumably refers to scoticus (A Murray pers comm). Identification (Plate 1) Scoticus is noticeably smaller (male — wing length 65.5 (64-67) mm cristatus; 63.3 (60-66) mm scoticus) and darker than nominate cristatus and mitratus; upper parts are duller and browner, underparts much duller, flanks and undertail coverts browner, less buff. White fringes of crown narrower and duller, also white of underparts duller and more dingy. Scoticus is closest to weigoldi (from the Iberian peninsula) in size and colour, though latter unlikely to cause confusion in Scotland. Weigoldi has pale fringes on forehead and crown broader and whiter than scoticus. Occurrences in Scotland of races other than scoticus must be considered unlikely. Even ‘paler and larger’ Scottish Crested Tits will probably be just that, rather than another race, nevertheless there must be some chance of Scandinavian vagrants and, presumably, these are likelier to be detected outwith normal scoticus areas, for example at coastal sites. All birds found outside Scottish Birds (2003) Scotland’s endemic subspecies Ze the normal range of scoticus should therefore be closely examined to determine race. Common Starling (Shetland Starling) Sturnus vulgaris zetlandicus Hartert 1918. Novitates Zoologicae 25: 329. (North Yell, Shetland) Distributionally this is perhaps the most complex of the Scottish endemic subspecies. Hartert’s type was from Yell and his series was separated on bill width (broader than nominate vulgaris, narrower than faroensis), wing length (intermediate between nominate vulgaris and faroensis) and colour of juvenile plumage (darker than nominate vulgaris). Birds from Fair Isle were classed as intermediate between zetlandicus and nominate vulgaris and Hartert suggested that St Kilda birds also showed intermediate characters. Contrary to Baxter & Rintoul’s (1953) claim that “In 1918 the Starling of the Outer Hebrides and Shetland was separated from the typical form”, Hartert actually makes no reference whatsoever to birds from the Outer Hebrides. Bannerman (1953a) similarly includes Outer Hebrides within the range of zetlandicus. Witherby (1920) lists zetlandicus as an addendum in his Practical Handbook as occurring in Shetland excluding Fair Isle. However, the Handbook of British Birds (Witherby et al 1938a) subsequently includes the Outer Hebrides, including St Kilda, in the distribution of this race. The author (HFW) comments “Birds [= juveniles] from Outer Hebrides are as dark as those from Shetland” though “Birds from South Uist and St Kilda have very much the same measurements as S v vulgaris.” It would appear that these statements formed the proposition that zetlandicus is the form occurring in the Outer Hebrides and that this view became more formally adopted through the works of Bannerman and Baxter & Rintoul referred to above. It is worth noting, however, that the BOU (1971) record zetlandicus as a ‘poorly defined’ subspecies. Baxter & Rintoul (1953) record the Shetland Starling being a common bird, even when starlings were generally scarce throughout Scotland around the mid 19th Century. This isolation of the Shetland population no doubt facilitated its differentiation. Hartert (1918) interprets the Fair Isle intermediates as the possible product of immigration from both north and south to that island. A current population estimate for Shetland is c29,000 breeding pairs, ranging from 19-39,000 pairs (Pennington et al in press). The Starling population in the Outer Hebrides is estimated at 10,000+ pairs (B Rabbitts pers comm, A Stevenson pers comm). To complete the picture, between 100-300 pairs are regularly estimated on St. Kilda (Murray 2002), with a proportion including some autumn migrants overwintering. The British breeding Starling population is largely sedentary, with c80% of movements of ringed birds being less than 20km (Wernham et al 2002) and there is little in the way of hard data to suggest that zetlandicus may be otherwise. Evidence from intermediate type birds ringed on Fair Isle shows a degree of juvenile dispersion eg recoveries in Shetland, Orkney, Caithness, Ross, Aberdeenshire, Angus and the North Sea Dogger Bank. Of 12 birds found dead on North Sea oil instal- lations (skins in NMS), 2 (both October) from the Beatrice C oil field (roughly midway between Wick and Banff) have bill widths of 8.1mm and 8.4mm, which are suggestive of zetlandicus. It is worth noting the somewhat curious report of 5 zetlandicus recorded on 5 October 1936 in Greenland (Bird & Bird 1941); a single bird was collected and is held at the Natural History Museum, Tring. 26 Identification As mentioned, Hartert described this subspecies as intermediate between nominate vulgaris and faroensis. The latter, in comparison with nominate vulgaris, is a large heavily built race, with stronger legs and feet and bill broader at the base. Birds from Fair Isle were considered intermediate between vulgaris and zetlandicus and therefore less distinctive. Plumage of juveniles (of both zetlandicus and faroensis) 1s more sooty black, with less white on chin and belly, the white making the dark spots on throat more obvious. Adults tend to be less glossy but more dark green than nominate vulgaris. Identification of zetlandicus is, therefore, a fairly subtle matter, with width of bill at the base being the significant factor (8-8.5mm cf. 7—8mm in nominate vulgaris) (Cramp & Perrins 1994). Although birds from the Outer Hebrides have been classed by some authors as zetlandicus due to darker juvenile plumage, it has been indicated more recently that dark juveniles are not uncommon within the range of nominate vulgaris, particularly in Scotland (Cramp & Perrins 1994). With biometrics also similar to vulgaris, the case for zetlandicus as the taxon in the Outer Hebrides must be fairly tenuous. Field identification of zetlandicus outwith Shetland is therefore highly problematic. Essentially, either ringing data or data on bill width would be necessary to substantiate claims for this subspecies. Common Linnet (Scottish Linnet) Carduelis cannabina autochthona Clancey 1946. Bulletin of the British Ornithologists’ Club 66: 84. (Carmunnock, Lanarkshire) Clancey described this subspecies on the skins from Renfrewshire, Lanarkshire and Fife, stating SB 24(1) of its distribution “imperfectly known at the present time”. Referring to a specimen from Fair Isle showing the appropriate characters, he added that it “is presumed to range throughout Scotland to the Orkneys and Shetlands”. Baxter & Rintoul (1953) add nothing to Clancey’s assertion. However, Bannerman (1953a) does add “perhaps also Ireland”, remarking that Irish resident birds are variable though “mostly dark and match those from Scotland”. Nevertheless, no comparative analysis of this observation seems to have been undertaken. The BOU Records Committee (BOU 1971) view the subspecies as “poorly distinguished”’. Our current understanding of the precise distri- bution of autochthona remains obscure. Although autochthona is believed to breed throughout most of Scotland, it grades into cannabina (which breeds in the rest of the British Isles and continental Europe) and no precise boundary can be drawn between the 2 races. At present there are no confirmed instances of autochthona breeding outwith Scotland although gradation within the species probably extends either side of the border. Clancey (1946), himself, noted that birds from Northumberland “show a leaning towards” autochthona. In The New Atlas of Breeding Birds, 520,000 Linnet territories were estimated for Britain in 1988-91 (Gibbons et al 1993). On the basis of occupied squares in the Atlas, we assess the Scottish breeding population to fall within the range 60,000—100,000 pairs. Natal site fidelity in Linnets appears to be high, and many Scottish breeding birds over winter in Scotland (Wernham et al 2002). It is also believed that Western European and Scandinavian birds, of the nominate race C cannabina cannabina, contribute to the higher Scottish Birds (2003) numbers that are evident around the east coast during winter (Bannerman 1953a; Lack 1986). Movements of English breeding birds to the continent are well established, though any pattern of movement for autochthona is more difficult to discern due to lack of data (Wernham et al 2002). There is only a solitary record of a Scottish bred Linnet (= autochthona) recorded outwith Scotiand. The bird, ringed as a chick near Sanquhar, Dumfries & Galloway on 18 June 1928 was recovered near Egremont, Cumbria on 8 November 1928. (BTO pers comm). Two other recoveries outwith Scotland were of birds ringed as adults. As no racial determi- nations of the birds were made, there is less certainty that these were genuine autochthona. One bird, ringed in Lothian on 26 August 1964 was recovered in Spain in 1 January 1965. The other bird was ringed on the Isle of May on 6 May 1953 and controlled 7 weeks later in Tyne & Wear (BTO pers comm). Scotland’s endemic subspecies If Ringing data indicate some movement from England to Scotland e g Orkney and Strathclyde (BTO pers comm) Fifty years ago Bannerman (1953a) considered Linnets in the Outer Hebrides, Orkney and Shetland to be passage migrants from the continent (= nominate cannabina). However, no real determination of subspecies has so far been made regarding the recent colonisation of the Outer Hebrides (Murray 2000). Whilst the race appears to be largely sedentary, we still have much to learn about the overall distribution and movements of autochthona. Identification This race is, on average, marginally darker with slightly more obvious striations on mantle. In a given sample, Scottish birds, males in particular, will tend to show a darker grey nape and duller, darker brown mantle and_ scapulars in comparison with nominate cannabina. Plate 1 Subspecies of Crested Tit Parus cristatus. Painted from skins held by NMS. scoticus mitratus 28 As these features are average differences, some individuals (a minority) from the darker Scottish population may actually be lighter and less streaked than some from the nominate cannabina population. Similarly, there may exceptionally be some birds from the nominate population which are darker than the average Scottish birds. On an individual basis, therefore, it is virtually impossible to categorise a particular bird as autochthona or nominate cannabina. At a_ given locality eg Northumberland, such a determination would have to involve the study of a representative sample of the local breeding population. Conclusion Our knowledge of Scotland’s endemic bird forms is exceedingly sketchy. Very little research has taken place in recent years to determine the current status of most of these forms and birders have made little effort to get to grips with their field identification. In particular, the actual size of the population and specific distribution of several of these races is poorly known. Further research is desirable and may result in a reappraisal of their status and validity. Linnet and Song Thrush are both UK Red listed species, as a result of rapid (50%+) declines in their overall UK breeding populations (Gregory et al 2002). The decline in the population of the Scottish race of Linnet autochthona, appears to be less than the UK average and there is some recent evidence of a partial recovery, but these conclusions are based on sparse monitoring information (R Gregory pers comm). Due to a lack of research, we know nothing about population changes within the hebridensis race of Song Thrush, although it may well not have declined to the same extent as clarkei (B Rabbitts pers comm). St Kilda Wren has probably the best credentials of any of the endemic races in Scotland for SB 24(1) possible future upgrade to species level. On the other hand, the status of Scottish Linnet as a subspecies is more tenuous. Crested Tit appears in Schedule 1 Part 1 and Schedule 4 of the Wildlife and Countryside Act 1981. Scottish Crested Tit is the only one of our endemic races to have specific protection. Under Section 3 of the Wildlife and Countryside Act 1981, there are ‘Areas of Special Protection’ (ASPs) and under the Birds Directive there are ‘Special Protection Areas’ (SPAs). We understand that St Kilda (as a World Heritage Site) and much of Fair Isle (the cliffs and hill area) are SPAs. As a result the entire population of St Kilda Wren and that part of the Fair Isle Wren population within the protected area, should have quasi Schedule 1 protection status, however this does not appear to have been fully tested in law. The majority of the population of all other endemic subspecies are at any one time outwith areas with either ASP or SPA status and therefore they have no special protection other than that afforded to the species. Common Starling appears in Schedule II Part 2 of the Wildlife and Countryside Act 1981 and as the Shetland race receives no special recognition it is therefore treated as a ‘pest’ species. There is certainly a strong case for subspecies to be afforded a higher level of overall protection than currently exists. We hope this paper raises awareness about the identification features, status and distribution of Scotland’s endemic subspecies and_ will stimulate further study. Scottish Birds (2003) Scotland’s endemic subspecies 29 Appendix 1 Original descriptions Text of the first published descriptions is given below. With one exception, the full citations for the following scientific names have appeared above in the main treatments of the subspecies. With regard to the name for Rock Ptarmigan, however, Hartert first used the subspecific name millaisi in 1923 (British Birds 17:106), although his description of the Scottish Ptarmigan originally appeared in Die Végel der paldark- tischen Fauna (1921 page 1868). Rock Ptarmigan (Scottish Ptarmigan) Lagopus mutus millaisi [Translation from German] The Scottish ptarmigan is clearly different from L mutus mutus in the following. [Adult] male in [breeding plumage] in L m mutus is rarely totally free of white feathers, on the upper tail covers, rump and back a few (often many) white feathers are [retained], [adult male mutus] always moulted really dark, without white spots. The old female in the same plumage [i e female breeding plumage] is [dorsally] and on the crop lively yellow brown, nearly orange-brownish. Autumn plumage is of a pure grey, with a lesser yellow brownish trace as in [nominate] mutus! This is really easy to see even in a limited series. The winter plumage, of birds from Scandinavia and Russia [is] always pure white, [usually] shows [in millaisi] some dark spots; on the colder east Scottish heights it [is said that it] can often go pure white. Wings of [adult] males 185-207 (most 192-203) mm. It is therefore easy to distinguish from L m mutus, but very similar to L m helveticus. After investigation of the specimens in the collections of John Millais, the British Museum, the Royal Scottish Museum in Edinburgh and the collection of Miss Jackson is it certain that it is dissimilar to L m mutus. Lives currently only in the mountain heights of Scotland, including the Inner Hebrides (Mull, Islay, Jura, Skye), but is said to be extinct on the outer Hebrides and Orkney islands, also on the heights of the lake district of north England. Winter Wren (Shetland Wren) Troglodytes troglodytes zetlandicus [Translated from German.] Definitely different from T t troglodytes by its significant size and in general only comparable to T t¢ islandicus, borealis and hirtensis - beak at least equally strong as in islandicus, culmen 15-16 mm, but wings are shorter: in 3 males 52—53.3, in one female 48.1mm. Different from T t borealis by the darker upper side, especially darker upper head, slightly darker wings and slightly darker under side, but especially by the stronger beak, [different] from T t hirtensis by the darker colouration, especially of the underside. Cross banding broad and dark. Tail male 37-39, female 34, foot 19mm. Type: female adult Dunrossness, Shetland Islands, north of Scotland, 12 Dec. 1906, collected by N B Kinnear, No. 59F in Royal Scottish Museum in Edinburgh. Often a resident on the rocky, stormy Shetland Islands, where he [the bird] fearlessly occurs and nests in rock crevices and under the grass cover over [hanging] cliffs. Winter Wren (Fair Isle Wren) Troglodytes troglodytes fridariensis Autumn plumage — paler, not so dusky, on breast and centre of belly as T t zetlandicus. Chin and throat whiter. Underparts suffused with buffish, but less strongly than in _hebridensis. Vermiculation on vent, flanks and sides of breast almost as extensive as on the Shetland form, and much more so than in the typical race and most Hebridean birds. Upper parts paler and brighter rufescent than in the Shetland race (which is dark reddish brown above) and having a greyish suffusion in fresh plumage. Worn breeding plumage — Five June birds from Shetland are darker rufous brown than 2 fridariensis indigenus y Scottish Birds (2003) Scotland's endemic subspecies al hebridensis clarkei philomelos MO | ES Mauss nh | ‘Plate 3 Subspecies of Song Thrush Turdus philomelos. Painted from skins held by NMS _ including syntypes of hebridensis. 32 comparable specimens from Fair Isle, which are brighter, more rufescent. Measurements — In wing, bill and tarsus measurements Fair Isle birds are intermediate between the Shetland and typical races. Winter Wren (Hebridean Wren) Troglodytes troglodytes hebridensis From T t troglodytes these differ in being darker and more heavily barred underneath. In size they are similar. From zetlandicus they do not differ in the colour of the upper parts, but on the whole the barring of the underparts is more confined to the abdomen, whereas in zetlandicus the barring usually extends from the abdomen to the breast. The bill in similar sexes is invariably smaller than in Shetland birds, and resembles the mainland form. Winter Wren (St Kilda Wren) Troglodytes troglodytes hirtensis Closely resembles T t pallescens [Commander Islands and Kamchatka] but much more distinctly barred on the back and head, and almost free from any traces of spots on the throat and breast. In general colour ... quite as pale and slightly greyer than examples ... from Algeria and Turkestan. The bill resembles that of borealis. The eye stripe is as distinct as in typical examples from Europe. Song Thrush (Hebridean Song Thrush) Turdus philomelos hebridensis Decidedly darker in plumage than any British or Continental representatives of the species. Mantle and wings are dark (clove) brown, the head is slightly redder, and the rump and upper tail coverts are dark olive. Thus the upper plumage of the Hebridean bird more resembles that of the Continental race than that of its British cousin. The most striking feature of the under surface, and indeed the plumage generally, is the multitude of intense black, ovate spots on the throat, chest and abdomen; while the buff, SB 24(1) which is confined to the throat and chest, is very pale as compared with song thrushes from other areas. The flanks are pronouncedly streaked with greyish brown, and show little of the buff, which is much in evidence on the flanks and breast of other song thrushes. The buff of the underwing coverts is richer (redder)... The wing measurements range from 116 to 120 mm, so that it is a mistake to say that the birds are smaller than their mainland representatives, which has been maintained by some writers, they are fully up to average. Crested Tit (Scottish Crested Tit) Parus cristatus scoticus [Translated from German.] Since finishing the manuscript of the current work I came to the conclusion, that rufescens is a well distin- guishable subspecies of the crested tit and is totally different from the eastern form. As I intend to return to these forms more thoroughly later on, here I give only short hints. Lophophanes |= Parus] cristatus falls into 3 geographically separated forms: 1. cristatus typicus (Linnaeus), entire upper side, 1 e back and rump isabell brown. [Range] Scandinavia, Eastern Prussia, Baltic provinces, Poland. 2. cristatus mitratus (Brehm), back wood brown grey, the rump more as the following [subspecies]. [Range] Central Europe. 3. cristatus rufescens (Brehm), white head colour more dull, belly sides more strongly [tinged with] a lively rust colour, crest usually longer and more upwardly curved, the red brown of the upper side, inclusive of the rump more intensive. Range Western Europe. If No 2 is perhaps very close to the typical form, No. 3 is totally different, as Brehm already knew very well. Namely the Scottish specimens are very dark brown and could maybe be called scotica. Common Starling (Shetland Starling) Sturnus vulgaris zetlandicus Nearest to faroensis, but bill not so wide and Scottish Birds (2003) long, though wider (only sometimes longer) than in vulgaris. First primary not as broad as in faroensis. [Series of wing lengths.] In 12 adult faroensis wings 133-136 mm, in 200 vulgaris only 128-132 mm, (exceptionally up to 134). Juv. As arule as dark as those of faroensis, much darker than those of vulgaris. [comparisons with Fair Isle specimens] — one must therefore say that these birds [Fair Isle] are on the whole intermediate between vulgaris and zetlandicus. Common Linnet (Scottish Linnet) Carduelis cannabina autochthona Male, autumn, differs from A c sejuncta [= nominate cannabina]| on account of the much darker grey brown of the crown and nape; darker and more heavily striated mantle; duller underparts with more strongly accentuated gular markings. Female, autumn, considerably darker and more strongly striated on upperparts. Underside with darker markings than [nominate cannabina]. Appendix 2 In addition to the endemic races (plus Hedge Accentor Prunella modularis hebridium) there are several other taxa, which have been described from specimens obtained in Scotland. These are:- Species Northern Gannet Morus bassanus Linnaeus 1758 Systema Naturae 10th edition, page 133. (Bass Rock) Roseate Tern Sterna dougallii Montagu 1813 Ornitholigical Dictionary Supplement (Cumbrae, Firth of Clyde) Scottish Crossbill Loxia (curvirostra) scotica Hartert 1904 Die Vogel der paldarktischen Fauna, page 120 (Easter Ross) Scotland’s endemic subspecies 35 Subspecies Willow Ptarmigan (Red Grouse) Lagopus (lagopus) scoticus Latham 1787 A General Synopsis of Birds. Suppl. 1. page 290 (Scotland) Winter Wren Troglodytes troglodytes indigenus Clancey 1937 Bulletin of the British Ornithologists’ Club 57: 143 (Carmunnock, Lanarkshire) Meadow Pipit Anthus pratensis whistleri Clancey 1942 Bulletin of the British Ornithologists’ Club 63: 6. (Dornoch, Sutherland) Yellowhammer Emberiza citrinella caliginosa Clancey 1940 Ibis 1940, page 94 (Dornoch, Sutherland) Acknowledgements We wish to thank the following for help and assistance with this paper: Lloyd Austin, Andrew Currie, Richard Gregory, Angus Murray, Brian Rabbitts, Roger Riddington, Andrew Stevenson, Ron Summers, Adam Watson and Phil Whitfield for supplying information about particular subspecies; Jorn Scharlemann for translating German text; Graham Appleton for prepublication access to the BTO’s Migration Atlas; The BTO Ringing Scheme, which is funded by a partnership of the British Trust for Ornithology, the Joint Nature Conservation Committee (on behalf of English Nature, Scottish Natural Heritage and the Countryside Council for Wales, and also on behalf of the Environment and Heritage Service in Northern Ireland), Duchas the Heritage Service — National Parks and Wildlife (Ireland) and the ringers themselves. References Bannerman D A 1953a. The Birds of the British Isles, Vol. 1. Oliver & Boyd, Edinburgh. Bannerman D A 1953b. The Birds of the British Isles, Vol 2. Oliver & Boyd, Edinburgh. 34 Bannerman D A 1954. The Birds of the British Isles, Vol 3. Oliver & Boyd, Edinburgh. Baxter E V & Rintoul L J 1953. The Birds of Scotland. 2 Vols. Oliver & Boyd, Edinburgh. Bird C G & Bird EG. 1941. The birds of Northeast Greenland. /bis 83:118—161. British Ornithologists’ Union 1971 The Status of Birds in Britain and Ireland. Blackwell, Oxford. British Ornithologists’ Union 1992. Checklist of Birds of Britain and Ireland 6th Edition. BOU, Tring, Herts. Clancey P A 1946. Two new races of Acanthis cannabina (Linnaeus) from the Western Palearctic region. Bulletin of the British Ornithologists’ Club 66:83—85. Clarke W E 1913. The Song Thrush of the Outer Hebrides — Turdus musicus hebridensis — anew racial form. Scottish Naturalist 1913:53-55. Clements J F 2000. Birds of the World : A Checklist. 5th Edition. Pica Press, Robertsbridge. Clugston D L, Forrester R W, McGowan R Y & Zonfrillo B 2001. The Scottish List — species and subspecies. Scottish Birds 22:33-49. Collinson M 2002. How Many Wrens? Birdwatch No 121: 28-32. Cook M J H 1982. Breeding Status of the Crested Tit. Scottish Birds 12:97-—106. Cramp S (ed) 1988. The Birds of the Western Palearctic, Vol V. Oxtord University Press, Oxford. Cramp S & Perrins C M (eds) 1993. The Birds of the Western Palearctic, Vol VU. Oxford. University Press, Oxford. Cramp S & Simmons K E L (eds) 1979. The Birds of the Western Palearctic, Vol IU. Oxford. University Press, Oxford. Gibbons D W, Reid J B & Chapman R A 1993. The New Atlas of Breeding Birds in Britain and Treland: 1985—1991.T & A D Poyser, London. Gregory R D, Wilkinson N I, Noble D G, Robinson, J A, Brown A F, Hughes J, Proctor D, Gibbons D W & Galbraith C A. 2002. The SB 24(1) population status of birds in the United Kingdom, Channel Islands and Isle of Man: an analysis of conservation concern. British Birds 95:410-448. Hartert E 1910. Die Vogel der paldarktischen Fauna. Friedlander & Sohn, Berlin. Hartert E 1918. Notes on Starlings. Novitates Zoologicae 25:327—337. Hartert E 1923. The name of the Scottish Ptarmigan. British Birds 17:106. King H 1969. Shetland Wren in Aberdeenshire. Scottish Birds 5:391. Lack P 1986. The Atlas of Wintering Birds in Britain and Ireland. Poyser, Calton. Macmillan A T 1969. [editorial comment] Scottish Birds 5:391. Mead C 2000. The State of the Nations’ Birds. Whittet Books, Stowmarket. Meinertzhagen R 1924. A note on Scottish wrens (Troglodytes) with characteristics of a newly defined Hebridean race. Scottish Naturalist 1924:135. Meinertzhagen R 1934. The relation between plumage and environment, with special reference to the Outer Hebrides. [bis 96:52-61. Murray R D (ed) 2000. 1998 Scottish Bird Report. SOC, Edinburgh. Murray S 2002. Birds of St Kilda. Scottish Birds Supp. Vol 23. Paton ER & Pike O G 1929. The Birds of Ayrshire. Witherby, London. Pennington M G, Osborn K, Harvey P V, Riddington R, Ellis PM, Heubeck M & Okill J Din press. Birds of Shetland Christopher Helm/A & C Black, London. Prazak J P 1897. Nachtraglische Bemerkungen. Journal fur Ornithologie 45:347-348. Ratcliffe D A 1990. Bird Life of Mountain and Upland. Cambridge University Press, Cambridge Rogers M J & the Rarities Committee 2002. Report on rare birds in Great Britain in 2001. British Birds 95:476-528. Scottish Birds (2003) Ruttledge R F 1966. Ireland’s Birds: their distribution and migrations. Witherby, London. Seebohm H 1884. On new species of British Wren. Zoologist 1884:333-335. Stone B H, Sears J, Cranswick P A, Gregory R D, Gibbons D W, Rehfisch M M, Aebischer N J & Reid J B 1997. Population estimates of birds in Britain and in the United Kingdom. British Birds 90:1—22. Summers R W, Mavor R A, Buckland S T & MacLennan A M 1999. Winter population size and habitat selection of Crested Tit Parus cristatus in Scotland. Bird Study 46:230-—242. Thom V M 1986. Birds in Scotland. T & AD Poyser, Calton. Vaurie C 1959. The Birds of the Palearctic Fauna. Passeriformes. Witherby, London. Vaurie C 1965. The Birds of the Palearctic Fauna. Non-passeriformes. Witherby. London. Waters E 1964. Observations on the St Kilda Wren. British Birds 57:49-64. Wernham C, Toms M P, Marchant J H, Clark J A, Sirtwardena G M & Baillie S R. 2002. The Migration Atlas: Movements of the birds of Britain and Ireland. T & AD Poyser, London. Scotland’s endemic subspecies 2/2 Williamson K 1951. The Wrens of Fair Isle. Ibis 93:599-601. | Williamson K 1958. Population and Breeding environment of the St Kilda and Fair Isle Wrens. British Birds 51:369-393. Williamson K 1965. Fair Isle and its Birds. Oliver & Boyd, Edinburgh. Witherby H F (ed). 1920. A Practical Handbook of British Bird, Vol 1. Witherby, London. Witherby H F, Jourdain F C R, Ticehurst N F & Tucker B W. 1938a. The Handbook of British Birds, Vol 1. Witherby, London. Witherby H F, Jourdain F C R, Ticehurst N F & Tucker B W. 1938b. The Handbook of British Birds, Vol Il. Witherby, London. Address for correspondence: Ronald W Forrester, Secretary, SBRC, The Gables, Eastlands Road, Rothesay, Isle of Bute PA20 9JZ Revised manuscript accepted February 2003 36 SB 24(1) Eurasian Reed Warblers in Scotland: a review of probable breeding records In Scotland, the Eurasian Reed Warbler Acrocephalus scirpaceus occurs regularly on migration in spring and autumn. Until the 1990s, it was thought to have bred only once in Scotland: in Shetland in 1973 (Thom, V M 1986, Birds in Scotland. Poyser, Calton). A number of birds have been seen in breeding habitat on the Scottish mainland since the early 1980s and the following account examines the extent of probable and possible breeding records in Scotland. In England and Wales it is a common breeding bird where the species is closely associated with reedbeds Phragmites australis since the favoured nesting habitat consists of reeds above standing water. This is the habitat indicated by all the possible breeding records in mainland Scotland although, in England, birds will also nest in dry herbage and arable crops (e g Kelsey, M 1993 in The New Atlas of Breeding Birds in Britain and Ireland. Poyser. Calton, London). The species avoids the colder, wetter, north and west of the UK. (Sharrock, J T R 1976 ed.) The Atlas of Birds in Britain and Ireland. T & A D Poyser. Calton, London). In recent years, however, the species has expanded its range to the north and west in Britain and northward in Scandinavia (Kelsey ibid). In Scotland, most records of Eurasian Reed Warblers are of birds at coastal sites and bird observatories where migrants occur on passage with peak numbers during autumn migration in August—October with some _ records in November and smaller numbers during spring migration in May—June. These records are often associated with weather patterns causing drift migration across the North Sea. D ROBERTSON Spring migration is protracted and some late birds are unlikely to return to breeding grounds in Western Europe before mid June (Dowsett- Lemaire, F and Dowsett, R J 1987. European Reed and Marsh Wablers in Africa: Migration patterns, moult and habitat. Ostrich 58: 65-85) by which time early nesters may have already fledged young. However, there are very few records from Scottish coastal sites in July and the first week in August, presumably because juveniles begin a partial post juvenile moult before migrating (Ginn, H B & Melville, D S 1983. Moult in Birds. British Trust for Ornithology, Thetford) and tend not to disperse far from their natal areas until some 51-60 days after fledging (Redfern C and Alker P in The Migration Atlas. Poyser. London: 548-551). Breeding was suspected at a site in south west Scotland in 1992 and confirmed in 1993. Since then, breeding has occurred there annually and a second breeding site was located in south west Scotland in 1996 (Bruce, K 1997. Reed Warblers breeding in south west Scotland. Scottish Birds. 19(2): 119-120). One pair bred at St Abb’s Head in 1997 and breeding was thought likely at the same site in 1998 but has probably not taken place since (Rideout in Jitt). In Fife, there are tantalising records of possible breeding at a number of sites. In 1998, a bird was heard singing through the summer at a suitable breeding site and a pair was present at another site where breeding was suspected (Fotheringham D, ed 1998 & 1999. Fife Bird Reports 1997 and 1998, Fife Bird Club). A bird was found dead at St Margaret’s Marsh, Rosyth, on 31 May 1993 where a bird was heard singing repeatedly in 1999 (Dewick pers comm). A bird Scottish Birds (2003) 24:36-39 A review of probable breeding records of Eurasian Reed Warblers 37, Table 1 Ages and year of capture of Eurasian Reed Warblers in the Tay Reedbeds. Retraps are not included. The single control is included. Year Juvenile Adult Age unknown 1981 1 1982 2 1983 1 1984 1985 1986 1 1987 pi 1988 2 1989 l 1990 l 1991 l 1992 1993 1994 1995 3 2 1996 l 1997 3 1998 1999 1 2000 2 2 was heard singing at the same site in 2000. In 1999 a breeding record was reported from yet another site in Fife where a used nest, thought to be of this species, was located. (Shuttleworth A 2000. Eurasian Reed Warbler, First Probable Breeding in Fife. Fife Bird Report 1999: 154). In Tayside, Eurasian Reed Warblers have been caught in the Tay Reedbeds at least since 1981 (Lynch in litt) and a minimum of 21 have been ringed there. In recent years they have been trapped there almost annually and birds have been heard singing during and, in some cases, throughout the breeding season ie June, July and early August in 1987, 1990, 1995 and 1996. Males with developed cloacal protruberances, and presumably in breeding condition, were caught in 1987, 1990, and 1991. A female with a well developed brood patch was caught in 1987 and, in 1988, a juvenile, with primaries still in pin and growing, was trapped and ringed (Robertson D 2001. Reed Warblers Breeding in the Tay Reedbeds. Tay Ringing Group’ Report 1998-2000). These data indicate that Eurasian Reed Warblers probably bred in the Tay Reedbeds numbers hk 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 Figure 1 Reed Warbler ccurrence between Ist July—7th August. 38 SB 24(1) Table 2 Observations at various sites from I July-7 August Year Tay Est St Abbs — \O oe) & wNoocoroocoocorr rR Were TD COcOrcocoocoococo°o WwWwooracncooodocdcoooc*nc”eoroocrooooooncoqc”jocto in 1987 and 1988. Breeding possibly occurred in 6 other years including 1999 and 2000 when further instances of breeding behaviour were observed. These are the first records of Eurasian Reed Warblers breeding in mainland Scotland. The species is probably under recorded in Scotland. It is a bird that is difficult to see in breeding habitat, especially if observers are not actively searching for it. Its song may be difficult to separate from that of Sedge Warbler Isle of May Fair Isle Fife oooorcorcocoeooc*coooqcoorqcocooocoooqo”tooqco PreWOOdOWrFcoeoceoeococoeocoeoeocoeocoocooqooqooooqoco SCNQUNHK-ROWWNKRODAAENE WE RDN HERE OOHNOW Acrocephalus schoenobaenus and can _ be difficult to hear in extensive reedbeds such as those on the Tay; the sort of habitat that colonising birds would favour. Thinly dispersed, colonising birds may sing less where they have no neighbours of the same species to compete with (Cramp, S and Simmons, K E L eds Birds of the Western Palearctic. Vol 6 Oxford University Press, Oxford) The recent confirmed breeding records in Scotland refer to ringing sites and particularly Constant Effort Sites where ringing Scottish Birds (2003) takes place throughout the breeding season. Consequently, it seems likely that a number of breeding attempts have been overlooked. Peak passage along the coast appears to be in October, at sites such as Fife Ness (J Cobb in litt), the Isle of May (Isle of May Computerised Logs), North Ronaldsay and Fair Isle Bird Observatories (Scottish Bird Reports). Records from the Isle of May and Fair Isle show very few ‘summer’ records before the end of the first week in August and there are no records of birds during this period on North Ronaldsay, whereas high number of birds in the Tay Reedbeds and inland Fife are caught in late July and early August. At Fife Ness, there is just one record of a bird caught in early August but, since the mid 1980s, birds have begun to appear at this site when there were no signs of drift migration (Cobb pers comm). This pattern of occurrence suggests that birds caught in late July and the first week of August may be Scottish breeding birds, perhaps local breeders or their young. Some 13 birds have been caught in the Tay Reedbeds during this “early” period and the first one of these was trapped in 1981. A total of at least 26 Eurasian Reed Warblers have been trapped in the Tay Reedbeds consisting of 11 juveniles, 11 adults and 4 of unknown age. There seem to be very few records of birds from other parts of Scotland during this ‘summer’ period. Potential breeding habitats such as the Insh Marshes in Highland and the reedbeds at Loch of Strathbeg in Aberdeenshire have no records of birds in the breeding season (RSPB pers comm) The exception to this seems to be Fife where, in addition to the sightings in this paper, birds have been caught in the first week of August in suitable habitat at St Margaret’s Marsh (Darling in litt) and Loch Gelly (Little in litt). Ringing recoveries indicate that a proportion of Scandinavian birds pass through Scotland on A review of probable breeding records of Eurasian Reed Warblers 39 southward migration, although none of these suggest that birds travel through Britain on their way to breeding grounds further north (Redfern C and Alker P ibid). Breeding records on the east coast of Scotland could be linked to expansion of the breeding population in Scandinavia. The expansion of range in Scotland may be an extension of the breeding population in northwest Europe as a whole, possibly involving birds from both the UK and Scandinavia and, perhaps, reflecting a change in climatic conditions brought about by global warming. The confirmed breeding records on the Tay in 1987 and 1988 and the breeding records at St Abbs coincide with peak numbers of birds on Fair Isle and 3 possible breeding records in Fife. While this suggests an influx of Scandinavian birds the picture is not clear. The records of birds at St Abbs and Isle of May in 1982 and 1994 and at Isle of May and the Tay in 1996, along with a scattering of other records when there were no large influxes in Fair Isle suggest that these birds might not have come in across the North Sea. Eurasian Reed Warbler is probably now best described as a sporadic and under recorded breeding bird in Scotland which has established itself at a handful of sites. It will be interesting to see if this extension to its breeding range continues. Acknowledgements Many thanks to all of the following for collating observations: Dave Arthur, Ken Bruce, Jim Cobb, Ian Darling, Tom Dewick, Les Hatton, Brian Little, Stephanie Little, Shirley Millar, Steve Moyes, Mark Oksien, Kevin Rideout, Deryk Shaw, Kevin Woodbridge, RSPB Edinburgh, Isle of May Bird Observatory, Fair Isle Bird Observatory, Fife Bird Report. Thanks also to Dr Stan da Prato who revised an earlier draft of this paper. Derek Robertson, Woodlands Studios, Bandrum nr Saline, Fife KY12 9HR Revised manuscript accepted February 2003 40 SHORT NOTES Unusual behaviour of Common Redshanks and Common Starlings towards dead Common Redshank At 11.45 am, on 23 September 2002, during a visit to North Ronaldsay, a Common Redshank Tringa totanus was seen to fly into overhead power cables traversing a grassland field being grazed by cows. The dead Common Redshank fell to the ground below the power cables. There was an immediate response by a nearby flock of feeding Common Redshanks. Many birds began alarm calling and 17 flew towards the corpse. Most of these 17 birds then moved in close to the dead Common Redshank, loosely forming a circle and looking intently with their necks Eurasian Oystercatcher apparently brooding young of Ringed Plovers There are many examples of Eurasian Oystercatcher Haematopus ostralegus displacement in conflict situations including ‘mock brooding’ but not, apparently, of a different species (Cramp & Simmons 1983, The Birds of the Wester Palearctic, vol 3. Oxford; Nethersole-Thompson & Nethersole- Thompson 1986, Waders, their breeding haunts and watchers. Calton). During 1975, on a small lowland heath (>1/2 acre) in West Galloway, the density of nesting waders was high including 4 Eurasian Oystercatcher nests, 3 Ringed Plovers Charidrius hiaticula nests and 4 Northern Lapwings Vanellus vanellus nests as well as 2 SB 24(1) outstretched and upwards, as if looking for a better view. The Common Redshanks were nervous, highly agitated and approached the corpse very cautiously. The Common Redshanks were then joined by a group of Common Starlings Sturnus vulgaris that behaved in an almost identical manner. This behaviour continued for about 2 minutes before all the birds began to lose interest. At this point a car passed by and flushed all the birds, none of which returned to the dead Common Redshank. The corpse was retrieved and the first winter Common Redshank was found to have had a broken neck and lower and upper mandibles. Peter Cosgrove and Ross McGregor, c/o North Ronaldsay Bird Observatory, North Ronaldsay, Orkney, KW17 2BE Revised manuscript accepted February 2003 nests of Common Terns Sterna hirundo. On 19 May 1975 an Oystercatcher was still brooding a mixed clutch of one Oystercatcher egg together with 2 Lapwing eggs (Dickson 1977, Oystercatcher hatching Lapwing eggs. British Birds 70:36). Most of the Ringed Plover eggs had hatched but only 1 chick was seen 10—15m from the Oystercatcher nest. An Oystercatcher, probably the mate of the mixed clutch, ran over to the Ringed Plovers’ chick and settled on top of the chick as if it was brooding it. One of the adult Ringed Plover circled the Oystercatcher in an ‘agitated’ manner and through binoculars I saw the Oystercatcher leave the brooded chick some time later. Oystercatchers may be unusually poor at distinguishing other eggs from their own eggs (Speakman 1987, Mixed Clutch of Oystercatcher and Lapwing eggs incubated by an Oystercatcher, Scottish Birds 14:184—185) Scottish Birds (2003) 24:40-42 or at distinguishing other wader chicks from their own, since they have also been recorded brooding Lapwing chicks from mixed clutches (Dickson op cit; French & Insley 2000, Oystercatcher usurping Lapwings nest. Scottish Birds 21:53-54). Perhaps the density of several species nesting close together may have added to the Oystercatchers’ behaviour. ia ron * *- ara Eurasian Oystercatcher by Dan Powell Female Eurasian Sparrowhawk caching prey Last Summer I had a similar experience to K Headle (Scottish Birds 2001 page 104). A female Eurasian Sparrowhawk Accipiter nisus killed a female Blackbird Turdus merula early in the morning in our garden and took it under the overhanging branches of a conifer hedge where she plucked and fed from it. After about 20 minutes she moved the remains further under the hedge and flew off. Several hours later she was seen again feeding from the Blackbird and after another 20 minutes moved what little was left further down the hedge and flew off. I examined 2 fs yp thé te ~s 4 Short Notes 4] R C Dickson, Lismore, New Luce, Newton Stewart, Dumfries & Galloway DG8 0AJ Revised manuscript accepted December 2002 MLE A \ 7 2 \ AT ANN SA ONY the remains, which were mainly the skeleton. That afternoon, I saw the hawk fly into the garden, grab the remains of the Blackbird from under the hedge and fly off with it. RN Cinderey, The Glebe, Church Road, Kirkpatrick-Durham, Castle Douglas DG7 3HF Revised manuscript accepted February 2003 We do not intend to publish any more examples of this behaviour. 42 Treecreeper feeding from a food cache in a brick wall On 12 February 2003 I was sitting in my office in the John Muir Building at about 1700hrs, facing a small side window looking out along the external wall, when I happened to notice a Treecreeper (Certhia familiaris) alight on the wall immediately outside. It moved up the wall in the manner characteristic of the species when foraging on tree trunks, but there was no obvious sign of it feeding. Then, when approximately level with the top of my window, it stopped and began to probe its bill energetically into a crack in the mortar at the corner of a chipped brick. It was clearly feeding, and occasionally a pale, shiny substance, which looked superficially like fat, was observed on its bill as it was withdrawn from the hole. The bird either licked this off with its tongue, or alternatively cleaned its bill on the brickwork before continuing its feeding. The bird continued to feed in this same spot for some 10 minutes, before cleaning its bill thoroughly on the brickwork and then moving off up the wall and out of view. SB 24(1) It seems probable that the bird was feeding at a food cache previously deposited in the brickwork, since it seems unlikely that it could have found the food so directly just by coincidence, particularly since this was my first observation of a Treecreeper on this wall in some 7.5 years. Whether the food had been deposited by the Treecreeper itself or by another species is uncertain, as I have previously observed Blue Tits (Parus caeruleus) and Great Tits (Parus major) storing food in crevices on this same wall. These food items were probably insects, which over time could conceivably break down into a fatty detritus, such as that observed here. I am not aware of previous records of Treecreepers utilising cached food stores, either deposited by themselves or in a kleptoparasitic fashion (if deposited by other birds). AR Lyndon, School of Life Sciences, John Muir Building, Heriot-Watt University, Riccarton, Edinburgh, EH14 4AS Revised manuscript accepted March 2003 Treecreeper by Dan Powell Scottish Birds (2003) 24:43-47 OBITUARIES C N L Cowper 27 April 1928-18 August 2002 An unexpected adventurer It doesn’t do to take people at face value. Charlie, with his old fashioned way of considering you and what you had to say, always seemed the archetypal Edinburgh professional man. Given his family background and education, this was perhaps to be expected, but the sometimes offputting exterior concealed an adventurous spirit, highly sociable inclinations, a dry, not always politically correct, sense of humour and a multitude of unsuspected interests and talents. Above all, a lifelong passion for the outdoors, for wildlife and especially for birds underlay most of his activities and adventures and many of the lifelong friendships he forged. Charlie was born in Edinburgh and educated at The Edinburgh Academy. As an early member of the SOC, he was a regular visitor at Aberlady, with contemporaries such as Keith Macgregor, Frank Hamilton, Jim Stewart and Bill Birrell. Charlie loved the Cairngorms and, after National Service in the Royal Signals, established a number of links there, keeping a family cottage at Newtonmore and making trips in search of Golden Eagle and other Highland birds with Doug Weir and Roy Dennis. When the Osprey project began, Charlie, with the expertise in electronics gained from his employment at Ferranti, was instru- mental in the design of the electronic alarm system and its installation in the nest tree at the Loch Garten site and renewed and tested the system every spring for many years. He loved the Osprey Camp in its early, tent and caravan days but did not approve of its later development, which he felt was out of keeping with the spirit of the place. Obituaries 43 Although a keen walker, Charlie was also addicted to unorthodox means of locomotion, whether to get where he wanted or just for the hell of it. He was an early skiing enthusiast, taking a carload of friends on winter Sundays over the first ferry at Queensferry and on up to Glenshee, returning the same evening. One New Year, in that era of very cold winters, first footing in Edinburgh, he startled Keith and Dorothy Macgregor by appearing on skis at their door in Merchiston. He built a canoe and used it for many years for trips on the Spey and Highland lochs. His enthusiasm for esoteric motor vehicles and for rallying combined with his keenness for birding sometimes led to awkward situations: once he had to winch his Landrover from an impassable track across the Lammermuirs and on another occasion to manhandle his MGB out of the dunes at Barns Ness. He also tried his hand at gliding and hot air ballooning. 1S ae Charlie on top of the Loch Garten nesting tree wiring up the alarm 44 Charlie’s membership of the SOC was a basis not only for friendships and excursions but also for much co operative activity. He was active in the (unauthorised) attempt in the late 50’s to create open pools on the saltmarsh at Aberlady. He took part over many years in the Wednesday night meetings at Regent Terrace, regularly tackled Threipmuir for Willie Brotherstone’s goose counts and organised and wrote up a survey of Edinburgh rookeries, (SB Vol 3, p177) and a survey of breeding Grey Wagtails, Dippers and Common Sandpipers on the Midlothian Esk (SB Vol 7, p302). The Isle of May was another of his great loves: he made annual visits there from the early 50’s and served as Treasurer to the Trust for many years. He worked at Ferranti in potentiometers, microelectronics and quality engineering from 1952 to 1990, and his technical interests ranged from electron microscopy through CB radio to astronomy. Charlie enjoyed overseas birding, making a pioneering trip in the early days to Sweden (in his Morris Minor) and to the Camargue. He had never much approved of things American, but, soon after retiring, relented and took an organised birding trip to Arizona... he was surprised to find the country and the people quite to his liking. Latterly, combining family and birding interests, he made regular visits to Australia, clocking up an outstanding list of exotic species. Charlie is survived by his wife, Rosemary, and children Gordon and Catriona. Tom Delaney Donald M Macdonald, MBE 1912-2002 Donald (Donnie) Macdonald passed away on 31 March 2002 aged 90 years. For almost 70 years he recorded the bird life around his native SB 24(1) Dornoch having been resident there all his life apart from one year in Edinburgh and 4 years in the Royal Navy during the war. His father, Captain Macdonald of the Seaforth Highlanders, had been killed in action when he was only 7 and it may be, in a strange way, that this gave him the strength of character to forge ahead with his hobby of bird watching when such a pursuit would have been regarded locally as almost slightly eccentric. He once told me how he would hide his binoculars under his jacket until he passed beyond the outskirts of the burgh. A meeting with Dr I D Pennie began a life long friendship and together they travelled to many SOC conferences. These conferences were major occasions in Donnie’s life and must have acted as a tremendous stimulant to his own research on Corn Bunting and Whitethroat. In his diaries conference programmes are displayed with each lecture and even the dinner being allocated a rating varying from superb to good. His diaries start in 1933 and form an extraordi- narily detailed account of the weather and bird life in SE Sutherland spanning a period of almost 70 years. His wish was that his journals would be deposited in the SOC library and it is easy to imagine some young researcher finding them a fascinating source of information in years to come, for in them are recorded arrival and departure dates of migrants along with changes in bird life over the years. Donnie was an all round naturalist and would enthuse just as much about a previously unrecorded billing display between male and female Blackbirds as he would about notable finds such as the first Stilt Sandpiper for Scotland. Along with many contributions to the Scotsman, Scottish Naturalist and Scottish Birds he wrote 2 major papers for Scottish Birds on _ the Whitethroat and the Corn Bunting. Neither bird was particularly common in SE Sutherland and NE EE-"”T~—”——————————— Scottish Birds (2003) this makes the data he gathered from finding nests and watching breeding behaviour all the more remarkable. Sadly, Corn Buntings are now extinct in his old stamping grounds and the subtle differences he detected between breeders in Sutherland and Southern England can no longer be examined in more detail. I sometimes sit on the scrubby slope above the golf course where Donnie studied Whitethroats and it is very easy to understand why he spent so many early mornings here watching his favourite bird. Before his death he had the distinction of being the longest contributor to the BTO nest record scheme. It always was a matter of great pride for him to obtain as full documentation of each nesting attempt as possible. Some of my most pleasant memories of Donnie are of winter nights, visiting him in his home, when the coffee cups would be put away and a particularly cherished book of bird paintings would be opened out on the dining room table. He was proud of the D Watson and G E Lodge paintings that adorned his walls. With infectious enthusiasm he would tell of his latest sightings and recount the challenges of the nesting season just gone, describing in exact detail the problems each nest presented him before he finally found it. A keen sportsman being both a fine tennis and badminton player in his younger days his tale of playing in remote village halls when the shuttlecock would get singed in the paraffin lamps hanging from the rafters reflect an era long gone. Later in life he enjoyed the challenge and social occasion provided by bowls. He had a good life, on the morning he died he was sweeping up leaves in his garden. Along with a very happy family life his interest in natural history gave him deep satisfaction. When Donnie died, Dornoch lost one of its most prominent citizens, a retired sheriff clerk, a Justice of the Peace and honorary sheriff and, for Obituaries 45 those of us who were fortunate enough to call him a friend, there is a feeling of gratitude that we had known a man of quality. Donald Bremner John Michael Stewart Arnott, OBE 1933-2002 I first met John early in 1955 as a fellow fresher at Cambridge. My main memory of him in his early career is of an attractive but somewhat solitary person with a remarkable fund of self reliance and incisive intellect. Add to these attributes his National Service as a Vampire pilot and his starting position of at least one rung on the ladder above most of us was fully merited. It was not that John competed and won; it was that he was already extraordinary, able to enjoy life as the rest of us but unusually primed to tackle personal challenges and true adventures. Who else but John would have decided that he would walk the whole way from Cambridge to Norway’s northernmost coast? And if such footslog struck one as a mite pig headed, then John would point with visible pride to an ashtray fashioned from a piece of the North Cape. It was inscribed ‘Operation Hogshead’. How we coveted that memento of an epic ‘one man power’ journey, and also John’s photograph of the Norwegian/Russian border taken from the Soviet side! Unlike the wholly bird obsessed contingent at Cambridge, John had other interests and skills and none was earlier honed than his ability to act. Within friendships, this facility was used sparingly to provoke laughter, but on the University’s stages in the mid 1950s it commanded wide attention and respect. To me, his Jaques in As you Like It remains a role model for all other actors — and at the time John led casts which included now great actors like David Buck and Derek Jacobi. Knowing that he found 46 his annual examinations in Law somewhat irksome, if not tedious, we rather expected John to spend his life treading boards to ever more acclaim. When I asked him why he had not done so, he just shrugged and said that his inheritance of a strong family work ethic had prevented such an indulgent occupation. In 1960, with his birdwatching still essentially recreational, John decided that the BBC offered him the best conduit for his creativity. He joined BBC Scotland, and it was from my mother that I learned that John was performing again — as a television newsreader. It was nerve wracking to witness the day’s events being broadcast in John’s wonderfully resonant but slightly sardonic voice and worry that at any moment he might raise a quizzical eyebrow at their inherent folly! Married to his delightful wife Lynne in 1965, John established a family home firstly at Balmore near Glasgow, then in Aberdeen, and finally at East Redford House, Edinburgh. This was to be the unchanged ‘base camp’ for his career in the BBC’s Edinburgh office and all his later adventures. Happily, work and personal interests often ran close together, and as a master of atmospheric tape recording, John could bring up the hairs on one’s hindneck. Who can forget the forthright admission of egg collecting that he elicited from Desmond Nerthersole-Thompson? It was a classical example of John’s candour producing a similar response. In the late 1970s John found time to become a major force in Scottish field ornithology. He took on a whole line of new duties, from Secretary and Chairman of the Isle of May Bird Observatory Trust, to chairman of the Fair Isle Bird Observatory Trust and on to President of the Scottish Omithologists’ Club. Among his contemporaneous physical achievements, John’s design and construction on the Arnott trap on the Isle of May has produced much pleasure among bird ringers. SB 24(1) Along his wider broadcasting remit, John’s viewpoint was often that of a distinctive human community on the edge of its ecosystem. This undoubtedly stemmed from his own preference for the Arctic, where he explored Spitsbergen, Siberia and Greenland several times, but it also enabled him to portray his particular brand of ‘vivid radio’ places as far apart as Scotland and the Falkland Islands and Peru and China. At home, John presented a calmer attitude to bird quests. I have no record of him ever chasing a rarity. But this did not stop the Presbyterian in him being occasionally seduced by such joys as the marsh birds of East Anglia. I remember well his delight in getting 3 lifers at Walberswick on 8th June 1955 — and their proper celebration in a sit down lunch at a nearby hostelry. John’s first membership of an expedition was in 1956 when he was one of the Edinburgh/ Cambridge 8 who survived 10 days on St Kilda with final provisions reduced to digestive biscuits, Gallagher’s cigarettes and, for him, pipe tobacco. His entries in the expedition’s log still read very well, being full of the warm perceptions that were to characterise his later broadcasting. Tan Wallace John’s keen interest in the countryside continued throughout his life. It took him to the Arctic, and it also took him into the world of countryside politics. Here the Nature Conservancy and the Countryside Commission for Scotland reflected a new thinking about the countryside in the second half of the twentieth century, and brought naturalists into the establishment. Visiting Greenland for the first time in 1979, John became fascinated with the arctic landscape and the wildlife. He also went to Spitsbergen, based on a boat which had variable Scottish Birds (2003) Obituaries 47 seamanship! Another year he explored Arctic Siberia, where flocks of Reindeer swam the rivers, and mosquitoes reached a new intensity around the tree line. In 195 he took part in an expedition to Peary Land. This was to be his last camping expedition, though he and Lynne later took a couple of boat cruises in the north. Interested in sharing experience with other Arctic enthusiasts, he served as Chairman of the Scottish Arctic Club from 1996 to 1999. In the public field, he became in 1983 a member of the Countryside Commission of Scotland, and later its Vice Chairman (1986-1992). His thoughtful and well informed consideration of matters such as access and conservation made an important contribution to thinking at that time. Building on the Commission’s reports, A Park System for Scotland (1974) and The Mountain Areas of Scotland (1990), he played an influential role in the establishment of National Parks in Scotland. In 1999 John became Chairman of the Scottish Wildlife Trust. This coincided with a period of great difficulty in the Trust’s affairs which took enormous toll of his deteriorating health, and he was forced to resign in June 2002. His steadfast commitment at this time was instrumental in making it possible for the Trust to go forward. Those of us who worked with him and were his friends greatly lament his passing and cherish our shared memories. He was a great companion, considerate and thought provoking wand with a gently astringent sense of humour — a person who served Scotland and its countryside with selfless devotion. Jean Balfour John Arnott ringing birds on the Isle of May, May 1991 48 Advice to contributors Authors should bear in mind that only a small proportion of the Scottish Birds readership are scientists and should aim to present their material concisely, interestingly and clearly. Unfamiliar technical terms and symbols should be avoided wherever possible and, if deemed essential, should be explained. Supporting statistics should be kept to a minimum. All papers and short notes are accepted on the understanding that they have not been offered for publication elsewhere and that they will be subject to editing. Papers will be acknowledged on receipt and are normally reviewed by at least 2 members of the editorial panel and, in most cases, also by an independent referee. They will normally be published in order of acceptance of fully revised manuscripts. The editor will be happy to advise authors on the preparation of papers. Reference should be made to the most recent issues of Scottish Birds for guidance on style of presentation, use of capitals, form of references, etc. Papers should be typed on one side of the paper only, double spaced and with wide margins and of good quality; 2 copies are required and the author should also retain one. We are also happy to accept papers on disk or by email at: mail @the-soc.org.uk, stating the type of word processing package used. If at all possible please use Microsoft Word . Contact the Admin Officer on 0131 653 0653 for further information. Headings should not be underlined, nor typed entirely in capitals. Scientific names in italics should normally follow the first text reference to each species unless all can be incorporated into a table. Names of birds should follow the official Scottish List (Scottish Birds 2001 Vol 22:33-49). Only single quotation marks should be used SB 24(1) throughout. Numbers should be written as numerals except for one and the start of sentences. Avoid hyphens except where essential eg in bird names. Dates should be written: ...on 5 August 1991...but not ...on the 5th... (if the name of the month does not follow). Please do not use headers, footers and page numbers. Please note that papers shorter than c700 words will normally be treated as short notes, where all references should be incorporated into the text, and not listed at the end, as in full papers. Tables, maps and diagrams should be designed to fit either a single column or the full page width. Tables should be self explanatory and headings should be kept as simple as possible, with footnotes used to provide extra details where necessary. Each table, graph or map should be on a separate sheet, and if on disc each table, graph, map etc should be on a separate document. Please do not insert tables, graphs and maps in the same document as the text. Maps and diagrams should be either good quality computer print out and in black and white (please do not use greyscale shading) or drawn in black ink , but suitable for reduction from their original size. Contact the Admin Officer on 0131 653 0653 for further details of how best to lay out tables, graphs, maps etc. The Scottish Ornithologists’ Club (SOC) was established by a group of Scottish ornithologists who met together in the rooms of the Royal Scottish Geographical Society in Edinburgh on 24 March 1936. Now, 67 years on, in 2003, the Club has 2200 members and 14 branches around Scotland. It plays a central role in Scottish birdwatching, bringing together amateur birdwatchers, keen birders and research ornithologists with the aims of documenting, studying and, not least, enjoying Scotland’s varied birdlife. Above all the SOC is a club, relying heavily on keen volunteers and the support of its membership. Headquarters provide central publications and an annual conference, and houses the Waterston Library, the most comprehensive library of bird literature in Scotland. The network of branches, which meet in Aberdeen, Ayr, the Borders, Dumfries, Dundee, Edinburgh, Glasgow, Inverness, New Galloway, Orkney, St Andrews, Stirling, Stranraer and Thurso, organise field meetings, a winter programme of talks and social events. The SOC also supports the Local Recorders’ Network and the Scottish Birds Records Committee. The latter maintains the “official” Scottish List on behalf of the Club. The Club supports research and sur- vey work through its Research Grants. The Club maintains a regularly updated web site, which not only contains much information about the Club, but is also the key source of information about birds and birdwatching in Scotland. www.the-soc.org.uk SOC Subscription Rates Annual membership subscription rates (as of August 1999) are as follows, with reduced rates for those paying by Direct Debit given in brackets: Adult £ 20.00 (£18.00) Family (2 adults and any children under 18 living at one address) £ 30.00 (£27.00) Junior (under 18, or student under 25) £ 8.00 (£ 7.00) Pensioner (female over 60, male over 65) or Unwaged (and claiming benefit) £ 11.00 (£10.00) Pensioner Family (both over 65) (2 adults living at one address) £ 16.00 (£14.50) Life £400.00 Life Family £600.00 All subscriptions are due on 1 October. They may be paid by Direct Debit and Gift Aided. Subscriptions paid by Direct Debit greatly assist the Club. Please ask for a Direct Debit form by contacting HQ or vis- iting our web site. Scottish Birds Volume 24 Part 1 June 2003 Contents Main Papers Mandarin Ducks in northern Scotland and the potential consequences for breeding Goldeneye - P Cosgrove 1 The distribution and status of the Red-billed Chough in Scotland in 2002 - § K Finney & D C Jardine 11 Scotland’s endemic subspecies - R Y McGowan, D L Clugston & R W Forrester on behalf of the Scottish Birds Records Committee 18 Eurasian Reed Warblers in Scotland: a review of probable breeding records - D Robertson 36 Short Notes Unusual behaviour of Common Redshanks and Common Starlings towards dead Common Redshank - P Cosgrove & R McGregor 40 Eurasian Oystercatcher apparently brooding young of Ringed Plovers - R C Dickson 40 Treecreeper feeding from a food cache in a brick wall - A R Lyndon 42 Female Eurasian Sparrowhawk caching prey - R N Cinderey 42 Obituaries C N L Cowper 1928-2002 - Tom Delaney 43 Donald M Macdonald, MBE 1912-2002 - Donald Bremner 44 John Michael Stewart Arnott, OBE 1933-2002 - Ian Wallace & Jean Balfour 45 Advice to contributors 48 Front Cover Mandarin Jim Dickson Published by the Scottish Ornithologists’ Club, Harbour Point, Newhailes Road, Musselburgh, EH21 6SJ. © 2003 in ty Ha Canoe and walking winter surveys on the River Spey, 1994-2003 The distribution of the Scottish Crossbill 1995-2003 Estimating breeding wader populations of Scottish uplands Origins of Common Starlings wintering in the Highlands Scottish Birds — The Journal of the SOC Editor: Dr S da Prato. Assisted by: Dr I Bainbridge, Professor D Jenkins, Dr M Marquiss, Dr J B Nelson and R Swann. Business Editor: Caroline Scott, Admin Officer, SOC, Harbour Point, Newhailes Road, Musselburgh EH21 6SJ. Scottish Birds, the official journal of the SOC, publishes original material relating to ornithology in Scotland. Papers and notes should be sent to The Editor, Scottish Birds, SOC, Harbour Point, Newhailes Road, Musselburgh EH21 6SJ Scottish Birds is published each year in June. Scottish Birds is issued free to members of the SOC, who also receive the quarterly newsletter Scottish Bird News, the annual Scottish Bird Report and the annual Raptor Round Up. Published by: The Scottish Ornithologists’ Club, Harbour Point, Newhailes Road, Musselburgh EH21 6SJ. Design by: Pica Design, 51 Charlton Crescent, Aboyne, Aberdeenshire AB34 5GN. Printed by: Meigle Colour Printers Ltd, Block 11, Units 1 & 2, Tweedbank Industrial Estate, Galashiels TD1 3RS Scottish Birds (2004) 24:1-10 Canoe and walking winter surveys on the River Spey, 1994-2003 I Canoe and walking surveys of wintering Goosanders, Red-breasted Mergansers, Great Cormorants and Common Goldeneyes on the River Spey, 1994-2003 P J COSGROVE, J RA BUTLER & R L LAUGHTON Winter surveys of Goosanders, Red-breasted Mergansers, Great Cormorants and Common Goldeneyes on the main stem of the River Spey were carried out in December 2002 and February 2003 using a novel survey methodology based on canoe counts. The River Spey holds nationally important wintering populations of Goosanders and Common Goldeneyes, but not Red-breasted Mergansers or Great Cormorants. These data were compared with historical survey data, which demonstrated that wintering Goosander densities have remained stable on the River Spey between 1994 and 2003. Recommendations are made for future survey methods of Goosanders, Red-breasted Mergansers, Great Cormorants and Common Goldeneyes on the River Spey. Introduction Recent work in north east Scotland has shown the national importance of rivers for some birds such as Goosander Mergus merganser, Red-breasted Merganser Mergus serrator (collectively known as ‘sawbills’) and Common _ Goldeneye Bucephala clangula (Duncan & Marquiss 1993, Marquiss & Duncan 1994a, Cosgrove 1996, 1997 and Watson et al 1998). Unfortunately, there have been very few estimates of whole river populations of sawbills and Common Goldeneye in Great Britain (GB), with most estimates made on standing waters and a few on sections of rivers. As a result, estimates of wintering populations of sawbills and Common Goldeneye have been compromised by a lack of systematic counts on rivers (Owen ef al 1986, Marquiss and Duncan 1994). On the River Spey, and elsewhere in Scotland, fishery managers often perceive sawbills and Great Cormorants Phalacrocorax carbo as a major threat to salmonid populations and fisheries because of the large number of fish that they consume (Spey Catchment Management Plan 2003). Sawbills are protected by law under the 1981 Wildlife and Countryside Act although Section 16 of the Act does make provision for the issue of licences to kill them to prevent serious damage to fisheries, where no other satisfactory solution can be found. However, this approach has been disputed by others who believe that there is no direct scientific evidence linking sawbills to damage of fisheries or fish stocks and that no licences should be issued until such evidence is available (Spey Catchment Management Plan 2003). Controversy has also surrounded estimates of sawbill and Great Cormorant abundance on Scottish rivers. The main River Spey stakeholders (eg local residents, fishery proprietors, conservationists, anglers, local authorities, government agencies etc) have agreed that further research on sawbills and Great Cormorants on the River Spey is necessary and that this should be taken forward at the earliest opportunity (Spey Catchment Management Plan 2003). Only when more information is available can informed decisions and progress be made on fishery management and perceived predation issues. ‘y oe & e@ Loch Insh a calda Figure 1. Map of the River Spey Historically, counts of sawbills and Great Cormorants on the River Spey have been undertaken for some sections of the river, but not for all of the main stem. Also, counts have been irregular, so few, if any comprehensive historical data sets exist. This situation has arisen for several reasons, including the practicalities, costs and logistics of adequately counting birds on Britain’s seventh longest river, which measures 157km from source to sea (Figure 1). The standard survey methodology developed for counting sawbills, Goldeneye and Great Cormorants on rivers is by means of walked transects along the main stem of the river. This is time consuming, labour intensive and needs to be undertaken during periods of similarly mild weather conditions to minimise the possible effects of weather related bird movements along the river. In a full survey of the River Deveron, SB 24(2) 40. lometres north east Scotland during the short daylight period in mid winter, it took 2 teams of 2 surveyors 4 days to survey the main stem of the river which was 87km long (Cosgrove 1996). The River Spey is almost double the length of the River Deveron and so would require consid- erable resources to regularly survey by means of a walked transect count. The aims of this study were (1) to census the wintering population of sawbill ducks, Common Goldeneyes and Great Cormorants on the main stem of the River Spey from Loch Insh to Spey Bay, using a new survey methodology; (2) to compare the recent census with relevant historical survey data; and (3) to make recommendations for future survey methods of sawbills, Common Goldeneyes and Great Cormorants on the River Spey. Scottish Birds (2004) Canoe and walking winter surveys on the River Spey, 1994-2003 3 Methods In December 2002 and February 2003, a novel method for surveying sawbills, Common Goldeneyes and Great Cormorants based on counts from Canadian canoes on the River Spey was trialled. On 4 December 2002 and 26 February 2003, 6 teams of trained surveyors simultaneously surveyed 6 contiguous sections of the main stem of the River Spey from Loch Insh to Spey Bay, a distance of 116km. The upper River Spey from Loch Insh to Spey Dam was not surveyed, so these data presented represent a minimum figure for the whole river. The teams drifted and paddled downstream to a predetermined location where the next team had started their survey and a vehicle waited to collect the personnel and canoes. In their River Dee study, Marquiss and Duncan (1984b) investigated the diurnal activity patterns of Goosanders. Birds spent most of the daytime foraging and loafing on rivers. The timing and arrival at communal still water roosts was noted, with most birds arriving from 50 minutes before to 10 minutes after sunset. Thus, the 2002 and 2003 river surveys were carried out during normal daylight hours avoiding the first light and dusk period, when birds may have moved to communal roosts away from the river. Each team consisted of one canoeist and one trained surveyor. The surveyor carefully scanned the river and sky ahead with binoculars and recorded every bird, its sex, the time it was seen and the location of each sighting on 1:25,000 OS maps. Occasionally the surveyor helped the canoeist paddle through more turbulent river sections. When a group of birds was sighted eg 3 male Common Goldeneyes and 2 female Common Goldeneyes, the surveyor counted these birds, noting the group composition and location and added them to the totals. To avoid double counting, if these birds flew downstream, the surveyor did not add another group of Common Goldeneyes to the totals until 3 males and 2 females were seen to fly back upstream. On 5 December 2002, 3 independent trained surveyors surveyed 3 randomly selected large sections of the River Spey, from Loch Insh to Spey Bay, using the standard walk transect method, which had been surveyed from canoes the day before. The bank based surveyors did not know of, or have access to, the canoe count data from the previous day. During the winters of 1994-95, 1995-1996, 1998-1999 and 2001-2002, 5 walked transect counts were undertaken on main stem sections of the River Spey from Loch Insh to Spey Bay by the Spey Fishery Board. This historical data is presented to allow for comparisons of winter sawbill and Great Cormorant densities to be made with those in winter 2002-2003. No comparative data for Common Goldeneye exists however. Results Goosander On 4 December 2002, 125 Goosanders were counted from canoe on the main stem of the River Spey up to Loch Insh at a density of 1.08 birds/km of river. On 26 February 2003, 90 Goosanders were recorded from canoe on the same section at a density of approximately 0.78 birds/km of river. Although Goosanders were spread throughout the river, the highest numbers occurred in the lowest reaches of the river, O- 40km upstream from the sea (Figure 2). However, there were some notable differences between the patterns of distribution recorded by the 2 counts. In particular, the highest Goosander counts occurred 30km from the sea on 04 December 2002, but none were recorded there on 26 February 2003. 4 SB 24(2) = | —— 04/12/02 | 26/02/03 Number of ) |. 94/12/02] “(+++ ++ 26/02/03 | =“ Number Distance from sea (km) Figure 3. Red-breasted Merganser numbers in relation to distance from sea on the River Spey, winter 2002-2003 —— 04/12/02 | [2+ 26/02/08 | Number of 90 110 Distance from sea (km) 04/12/02, | 26/02/03 | — e) ao © pl = ae | = Distance from sea (km) Figure 5. Goldeneye numbers in relation to distance from sea on the River Spey, winter 2002-2003 Scottish Birds (2004) Canoe and walking winter surveys on the River Spey, 1994-2003 5 The density of Goosanders recorded in winter by walked transect counts between December 1994 and January 2002 shows a_ remarkable consistency, with between 1.19 and 1.40 birds/km of River Spey (Table |) in each survey. The density derived from the canoe count for December 2002 (1.08 birds/km) was _ also similar. The counts of Goosander made by canoe in December 2002 were similar to the walk transect counts undertaken 24 hours later for the same 3 sections of river (Table 2). Red-breasted Merganser Twenty two Red-breasted Mergansers were counted from canoe on the main stem of the River Spey up to Loch Insh on 04 December 2002, at a density of approximately 0.19 birds/km of river. Only 4 Red-breasted Mergansers were counted from the same section on 26 February 2003, at a density of 0.03 birds/km of river. Although the numbers of birds counted during both surveys were markedly different, Red-breasted Mergansers were mainly recorded in the lowest reaches of the River Spey, 0-20km upstream from the sea (Figure 3). There was considerable variation in the density of Red-breasted Mergansers recorded by walked transect counts between December 1994 and January 2002, with between 0.03 and 0.52 birds/km of River Spey (Table 1) in each survey. The counts of Red-breasted Mergansers made by canoe in December 2002 were similar to the walk transect counts undertaken 24 hours later for the same 3 sections of river (Table 2). Great Cormorant Forty Great Cormorants were counted from canoe on the main stem of the River Spey up to Loch Insh on 04/12/02, at a density of approxi- mately 0.34 birds/km. Only 5 Great Cormorants were counted on the same section on 26 February 2003, at a density of 0.04 birds/km of river. Although there was a large difference between the numbers of birds recorded in the 2 count periods, there were some similarities in the overall distribution of the birds on the River Spey, with most birds occurring 0-10km from the sea, and others 90km from the sea (Figure 4). The density of Great Cormorants recorded by walked transect counts between December 1994 and January 2002, varied considerably with between 0.04 and 0.84 birds/km of River Spey (Table 1) in each survey. There was a 50% difference in the canoe and walked transect counts of Great Cormorants for the 3 sections of river surveyed in December 2002, with 31 counted from canoe and only 14 counted from the walk transect the following day (Table 2). However, much of this difference was attributed to a single flock of 20 Great Cormorants counted by the canoe survey on a shingle bar at the mouth of the Spey. Human disturbance at the mouth of the Spey or changes in the tide could easily have moved the birds elsewhere overnight and this one roosting flock, which may have used the sea and not the river, could account for the difference in the canoe and walk transect counts. Similarly high densities of Great Cormorants have been recorded during previous winter surveys (Table 1). Common Goldeneye There was remarkable similarity in the counts of Common Goldeneye between the December 2002 and February 2003 surveys. Two hundred and five Common Goldeneyes were counted from canoe on the main stem of the River Spey up to Loch Insh on 4 December 2002, at a density of 1.8 birds/km of river. Two hundred Common Goldeneyes were then recorded from canoe on the main stem of the River Spey up to Loch Insh on 26 February 2003, giving a density of 1.7 birds/km of river. Although Common Goldeneyes were spread throughout the length of the Spey surveyed, the highest numbers 6 SB 24(2) Table 1. The density of Goosanders, Red-breasted Mergansers and Great Cormorants recorded in winter between 1994 and 2002 on the main stem of the River Spey. Date Distance Goosanders R-b Mergansers Great of count surveyed surveyed surveyed Cormorants (km) (km) (km) (birds/km) December 1994 92km 22 0.24 0.84 November 1995 150km 1.19 0.11 0.07 January 1996 116km 1.40 0.40 0.30 December 1998 100km 1.34 0.24 0.30 January 2002 105km 1.19 0.52 0.32 December 2002* 116km 1.08 0.19 0.34 February 2003* 116km 0.78 0.03 0.04 * Canoe count data Table 2. The number of Goosanders, Red-breasted Mergansers, Great Cormorants and Common Goldeneyes recorded using different methods on consecutive days on the main stem of the River Spey in December 2002. Goosander canoe/walk count River Spey section R-breasted Merganser Common Goldeneye Great Cormorant canoe/walk count canoe/walk count canoe/walk count Spey Bay to Fochabers 9/10 Fochabers to Boat 0’ Brig 15/10 Rothes to Aberlour L1y9 Total 35/29 during both counts occurred in the lower reaches of the river 0-20km from the sea, and also around 80-100km upstream (Figure 5). The counts of Common Goldeneye made by canoe in December 2002 were very similar to the walk transect counts undertaken 24 hours later for the same 3 sections of river (Table 2). Discussion Goosander The lack of river population estimates for Goosander makes it difficult to put the River Lii2 14/12 20/6 22/17 2/0 9/8 17/19 1/0 2/0 8/9 31/14 47/45 Spey count data into a national context. Nevertheless, the GB wintering Goosander importance threshold of 90 birds (Musgrove et al 2001) has been reached on the 2 most recent counts in December 2002 and February 2003. This indicates that the River Spey is a nationally important wintering site for Goosander, holding at least 1% of the estimated population in GB. Although comprehensive historical sets of count data for the River Spey are unavailable, comparable density estimates for long sections over the last decade suggest that the river’s importance for wintering Goosander is long Scottish Birds (2004) Canoe and walking winter surveys on the River Spey, 1994-2003 Table 3. Comparisons of the logistical considerations between walk transect counts and canoe counts of sawbills, Great Cormorants and Common Goldeneyes on the River Spey. Variable/logistics Number of person days to survey main stem Number of competent surveyors needed Importance of weather Access arrangements Health and safety considerations Additional equipment necessary Counting efficiency Walk transect count 32 person days (4 teams of 2, 4 days). 8 surveyors. High (4 stable days). Difficult — almost impossible to get permission from all riparian Owners. High, although surveyors use a ‘buddy system’ of reporting when working alone. None. High, most birds (c75%) are Canoe count 12 person days (6 teams of 2, | day). 6 surveyors. Low (1 stable day). Easy, 7 entry and exit points. Sunday counts preferable in fishing season. High, trained canoeists needed. Working safely on water requires training. 6 canoes. High, most birds (c90%) are seen and flushed, despite efforts not to disturb them. standing with densities remaining stable since 1994. This is surprising as WEBS count data have shown a decline in the number nationally by around two thirds, since the mid 1990s, which coincided with a fall in numbers on the Inner Moray Firth, for years the most important site in the country for this species (Pollitt et al 2003). The pattern of birds spread throughout the river, with the highest numbers in the lowest reaches of the river is similar to data for other rivers in north east Scotland, such as the River Dee (Marquiss and Duncan 1994a) and River Deveron (Cosgrove 1997). Marquiss and Duncan (1994a) attributed this to the water in the upper river being very cold and the smaller fish living there being largely unavailable to Goosanders, at least during the daytime, because they are buried in the substrate. seen and flushed on all but the widest river sections. The relatively constant densities derived from the River Spey winter counts are interesting in light of similar data from the River Dee. Marquiss et al 1998 showed large annual variation in the numbers of breeding Goosanders and their production of ducklings on the River Dee, but the midwinter population was remarkably stable year on year. Marquiss and Duncan (1994a) found that Goosander abundance varied between seasons and years reflecting their needs for food, security from predators, pairing and nesting, so it is strongly recommended that future work on the River Spey looks at autumn, winter, spring and summer populations. Goosanders breed along the River Spey, but the numbers of pairs currently involved is unknown. In April 1984, 30-50 pairs were estimated to be breeding on the River Spey (Dennis 1984). Red-breasted Merganser The lack of river population estimates makes it difficult to put the River Spey count data into a national context. The population and density estimates have varied considerably over time and suggest that the River Spey does not hold a nationally important wintering population of Red-breasted Mergansers. Small, but variable numbers of Red-breasted Mergansers have used the lowest sections of the River Spey during the winter over the last decade and this fits into the general pattern seen during intensive studies on the River North Esk (Marquiss and Duncan 1993). Red-breasted Mergansers breed along the River Spey, but the numbers of pairs currently involved is unknown. In 1984, 10-15 pairs were estimated to be breeding on the River Spey (Dennis 1984). Great Cormorant The population and density estimates have varied considerably over time and suggest that the River Spey does not hold a_ nationally important wintering population of Great Cormorants, being well below the threshold of 130 birds (Musgrove et al 2001) during counts in December 2002 and February 2003. The distribution of Great Cormorants on the River Spey appears to reflect their choice in preferred feeding areas, with most birds in the lower river, close to the sea. Common Goldeneye As with the sawbills counted, the lack of river population estimates for Common Goldeneye makes it difficult to put the River Spey count data into a national context. Nevertheless, the GB wintering Common Goldeneye importance threshold of 170 birds (Musgrove et al 2001) has been reached on both of the recent counts in December 2002 and February 2003. In April 1984, 246 Goldeneye were estimated to be present on the River Spey (Dennis 1984). This SB 24(2) indicates that the River Spey is a nationally important wintering site for Common Goldeneye, holding at least 1% of the estimated population in GB. Duncan and Marquiss (1993) found high numbers of wintering Common Goldeneyes on whole river counts in north east Scotland and estimated that 62% of the region’s Common Goldeneyes were on rivers, although numbers did vary between years. The River Spey population, along with those of the Rivers Deveron, Don and Dee, confirms that the north east of Scotland is a nationally important area for wintering Common Goldeneyes. However, along with Goosanders, the national population of wintering Common Goldeneye has been estimated without compre- hensive rivers surveys, so that the figures and levels qualifying for national interest need to be reassessed following national surveys of all Common Goldeneye habitats. After nesting for the first time in Britain in 1971, Badenoch and Strathspey has become the main British breeding site for Common Goldeneyes (Dennis 1995). In 2002 a concerted effort was made to monitor the Scottish breeding population and this survey revealed that at least 91 clutches were laid and only 2 of these were outside Badenoch and Strathspey (Goldeneye Study Group 2002 Newsletter). However, recent unpublished studies have revealed that as many as 2/3 of these clutches contained eggs laid by more than one female, making a firm breeding population estimate problematical (Begg 2002). Thus, the River Spey catchment is unique as it holds almost all of Britain’s breeding Common Goldeneye and a significant nationally important wintering population as well. Comparison of count methods Although based on comparisons of counts from only 3 sections of the River Spey, the data collected suggests that the canoe count Scottish Birds (2004) methodology is comparable to the walk transect count methodology. This comparison should be tested again elsewhere, but from our initial findings it would appear that counting from a canoe is an efficient and effective method of surveying sawbills on a large river like the River Spey. It is possible to see and count birds using both methods, but the window of vision provided by canoes may allow more complete counts, particularly in areas of dense riparian vegetation which obscure visibility for walk transects. Field experience shows that most birds are disturbed by both methods, despite attempts not to do so, suggesting that the risks of double counting flushed birds are similar in both cases. Both methods have their advantages and disadvantages and some similar strengths and weaknesses (see Table 3), but by far the most efficient method is the canoe count. In terms of staff time and resources, the walk transect uses almost 3 times the number of ‘person days’ to complete the survey than the canoe count. This would provide a considerable saving in terms of direct staff costs (wages) and indirect costs (travel and overnight subsistence) for the canoe count method. The canoe count also requires 20% fewer competent surveyors than the walk transect count. Formal access arrangements from the walk transect count are highly problematic and if carried out properly, add considerable logistical and staff costs. Very few access points are needed if using a canoe, but visits may need to be restricted to early mornings or Sundays during the fishing season, providing a time constraint. Both methods take account of health and safety considerations but the necessary presence of two personnel at all times in a canoe provides greater security for both surveyors. If the canoe count method was adopted on other large rivers, which we recommend, it would Canoe and walking winter surveys on the River Spey, 1994-2003 Y significantly improve the efficiency and therefore feasibility of regular sawbill monitoring on rivers. There is a third well tried method that was not considered here. Aerial surveys have been used for a variety of bird monitoring purposes and could be trialled. In 1984, Dennis undertook an aerial survey of the River Spey from the sea to the source. It took 2 hours, one surveyor, one pilot, good weather, no access problems, high health and safety consid- erations and no additional equipment other than hiring a light plane. Although the counting efficiency of aerial river surveys 1s unknown, the costs of hiring a plane for such a survey appear to be comparable or perhaps even cheaper than the canoe count and, therefore, it is worthy of further consideration. An increase in sawbill, Great Cormorant and Common Goldeneye monitoring data would be beneficial to all the main River Spey stakeholders and it would allow further investi- gations into the birds’ biology and eventually into assessments of their potential impact on salmonid populations, as well as the relative conservation importance of populations. Acknowledgements This paper would not have been possible without the help of a large number of people and consequently we are very grateful to all the surveyors and canoeists who helped with the fieldwork. In particular, we would like to thank Tim Walker and Stuart Upward of Glenmore Lodge, Andy Wells, Dave Craig and Squadron Leader Malcolm Lee for their help with canoes. Many thanks to Mick Marquiss, Carl Mitchell, Keith Duncan, Kristin Scott and Robin Cosgrove who all made useful comments on an early draft of this paper. References Begg T 2002. Aspects of the breeding biology of the Goldeneye Bucephala clangula in 10 Scotland. Unpublished MSc University of Glasgow. Cosgrove P J 1996. A winter survey of Common Goldeneyes on the River Deveron, north east Scotland. Scottish Birds 18: 242-246. Cosgrove P J 1997. A winter survey of sawbill ducks and Great Cormorants on the River Deveron, north east Scotland. Scottish Birds 19: 93-100. Cranswick P, Pollitt M, Musgrove A & Hughes B 1999. The Wetland Bird Survey 1997-1998 Wildfowl and Wader Counts. BTO, WWT, RSPB and JNCC. Dennis R. 1984. Goosander Survey. /nternal RSPB report. Dennis R 1995. The Birds of Badenoch and Strathspey. Colin Baxter Photography Ltd. Duncan K & Marquiss M 1993. The sex/age ratio, diving behaviour and habitat use in Goldeneye Bucephala clangula wintering in northeast Scotland. Wildfowl 44: 11-120. Goldeneye Study Group 2002. Breeding season newsletter. RSPB Insh Marshes. Little B & Marchant J 2002. Goosander in The Thesis, Migration Atlas: Movements of the Birds of Britain and Ireland. ed Wernham C, Toms M, Marchant J, Clark J, Siriwardena G and Baillie S, BTO, Poyser. Marquiss M & Duncan K 1993. Variation in the abundance of Red-breasted Mergansers Mergus serrator on a Scottish river in relation to season, year, river hydrography, salmon density and spring culling. /bis 135: 33-41. Marquiss M & Duncan K 1994a. Seasonal switching between habitats and changes in abundance of Goosanders Mergus merganser within a Scottish river system. Wildfowl 45: 198-208. Marquiss M & Duncan K_ 1994b. activity patterns of Goosanders merganser on a_ Scottish river Wildfowl 45: 209-221. Marquiss M, Carss D, Armstrong J & Gardiner R 1998. Fish eating birds and salmonids in Diurnal Mergus system. SB 24(2) Scotland. The Scottish Office Agriculture, Environment and Fisheries Department. Musgrove A, Pollitt M, Hall C, Hearn R, Holloway S, Marshall P, Robinson J & Cranswick P 2001. The Wetland Bird Survey 1999-2000 Wildfowl and Wader Counts. BTO, WWT, RSPB and JNCC. Owen M, Atkinson-Willes G L & Salmon D G 1986. Wildfowl in Great Britain. 2nd Edition. Cambridge University Press. Pollitt M, Cranswick P, Musgrove A, Hall C, Hearn R, Robinson J & Holloway S 2000. The Wetland Bird Survey 1997-1998 Wildfowl and Wader Counts. BTO, WWT, RSPB and JNCC. Pollitt M S, Hall C, Holloway S J, Hearn R D, Marshall P E, Musgrove A J, Robinson J A & Cranswick P A 2003. The Wetland Bird Survey 2000-01: Wildfowl and Wader Counts. BTO/WWT/RSPB/JNCC, Slimbridge. River Spey Catchment Management Plan 2003. Spey Catchment Steering Group, Achantoul, Aviemore. Toms M 2002. Red-breasted Merganser in The Migration Atlas: Movements of the Birds of Britain and Ireland. ed Wernham C, Toms M, Marchant J, Clark J, Siriwardena G and Baillie S, BTO, Poyser. Watson A, Marquiss M & Cosgrove P 1998. North east Scottish counts of Goldeneye, Goosander, Red-breasted Merganser and Cormorant in 1944-50 compared with 1988-97. Scottish Birds 19: 249-258. Peter Cosgrove, The Cairngorms National Park Authority, 14 The Square, Grantown on Spey, Morayshire, PH26 3HG James Butler and Bob Laughton, Spey Fishery Board Research Office, 1 Nether Borlum Cottage, Knockando, Aberlour, Morayshire, AB38 7SD Revised manuscript accepted August 2003 Scottish Birds (2003) 24:11-16 The distribution of the Scottish Crossbill, 1995-2003 Il The distribution of the Scottish Crossbill, 1995-2003 RW SUMMERS, DC JARDINE & RJG DAWSON The breeding distribution of the Scottish Crossbill was investigated during January to April from 1995-2003. Identification was based on an excitement call, believed to be characteristic for this species. Woods with Scottish Crossbills are listed and mapped. Scottish Crossbills were found from the Caithness Flow Country in the north to Stirlingshire in the south, and from Glen Garry in the west to Fetteresso Forest near Stonehaven in the east. Additional records from outside the breeding season included one from Fife. Thus, their range is greater than previously thought. It is recognised that the list of woods is incomplete and that the range in any one year will vary depending on cone crops of different conifer species. Introduction The Scottish Crossbill Loxia scotica is regarded as Britain’s only endemic bird species (BOURC 1980). Prior to this, it was alternately classed as a subspecies of either the Common Crossbill L curvirostra or the Parrot Crossbill L pytyopsittacus (see Knox 1975 for a review). The population size of Scottish Crossbills was thought to be around 1500 adults in the 1970s (Nethersole-Thompson 1975). Also, it has been assumed that Scottish Crossbills are associated primarily with the remaining fragments of the Caledonian pine forest, a small and threatened habitat (Nethersole- Thompson 1975, Knox in Gibbons et al 1993, Anon 1995). Because of the small population size and small area of its main habitat, the Scottish Crossbill was classed as a species of global conservation concern (Tucker & Heath 1994). However, this classification acknowledged a lack of information and recognised that the formal conservation status could change when more data became available on its distribution and population size. These aspects of the status of the Scottish Crossbill are key objectives for this species under the UK’s Biodiversity Action Plan (Anon 1995). The key difficulty about obtaining data on the status of the Scottish Crossbill was that a reliable method of field identification was lacking. In addition, Parrot Crossbills are now known to be nesting in Scotland and may breed alongside Common and Scottish Crossbills at the same site in a given year (Summers 2002), thus exacerbating the identification problem. There are no distinct plumage differences among the 3 species, and there are only small differences in overall size. The size of the bill is often mentioned as a defining character (Svensson 1992); the Common Crossbill has the smallest, Parrot Crossbill has the largest, while Scottish Crossbill is intermediate between these. However, overlap between the ranges of bill sizes of Common and Scottish Crossbills and between Scottish and Parrot Crossbills makes identification difficult (Knox 1976, 1990). Even when captured and measured, it may not be possible to identify some birds. Crossbills are also difficult to catch, so with the exception of museum specimens, in hand identification has been restricted largely to captures of irrupting birds, usually at coastal localities (Davis 1964) and to dedicated projects working in breeding areas (Marquiss & Rae 2002, Summers et al 2002). 1G SI t Wat r¢ we a 1g distrib Figure 1. The ution of woods occupied by Scottish Crossbills (in black) in northern Scotland SB 24(2) ‘| Se (north of 56°50’N) during 1995-2003. Records south of this are not shown Crossbills breeding in Scotland have a range of distinct flight and excitement calls. However, individuals consistently give a particular flight and excitement call and combination of these calls. In addition, these calls are shared among birds with similar bill size (Summers ef al 2002). This discovery provides the basis for a field method of species identification that distinguishes Scottish, Common and Parrot Crossbills. For reference, the different flight calls have been given numbers and _ the excitement calls letters. The combination of flight and excitement calls that characterises the Scottish Crossbill is flight call 3 and excitement call C. Although the different calls may be distinguished by ear with practice, the only way to confirm identification is to make a good quality sound recording and create a sound spectrogram (sonogram). This procedure allowed a description of the range of the Scottish Crossbill to be made, based on 10km squares occupied (Summers et al 2002). This paper presents the results in terms of woods occupied and updates the known distribution. Methods The study was carried out from 1995-2003, during January to April when it can be reasonably assumed that Scottish Crossbills are either breeding or on territory (Nethersole- Thompson 1975). All 10km squares north of 7 Scottish Birds (2004) 56°50’N with more than about 20% woodland were visited, and generally the larger woods were searched until at least one recording of a crossbill was made. Multiple recordings from 10km squares were restricted to only a few squares. Crossbills were lured using excitement call type D broadcast from mini speakers. Lured birds are usually vocal, and both flight and excitement calls can be tape recorded. For mapping, only the excitement call (type C) was used to indicate the presence of Scottish Crossbill, as this is regarded as characteristic. Birds that responded were recorded onto magnetic tape using a directional microphone. Sonograms were made from the tape recordings allowing species identification (see Summers ef al 2002 for further details). Boundaries of woods were taken from a 1:250 000 Ordnance Survey map. In some cases, a main road divided the wood, and the sections on either side of the road were _ treated independently. Any wood where at least one recording of the type C excitement call was made during 1995 to 2003 was regarded as occupied. Results Scottish Crossbills were recorded from 83 x 10km squares in 12 counties, representing 94 woods. The occupied woods are listed in Table 2. The highest number of woods (24, 26% of the total) was in Inverness shire (Table 1), while 4 counties (Sutherland, Ross shire, Inverness shire and Aberdeenshire) had 66 of all the woods (70%). Of the 94 woods, 9 (9.6%) were native pinewoods, the remainder being plantations, mainly planted after 1900. Scottish Crossbills were found mainly in the eastern Highlands. The limits were from Caithness in the north to Stirlingshire in the south, and from Glen Garry (Inverness shire) in the west to Fetteresso Forest (Kincardineshire) in the east (Fig 1). The single record of a Scottish Crossbill from Stirlingshire was of one bird associating with The distribution of the Scottish Crossbill, 1995-2003 ig about 70 Common Crossbills in a spruce Picea sp and larch Larix sp plantation. No breeding activity was noted in this area and crossbill numbers declined over the following weeks presumably due to the dispersal of the winter flock. Outside our study period (January to April), there was a December record of 3 birds in Fife. The only other Fife records are specimens collected in 1897 (now in the National Museums of Scotland, Edinburgh). Discussion The earlier mapped distributions of the Scottish Crossbill (Nethersole-Thompson 1975, Lack 1986, Gibbons et al 1993) were incomplete but demonstrated the perception that Scottish Crossbills were concentrated in Strathspey and Deeside. The present survey confirms this but also shows that they are more widespread in northeast Scotland. Although Scottish Crossbills were recorded from _ several Caledonian pinewoods, supporting the earlier perceived association with this habitat, most records came from plantations. Scots Pine Pinus sylvestris is often a component of plantations, but further ecological work is required to understand the associations with different conifer woods. The sampling unit was the 10km square, and often tape recordings were made from only one crossbill. However, repeated visits to some woods showed that all 3 species were present (eg Abernethy Forest, Summers 2002). In addition, excitement calls were not tape recorded in all cases. On some occasions, flight calls likely to be of Scottish Crossbills were recorded, but because this call is occasionally made by Parrot Crossbills, it is not diagnostic (Summers et al 2002). Therefore, by restricting our study to excitement calls, and often from a single bird, the distribution is likely to be incompletely described. Further, 14 SB 24(2) Table 1 Counties where Scottish Crossbills were recorded between 1995 and 2003. County Caithness 2 Sutherland 15 Ross shire 9 Inverness shire 18 Nairnshire 2 Morayshire 9 Banffshire 7 Aberdeenshire i) Kincardineshire 3 Angus | Perthshire l Stirlingshire l Total 83 Number of 10km squares Number of woods 24 — aS RS KS WN CO WO W ‘oe = Table 2 A list of 10km squares and locations where Scottish Crossbills were recorded in different counties during 1995-2003. Names in bold refer to woods classified as native pinewoods (Forestry Authority) and where Scottish Crossbills had been recorded in native stands (some sites have mixtures of stand types). Caithness NDO4 Blar Geal ND23 Rumster Sutherland NCS5O Raemore Wood NCS3 Altnaharra NCS4 Loch a’ Mhoid NC61 Dalchork NC65 Borgie Forest NCS&1 Balnacoil NC8&5 Dyke NC91 Strath Ullie NH39 Glen Einig NH48 Amat NH49 _— Inveroykel NH58 _ Ardgay, Kincardine Hill NH68 Kincardine Hill NH69 Maikle Wood NH79 ~~ Clashmore Wood, Ospisdale Ross shire NH35 NH36 NH46 NHS56 NH65 NH66 NH67 NH77 NH7& Inverness NH22 NH33 NH4]1 NH44 NHS53 NHS54 Loch Meig/Strathconon Longart Forest Strath Sgitheach Glen Glass Mount Eagle, Millbuie Forest Millbuie Forest Stittenham Wood, Wallace Hill, Kinrive Wood, Morangie Forest, Strath Rory West Morangie Forest Morangie Forest shire Glen Affric Cannich Loch Knockie Rheindown Wood Loch Laide, Abriachan Blackfold Scottish Birds (2004) NH63 Carr Ban, Farr Loch NH72 Strathdearn NH73 ~~ Moy, Meall Mor Forest NH82_ _ Sluggan NH90 ~~ Glenmore Forest, Rothiemurchus Forest NHQ91 Glenmore Forest, Abernethy Forest NH92 Carrbridge Woods, Curr Wood, Abernethy Forest NJOI Abernethy Forest NJO2 Grantown-on-Spey, Craigmore Wood NJO3 Tomvaich NN29 Glen Garry NN89 — Glen Feshie, Inshriach Forest Nairnshire NHS4 Carn na Caillich, Assich NH94 _ Ferness Forest Morayshire NH95_———-Darnaway Forest NJO5 Altyre Woods NJO6 Culbin Forest NJ13 Hill of Dalnapot NJ14 Knockando NJI5 Hill of the Wangie NJ24 Elchies Wood NJ25 Teindland Wood NJ35 Whiteash Hill Wood Banffshire NJI1 Glenmullie, Tomintoul NJ12 Bridge of Brown NJ23 Morinsh Wood the 10km squares sampled were those that contained at least 20% conifer woodland and were situated above 56°50’N. Thus, smaller woods and woods south of this line are poorly represented. However, it can be reasonably assumed that by visiting the bulk of larger woods, we sampled a significant proportion of the habitat available for Scottish Crossbills. Also, tape recordings made in southern Scotland yielded just 3 records of Scottish The distribution of the Scottish Crossbill, 1995-2003 iB) NJ34 Ben Aigan Forest NJ44 Balloch Wood NJ45 Corskell Moss NJ46 Bin of Cullen Aberdeenshire NJ20 Gairnshiel NJ31 Fordbridge Hill NJ40 Loch Davan NJ42 Clayshot Hill NJ43 Strathbogie NJ5O Tarland NJ52 Correen Hills NJ54 The Bin NJ60 Torphins NJ62 Bennachie NJ70 Midmar NO19 — Glen Quoich NO39 — Alltcailleach Forest NO49 Glen Tanar NOS9 _ Bogshiel Lodge Kincardineshire NO68 — Craigangower NO69 — Blackhall NO7S Fetteresso Forest Angus NOS5 Montreathmont Forest Perthshire NN55 — Blackwood of Rannoch Stirlingshire NS68 Carron Valley Forest Crossbills, despite regular encounters with Common Crossbills (Summers et al 2002). Numbers of crossbills in Strathspey pinewoods fluctuate in response to annual variations in Scots Pine cone crops (Nethersole-Thompson 1975, Summers 1999). In addition, ringing recoveries have shown that Scottish Crossbills move between Deeside and _ Strathspey (Marquiss et al 1995). Therefore, it is likely 16 that the range in a given season or year will be smaller than the composite map derived from several years work (Fig 1). Such movements in response to food supplies probably play a role in the colonisation of new areas. This is demonstrated by the records in the Lodgepole Pine P contorta plantations of the Flow Country, an area previously unforested. The more southerly records also show their dispersive capability. Further population and ecological studies are required to understand the dynamics of the distribution and abundance of Scottish Crossbills. In this paper, we _ have demonstrated the use of a distinctive call for field identification and mapping of Scottish Crossbills. We have confirmed the importance of northeast Scotland for this species, and shown that it is widespread in plantations. Acknowledgements Drs DW Gibbons, M Marquiss and J Wilson commented on the draft. References Anon 1995. Biodiversity: the UK Steering Group Report. Vol. 2: Action Plans. HMSO, London. BOU Records Committee 1980. Records committee: tenth report. /bis 122: 564-568. Davis P 1964. Crossbills in Britain and Ireland in 1963. British Birds 57: 477-501. Forestry Authority 1994. Caledonian Pinewood Inventory. Forestry Commission, Glasgow. Gibbons DW, Reid JB & Chapman RA 1993. The New Atlas of Breeding Birds in Britain and Treland: 1988-1991. Poyser, London. Knox AG 1975. Crossbill taxonomy. In, Nethersole-Thompson D. 1975. Pine Crossbills. pp 191-201. Poyser, Berkhamsted. Knox AG 1976. The taxonomic status of the Scottish Crossbill Loxia sp. Bull. BOC 96: 15-19. Knox AG 1990. Identification of Crossbill and SB 24(2) Scottish Crossbill . British Birds 83: 89-94. Lack P 1986. The Atlas of Wintering Birds in Britain and Ireland. Poyser, Calton. Marquiss M & Rae R 2002. Ecological differen- tiation in relation to bill size amongst sympatric, genetically undifferentiated crossbills Loxia spp. Ibis 144: 494-508. Marquiss M, Rae R, Harvey P & Proctor R 1995. Scottish Crossbill moves between Deeside and Strathspey. Scottish Bird News 37: 2-3. Nethersole-Thompson D 1975. Pine Crossbills. Poyser, Berkhamsted. Summers RW 1999. Numerical responses by crossbills Loxia spp. to annual fluctuations in cone crops. Ornis Fennica 76: 141-144. Summers RW 2002. Parrot Crossbills breeding in Abernethy Forest, Highland. British Birds 95: 4-11. Summers RW, Jardine DC, Marquiss M & Rae R. 2002. The distribution and habitats of crossbills Loxia spp. in Britain, with special reference to the Scottish Crossbill Loxia scotica. Ibis 144: 393-410. Svensson L 1992. Identification Guide to European Passerines, 4th edition. Lars Svensson, Stockholm. Tucker GM & Heath MF 1994. Birds in Europe: their conservation _ status. BirdLife International, Cambridge. Ron Summers, Royal Society for the Protection of Birds, Etive House, Beechwood Park, Inverness, IV2 3BW E mail: ron.summers @rspb.org.uk David Jardine, 49 Bellfield Road, North Kessock, Inverness, IV1 3XX. Robert Dawson, Royal Society for the Protection of Birds, Etive House, Beechwood Park, Inverness, IV2 3BW Revised manuscript accepted October 2003 Scottish Birds (2003) 24:17-28 Estimating breeding wader populations of Scottish uplands 17 Kstimating the breeding wader populations of Scottish uplands and salt marshes M O’BRIEN & C S WHITE Recent surveys have generated population estimates for 5 species of breeding waders on farmed land in Scotland. There have been no comparable surveys of the Scottish uplands, even though this habitat represents around 50% of total land area in Scotland. Surveys of selected sites have, however, been undertaken. The statistical relationship between wader densities from these upland surveys and information on the frequency of occurrence of the same species in upland habitats, as derived from data collected for the New Atlas of Breeding Birds, have been used as the basis for estimating population sizes for the whole of the Scottish uplands. These estimates are 2,500 pairs of Eurasian Oystercatcher, 4,500 Northern Lapwing, 9,100 Common Snipe, 13,100 Eurasian Curlew and 500 Common Redshank. These are compared with current farmland population estimates in Scotland. Introduction The populations of 5 breeding wader species (Eurasian Oystercatcher Haematopus ostralegus, Northern Lapwing Vanellus vanellus, Common Snipe Gallinago gallinago, Eurasian Curlew Numenius arquata and Common Redshank Tringa totanus) on the farmed land of Scotland have recently been estimated from a stratified random survey of c450 Ikm squares (O’Brien 1996, O’Brien et al 2002). These estimates exclude an unknown population of each of these species breeding in upland Scotland. There have been a number of wader surveys of substantial areas within the Scottish uplands over the last 20 years, many of which have remained as unpublished reports within the offices of the Nature Conservancy Council (now Joint Nature Conservation Committee or Scottish Natural Heritage) and the Royal Society for the Protection of Birds. These provided a range of density estimates for a variety of species in a variety of selected areas of uplands. The fact that the areas surveyed had often been selected as ‘good’ for waders suggested that they were unlikely to be representative of wader densities in the uplands as a whole. Accordingly these figures have not previously been used to provide estimates of upland Scottish wader populations. The New Atlas of Breeding Birds in Britain and Ireland was undertaken between 1988 and 1991 (Gibbons eft al 1993). Fieldwork for the Aflas was based on surveys of 2x2 km squares (tetrads). Between 8 and 25 tetrads were surveyed in each 10km square and each species was scored as present or absent in each of the tetrads surveyed. The proportion of visited tetrads in each 10km square in which a species was recorded provides an approximate abundance (strictly, frequency of occurrence) index for each 10km square, and this has been used to derive the Atlas density maps for each species. In this study, we correlate Aflas frequency indices and density estimates derived from intensive surveys of selected areas for Eurasian Oystercatcher, Northern Lapwing, Common Snipe, Eurasian Curlew and Common 18 Redshank. Where a significant correlation exists, this is used as a basis for extrapolation to generate national density and, hence, population estimates for these species in the Scottish uplands. Slightly different methods were required for some Scottish island groups, these are outlined in the methods section. Methods Definition of upland Land was classified as “upland”, “lowland” and “other”, according to the MacAulay Land Use Research Institute Land Capability classification (The MacAulay Institute for Soil Research Figure 1. Relationship between the distribution of the uplands and the distribution of areas surveyed for breeding waders in upland habitats. The cross hatched squares are all 10km squares that include upland habitats and that have bird frequency index scores derived from the New Breeding Atlas. The shaded squares are those areas where detailed breeding wader surveys have been undertaken using appropriate methods and within the given time period (see Methods). SB 24(2) 1982). The breeding wader survey of Scotland defined land classes | to 5.3 as farmed land (O’Brien 1996, O’Brien et al 2002), and so, by default, the uplands can be defined as land classes 6 and 7. These land classes are described as land suitable only for rough grazing (land class 6), or land of very limited agricultural value (land class 7). Saltmarsh is a scarce habitat, totalling approxi- mately 60km? of saltmarsh vegetation in Scotland, that has been identified as being an important habitat for breeding waders, in particular Common Redshank (Allport et al 1986, Brindley et al 1998). Saltmarsh habitats are included in the above definition of upland. The number of pairs of Common Redshank breeding on saltmarshes in Scotland has previously been estimated (Brindley et al 1998), while the numbers of other breeding waders recorded on a sample of Scottish saltmarshes were presented in Allport et al (1986). Using this data to estimate the saltmarsh population, together with confidence intervals, of these species was undertaken in the manner described in Brindley et al (1998). This definition of the uplands represents land totalling 39,817 km?, or 51.6%, of all land in Scotland together with a further 60km? of saltmarsh. Estimating populations in the uplands of mainland Scotland For the 5 breeding wader species considered, we determined which of the tetrads, surveyed for the Atlas, were predominantly upland (ie 80% or more of the land in the tetrad in land classes 6 and 7), and used only these tetrads to estimate an Upland Frequency Index for each species across all upland 10km squares in mainland Scotland. For the purposes of this analysis mainland Scotland includes the Inner Hebrides, but excludes Shetland, Orkney and the Outer Hebrides. Scottish Birds (2004) Only those Scottish upland surveys that were conducted between 1985 and 1995 (see Figure | for the distribution of these studies) to coincide approximately with the period of Atlas fieldwork were used. Also, only those surveys that had employed either the Brown and Shepherd (1993) or 200m transect (Stroud et al 1987) methods were used. Each of these methods cover ground in a manner similar to the farmed land breeding wader surveys, ie they get to within 100m of all parts of the surveyed area and map all birds observed. Other methods, such as 500m transects, do not aim to record all birds in the study area. Information on the location of breeding waders was obtained from survey maps from these studies and assigned to appropriate 10km squares. Any survey work in areas of these squares that would be classed as lowland, using the definition defined above, was excluded. Note that the land class based definition of lowlands does not exclude areas of moorland if these areas are deemed to be potentially suitable for agricultural improvement. Consequently, some areas that were intensively surveyed as upland habitats will be classed for the purposes of this analysis as lowland. These areas have been excluded from this analysis, but will have been included as part of the survey of breeding waders on farmed land. We used a regression analysis to model density estimates (pairs km?) derived from intensive surveys as a function of Upland Frequency Indices for all upland 10km squares where both data were available (see Appendix for details). A null model (ie mean density over all intensively surveyed 10km squares) was compared with a model that measured the extent to which the upland frequency indices explained variation from this null model. If the frequency index scores significantly improved the model (at p<0.05) then this latter model was used to derive population estimates (and 95% confidence intervals). If the frequency index score did not Estimating breeding wader populations of Scottish uplands 19 significantly improve the model then the best estimate was derived from the mean of the density estimates from the intensive surveys. Population estimates for the Scottish mainland were then derived as follows: 1. Population density over the upland area covered by intensive surveys was converted into a total population estimate for that area. 2. The mean upland frequency index was substituted into the regression model in order to estimate an upland population density. This value was multiplied by the remaining upland area (ie that not included in 1, above) to generate a population estimate for upland area not covered by intensive surveys. 3. The standard errors derived from _ the regression model were used to derive upper and lower confidence intervals (p=0.05) around the generated population estimates for the upland area not covered by intensive surveys. 4. The overall population estimates were calculated by summing the population estimate from | and the estimate from 2. The upper and lower confidence intervals were estimated by adding the population estimate from 1 to each of the upper and lower confidence intervals from 3. Estimating upland populations on the islands of Scotland Breeding wader densities in parts of the Northern and Western Isles considerably exceed those on the Scottish mainland (Galbraith et al 1984). Consequently, where possible, island specific analyses were undertaken. The method used varied depending on the amount of information available on upland populations in the island groups. Comprehensive upland wader surveys have been undertaken on both Shetland and Lewis & Harris, while there have been no recent surveys of the uplands on the Uists, Benbecula and Barra or in Orkney. The total area of uplands within Shetland has been estimated as 1,019km’, 71% of the total area of land within Shetland (Macaulay Institute for Soil Research 1982). On Shetland a represen- tative sample of these uplands was surveyed by Rothwell et al (1986); 29 of the 32 sites that they surveyed were classed as upland areas according to the definition used in the current paper. A number of additional areas known to have high wader densities were surveyed by Peacock et al (1985). Data from these 2 studies were considered to provide absolute counts for the survey areas covered (182 and 114 km”, respec- tively). Mean wader densities from the sites surveyed by Rothwell et al (1986) were extrap- olated to estimate the number of breeding waders on the remaining (723km?) upland areas of Shetland. Confidence intervals were derived from the sample of 29 site specific density estimates for each species, by bootstrapping. 999 sets of 29 density estimates from the 29 sites were randomly resampled with replacement, a mean density was calculated for each set, and this was extrapolated to calculate a total population estimate for the unsurveyed area. The (known) population of the surveyed areas was then added to give a total population estimate for Shetland. All 1000 estimates were then ranked and the 25th and 975th ranked estimates were used as the 95% confidence intervals. A survey of the number of breeding Eurasian Curlew on 24 randomly selected upland 1km squares on Orkney was undertaken in 1995 (Cadbury and Lambton 1995). This estimated the upland breeding Eurasian Curlew density at 4.98 pairs/km* on Orkney. This density estimate was used to calculate the total population of Eurasian Curlew on the uplands in Orkney outside of RSPB reserves. Information from the upland RSPB reserves on Orkney was included. Data for the RSPB reserves on Hoy, at Hobbister and Trumland and Birsay moors were extracted from the reserve management SB 24(2) plans. Estimates for other waders on the uplands of Orkney were obtained by applying densities calculated for the Scottish mainland to the area of Orkney upland, excluding RSPB reserve areas. Population estimates were obtained by summing the count of waders on upland RSPB reserves and the estimates for the remaining upland areas on Orkney. A detailed analysis of the upland wader populations of Lewis and Harris has recently been undertaken by Scottish Natural Heritage (Bates et al 1994). Surveys covering 259km2 of the 1965km2 of uplands in Lewis and Harris were extrapolated to areas of unsurveyed but suitable habitat as assessed by satellite imagery and LCS88 data. Population estimates for Common Snipe, Eurasian Curlew and Common Redshank were derived from this source. Population estimates were not attempted for either Northern Lapwing or Eurasian Oystercatcher by this method as these species were recorded less frequently than the other 3 species. Accordingly, the regression model derived from mainland Scotland was used to estimate these wader populations based on the mean New Atlas upland frequency index score for Lewis/Harris and the total area of uplands in the islands. There have been no detailed wader surveys of the upland areas of the southern half of the Outer Hebrides (the Uists, Benbecula and Barra); 66 predominantly upland tetrads were surveyed for the New Atlas. These provide wader upland frequency indices specific to these islands. These, together with the total area of uplands in these islands (543km7), were incorporated into the regression model derived for the Scottish mainland and used to produce total population estimates together with confidence intervals. Combined Scottish population estimates The breeding wader estimates for Scottish uplands can be calculated by summing the Scottish Birds (2004) individual Scottish upland region estimates. Similarly, the overall Scottish population estimate can be calculated by summing the upland and lowland estimates. It is incorrect, however, to sum the 95% confidence intervals around these estimates to derive overall confidence intervals around the total estimate. Accordingly we used a randomisation approach to derive confidence intervals around the final population estimate (See Appendix for details). Results Identifying the relationship between wader density and upland frequency index Upland breeding wader densities and upland frequency indices were jointly available for 76 10 km squares in mainland Scotland. The models that best fitted the data are presented in Table 1. Adding the quadratic term for upland frequency index did not significantly improve (at a P<().05 level) the relationship for any of the 5 species. For 4 of the 5 species the best fit model included a linear and positive relationship between wader density and the upland frequency index. Only for Common Snipe was there no relationship. These models are used to estimate the wader populations in all subsequent analyses except on Shetland and, for 3 of the 5 species, on Lewis and Harris (see above). Estimating breeding wader populations of Scottish uplands 21 Scottish mainland The total area of upland on mainland Scotland has been calculated as 36,123 km”. This is the total area of land capability classes 6 and 7 in Scotland less the area of land class 6 and 7 on Shetland, Orkney and the Outer Hebrides (The MacAulay Institute for Soil Research 1982). Of this, 719km? were covered by intensive surveys. For each of the 5 species, upland frequency indices for the intensively surveyed squares, and for all upland squares, are shown in Table 2. This indicates that, for Northern Lapwing, Eurasian Curlew and Common Redshank, the surveyed areas tended to be in squares with higher frequency index scores than upland squares in general. Correcting for inter square bias was therefore desirable. The reverse was true for Eurasian Oystercatcher, probably indicating that survey areas were chosen on the basis of a focus on upland wader species, whose habitat requirements will differ from those of Eurasian Oystercatchers. Survey areas may therefore have represented rather poor quality habitat for this species. The only species where there was no significant relationship between density and frequency index, Common Snipe, is one of the 2 species where the survey and upland frequency index scores are very similar. Simply extrapolating mean densities in survey areas therefore seems reasonable. Table 1. Estimating the populations of breeding waders in the uplands of mainland Scotland. The intercept and slope coefficients are derived from a regression model with a Poisson error function of the form log(pairs km‘?)= Intercept + slope * Upland Frequency Index. The analysis is derived from estimates of the density and upland frequency index on 76 separate 10 km squares. Intercept Slope coefficient Standard error _ coefficient Standard error Eurasian Oystercatcher -4.248 0.405*** 1.863 0.733" Northern Lapwing -2.498 0.2627 ** 1.508 OA4735* Common Snipe -1.608 O.147*** 0.000 Eurasian Curlew -1.805 0.342 *+* 1.966 0).409*** Common Redshank -5,.332 O.S76*+* 4.765 0.991 *** sd SB 24(2) Table 2. Upland frequency indices. UFI Survey UFI Upland Eurasian Oystercatcher 0.195 0.203 Northern Lapwing 0.164 0.110 Common Snipe 0.179 0.174 Eurasian Curlew 0.324 0.246 Common Redshank 0.081 0.050 UFI Survey = upland frequency index for survey squares. UFI Upland = upland frequency index for all upland squares. Table 3. Population estimates for the uplands of mainland Scotland. Population in Estimated population Total upland survey areas in remaining upland areas mainland population Eurasian Oystercatcher op Northern Lapwing 116 Common Snipe 160 Eurasian Curlew 468 Common Redshank 1] For each of the 5 wader species the total number of breeding pairs in the 719km/2 of the surveyed areas, the total population estimated for the remaining upland area (36,063-719 = 35,344km7) based on the relevant regression model from Table 1, and the total upland population estimate for the Scottish mainland are presented (Table 3). Scottish islands Tables 4-7 present population estimates for the 5 key wader species for the uplands of Shetland (Table 4), Orkney (Table 5), Lewis & Harris (Table 6), the Uists, Benbecula and Barra (Table 7) and saltmarsh (Table 8). Total population estimates The Scottish upland population of the 5 wader species has been estimated by summing the totals for each of the regions (Table 9). A randomisation 737 (326-1,686) 768-357-1717) 3434 (2031-5644) 3,550 (2,147-5,760) 7080 (5233-9162) 7,240 (5,393-9,322) 9437 (4477-18510) 9,905 (4,945-18,978) 217 (67-710) 228 (78-721) test was used to estimate the confidence intervals around these estimates (see Methods). Combining the saltmarsh, upland and farmed land (O’Brien et al 2002) estimates gives the total population of the 5 species of breeding waders in Scotland (Table 10). For all species, the upland population represents only a small proportion of the Scottish total, for 3 species this is less than 5% of the total. The uplands represent an important habitat for just 2 of the species, Eurasian Curlew and Common Snipe. Discussion The upland wader population estimates presented here complement the results of the lowland wader surveys of 1992-93 and 1997-98 (O’Brien, 1996, O’Brien et al 2002). Estimates of the Scottish populations of Eurasian Oystercatcher and Scottish Birds (2004) Estimating breeding wader populations of Scottish uplands 23 Table 4. Estimates of breeding wader populations in the uplands of Shetland. Number of pairs recorded on upland sites Eurasian Oystercatcher 276 [52 Northern Lapwing 102 0.56 Common Snipe 139 0.76 Eurasian Curlew 313 Lv2 Common Redshank ga 0.12 Density estimate (pairs km-2) Total number Upland Population of birds Estimate (95% surveyed Confidence Intervals) 597 1,694 (1,351-2,136) 221 626 (496-766) 241 793 (646-959) 487 1,731 (1,362-2,161) 40 123 (88-171) Table 5. Estimates of breeding wader populations in the uplands of Orkney. Total number recorded on RSPB reserves (75km7) Eurasian Oystercatcher 20 Northern Lapwing 28 Common Snipe 140 Eurasian Curlew 418 Common Redshank 14 Eurasian Curlew exceed previously published UK totals by 2.4 and 1.5 times respectively (based on middle of range data summarised in Stone ef al (1997)). The importance of the populations of breeding waders in Scotland within a European context has previously been highlighted (O’ Brien et al 2002). A significant proportion of the population of 2 of the species, Common Snipe and Eurasian Curlew, occurred in the uplands. These, combined with the large populations of Golden Plover Pluvialis apricaria, Dunlin Calidris alpina and, in parts of the uplands, Greenshank Tringa nebularia indicate that upland habitats provide an important resource for breeding waders in Scotland. There are a number of assumptions in the analysis that need to be considered when assessing the Upland Frequency Index scores for Orkney (from 26 tetrads) Total upland estimate (95% confidence intervals) (216 km7) 0.42 24 (21-29) 0.27 31 (29-35) 0.88 157 (147-191) 0.88 1120 (1,003-1,314) O19 15 (14-16) accuracy of the population estimates. M@ The potential for bias in tetrad selection from within a 1Okm square is considered in the New Atlas. The New Atlas includes a comparison of measures of abundance obtained from the main Atlas survey with those from a survey of ‘key squares’. The selection of tetrads in the Key Square Survey was random. The results of the comparison indicate that the Atlas survey contained a bias towards tetrads with Eurasian Oystercatchers (P<0.05 in Scotland), but no significant biases associated with any of the other target species (Gibbons et al 1993). M@ The patchy distribution of the survey squares that provided the density estimates (as shown by Figure 1) to compare with the frequency indices may bias the relationship. It is possible that the 24 SB 24(2) Table 6. Estimates of breeding wader populations in the uplands of Lewis and Harris. Total pairs surveyed on Lewis/Harris Eurasian Oystercatcher 2 Northern Lapwing 33 Common Snipe 140 Eurasian Curlew 32 Common Redshank pA Upland Frequency Index (from 199 tetrads) Lewis/Harris population estimate 0.18 0.20 32 (15-77) 221 (136-359) 800 (649-992) 200 (105-390) 100 (52-175) Table 7. Estimates of breeding wader populations in the uplands of the Uists, Benbecula and Barra. Upland Frequency Index (from 66 tetrads) Eurasian Oystercatcher 0.27 Northern Lapwing 0.36 Common Snipe 033 Eurasian Curlew 0.08 Common Redshank 0.32 relationship varies within the Scottish mainland, for example between the west and east coasts of Scotland or between the Flow Country and other parts of the Scottish uplands. M Frequency Indices apply to the whole of the upland area within a 10 km square, whereas the detailed bird surveys covered only part of any 10km square (at times the densities were based on a survey of a single 1km square). Densities derived from these detailed surveys were then applied to the whole 10km square, with the assumption that survey plot densities are representative of densities in the 10km square as a whole. Any intra square bias caused by positively selecting areas from within the 10km square may inflate the apparent wader density for the given frequency index. Note that many of the density estimates in Eastern Scotland were derived from 1 km squares that had been randomly selected from within a 15km by 15km Uists, Benbecula and Barra population estimate 13 (5-32) 77 (41-146) 109 (81-144) 101 (52-211) 12 (3-42) square so the bias associated with selecting the areas for survey would not be applicable to these sites (Shepherd et al 1989). @ It has been assumed that wader populations in the uplands were reasonably constant between the years that the upland surveys were undertaken and the time of the Atlas fieldwork. The Atlas fieldwork was spread between the years 1988 and 1991. Frequency Index data does not identify the year in which data was gathered for a particular 10km square. The upland surveys analysed took place between 1985 and 1990. Any rapid change in wader numbers between the atlas and the survey years will reduce the level of accuracy. It should be noted, however, that a previous study of upland breeding wader populations found that there was very low inter annual within plot variation in wader densities (Langslow and Reed, 1985). Scottish Birds (2004) Estimating breeding wader populations of Scottish uplands 25 Table 8. Estimates of breeding wader populations on Scottish saltmarshes. Counts of breeding waders are derived from the 1985 survey (Allport et al 1985) except for Common Redshank which were resurveyed in 1996 (Brindley et al 1998). Total area surveyed was 9.81 km? — 88.89% of which was vegetated (Brindley et al 1998) Number recorded Population estimate (8.72km7) (c60km7) Eurasian Oystercatcher 107 702 (520-888) Northern Lapwing 49 348 (141-623) Common Snipe 0 0 Eurasian Curlew 17 114 (55-183) Common Redshank 210 811 (397-1246) Table 9. Breeding wader population estimates for the uplands of Scotland. The 95% confidence intervals around the population estimates were derived using a randomisation test (see Methods). Main- Lewis Uists, Scottish upland land and_ Benbecula Population Harris and Barra Orkney Shetland Estimate Eurasian Oystercatcher 768 35 13 24 1,694 2,534 (1,950-3,600) Northern Lapwing 3,550 221 Ta 31 626 4,505 (3,087-6,782) Common Snipe 7,240 800 LO9 157 793 9,099: (7,282=11,212) Eurasian Curlew 9,905 200 101 1,120 1,731 13,057 (8,209-22,521) Common Redshank 228 100 Jp 15 123 478 (302-964) Table 10. Total population estimates for farmland breeding waders in Scotland. The 95% confidence intervals have been derived using a randomisation test (see Methods). Upland Saltmarsh Farmed Scottish % in Land Total uplands Eurasian Oystercatcher 2,534 G02 91,102 94,338 (80,152-107,170) 3% Northern Lapwing 4,505 348 86,654 91,507 (71.801-105,842) 5% Common Snipe 9:099 0 32,977 42,088 (36,751-51,181) 22% Eurasian Curlew 13;057 114 45,627 58,814 (50,387-71,510) 22% Common Redshank 478 811 13,431 14,720 (11,709-17,433) 3% 26 M The relationship between wader density and frequency index may not be linear. Frequency -Indices vary between zero and one while bird densities vary between zero and infinity. A frequency index of one (all surveyed tetrads being occupied) could have a minimum density of 0.25 pairs km*? whereas the maximum densities recorded for any one 10km square in the current survey was 14 pairs km? (Northern Lapwing on Lewis). Attempts to fit curvilinear plots (x+x2) failed to significantly improve the fit of the models, perhaps because there was insufficient data to indicate a curvilinear relationship. This may be a particular issue in studies where a high proportion of the frequency indices are close to the maximum possible. M The density data for this analysis were derived from either 200m transect surveys or the Brown and Shepherd method (Brown and Shepherd, 1993). There have been few comparisons between the numbers of breeding waders recorded by the 200m transect surveys and the number recorded by intensive surveys (Jackson and Percival 1983, Stroud et al 1987). Jackson and Percival (1983) compared the single, June, survey on machair sites with intensive survey work and found that the single survey underestimated Ringed Plover and Dunlin by 16% and 21% respectively, but found no difference for Oystercatcher and Redshank. Stroud ef al, working in the Flow Country, suggested that the number of breeding waders recorded by the transect method was between 65% (Dunlin) and 90% (Golden Plover) of the actual number of pairs present. The Brown and Shepherd method was found to record around 60% of all Eurasian Oystercatchers, 53% of Northern Lapwing, 73% of Common Redshank and 86% of Eurasian Curlew on a site when compared with intensive nest finding methods. The intensively studied sites used for the comparison with the species considered in the present analysis tended to be sites on SB 24(2) agricultural sites. The applicability of these estimates to upland, and in particular low density, sites have not been determined. It is likely that the population estimates presented here represent minimum figures for each of the species. The estimates presented here are currently the best available, based as they are on the use of Atlas frequency index data to compensate for the bias associated with site selection in previous upland surveys. They are, however, already more than 10 years out of date. It is unlikely that a more robust survey will be undertaken to specifically count the upland wader population in the near future. If an updated atlas provides similar opportunities for comparison across the uplands then undertaking upland surveys to calibrate atlas information may provide the best option for revising these population estimates. The only alternative way of improving on the accuracy of the current estimates may be to undertake a detailed analysis of the habitat requirements of breeding waders in the uplands, and use this to extrapolate to areas not previously surveyed. This type of approach has been attempted for Golden Plover, Greenshank and Dunlin in the flow country (Stroud et al 1987, Avery and Haines- Young 1990). Acknowledgments Many thanks to staff at the BTO, in particular Deborah Lang, Nick Carter, Rob Fuller and Carl Evans who provided the atlas derived data. Phil Whitfield at SNH, Andy Brown at EN and David Stroud at JNCC_ provided unpublished information. Within RSPB Ian Fisher, Mark Hancock, Murray Grant, James Pearce-Higgins and Innes Sim have all stimulated much discussion about the methods used while Ian Bainbridge, David Gibbons and Jeremy Wilson improved previous drafts of this paper. Scottish Birds (2004) References Allport G, O’Brien M & Cadbury C J 1986. Breeding Redshanks Survey On Saltmarshes 1985. Unpublished report, RSPB, Sandy. Avery M I & Haines-Young R H_ 1990. Population estimates for the dunlin Calidris alpina derived from remotely sensed satellite imagery of the Flow Country of northern Scotland. Nature 344: 860-862. Bates M A, Shepherd K B, Whitfield D P & Arnott D A. 1994. A breeding wader and upland bird survey of selected sites in Lewis and Harris. - A report to SNH RASD contract 1994/5 -001. Brindley E, Norris K, Cook T, Babbs S, Forster Brown C, Massey P, Thompson R & Yaxley R 1998. The abundance and conservation status of Redshank Tringa totanus nesting on saltmarshes in Great Britain. Biological Conservation 86: 289-297. Brown A F 1991. An annotated bibliography of moorland breeding bird and breeding wader surveys, 1970-1990. Joint Nature Conservation Committee, Peterborough. Brown A F & Shepherd K B 1993. A method for censusing upland breeding waders. Bird Study 40: 189-195. Cadbury C J & Lambton S 1995. Breeding bird survey of Orkney moorland and Sutherland flows. Unpublished report, RSPB, Sandy, Beds. Galbraith H, Furness R W & Fuller R J 1984. Habitats and distribution of waders breeding on Scottish agricultural land. Scottish Birds 13(4): 98-107. Gibbons D W, Reid J B & Chapman R A 1993. The New Atlas of Breeding Birds in Britain and Ireland: 1988-1991. T and AD Poyser, London. Hayhow S J, Scotchford C A & Campbell L H. 1987. Moorland breeding bird survey, Wester Ross, 1987. Unpublished report, RSPB, Sandy, Beds. Jackson D B & Percival S M 1983. The breeding waders of the Hebridean machair: a validation check of the census method. Wader Study Estimating breeding wader populations of Scottish uplands Df Group Bulletin 39: 20-24. Langslow D R & Reed T M 1985. Inter-year comparisons of breeding wader populations in the uplands of England and Scotland. Pages 165-173 in K Taylor, RJ Fuller and PC Lack (eds). Bird census and atlas studies. Proceedings of the VIII International Conference on Bird Census and Atlas Work. British Trust for Ornithology, Tring, Herts. Macaulay Institute for Soil Research 1982. Soil Survey of Scotland. 1:250 000 maps. Land Capability for Agriculture, Sheets 1 to 7. University Press, Aberdeen. O’Brien M 1996. The numbers of breeding waders in lowland Scotland. Scottish Birds 18: 231-241. O’Brien M, Tharme A & Jackson D 2002. Changes in breeding wader numbers on Scottish farmed land during the 1990s. Scottish Birds 23: 10-21. Peacock M A, Beveridge F M & Campbell L H 1985. Shetland Moorland Breeding Birds Survey, Unpublished report, RSPB, Sandy, Beds. Rothwell A, Stroud D A & Shepherd K B. 1986. Shetland Moorland Bird Surveys 1986. Unpublished NCC report No 775, Peterborough. Shepherd K B & Brown A F. 1989. A survey of breeding birds of some moorland areas in Angus in 1989. - Chief Scientist Directorate Commissioned research report no 950, Peterborough. Shepherd K B, Brown A F & Harding N J. 1989. Moorland bird surveys in Grampian, Tayside and Central Regions in 1989 - site descriptions and preliminary results. - NCC Chief Scientist Directorate report 965 (GTC), Peterborough. Stone B H, Sears J, Cranswick P A, Gregory R D, Gibbons D W, Rehfisch M M, Aebischer N J & Reid J B 1997. Population estimates of birds in Britain and in the United Kingdom. British Birds 90: 1-22. Stroud D A, Reed T M, Pienkowski M W & Lindsay R A 1987. Birds, bogs and forestry. The 28 peatlands of Caithness and Sutherland. Nature Conservancy Council, Peterborough, UK. Whitfield D P 1997. Waders Charadrii on Scotland’s blanket bogs: recent changes in numbers of breeding birds. In L Parkyn, RE Stoneman and HAP Ingram (eds). Conserving Peatlands. Centre for Agriculture and Biosciences International, Wallingford, Oxon. Mark O’Brien, RSPB, Dunedin House, 25 Ravelston Terrace, Edinburgh, EH4 3TP. Tel 0131 311 6500. Email: mark.obrien @rspb.org.uk. Craig Whyte, 5 Queen’s Gardens, Aberdeen, ABI5 4YD Appendix We used a regression model with a poisson error function and a logarithmic link function to relate the number of birds surveyed in a square to the frequency index for that square. The number of birds surveyed will be dependent on the area surveyed, so we included log(area surveyed) as an offset in the model. This is equivalent to comparing the density of birds surveyed in the square with the frequency index. The advantage over using density as a dependent variable 1s that this approach weights the density by the area over which the density has been determined. The equation for the analysis is Log(No of birds in survey) = intercept + log(area) [+ slope*frequency index]. SB 24(2) We compare the equation with and without the term in square brackets by using change in the residual deviance, tested against a ~2 distribution with the appropriate degrees of freedom. If adding the frequency index component significantly improved the model (p<0.05) then we can use the frequency index scores derived for all upland squares to estimate the average density of each of these species across all uplands in Scotland. Summing the upper and lower confidence intervals for each of the areas is an inappropriate way of determining the 95% confidence intervals for the population as a whole. Rather, we generated a set of 999 standardised normal values at random from a normal distribution with mean=zero and standard deviation=1. We then multiplied each of these by the standard error derived from the regression model for the given species and used this figure to calculate a ‘pseudo population’ estimate for the species in the given region. We repeated this for all 999 values. We generated a new set of standardised normal values for each of the regions where we used the regression model to derive a population estimate. We added the actual population estimates for each region to this ‘pseudo population’ estimate, making 1000 estimates. We then combined this, in random order, with the bootstrapped estimates for each of the Shetland and the Lewis and Harris populations. We then summed the pseudo population estimates across each of the regions to provide 1000 estimates of the total wader population for each species. This is then sorted and the 25th and 975th value used as an estimate of the 95% confidence intervals around the total population estimate. ~ Morris Scottish Birds (2003) 24:29-42 Origins of Common Starlings wintering in the Highlands 29 Origins and movements of Common Starlings wintering in the Highlands HUGH CLARK & ROBIN M SELLERS This paper presents an analysis of the origins and movements of Common Starlings wintering in the Highlands of Scotland based on 391 ringing recoveries generated over the past half century. They show that the midwinter population is made up of birds from 4 main areas, the breeding population of the Highlands, migrants from the continent, birds dispersing or in some cases migrating from Orkney and Fair Isle, and a small number of birds from southern Scotland and northern England. Information on the approximate percentage of birds from these different areas and their age and sex are tabulated. The recoveries show that many birds depart in late winter or spring, with substantial numbers returning to the continent to breed, and much smaller numbers to Fair Isle. They reveal also a distinct and unusual series of movements of birds to the south of the Highlands. These movements take the birds to northeast, central and southern Scotland as well as northern England. For movements over 50 km, the frequency distribution of distances moved is approximately exponential with a ‘half distance’ of 193 + 22 km. Most of the birds undertaking these long distance southward movements are first years, though some adults are also involved. The origins of these birds are discussed. Introduction Most mainland European populations of the Common Starling Sturnus vulgaris migrate along a broadly northeast/southwest axis, with substantial numbers moving into Britain to winter (eg Goodacre 1959, Dorst 1962, Feare 1984, 2002). The British breeding population, by contrast, has been viewed as _ largely sedentary, with movements generally confined to a few tens of kilometres from the natal site (Feare 1984, 2002, Thom 1986), though Rae & (1978) and Duncan (1984) have presented some preliminary observations based on the ringing of Common Starlings in Aberdeen suggesting that more extensive movements orientated to the south may also occur. During the winter months Common Starlings are largely absent from high ground in Scotland, and appear in greatest numbers in Orkney, Caithness, Aberdeenshire and the Central Lowlands, according to the 1981-84 Winter Atlas Survey (Lack 1986). In Caithness they are one of the commonest winter birds, feeding by day mainly on farmland and congregating at night in large communal roosts in or on buildings, conifer plantations and other woodland, and sea caves. As a breeding species Common Starlings are very unevenly distributed in the Highlands. The 1988-91 Breeding Atlas Survey shows them to be concentrated in Caithness (excluding the Flow Country), around Dornoch in Sutherland, between the Dornoch Firth and the Cromarty Firth in Easter Ross, and in the Black Isle (Gibbons, Reid & Chapman 1993). In these areas winter numbers are appreciably greater than in the breeding season, typically building up in October and November, and decreasing 30 between late February and the end of March. We report here the results of an investigation into the origins of Common Starlings wintering in the north of Scotland and the movements associated with the build up and decline of winter numbers. Material and methods For the purposes of this study the Highlands are taken as the former Highland Region (that is Caithness, Sutherland, Ross & Cromarty, Invernessshire, Nairn, and Argyll north of the Sound of Mull including the Glencoe area). The analysis 1s based on 391 recoveries, details of which were kindly supplied by the British Trust for Ornithology. Of these, 191 (49%) were birds ringed and recovered in the Highlands, 143 (37%) birds ringed in the Highlands and recovered elsewhere, and 57 (15%) birds marked elsewhere and recovered or recaptured in the Highlands. For those marked in the Highlands, the distribution of ringing sites showed a strong bias towards low lying land especially that between Wick and Inverness. This reflects primarily the distribution of ringers, but in midwinter, when Common Starlings become regular visitors to gardens, it seems to be a fair reflection of the distribution of birds. The principal ringing sites (or groups of sites) in the Highlands in terms of the number of recoveries generated were Wick (121 recoveries), Inverness (72), Golspie (22), Thurso (16), Beauly (13) and Invergordon (12). In Caithness the vast majority of the birds were caught in elastic propelled ‘whoosh’ nets baited with household scraps and a mixture of fish, batter and fat waste. We have no direct information on the catching methods used elsewhere in the Highlands, but many of the recoveries originate from before the advent of mist nets and presumably involve clap nets and static traps such as Potter traps or chardonnerets. SB 24(2) The biases inherent in these catching methods as concerns Common Starlings are unknown but the possibility of some bias needs to be borne in mind when considering the age and sex ratios described below. The bulk of the recoveries were generated from ringing during the past half century, but there was some variation in the numbers marked in different parts of the Highlands at different dates. In Caithness most were caught between 1985 and the present, with especially large numbers of recoveries being produced from intensive ringing in the winters of 1985-86 and 1986-87. Ringing elsewhere in the Highlands was more evenly spread, but with a slight bias towards the 1950s. The majority (67%) of the recoveries refer to birds caught in the winter months (November - March inclusive). The 391 recoveries include 43 (11%) of birds ringed as nestlings. Many of the recoveries were generated from birds found dead, but in the majority of such cases the cause of death was unknown and the birds were simply reported as found dead or dying (195 recoveries, 50%). Of those in which the cause of death was known, 46 (11.8%) recoveries resulted from birds killed by cats, 16 (4.1%) from birds found dead after becoming trapped in buildings, 13 (3.3%) from birds which had been shot, 9 (2.3%) were road casualties, 6 (1.5%) were from birds which had been predated, 2 (0.5%) were from birds which had flown into windows, 2 (0.5%) from birds which had collided with wires, 2 (0.5%) birds which had drowned and | (0.3%) was from a bird killed by a dog. In addition, 63 (16%) were birds caught (‘controlled’) by other ringers and 31 (8%) were birds which had been caught alive by non ringers and released (mostly birds trapped in buildings). In the remaining 5 (1.3%) the recovery circumstances were unknown. There is an obvious association with man and his world. Scottish Birds (2004) Sexing Common Starlings is_ relatively straightforward but ageing techniques have improved considerably in the half century or so during which the recoveries were generated. Nevertheless some birds remain difficult to age (cf Williams 1991, Svensson 1992) and the possibility that some of the recoveries included in this study result from birds that were incorrectly aged on ringing needs to be borne in mind. Statistical analysis was undertaken using the Minitab Statistical Software Package, release 13, or following procedures described in Bailey (1981) or Batschelet (1981). Results Common Starlings marked in the Highlands have been recovered in 4 main areas: (1) northern Europe, (1i) the United Kingdom south of the Highlands, (111) the Northern Isles and (iv) the Highlands themselves. Movements of birds marked outside the Highlands and recovered there corresponded with categories (i), (11) and Origins of Common Starlings wintering in the Highlands | (iii). These 4 categories form the basis of the analysis presented here. In none of them did the sex ratio differ significantly from 1:1 (Table 1). Movements between the Highlands and northern Europe A total of 66 recoveries relate to movements between the Highlands and Northern Europe, and include 2 from ships in the North Sea. Of these 56 (85%) were birds marked in the Highlands, and 10 (15%) on the continent. Fig 1 shows the distribution of ringing or recovery sites on the continent. The majority (85%) were in Norway, with a marked bias towards the Bergen and Stavanger areas in SW Norway and around Trondheim in south central Norway, the provinces of Rogaland, Hordaland, More og Romsdal and Sér-Trondelag accounting for 66% of movements to or from Norway. In addition, there were movements to or from Denmark (5), Finland (2) and Estonia (1). The most northerly movement was of a bird recovered in Troms in N Norway (69°N) within the Arctic Circle, the most southerly one in Fyn, Denmark (56°N) and the most easterly a bird from Parnu, Estonia Table 1. Age and sex composition of birds moving to and from the Highlands Category Number of recoveries Total no. females* males* n_ first-years adults n of birds? to/from Continent 26 (43%) 34(57%) 60 36(67%) 18 (33%) 54 66 (age on handling in Highlands) to South 37 (49%) 38(51%) 75 51 (80%) 13 (20%) 64 80 from South 4 (57%) 3(43%) 7 8 (80%) 2 (20%) 10 12 to Northern Isles 4 (67%) 2 (33%) 6 6(100%) O0(0%) 6 q from Northern Isles 13 (57%) 10(43%) 23 31(94%) 2 (6%) 33 35 within Highlands 65 (44%) 82(56%) 147 109 (69%) 48 (31%) 157 191 (age on ringing) a Sex ratio not significantly different from 1:1 in any case (P > 0.1) | b Note not all birds aged and sexed. | 32. (24c0E). Ignoring the 2 recoveries in the North Sea, about 50% of birds had moved between 480 km (which represents the shortest sea crossing between the North of Scotland and the continent) and 800 km, and 75% between 480 km and 1100 km. The longest recorded movement was the bird ringed in Estonia, already referred to, which travelled to Wick, a distance of 1614 km. Movements (in terms of | recovery @o @ © 2 recoveries @oO 3 recoveries SB 24(2) those from Scotland to the continent) were nearly all orientated between northeast and east. The dates of handling of these birds are shown in Fig 2a. A very clear separation is apparent, with all the handlings in the Highlands falling in the winter months (November - March) with just one exception (a bird ringed in More og Romsdal, Norway in August and recovered in @ Ringed in the Highlands © Ringed on the Continent Scottish Birds (2004) Origins of Common Starlings wintering in the Highlands 5) (a) Movements between the Highlands and N Europe (n = 66) (b) Movements from the Highlands to the south (n = 80) (c) Movements from the Northern Isles to the Highlands (n = 35) No. recoveries Nw © (d) Movements in the Highlands over 10 km of birds ringed in the breeding season (n = 26) (ec) Movements in the Highlands over 10 km of birds ringed in the winter (n = 39) jeden Mee Mae nk Do Ae- SS Ob. N+ 2D Figure 2. Dates of ringing and recovery of movements into and within the Highlands. (a) - (c) open columns, dates of handling in Highlands; solid columns, dates of handling out of the Highlands; (d), (e) open columns, ringing dates; solid columns, recovery dates 34 Caithness the following July), and those on the continent being primarily in the breeding season (March - October) with just 3 exceptions. The latter were all birds ringed in Scotland in the winter months and recovered in Norway (2 birds) or on board a ship in the North Sea (one bird) in January in a subsequent season. The ages and sexes of the birds involved in these movements are summarised in Table 1. First year birds outnumbered adults by about 2:1. Movements between the Highlands and the South Of the 92 recoveries involving movements between the Highlands and the South, 80 (87%) concerned movements from north to south, and 12 (13%) ones in the opposite direction. In the former group there were 56 (70%) recoveries widely distributed in Scotland with particular concentrations in Grampian and _ ‘Tayside Regions, 23 (29%) in England, mainly in northern England but extending as far south as Avon, and one (1%) in N.Ireland. As Figure 3 demonstrates, most of these movements were orientated between ESE and south, with a substantial subset of movements to the east, mostly birds from near Inverness moving to NE Scotland and the Aberdeen area. The majority of recoveries were found between 50 and 150 km, with mostly decreasing numbers in each subsequent 100 km interval above this (Fig 4a). A plot of the logarithm of the number of recoveries in each 100 km interval against distance was close to linear (Fig 4b), showing the frequency distribution of distances moved to be approximately exponential. From the slope of the line in Fig 4b, the ‘half distance’1 can be estimated as 193 + 22 km. The longest movement recorded was from Nairn to Bath, Avon, a distance of 695 km. Among birds that were aged, first year birds outnumbered adults by about 4:1. Two of the SB 24(2) recoveries involved birds Highlands as nestlings. ringed in_ the The dates of handling and recovery of birds moving out of the Highlands to the south are shown in Fig 2b. The majority of these were birds marked in the winter months (December - February inclusive), and, although there were recoveries in all months, most were in late winter and spring (February - June inclusive), suggesting that the movements had taken place in late winter. The timing of ringing and recovery of 34 birds found within a year of being marked are illustrated in Fig 5, and showed 2 distinct phases of movements, one in the summer months (mostly July - October inclusive) and a second from late winter into spring (mid-January - April inclusive). The former involved only first year birds, with movements not clearly orientated in any particular direction (movements fell between NNE and SW, cf Fig 6a) and distances moved of mostly 10 recoveries Dit =e / x %, s . < \ Figure 3. Movements from the Highlands to other parts of Britain and Ireland. percentage of recoveries to the south over 50 km varied between 0% and 100%, with an overall mean of 43%. Sample sizes in any one period were generally small, however, and the results mostly not suitable for statistical analysis. Low numbers moved more than 50 km south in 1976/77 (1 of 8 recoveries), 1984/85 (1 of 8) and 1990/91 (none of 5), and each of these were significantly different from each of 4 other winters which gave rise to at least 10 recoveries (Fisher Exact Probability Test, P < 0.05 in all cases). It appears that the numbers moving south do vary between one year and the next to some extent, but this conclusion must remain tentative until a larger data set becomes available. wy) jo) S ) N (=) L Number of recoveries (N, io) Pm) a f © (b) | ca = e ° Ee Be ge), | ee = | u ; ae i 0 200 400 600 800 Distance (km) Figure 4. Frequency distribution of distances moved over 50 km for birds moving from the Highlands to the South. (a) Plot of number of recoveries against distance. (b) Plot of In(number of recoveries) against distance. There were too few recoveries of birds moving into the Highlands from the south to provide any detailed characterisation of such movements. Of the 12 such recoveries, ringing locations were equally divided between Scotland and England and involved 5 movements in the range 100-200 km, 2 between 200 and 400 km and 5 between 400 and 700 km. Three of the birds were marked as nestlings, 4 others as first years and 2 as adults. The 3 birds ringed as nestlings were all marked in May and recovered (1) 136 km away in the following month, (11) 151 km away 7 months later, and (iii) 8 months later, having moved 697 km from Slough, Berkshire to Inverness, the longest south to north movement recorded. 36 Movements between the Highlands and the Northern Isles There were 42 recoveries of birds moving between the Highlands and the Northern Isles, 7 (17%) from south to north, and 35 (83%) in the opposite direction. The northward movements included 3 from Caithness to Orkney, 3 from Caithness to Fair Isle and one from Nairn to Orkney, the latter at 169 km was the longest of these 7 movements. All bar one of these were birds ringed in their first year of life. They included one marked as a nestling in Caithness and found on South Ronaldsay, Orkney, 44 km away, 2 months after ringing, and another marked as a first year female in Wick on 26 February 1994, caught on Fair Isle on 17 May 1995 and retrapped back in Wick on 6 December 1998. The ringing and recovery locations of the 35 movements from the Northern Isles to the Highlands are summarised in Table 2. Most (> 100-200 km a Z oO ass Vv O SB 24(2) common amongst these were birds ringed on Fair Isle and found in Caithness, a distance of around 150 km SW. Nearly all the birds making these movements were in their first year of life having been ringed either as nestlings (8 birds) or first year birds (23 birds); just 2 were aged as adults on ringing and 2 were unaged (cf Table 1). These movements appear to take place mainly in late summer, autumn and early winter. Thus the majority of the birds were marked in either June or July and recovered or recaptured between December and March (Fig 2c). This is supported by the timing of 13 birds recovered or recaptured in the same season as they were ringed. Movements within the Highlands The largest single group of recoveries were those in the Highlands, comprising some 191 in total. Of these 114 (60%) were within 5 km of the place of ringing, 160 (84%) within 25 km and 179 (91%) within 50 km. The general Mea! AS ONDIJFMAMJJ ASO Figure 5. Dates of ringing and recovery of birds moving south from the Highlands and recovered in the same season Scottish Birds (2004) é (a) Birds ringed June - November (Rayleigh test, z= 1.071, n= 12. P>0,05, NS.) Figure 6. Orientation of same season movements over 50 km of birds ringed in the Highlands and recovered in Britain. (a) Birds ringed June - November (Rayleigh test, z = 1.071, n = 12, P > 0.05, N.S.); (b) Birds ringed December - April (Rayleigh test, z = 16.06, n = 22, P < 0.01). pattern of distances moved (the majority near the place of ringing, with declining numbers as the distance from the place of ringing increases) was the same regardless of whether birds had been ringed in the breeding season (May to September inclusive) or during the winter (November to March) (Fig 7). Movements of 10 to 50 km showed no particular orientation, regardless of when the birds were ringed (Fig 8a & 8c), whilst those over 50 km were orientated mainly to the SSW/SW if the birds were ringed in winter (Fig 8d). There was some clustering in the opposite direction for birds marked in the breeding season, but the number of recoveries was small and the results were not (quite) statis- tically different from random (Fig 8b). _ Of birds marked in the non-breeding season (November - March) and moving more than 10 km, 61% were in their first year of life compared with 65% for those recovered within 10 km. By contrast of those ringed in the breeding season Origins of Common Starlings wintering in the Highlands 37, (b) Birds ringed December - April (Rayleigh test, z = 16.06, n = 22, P< 0.01) and moving more than 10 km, 88% were first year birds, compared with 59% in _ birds remaining within 10 km. Some 23 of these recoveries involved birds ringed as nestlings, that is were drawn from the local breeding population. Of these, 7 (30%) were found within 10 km, and 16 (70%) beyond this. The dates of handling and recovery of birds remaining in the Highlands but moving at least 10 km are shown in Figs 2d and 2e, but give little indication of any seasonality in the movements. Birds recovered within the same season as they were ringed also show that movements of 10 km or more take place at almost any time between fledging and the following spring. Some birds ringed as nestlings moved within a few months of fledging, indeed in some instances within a few weeks, and included, for example, a bird ringed as a nestling at Dores, Invernessshire which was found 11 weeks later in Thurso, Caithness, 150 km away. 38 SB 24(2) Table 2. Ringing and recovery locations of birds moving from the Northern Isles to the Highlands Ringing location* Recovery location No. of recoveries (%) Fair Isle Caithness pA | (60%)+ Sutherland 2 (6%) Ross & Cromarty l (3%) Inverness-shire l (3%) Orkney Caithness 5 (14%) Sutherland 2 (6%) Inverness shire 3 (9%) * There were no movements from Shetland excluding Fair Isle to the Highlands. + Most of these were birds caught (‘controlled’) as part of our ringing programme. | ‘Half distance’ (d_) is related to the slope (a) of plots such as Fig 4b by d_ = (/n2)/a (cf Boddy & Sellers 1983) and is the distance by which half the birds moving beyond a particular distance will have stopped moving. Thus 50% of birds moving more than 50 km will have stopped within 243 km (50 + 193), 75% within 436 km (243 + 193), 87.5% within 629 km, and so on. 70 60 50 40 % of recoveries 0 5 10 15 20 25 30 35 40 45.50 >50 Distance/km Figure 7. Frequency distribution of distances moved for movements within the Highlands. (Open columns, birds ringed November - March; solid columns, birds ringed May - September) Scottish Birds (2004) Origins of Common Starlings wintering in the Highlands (a) Movements of 10-50 km of birds ringed April - October (Rayleigh test, z = 0.353, n=22, P>0.05, N.S.) & tof Soon O O ef ® o§ O (c) Movements of 10-50 km of birds ringed November - March (Rayleigh test, z = 0.569, n= 28, P> 0.05, N.8.) 30 . (b) Movements over 50 km of birds ringed April - October (Rayleigh test, z = 2.616, Me Oe 0 5..N eS.) (d) Movements over 50 km of birds ringed November - March (Rayleigh test, z = 3.334, n=11,P<0.05) Figure 8. Orientation of movements over 10 km of birds ringed and recovered in the Highlands. | Recovered before or during the next breeding season; ° Recovered after next breeding season. Movements of 10-100 km within the Highlands occurred more or less throughout the winter months, and included an adult male ringed in Wick, Caithness in February and recovered in Fort William, Invernessshire, 217 km away, the following May. We conclude that movements within the Highlands occur more or less throughout the summer, autumn and winter with no obvious seasonality 40 SB 24(2) Discussion Many of the birds ringed in the Highlands during the breeding season, that is, those drawn from the local breeding population, remained within a short distance of the site of ringing. Some, however, do move and these seem dispropor- tionately to involve first year birds. Movements of 10-50 km showed no particular orientation and appear to have taken place at almost any time during the year except spring. Although the evidence is not conclusive, we suspect that these movements are primarily dispersive in nature, reflecting not only post fledging dispersal, but also movements associated with the formation of post fledging flocks in the late spring and summer, and, in autumn and winter, those connected with the establishment of winter roosts and movements between these and daytime feeding areas, which may typically be up to 30 km away (eg Wynne-Edwards 1930, Eastwood et al 1962, Hamilton & Gilbert 1969, Bray et al 1975). The growth in numbers in late autumn/early winter 1s a result of influxes into the Highlands from at least 3 distinct areas. Firstly, and most obviously, there appears to be a major influx from Northern Europe, primarily from Norway, but also from other countries around the Baltic. Most birds arrive in the Highlands in October/November and return in late February and March, though it is evident from the Norwegian ringed bird recovered in Caithness in July that some, possibly sickly, birds may occasionally remain in Britain during the breeding season. A second group of birds arrives in the Highlands from the Northern Isles. Mostly these are birds from Fair Isle. Some may have been caught on passage from Northern Europe, but from the dates of ringing probably the majority originate from Fair Isle itself (and certainly those ringed as nestlings are Fair Isle birds). The precise nature of the movements beginning in Fair Isle is unclear, but we suspect that they are true migrations rather than dispersals. The bird moving from Wick to Fair Isle and back to Wick is consistent with this. It could conceivably have been a continental bird, though the handling date in Fair Isle (mid May) argues against this. The Common Starlings of Shetland are a separate subspecies, S v zetlandicus, and it is difficult to see how such subspecifc differences could be maintained if there were permanent movements in either or both directions. A third group of birds wintering in the Highlands originates from southern Scotland and northern England. Such movements are, however, represented by only 12 recoveries in the dataset available to us, and with such a limited sample it is not possible to reach any definite conclusions about the nature of the movements concerned. Movements from the Highlands to the South are represented by a much larger number of recoveries and have the following principal features: (i) they predominantly involve birds in their first year of life, (41) the movements principally take place in late winter and early spring, (111) the movements are in approximately the same, seasonally inappropriate, direction (S/ESE), (iv) the distances moved are approxi- mately exponentially distributed (with a half distance of around 200 km) and, more tentatively, (v) the number of birds undertaking such movements varies between seasons. There remains the question of where these birds begin their lives. The pattern of movements, described in the previous paragraph mirror some movements described by Rae & Morris (1978) and by Duncan (1984) for birds ringed in the Aberdeen area. Rae & Morris proposed that these were continental birds arriving in Britain Scottish Birds (2004) Origins of Common Starlings wintering in the Highlands 4] across the southern North Sea and then moving north through Britain into NE Scotland, returning by the same route at the end of the winter. Duncan (1984) noted that this was not consistent with the results which had subsequently become available, especially the dates of ringing and recovery, which clearly show the birds to be in northern Scotland in winter and further south in Britain in the spring and summer and where, presumably, they were breeding, and the absence of recoveries on those parts of the continent adjacent to the southern North Sea (Belgium, Netherlands, Germany etc). We agree with these conclusions; the birds which move south from the Highlands and the Aberdeen area are not of continental origin. The only other reasonable options for the origin of these birds are that they start their lives in the Highlands, or that they are birds from northern England and southern Scotland returning there at the end of winter. There is not much to choose between these 2 alternatives. Some (2 of 80 recoveries) definitely originated in_ the Highlands itself, where they were ringed as nestlings. There is a marked disparity between the number of northward movements (12) and those in the opposite direction (80), but we feel that too much should not be read into this as it may simply reflect differences in the finding rate between the Highlands and further south. Some of those going south are certainly Highland birds (where they were ringed as nestlings), but the possibility that a proportion are southern birds returning south cannot be ruled out. Whichever explanation is correct, it is clear that these southward movements are of a highly unusual nature, particularly as regards their orientation and late winter timing, and bear little or no resemblance to the various kinds of movements found in European Common Starling populations (eg Perdeck 1958, Cramp & Perrins 1994). Exponentially distributed and orientated movements have been recorded in a number of species including the Cormorant Phalacrocorax carbo (Coulson & Brazendale 1968), Shag P aristotelis (Potts 1969, Swann & Ramsay 1979), Guillemot Uria aalge (Birkhead 1974), Western Gull Larus occidentalis (Coulter 1975) and Greenfinch Carduelis chloris (Boddy & Sellers 1983), but in all these cases the birds move in summer or early autumn, retrace these journeys the following spring, and appear to repeat these out and back movements annually throughout the rest of their lives. The movements of the Rosy Starling S roseus, especially its periodic spring ‘irruptions’, which appear to be due to food shortages, have some similarities with what we describe here for S$ vulgaris in terms of timing and the long distances involved, but are not sufficiently well characterised to enable a proper comparison to be made (Cramp & Perrins 1994). Acknowledgements A large number of people, many unknown to us, have contributed to the generation of the ringing recoveries on which this study was based and we owe a tremendous debt to them all. In particular we would like to thank Keith Banks, Stuart Mackay and the late Iain Mackay for their help in catching and ringing Common Starlings in Caithness which resulted in about half of the total sample of recoveries available. A listing of all ringing recoveries of Highland Common Starlings was kindly made available through the BTO Ringing Scheme which is supported by the BTO, and the Joint Nature Conservation Committee (on behalf of the Countryside Council for Wales, the Department of the Environment (Northern Ireland), English Nature and Scottish Natural Heritage, the National Parks and Wildlife Service (Ireland), and the ringers themselves. References Bailey N T J 1981. Statistical Methods in 42 Biology. 2nd ed. Hodder and Stoughton, London. Batschelet E 1981. Circular Statistics in Biology. Academic Press, London. Birkhead T R 1974. Movement and mortality rates of British Guillemots. Bird Study 21: 241-254. Boddy M & Sellers R M 1983. Orientated movements by Greenfinches in Southern Britain. Ringing & Migration 4: 129-138. Bray OE, Larsen K H & Mott D F 1975. Winter movements and activities of radio-equipped Starlings. Journal of Wildlife Management 39: 795-801. Coulson J C & Brazendale M G 1968. Movements of Cormorants ringed in the British Isles and evidence of colony-specific dispersal. British Birds 61: 1-21. Coulter M C 1975. Post-breeding movements and mortality in the Western Gull, Larus occidentalis. Condor 77: 243-249. Cramp S & Perrins C M 1994. The Birds of the Western Palearctic, Vol.VIII. Oxford University Press, Oxford. Dorst J 1961. The Migrations Heinemann, London. Duncan R_ 1984. Wintering Starlings in Aberdeen. Grampian Ringing Group Report No.4. 1-11. Eastwood E, Isted GA & Rider G C 1962. Radar ring angels and the roosting behaviour of Starlings. Proceedings of the Royal Society 156B: 242-267. Feare C 1984. The Starling. Oxford University Press, Oxford. Feare C 2002. Common Starling. The Migration Atlas - Movements of the Birds of Britain and Ireland, ed C Wernham, M Toms, J Marchant, J Clark, G Siriwardena and S Baillie. T& A D Poyser, London, pp 629-632. Gibbons D W, Reid J B & Chapman R A 1993. The New Atlas of Breeding Birds in Britain and Ireland: 1988-1991. T & A D Poyser, London. of Birds. SB 24(2) Goodacre M J 1959. The origin of winter visitors to the British Isles. 4. Starling (Sturnus vulgaris). Bird Study 6: 180-192. Hamilton W J & Gilbert W M 1969. Starling dispersal from a winter roost. Ecology 50: 886-898. Lack P 1986. The Atlas of Wintering Birds in Britain and Ireland. T & A D Poyser, Calton. Perdeck A C 1958. Two types of orientation in migrating Starlings, Sturnus vulgaris L., and Chaffinches, Fringilla coelebs L., as revealed by displacement experiments. Ardea 46: 1-37. Potts G R 1969. The influence of eruptive movements, age, population size, and other factors on the survival of the Shag. Journal of Animal Ecology 38: 53-102. Rae R & Morris J 1978. Scandinavian Starlings wintering in Aberdeen. Grampian Ringing Group Report No.2. Svensson L 1992. Identification Guide to European Passerines. Stockholm. Swann R L & Ramsay A D K 1979. An analysis of Shag recoveries from north east Scotland. Ringing & Migration 2: 137-143. Thom V M 1986. Birds in Scotland. T & A D Poyser, Calton. Williams T D 1991. Ageing criteria in the Starling Sturnus vulgaris. Ringing & Migration 12: 113-117. Wynne-Edwards V C 1930. The behaviour of Starlings in winter, Part Hl. British Birds 24: 346-353. H Clark, 3 Lindsay Place, Wick, Caithness, KWI1 4PF R M Sellers, Crag House, Ellerslie Park, Gosforth, Cumbria CA20 IBL Revised manuscript accepted November 2003 Scottish Birds (2003) 24:43-50 SHORT NOTES Sunning behaviour by a fledgling Merlin In the Merlin Falco columbarius the subject of sunning behaviour has_ been — generally overlooked in Britain and the only documented records are of a female/juvenile Merlin standing in a loose spread wing posture in winter and an adult female sun basking and sunning in the spreadeagled position on the ground in summer (Dickson 1998. Merlin’s sunning behaviour in winter. Scottish Birds 19:176; Rollie 1999, Merlin’s sunning behaviour in summer. Scottish Birds 20:-39). On 7 July 1992, from 0915 to 1200 hours on a warm and sunny day, I watched a fledgling male Merlin, aged about 4 weeks old and less than a week out of the nest, standing on a stone wall about 5m from my hide in west Galloway. At 1047 hrs the fledgling yawned, preened his rump, lay flat on the wall facing the sun and covered his eyes with his nictitating membranes, and apparently sun basking. After preening at 1100hrs the young Merlin looked up into the sky, facing the sun, with his nictitating membranes again drawn across his eyes and sat erect, sun basking for 4 minutes. An early record of a Parrot Crossbill in Scotland The Parrot Crossbill Loxia pytyopsittacus is the largest of the crossbills. It is distributed across FennosScandianavia and northern Russia where it is primarily associated with Scots Pines Pinus sylvestris whose seeds are its major food source (Cramp and Perrins, 1994. The Birds of the Western Palearctic, Vol 8, Oxford). During the 19th century, it was clearly regarded as rare Short Notes 43 Merlin Most sunning birds often stare into the sun while sunning (Heinroth & Heinroth 1924-33. Die Vogel Mitteleuropas, Berlin) but it would appear that by drawing its nictitating membranes across its eyes, the young Merlin could have used them as protection from strong sunlight. Like a fledgling Hen Harrer Circus cyaneus (Dickson 2002. Hen Harrier’s sunning behaviour in summer and winter. Scottish Birds 23:48-49) it would seem that the sunning behaviour by a young Merlin at a very early age was an unconditioned (innate) response to sunlight (Simmons 1986, The Sunning Behaviour of Birds, Bristol). R C Dickson, Lismore, New Luce, Newton Stewart, Dumfries and Galloway DG8 0AJ Revised manuscript accepted June 2003 within Scotland, as it is rarely mentioned in the ‘Vertebrate Fauna’ series. Harvie-Brown and Buckley (18897. A Vertebrate Fauna of Sutherland, Caithness, and West Cromarty, Douglas, Edinburgh) give records from a fishing boat off Caithness and at a plantation at Lyth (Caithness). Harvie- Brown (1906. A Fauna of the Tay Basin and Strathmore, Douglas, Edinburgh) also notes collections of Parrot Crossbills at Murthly (see below) and Stanley (Perthshire), and suspected breeding in East 44 Ross. In addition, Gray (1871. Birds of the West of Scotland, including the Outer Hebrides, Murray, Glasgow) records them from Wemyss Bay (Firth of Clyde), Ross-shire and Lochend (Inverness). Once the Scottish Crossbill L. scotica was described (Hartert, 1904. Die Végel der Paldarktitschen Fauna, Freidlander, Berlin) (initially as a subspecies of the Common Crossbill L curvirostra), ornithologists were presented with further problems when identifying Parrot Crossbills in Scotland, given the similarity in plumage and the apparent overlap in size between Scottish and Parrot Crossbills (Knox, 1976. The taxonomic status of the Scottish Crossbill Loxia sp Bull BOC 96: 15- 19). There were, in fact, calls to classify the Scottish Crossbill as a subspecies of the Parrot Crossbill, prior to the former becoming recognised as a species in its own right (Knox, 1975. In Nethersole-Thompson, Pine Crossbills, Poyser, Berkhamsted). As a result, in the first major review of the status of Scotland’s birds, Baxter and Rintoul (1953. The Birds of Scotland. Oliver & Boyd, Edinburgh) found no compelling evidence for Parrot Crossbills occurring in Scotland. They concluded that all records were ‘open to doubt’, but did list the counties in which it had been recorded: Dumfries, Wigtown, Renfrew, Perth, Inverness, Ross and Caithness. With the establishment of coastal bird observa- tories during the early to mid 1900s, it was not long before a migrant Parrot Crossbill was captured. A bird trapped on the Isle of May in 1953 was identified as a Parrot Crossbill. During the following decades, there were further records, on Fair Isle and other northern isles, associated with periodic invasions from northern Europe (Thom, 1986. Birds in Scotland, Poyser, Calton). In addition, there are now inland records from native pinewoods and plantations, including breeding records (Summers, 2002. Parrot Crossbills breeding in SB 24(2) Abernethy Forest, Highland. British Birds 95: 4-11. Marquiss and Rae, 2002. Ecological differentiation in relation to bill size amongst sympatric, genetically undifferentiated crossbills Loxia spp Ibis 144: 494-508). In a review of museum skins of crossbills collected in Scotland, Summers et al (2002. The distribution and habitats of crossbills Loxia spp in Britain, with special reference to the Scottish Crossbill Loxia scotica. Ibis 144: 393-410) found 8 whose bill depths indicated that they could have been Parrot Crossbills. These included 2 19th century specimens in the Perth Museum that the collector had obtained from Murthly and identified them as Parrot Crossbills (Millais, 1884. Notes on the occurrence of the Parrot Crossbill in Perthshire, and probable nesting. Proceedings of the Perthshire Society of Natural Science 1881-18862: 182). However, although the bills of all these museum specimens were within the range of Parrot Crossbills, they were not particularly large, so they could have been large Scottish Crossbills. Therefore, the 1953 record on the Isle of May remains as the first accepted record in Scotland. Since the publication of our review (Summers et al 2002), another specimen has come to light. It refers to an adult male collected by Frank Nisbet near Pitlochry in August 1928 and sold to HF Witherby. At this stage, it was identified as a Scottish Crossbill Loxia curvirostra_ scotica. However, presumably in an exchange with a Dutch collector (PA Hens), the skin made its way to The Netherlands and was renamed Loxia pityopsittacus [sic] scotica. Latterly, the Hens collection was obtained by the National Museum of Natural History, Leiden and the bird (catalogue number 56978) was renamed again, by Kees Roselaar as Loxia p pityopsittacus [sic]. This latest identification was confirmed on measurements. It had a bill depth of 13.9 mm, Scottish Birds (2004) which is at the upper end of the range for Parrot Crossbills (Summers et al 2002). It was, therefore, very unlikely to be a Scottish Crossbill. The bill length (20.0 mm), bill width (13.5 mm), and wing length (105 mm) were also consistent with it being a Parrot Crossbill (Knox 1976). This new observation puts the first record of a Parrot Crossbill in Scotland back 25 years. The fact that it was an adult inland in August suggests that it might have been breeding there, because invasions of this species, unlike Common Crossbill, typically occur in late autumn. The Parrot Crossbill is known for its invasions in mainland Europe, though these are not as large or as and frequent as those of Common Crossbills (Newton, 1972. Finches, Collins, London). Given that Common Crossbill invasions reach Britain, it therefore seems likely that Parrot Crossbills have also been invading Britain for many decades. It is only their small Peregrine Falcon predating Slavonian Grebe On 29 May 2003 Hugh Insley and Pete Mayhew were ringing a brood of 4 Peregrine Falcon Falco peregrinus chicks at a cliff nest site in Inverness- shire. Among the prey remains in the eyrie was part of a bird’s wing which we did not immediately recognise. This wing was grey brown on the upperside and creamy white below. Enough remained of the wing to enable us to measure a maximum chord which was 150mm. The most significant feature was the oily nature and slightly fishy smell of the feathers from which we concluded that it had belonged to a water bird. Apart from the main wing bones, all that remained were some of the coverts, the noticeably curved primaries (11 plus a greatly reduced 12th) and a single white secondary feather. This strongly suggested that the bird was Short Notes 45 numbers, infrequency and similarity to Scottish Crossbills that have made them difficult to detect. However, it is impossible to say whether they have been breeding continuously or only sporadically after invasions. I thank René Dekker of the National Museum of Natural History, Leiden and Bob McGowan of the National Museums for Scotland for arranging the loan of the crossbill from The Netherlands, and Kees Roselaar of Zoological Museum, University of Amsterdam who alerted me to this specimen. Bob Dawson, René Dekker, David Gibbons, David Jardine, Bob McGowan, Kees Roselaar and Jeremy Wilson commented on the draft. RW Summers, RSPB, Etive House, Beechwood Park, Inverness, IV2 3BW Revised manuscript accepted July 2003 a grebe and the wing length indicated that it fell only within the size range for Slavonian Grebe Podiceps auritus. This conclusion’ was subsequently confirmed by comparison with reference material in the National Museums of Scotland by Robert Y McGowan. Feather size, shape and colouration of the wing were fully compatible with Slavonian Grebe skins and the humerus length (76.9mm) matched the mean (77.3mm) for 3 skeletal specimens. No grebes of any species are listed in the analysis of prey taken by British Peregrines given by Ratcliffe (1993 The Peregrine Falcon, London), although Ratcliffe pers comm is of the opinion that Peregrine Falcons will take any bird species within the constraints of size and opportunity. BWP gives prey size for Peregrine Falcon as ranging from Goldcrest Regulus regulus to Grey Heron Ardea cinerea and goose Anser ; weight 46 range 10 to 1800 g, with females taking larger prey items than males. The weights for Slavonian Grebe of 415 to 470g (Iceland and Norway, during summer, BWP) fall well within this range. This appears to be the first time that Slavonian Grebe or a grebe of any species has been recorded being taken as prey by Peregrine Falcon in Britain. Forest nesting Merlin apparently specialising on Barn Swallows As part of a long term study of the breeding ecology of Merlin Falco columbarius in north- east Scotland (Rebecca et al 1992, Scottish Birds SB 24(2) Hugh Insley, Pete Mayhew and Robert Y McGowan c/o RSPB North Scotland Office, Etive House, Beechwood Park, Inverness IV2 3BW Revised manuscript accepted August 2003 16: 165-183; Rebecca & Cosnette 2003, in Birds of Prey in a Changing Environment, chapter 14, eds D B A Thompson, S M Redpath, A H Fielding, M Marquiss & C A Galbraith, HMSO, Edinburgh) prey remains at breeding areas are collected and identified. When more than one Table 1 Bird prey identified at a forest nesting Merlins site in north east Scotland in 1991. Species Adult Barn Swallow Hirundo rustica aD Meadow Pipit Anthus pratensis 1] Chaffinch Fringilla coelebs 9 Common Linnet Carduelis cannabina 7 Pied Wagtail Motacilla alba 6 Goldcrest Regulas regulas 5 Willow Warbler Phylloscopus trochilus + Eurasian Siskin Carduelis spinus 3 Yellowhammer Emberiza citrinella 3 Common Starling Sturnus vulgaris pe Northern Wheatear Oenanthe oenanthe Sky Lark Alauda arvensis Whinchat Saxicola rubetra House Sparrow Passer domesticus Hedge Accentor Prunela modularis European Robin Erithacus rubecula Great Tit Parus major Coal Tit Parus ater Blue Tit Paris caeruleus Long-tailed Tit Aegithalos caudatus Oe ee ee ee ee ee ee es Juvenile Fledgling Total 16 a! 7 18 9 i 6 5 4 3 3 D I 1 l 1 1 ] l | 1 1 Total 118 Note: 22 Northern Eggar moths Lasiocampa quercus were also found. Short Notes 47 Scottish Birds (2004) n ” SOR a = ” oO 10 © Ba. 3. Soe = —- OF gee ——, (oe) ea af oR ! Noy & Sy 2 uk ee = to wm S/S Se oo S i —= — OTI-1OI ‘Le Apmis Pg ‘0661 TIO % Sua 817-607 ‘SE APMIS PALIT “8861 JOP 76-88 :81 SPA1G YSUIOIS “C66| 12 12 SopIsIAvaH] 8ST-EST :9T APIS PAT “6L6T UOSIEM OPI-E€I :6L JUaumasvuvpy XP ASOJOIT ISAO] ‘SH6] 1D 12 ANdd CLO PUBPOIS MS 76-1661 PLIQUINYLON ‘IS210,J JOp[aryy I9T /0 088/17 a0) v0 oo 9S-6r :1E APMIS PAlg ‘p86] 1D 12 UOIMON Z8-PL6l PLIQuINYyVON 100c /8 99¢9 / OI P8-T86I “Sele 68-CL6I ‘SelM AN OL-79 ‘PE Apis pllg “L861 Sqqtg 801-88 °C SPL YS]2M “6661 SAUOL 29 SIqGOY SPOT / OS suia}t Avid [v0], / uleg % MOT[BPMS Weg JO ‘ON MOT[EMS sivak pue svarly I0U9.19JOY “UIVJLIG Ul SIIPN)S JIIP IAISUI}X9 WOAJ SUID, SUIPIe1g Aq Udye} SMOT[VMS Ueg JO JaquINU ayy, Z aqeL individual of the same species is found at a plucking place distinct primary or tail feathers are counted giving a minimal number for that species. In 1991, I located a successful repeat breeding attempt within Fetterresso Forest. This was only the second Merlin nest located in a mature conifer plantation in north east Scotland (Rebecca 1992, North-East Scotland Bird Report 1991: 61-62, Aberdeen). The site was an old Carrion Crow Corvus corone nest, about 12 m from the ground, on the edge of an unthinned 38 year old Sitka Spruce Picea sitchensis block bordering a 7 year old Lodgepole Pine Pinus contorta plantation. The first breeding attempt was approximately 500 m away in the pine plantation. After the 2 young fledged in mid July prey remains were collected at plucking places up to 150 m from the nest. A total of 118 birds were identified including 35 adult and 16 juvenile Barn Swallow Hirundo rustica (43%). This was an exceptionally high number of Barn Swallow in comparison to prey identified at this breeding area in 1980, 1983, 1984, 1986-88 and 1990 when only one was found in 232 items (unpublished data). The remainder of the prey items in 1991 were typical for breeding Merlins (Table 1). In Britain, Barn Swallows usually represent a small proportion of the breeding season diet of Merlin (Table 2) and in north east Scotland they are normally found in similarly low numbers annually (unpublished data). It is interesting to reason where these Barn Swallows might have been taken. There was no known or potential Barn Swallow nest sites within a 2 km radius of the Merlin nest and the main habitats in that area were coniferous plantations; estimated at 50% aged 5-10 years and 40% aged at least 35 years, with the remaining 10% unplanted Ling heather Calluna vulgaris. At 2-4 km radius there was a more complex habitat mix of various aged coniferous 48 plantations, heather moor, rough hill grazing and intensive farmland, with 8 farms as potential Barn Swallow nesting sites. The Merlin using this breeding area in 1984 and 1988 took unfledged waders on rough farmland at 3.8 km from their nests (Rebecca et al 1990, Scottish Birds 16: 38-39) and Merlin in Wales, tracked by radio telemetry, were known to hunt up to 4 km from the nest (C J Bibby pers comm). Some of these Barn Swallows may have been caught near the farms but it is unlikely that the majority were. Other typical farmyard species such as Common Starling and House Sparrow only featured in small numbers (Table 1) and breeding Merlin are not usually associated with occupied farms or intensive farmland (BWP2). Ring Ouzel killing and feeding shrew to nestlings As part of a study of the breeding ecology of Ring Ouzels Turdus torquatus in north east Scotland nests are located, and nestlings and adults are metal and colour ringed (Sim et al 2003, Scottish Bird News 69: 8-9). The study area, in Glen Clunie upper Deeside Aberdeenshire, holds a stable and relatively high density population of Ring Ouzels (Rebecca 2001, Scottish Birds 22: 9- 19). In 2003 detailed nest and foraging watches were carried out as part of an environmental science MSc degree (Prigmore J, 2003, The foraging behaviour of breeding Ring Ouzels at Glen Clunie, north east Scotland. Unpublished dissertation, University of Aberdeen). On 5 June 2003, we were watching a nest with 4 x 7 day old nestlings situated on a 4 m high crag on a tributary of the River Clunie. A Ring Ouzel was alarm calling which we initially attributed to our movements. However, the female of the pair was still alarm calling when we were about 60 m from the nest, well outside the normal range for SB 24(2) An alternative, and possibly more realistic, explanation is that they were caught as they foraged over the forest canopy. Hirundines are commonly seen hawking insects’ over Fetterresso Forest and adjoining coniferous plantations, and perhaps 1991 was a particularly good year for insects and Barn Swallows. I thank the Forestry Commission and Fountain Forestry for access to study the Merlins and Mick Canham for the details of the Sitka Spruce plantation Graham W Rebecca, RSPB, 10 Albyn Terrace, Aberdeen, AB10 1YP Revised manuscript accepted October 2003 this behaviour. Simultaneously, we watched the female, and set up a tripod and telescope (20-40 X ZOOM). The female was seen attacking something on the ground about 10 m from the nest. It repeatedly pecked and then jumped back quickly for about 4 to 5 minutes in an area of about one m7. It had obviously killed the prey as it then held it motionless in its beak. The female proceeded to peck and bash the prey from side to side for 2 to 3 minutes. By this stage we had identified the prey as a shrew, later judged by size and colour to have probably been a Pygmy Shrew Sorex minutus (Corbet, G B & Southern H N 1977, The Handbook of British Mammals, Blackwell, Oxford). The female then stood with the shrew in her beak for about a minute, flew to the nest and was seen, through the telescope, to feed the nestlings. When the female was alarm calling and killing the shrew the male perched nearby and appeared calm and quiet. However, when the female had the dead shrew in her beak, the male’s behaviour changed. He became very agitated, and flew around alarm calling. Scottish Birds (2004) Short Notes 49 The breeding season diet of Ring Ouzels is reasonably well known consisting almost entirely of insects, earthworms and small snails or slugs, with small lizards sometimes recorded; at other times their diet is mainly berries (BWP, volume 5). This would appear to be the first documented record of a Ring Ouzel preying on a mammal. We could not determine which species began the encounter and the Ring Ouzel may have reacted in defence. In 2003 we witnessed Ring Ouzels successfully defend their nests European Robin feeding on dead road casualties On 13 December 1975, a very cold day with hard frost, I saw a European Robin Erithacus rubecula peck at the exposed flesh of a Brown Hare Lepus europaeus carcase that was lying on a minor road on low ground in west Galloway for over 5 minutes. On 2 January 1977, a very cold frosty day, on the same road, I again saw a Robin pecking at the exposed flesh of a dead female Common Blackbird Turdus merula. Robins’ food in hard weather is known to Soaring behaviour by Merlins According to Cramp & Simmons (1980, The Birds of the Western Palearctic, Vol 2, Oxford) the Merlin Falco columbarius soars infrequently while Feldsine & Oliphant (1985, Raptor Research 19:60-67) considered that high circling/soaring was a mild territorial display by F c richardsonii in an urban population in Canada. During studies of Merlins in Galloway, I recorded soaring flights, on set wings, in summer, in winter and at communal winter roosts. Most were associated with birds and animals, including man, and all were made at against Stoats Mustela erminea by distraction, on 3 out of 4 occasions. We thank Invercauld Estate for cooperation with the study. Justin Prigmore & Carina Prigmore, 9 Kennerty Court, Peterculter, Aberdeen, AB14 OLU Graham W Rebecca, RSPB, 10 Albyn Terrace, Aberdeen, AB1I0 1YP Revised manuscript accepted November 2003 include carrion and once, meat in a butcher’s ship (Cramp 1988, The Birds of the Western Palearctic, vol 5, Oxford), and there is a record of one feeding on a Eurasian Collared Dove Streptopelia decaocto at a frozen pond (Waters 1990, British Birds 83:123). This note appears to be the first record of Robins feeding on road casualties. R C Dickson, Lismore, New Luce, Newton Stewart, Dumfries & Galloway DG8 0AJ Revised manuscript accepted January 2004 heights up to 150+m and usually initiated from a perch. My data suggest that aggressive soaring behaviour by Merlins may not be _ that uncommon. Summer Between 1973 and 1997, of the 23 occasions on which I observed Merlins soaring over breeding areas, 10 were by males, 11 by females and 2 by young fledglings. Four soaring flights by males and one by a female occurred above humans and a dog that had intruded into their nest areas, although one person was .09km away walking up a forestry track. A male soared above his nest 50 on a steep cliff face after a pair attacked a Golden Eagle Aguila chrysaetos; another swooped on a Common Wood Pigeon Columba palumbus then soared to 50m; another chased a non resident female Merlin away from his breeding area then soared to 100m. A male soared over moorland and attacked a female Hen Harrier Circus cyaneus and a Peregrine Falcon Falco peregrinus, and thereafter carried out a series of ‘switch backing’ flights. A female soared above a Peregrine Falcon and Common Kestrel Falco tinnunculus, and another soared above a Carrion Crow Corvus corone; another soared above a Short-eared Owl Asio flammeus and stooped on it; another soared to 50m directly above a Hen Harrier’s nest; another soared above a loch and called. On 2 occasions after food passes by pairs they soared together to 70m and 100m. Another male soared after a nest relief and on 3 occasions a female soared above her nest during reliefs (Dickson 1995, Scottish Birds 18:20-23). On 24 July 1992 2 young fledglings rose from a moor and soared together to a height of about 150m, sometimes touching, departed their natal area for the last time and landed about 3-4 km away; another fledgling soared above its nesting wood and landed on a stone wall. The only recent documentation of soaring in the breeding season was that by Orton (1980, The Merlins of the Welsh Marches. London) who recorded a male soaring and stooped closely on 2 sheep. Winter Between 1970 and 1997 6 soaring flights by Merlins were observed in winter in west Galloway, involving birds, 3 by males and 3 by females/juveniles at heights of 20m-70m. A male soared above another hunting male; another male soared repeatedly above moorland attacking 2 Carrion Crows for 20 minutes (Dickson 1991, Country-Side 27:14-15); another soared 70m above hunting a male Common Kestrel. A female/juvenile Merlin soared above SB 24(2) a female/juvenile Common Kestrel and stooped on the other falcon; another female, near a breeding area, soared above 2 Common Buzzards Buteo buteo, one _ Eurasian Sparrowhawk Accipiter nisus and a Common Kestrel over moorland; another female/juvenile soared with a Peregrine Falcon, 2 Sparrowhawks and a female/juvenile Common Kestrel over moorland. Winter roosts On 4 occasions soaring flights were observed at a communal roost, twice by female/juveniles before going to roost in sallows; another female/juvenile rose from sallows at midday and began to soar before swooping down out of sight. A male and 2 females/juveniles soared together and were joined later by a soaring male Hen Harrier before roosting (Dickson 1973, Scottish Birds 7:288-292). R C Dickson, Lismore, New Luce, Newton Stewart, Dumfries & Galloway DG8 9AJ. Revised manuscript accepted January 2004 Scottish Birds (2003) 24:51-52 Obituaries 5] OBITUARIES Gordon Booth (1905-2003) Gordon Booth, well known to birdwatchers visiting Islay in the 1970s and 1980s, died in November 2003. After a lifetime working for a family textile firm in Yorkshire, he and his wife, Phyllis, retired in the mid 1960s to a lovely house in the Yorkshire Dales. However, in winter 1967-68, Gordon accompanied me on a goose counting visit to Islay. He was so struck by the island that he took Phyllis there in the summer of 1968 and within a year the house in Yorkshire had been sold and he and Phyllis had moved to Islay. Here they stayed until the end of 1985 when they moved again, to Comrie, to be near their 2 daughters for what, they fully expected, would be the last few years of their lives, only for both to survive well into their 90s, Phyllis dying in July 2003 aged 95 and Gordon reaching 97, both thankfully retaining all their faculties until the end. Gordon’s interest in birds and other wildlife soon led him to become bird recorder for Islay and, being a very well organised person, he began to collect and collate all the bird records for the island and to systematically transfer them to an extensive card index file. At the same time, he recorded his own observations in meticulous detail. This culminated in the publication of his Birds in Islay in 1975, with a second edition in 1981. Gordon was always pleased to be called on by visiting birdwatchers and was both interested in what they saw and ready with advice. He was a member of the SOC for over 30 years and attended a number of annual conferences. When the Islay Natural History Trust for founded in 1984, Gordon, with typical generosity, passed over not just his invaluable card index, but a great many books, papers and other material which formed the basis of both the Trust’s database of natural history records Gordon Booth and its library. Birds were not Gordon’s only interest on the island, as he was largely instru- mental in the creation of the very successful Museum of Islay Life in Port Charlotte. Gordon retained his interest in Islay and in birds right to the last, and each time I called to see them both in Comrie he would present me with magazines, journals and cuttings that he thought I should have for the libraries of either the Natural History Trust or the Museum. Malcolm Ogilvie Peter Webster (1950-2003) Many SOC members will be sad to learn of the death at the end of December of Peter Webster, RSPB Scotland’s Reserve manager at Loch of Strathbeg since 1994. During his career with RSPB, Peter worked efficiently and with great enthusiasm at the reserve improving its bird habitats and _ visitor facilities plus the infrastructure. At the same time he was always aware of the importance of high standards of health and safety. D2 In his 10 years at Loch of Strathbeg he oversaw the introduction of a very successful arable farming regime that was of great benefit to seed- eating passerines, improving wetlands and reedbeds, creating tern islands, a new hide and walkway and renovating the visitor centre. He was fiercely proud of “his” reserve and worked hard to ensure that it remained the premier coastal bird site in eastern Scotland. Peter will be greatly missed by many but none more than his wife Kate, son Peter and daughter- in-law Sarah to whom we offer our sincere condolences for their loss. Tan Francis Advice to contributors Authors should bear in mind that only a small proportion of the Scottish Birds readership are scientists and should aim to present their material concisely, interestingly and clearly. Unfamiliar technical terms and symbols should be avoided wherever possible and, if deemed essential, should be explained. Supporting statistics should be kept to a minimum. All papers and short notes are accepted on the understanding that they have not been offered for publication elsewhere and that they will be subject to editing. Papers will be acknowledged on receipt and are normally reviewed by at least 2 members of the editorial panel and, in most cases, also by an independent referee. They will normally be published in order of acceptance of fully revised manuscripts. The editor will be happy to advise authors on the preparation of papers. Reference should be made to the most recent issues Of Scottish Birds for guidance on style of presentation, use of capitals, form of references, etc. Papers should be typed on one side of the paper only, double spaced and with wide margins SB 24(2) and of good quality; 2 copies are required and the author should also retain one. We are also happy to accept papers on disk or by email at: mail @the-soc.org.uk, stating the type of word processing package used. If at all possible please use Microsoft Word . Contact the Admin Officer on 0131 653 0653 for further information. Headings should not be underlined, nor typed entirely in capitals. Scientific names in italics should normally follow the first text reference to each species unless all can be incorporated into a table. Names of birds should follow the official Scottish List (Scottish Birds 2001 Vol 22:33-49). Only single quotation marks should be used throughout. Numbers should be written as numerals except for one and the start of sentences. Avoid hyphens except where essential eg in bird names. Dates should be written: ...on 5 August 1991...but not ...on the 5th... (if the name of the month does not follow). Please do not use headers, footers and page numbers. Please note that papers shorter than c700 words will normally be treated as short notes, where all references should be incorporated into the text, and not listed at the end, as in full papers. Tables, maps and diagrams should be designed to fit either a single column or the full page width. Tables should be self explanatory and headings should be kept as simple as possible, with footnotes used to provide extra details where necessary. Each table, graph or map should be on a separate sheet, and if on disc each table, graph, map etc should be on a separate document. Please do not insert tables, graphs and maps in the same document as the text. Maps and diagrams should be either good quality computer print out and in black and white (please do not use greyscale shading) or drawn in black ink , but suitable for reduction from their original size. Contact the Admin Officer on 0131 653 0653 for further details of how best to lay out tables, graphs, maps etc. The Scottish Ornithologists’ Club was formed in 1936 to encourage all aspects of ornithology in Scotland. It has local branches which meet in Aberdeen, Ayr, the Borders, Dumfries, Dundee, Edinburgh, Glasgow, Inverness, New Galloway, Orkney, St Andrews, Stirling, Stranraer and Thurso, each with its own programme of field meetings and winter lectures. The Waterston Library at the Club’s headquarters at Harbour Point, Newhailes Road, Musselburgh, EH21 6SJ is the most comprehensive ornithological library in Scotland and is available for reference during office hours (Monday to Friday 0930hrs - 1630hrs - but please phone beforehand). A selection of Scottish local bird reports is held at headquarters and may be purchased by mail order. Check out our website for more information about the SOC and other bird related organisations: www.the-soc.org.uk An annual issue of Scottish Birds is published each year, in June. Scottish Bird News is published quarterly in March, June, September and December and the Scottish Bird Report is annual. The Scottish Ornithologists’ Club publishes the annual Raptor Round Up on behalf of the Scottish Raptor Study Groups with grant aid from Scottish Natural Heritage. It is sent to all members. Bird names used in publications follow those used in the Scottish List (Scottish Birds Vol 22(1): 33-49). Copies of these reports may be purchased by non members on application to the SOC. SOC annual membership subscription rates Adult £ 25.00 Family (2 adults and any juniors living at the same address) £ 35.00 Junior (aged 17 or under) £ 10.00 Student (aged 18 to 24 and in full-time education) £ 10.00 Concession (Pensioner) £ 16.00 Concession (Unwaged) £ 16.00 Joint Concession £ 21,00 Life £500.00 Life Family £700.00 Overseas supplement £ 10.00 All subscriptions may be paid by Direct Debit and Gift Aided. Subscriptions paid by Direct Debit greatly assist the Club. Please ask for an application form by telephoning the Admin Officer at the above address or visiting the clubs website at www.the-soc.org.uk Scottish Birds Volume 24 June 2004 STITUTION LIBRARIES iii 01075 4505 Contents Main Papers Canoe and walking surveys of wintering Goosanders, Red-breasted Mergansers, Great Cormorants and Common Goldeneyes on the River Spey, 1994-2003 - P J Cosgrove, JRA Butler & R L Laughton 1 The distribution of the Scottish Crossbill, 1995-2003 - RW Summers, D C Jardine & R J G Dawson 11 Estimating the breeding wader populations of Scottish uplands and salt marshes - M O’Brien & C S White Wi, Origins and movements of Common Starlings wintering in the Highlands - H Clark & RM Sellers 29 Short Notes Sunning behaviour by a fledgling Merlin - R C Dickson 43 An early record of a Parrot Crossbill in Scotland - R W Summers 43 Peregrine Falcon predating Slavonian Grebe - H Insley, P Mayhew & R Y McGowan 45 Forest nesting Merlin apparently specialising on Barn Swallows - G Rebecca 46 Ring Ouzel killing and feeding shrew to nestlings - J & C Prigmore & G Rebecca 48 European Robin feeding on dead road casualties - R C Dickson 49 Soaring behaviour by Merlins - R C Dickson 49 Obituaries Gordon Booth (1905-2003) - Malcolm Ogilvie Si Peter Webster (1950—2003) - Jan Francis Si Advice to contributors a7 Front Cover Goosander Edmund Fellowes Published by the Scottish Ornithologists’ Club, Harbour Point, Newhailes Road, Musselburgh, EH21 6SJ. © 2004 q SEE The 2002 census of the Mute Swan in Scotland The eradication of Brown Rats from Handa Island Eagle behaviour before and after construction of a windfarm The biology of a population of Willow Warblers in East Fife ipa Rip ia Scottish Birds — The Journal of the SOC Editor: Dr S da Prato. Assisted by: Dr I Bainbridge, Professor D Jenkins, Dr M Marquiss, Dr J B Nelson and R Swann. Business Editor: Kate Walshaw, Admin Officer, SOC, The Scottish Birdwatching Resource Centre, Waterston House, Aberlady, East Lothian EH32 OPY. Scottish Birds, the official journal of the SOC, publishes original material relating to ornithology in Scotland. Papers and notes should be sent to The Editor, Scottish Birds, SOC, The Scottish Birdwatching Resource Centre, Waterston House, Aberlady, East Lothian EH32 OPY. Scottish Birds is published each year in June. Scottish Birds is issued free to members of the SOC, who also receive the quarterly newsletter Scottish Bird News, the annual Scottish Bird Report and the annual Raptor Round Up. Published by: The Scottish Ornithologists’ Club, The Scottish Birdwatching Resource Centre, Waterston House, Aberlady, East Lothian EH32 OPY. Design by: Pica Design, 51 Charlton Crescent, Aboyne, Aberdeenshire AB34 SGN. Printed by: Meigle Colour Printers Ltd, Block 11, Units 1 & 2, Tweedbank Industrial Estate, Galashiels TD1 3RS Scottish Birds (2005) 25: 1-16 The 2002 census of the Mute Swan in Scotland | The 2002 census of the Mute Swan in Scotland ALLAN W BROWN & LYNDESAY M BROWN A total of 7,028 swans was recorded in Scotland during the spring of 2002, including 1,012 pairs which bred. The total population had increased by 41% since 1990 but this varied between the north, centre and south of the country. Whilst the majority of birds occurred on still waters, use was made also of rivers, canals and marine shores. The mean altitude at which pairs bred increased from 45m in 1983 to 64m in 2002. Although a few pairs bred at an altitude in excess of 300m in the Borders, 63% of pairs bred at an altitude of 60m or lower. Six flocks which comprised 100 or more non territorial swans were recorded, the largest being 302 on the Outer Hebrides. Introduction Almost half a century has passed since the first national census of the Mute Swan Cygnus olor was undertaken in Scotland and across Britain in 1955-56 (Rawcliffe 1958, Campbell 1960). Subsequent censuses were held in Scotland in 1978 (Ogilvie 1981), in 1983 (Brown & Brown 1985, Ogilvie 1986), in 1990 (Delany et al 1992, Brown & Brown 1993) and the fifth national census was held in 2002. The purpose of the 2002 census was to enable the current size of the population to be quantified, but additionally to enable any changes in its structure and distribution to be identified and long term trends to be monitored. Initially the census was to be held in the spring of 2001. However, due to the widespread ban on public access to the countryside as a consequence of the Foot and Mouth epidemic amongst farm animals, a decision to postpone the census was taken in February 2001 with it being deferred until spring 2002. As a signatory to international conservation conventions Britain is legally bound to conserve waterfowl, including the Mute Swan, and their habitats. In order to meet those obligations regular species monitoring is necessary and consequently censuses contribute in part to the ongoing, and wider, programmes of waterfowl monitoring in Britain (Pollitt et al 2003). There had been an increase in the population in some areas of Scotland (Murray et al 1996, Brown & Brown 2002) since the last census held in 1990 and an increase had been identified in the winter population (Pollitt et al 2003). Therefore, an update of the 1990 results was deemed necessary to facilitate a review of the national threshold for the Mute Swan which is 1% of the British population. Within a Scottish context a decrease occurred in the number of swans in Scotland between 1955- 56 and 1983 and an increase between 1983 and 1990 (Rawcliffe 1958, Brown & Brown 1985, Brown & Brown 1993). This census was undertaken specifically to determine both recent and long term trends in the population by determining the size of the population with regard to the numbers of territorial and breeding pairs, and the number of non_ breeding individuals. Moreover, expansion or contraction in specific regions would be investigated in addition to any variation in habitats occupied. Methods The census was organised jointly by The Wildfowl and Wetlands Trust, the British Trust for Ornithology, the Scottish Ornithologists’ Club and the Swan Study Group, while the authors coordinated the census in Scotland. As with previous censuses, a team of local organisers coordinated local field workers who voluntarily undertook fieldwork primarily during April and May. A sample of randomly selected 10km squares of the Ordnance Survey national grid for which coverage was essential was allocated to each local organiser. However, in Scotland most local organisers proved extremely supportive of the project and agreed to obtain full coverage in their respective areas to enable a full census of the species to be achieved once again in Scotland. Field workers were requested to locate territorial and breeding pairs of swans, to count all non territorial birds and record the grid reference of each site. Repeat visits to determine whether or not territorial pairs actually nested were encouraged. Whilst non territorial birds were normally counted in mid April some flexibility was necessary due to the relatively short notice regarding commencement of the census following its postponement from the previous year. Additionally, the difficulty in obtaining coverage of all sites at the appropriate time in more remote areas and additional commitments of fieldworkers also necessitated some flexibility. Observers were requested to identify the type of habitat occupied by swans. This together with the grid reference for each site meant that it was possible to categorise the wetland habitat occupied by swans as canal, river, still water or marine open shore and to estimate their altitude from Ordnance Survey maps scale 1:25,000. In order to facilitate comparison of 2002 results with those for 1983 (Brown & Brown 1985), the habitat occupied by territorial and breeding pairs in 1983 was retrospectively reanalysed and categorised as canal, river, still water or marine open shore and the altitudes determined also. To enable further direct comparison of results with those from previous Scottish censuses, the data were grouped according to the old county boundary system. SB 25 For analytical purposes the country was divided into north, centre and south of the country equating, as far as county boundaries allowed, generally with the Highlands and Islands, Central Lowlands and Southern Uplands. The north comprised Orkney, Outer Hebrides, Shetland, Sutherland, Ross and Cromarty, Caithness, Kincardineshire, Angus, Inverness shire, Nairn, Morayshire, Banffshire, Aberdeenshire, Perthshire and Argyll and islands; the centre comprised Stirlingshire, Clackmannan, Kinross, Fife, Midlothian, West Lothian, East Lothian, Arran and Bute and Cumbrae, Dunbartonshire, Renfrewshire and Lanarkshire; the south comprised Ayrshire, Peeblesshire, Roxburghshire, Selkirkshire, Berwickshire, Dumfriesshire, Kirkcudbrightshire and Wigtownshire. Data analyses were undertaken using Microsoft Excel. Due to the skewed distribution of altitude data it was analysed using the heteroscedastic Type 3 t test (Dytham 1999). Results Fieldwork was undertaken in all counties during the census period. Coverage was generally good and for most counties it was considered to be complete. Many local organisers provided an assessment of coverage within their area and their comments are summarised in Appendix 1. Spring 2002 was rather wet and high water levels may have impacted on the number of territorial pairs which progressed to nest but should not have affected the total number of swans observed. Total number of swans The total number of swans recorded in Scotland in 2002 was 7,028, a 41% increase since 1990 (Table 1). Notwithstanding, a decrease occurred in 11 counties, whilst an increase occurred in 22, with Kirkcudbrightshire being the only county in which no change was recorded. Although the Scottish Birds (2005) north held a total of 3,186 swans or 45% of the Scottish population (Table 2) a 2% decrease occurred there between 1990 and 2002 (Table 3). In contrast a substantial increase occurred in the centre with the total population increasing by 210% to reach 2,188 in 2002. Similarly, in the south the total population increased by 57% and 1,654 swans were recorded there in 2002. The number of swans recorded in Scotland during the last century was particularly low in 1983, however, numbers recovered by 1990 and subsequently continued to increased to reach their highest recorded total by 2002 (Figure 1). The total population comprised territorial and non territorial swans and as the total number of swans increased from 1983 so the percentage of the population which was territorial decreased from 49% in 1983, to 43% in 1990 and to 39% in 2002. Territorial pairs A total of 1,375 pairs held a territory in 2002, a 29% increase since 1990. The total number of territorial pairs comprised pairs which held a territory but did not breed and pairs which held a territory and bred (ie built a substantial nest and probably laid eggs, or cygnets were observed) (Table 1). A 53% increase was recorded between 1990 and 2002 amongst the number of pairs which held a territory but did not breed, but as a percentage of the total population this remained little changed at 10% compared with 10% in 1990 and 11% in 1983. Whilst an increase of 7% occurred in the number of swans which only held a territory in the north between 1990 and 2002, this was low when compared with 157% in the centre and 111% in the south (Table 3). Although the number of pairs which only held a territory decreased from 43 pairs in 1990 to 19 pairs in 2002 on the Outer Hebrides, the most westerly pair recorded in Scotland was found there on the Monach Isles in 2002. The 2002 census of the Mute Swan in Scotland 3 During the censuses conducted in 1955-56 and 1978 there was no attempt to disaggregate the swans which did not breed into those which did or did not hold a territory. Consequently, the number of pairs which held a territory but did not breed was available only from 1983. Figure 2 illustrates the increase since 1983 in the number of pairs which held a territory but did not breed. Pairs which bred The breeding population increased by 22% between 1990 and 2002, reaching 1,012 pairs in 2002, the first time that over a thousand pairs has been recorded in Scotland (Figure 2). Whereas no pairs bred in Shetland in 1990, 7 were recorded in 2002 and yet numbers in neighbouring Orkney declined from 163 pairs to 126 pairs during the same period. Scotland’s most northerly breeding pair set up home on the Loch of Benston on Mainland, Shetland. The islands of Arran, Bute and Cumbrae held no pairs in 1990 but 16 pairs bred in 2002. However, the greatest change occurred in Fife with an increase from 15 pairs in 1990 to 66 pairs in 2002. Amongst the 34 Scottish counties an increase occurred in 22, a decrease in 10 with no change being recorded in only 2. Whilst the number of breeding pairs decreased by 10% in the north, a substantial increase occurred in both the centre and in the south (Table 3). Since 1983 both the total population and number of pairs which bred increased in number but the percentage of the total population which bred decreased and at 29% in 2002 was similar to that in 1955-6 (Figure 3). Non territorial birds Non territorial swans occurred singly and in flocks and comprised birds which may have been too young or too old to breed, or birds which had not repaired following the loss of a mate. A total of 4,280 was counted in 2002, an increase of 50% since 1990. The number of non territorial swans increased between 1983 and 1990 and again between 1990 and 2002 4 SB 25 Table 1 Counts of Mute Swans in each old Scottish county in 2002. Old No of pairs which Noof Noofnon Total % Change in total County held a territory pairs territorial number number of swans but did not breed whichbred swans’ ofswans_ from 1990-2002 Shetland 1 7 18 34 330 Orkney 69 126 at 701 -13 Outer Hebrides 19 93 404 628 -24 Caithness 2 16 90 126 -25 Sutherland () I 0 2 -88 Ross & Cromarty 4 55 184 302 13 Inverness shire 6 5 40 62 -32 Nairn l 4 33 43 258 Moray 6 15 36 78 22 Aberdeenshire 18 36 152 260 -20 Banffshire 0 0 0 0 -100 Kincardineshire 2 4 l 13 63 Angus 1] pa 235 JL 88 Perthshire 16 ae) 323 463 94 Stirlingshire 8 39 a7 151 251 Clackmannanshire 0 ls) 28 58 480 Kinross 3 9 2135 243 406 Fife 19 66 120 290 174 West Lothian 8 3] 167 245 240 Midlothian 6 2) 228 294 QY7 East Lothian 3 22 40 90 -28 Ayrshire 10 34 bS2 240 45 Arran & Bute & Cumbrae 5 16 39 81 913 Peeblesshire 8 7 6 36 260 Berwickshire 13 26 102 180 62 Roxburghshire 36 20 22 382 94 Selkirkshire 3 [3 8 40 -11 Argyll and Islands 6 Zi 109 163 -29 Dunbartonshire 12 28 186 266 171 Renfrewshire 14 32 49 141 244 Lanarkshire pa 7] 480 676 271 Dumfriesshire 5 26 51 143 -19 Kirkcudbrightshire 14 42 103 215 0 Wigtownshire 6 14 61 101 58 Total 363 1012 4280 7028 % of Total population in 2002 10% 29% 61% % of Total population in 1990 10% 33% 57% % of Total population in 1983 11% 38% 51% % Increase between 1990-2002 53% 22% 50% 41% Scottish Birds (2005) The 2002 census of the Mute Swan in Scotland 5 Table 2 Counts of Mute Swans in the north, centre and south of Scotland in 2002. North Centre South Number % of Number % of Number % of Scottish Total Scottish Total Scottish Total Pairs which held a territory but did not breed 161 44 90 25 1Lt2 31 Pairs which bred 465 46 312 31 239 25 Non territorial individuals 1936 45 1384 32 960 ps6 Total swans 3186 45 2188 31 1654 24 Table 3 Percentage change in the numbers of Mute Swans in Scotland between 1990 and 2002. Area North Centre South Scotland Pairs which held a territory but did not breed f) bs] me ae) Pairs which bred -10 131 34 22 Non territorial individuals 2 279 62 50 Total swans ~2 210 a7 41 8000 # 6000 Cc i) > (Fed © 4000 @ 2 = = = 2000 0 1955-56 1978 1983 1990 2002 Year Figure 1. Total numbers of Mute Swans recorded in Scotland during national censuses held between 1955-66 and 2002. 1200 1000 800 600 400 Number of pairs 200 oO 1955-56 1978 SB 25 1983 Year 1990 2002 Pairs which held a territory only m Pairs which bred Figure 2 The number of pairs of Mute Swans recorded in Scotland during national censuses held between 1955-56 and 2002. 50 40 30 “fo 20 10 1955-56 1978 1983 Year 2002 1990 Figure 3 Percentage of the total population of Mute Swans which bred in Scotland between 1955-56 and 2002. (Figure 4). The percentage of the total population which was non territorial increased from 51% in 1983 to 57% in 1990 and 61% in 2002. At the county level an increase occurred in 23 counties and a decrease in 11 with notable changes occurring in Kinrosshire where numbers increased from 16 in 1990 to 215 in 2002, in Lanarkshire up from 112 to 480 and the Outer Hebrides down from 583 to 404. Only a 2% increase in numbers occurred between 1993 and 2002 in the north but increases were substantially greater in both the centre and in the south (Table 3). The most northerly swans recorded in Scotland were located on Kirkhouse Loch, Mainland, Shetland where 3 non territorial individuals were sighted. Scottish Birds (2005) Number of swans w a) oS o The 2002 census of the Mute Swan in Scotland 7 1955-56 1978 1983 2002 Year 1990 @ Non breeding swans” Non territorial swans Figure 4 Numbers of non breeding and non territorial Mute Swans in Scotland between 1955-56 and 2002. A total of 27 flocks which comprised 30 or more swans were located and these accounted for 50% of the non territorial population (Table 4). Whilst 11 of these flocks occurred in the north and 11 in the centre, only 5 occurred in the south. Additionally, of the 6 flocks comprising over 100 birds 3 occurred in the north and 3 in the centre but none in the south. The largest recorded flock of 302 was recorded on Loch Bee in the Outer Hebrides. Between 1990 and 2002 there was a small increase in the number of flocks which held at least 30 swans but, as was the case in 1983, the majority of those flocks continued to hold between 30 and 100 swans (Figure 5). During the early censuses a figure for the number of non breeding swans was obtained by summing the number of swans which held a territory but did not breed and the number of non territorial swans. In order to facilitate a long term comparison of this data a similar value has been determined for more recent censuses (Figure 4). Habitat Over half of all swans recorded were located on still waters and the percentages of territorial and non territorial swans recorded on such habitat, rather than marine shores or canals or rivers, increased since 1983 (Table 5). Additionally, over 50% of territorial, breeding and non territorial swans in each of the north, centre and south of the country were recorded on still waters (Table 6). These findings suggest that availability of suitable unoccupied still water habitat may have contributed to the growth in the size of the swan population. Large numbers of non territorial swans frequented marine shores in both the north (n=620) and centre (n=146) but they were recorded in greater numbers on river habitat in the south. Whilst the total numbers of territorial, breeding and non breeding swans had increased in those areas since 1983, and their distribution between the different types of habitat had altered, overall they continued to favour still waters. In the north there was a decrease since 1983 in the percentage of territorial pairs, breeding pairs and non territorial birds on marine shores and an increase on still waters (Table 6). Although the number of non territorial birds on marine shores increased from 342 to 620 there was a compara- tively greater increase on still waters. Over half of northern non territorial swans were recorded Table 4 Flocks of 30 or more non territorial Mute Swans recorded in Scotland during the census SB 25 in 2002. Old County Location Habitat Count Outer Hebrides Loch Bee Still water 302 Kinross The Cut Still water 186 Orkney Harray Loch Still water 172 Angus Montrose Basin Marine shore 144 West Lothian Linlithgow Loch Still water 144 Lanarkshire Strathclyde Loch Still water 106 Lanarkshire Hogganfield Loch Stil water 103 Ross & Cromarty Kyle of Sutherland Marine shore 99 Ayrshire Ayr Harbour River 89 Midlothian R Esk Mouth Marine shore TI Kirkcudbrightshire Milton Loch Still water 66 Lanarkshire Lochend Loch Still water 56 Dunbartonshire River Leven River 50 Aberdeenshire Ythan Estuary Marine shore 49 Argyll and Islands Oban Harbour Marine shore 47 Caithness Loch Watten Still water 46 Lanarkshire Bingham's Pond Still water 43 Ayrshire Irvine Harbour River 43 Perthshire R Tay Balhepburn River 4] Aberdeenshire Loch of Strathbeg Still water 39 Ross & Cromarty Nigg Bay Marine shore 38 Fife Kincaple River 38 Midlothian Cramond Marine shore 36 Midlothian Inverleith pond Still water 35 Renfrewshire Castle Semple Loch Still water 35 Inverness shire Inverness Firth Marine shore 32 Roxburghshire Rosebank Kelso River 30 Total number of swans 2140 on still waters in 2002 rather than marine shores as in 1983. Although both numbers and percentages on still waters increased, the increase was smallest amongst breeding pairs. However, still waters continued to be the most favoured habitat for breeding swans in the north. In contrast to territorial and breeding pairs, non territorial birds on rivers increased in number. The numbers of swans recorded in the centre increased substantially between 1983 and 2002. The numbers of territorial birds on both still waters and on rivers increased, additionally marine shores and canals also held territorial birds in 2002. The diversification into a wider range of habitats caused the percentage of territorial swans on still waters to decrease since 1983 while that on rivers, marine shores and canals increased. The increasing size of the territorial population may have caused occupation of marginal territories on rivers which had remained unoccupied until recently. Although the number of breeding birds increased in each habitat still waters continued to be the Scottish Birds (2005) The 2002 census of the Mute Swan in Scotland 9 Table 5 Percentage occupation of different types of habitat by territorial, breeding and non territorial swans in Scotland in 1983 and in 2002. Habitat Marine Canal Still Waters River Territorial only 1983 16 0 59 25 2002 5 1 Te a2 Breeding 1983 10 | 80 9 2002 i) 3 82 8 Non territorial* 1983 47 0) 39 14 2002 gM| | 59 18 * Excludes Outer Hebrides: 1983 data not disaggregated by habitat type 25 20 15 10 Number of flocks 101 - 200 201 - 300 300 + Flock size 1983 = 1990 m 2002 Figure 5 The number of flocks of 30 or more Mute Swans recorded during the national census in 1943, 1990 and 2002. 10 SB 25 Table 6 Numbers and percentages of Mute Swans which occupied 4 different types of habitats in north, centre and south of Scotland in 1983 and 2002. Marine Canal Area Year Number % Number % North Territorial 1983 24 23 0 0 2002 10 3 l 0 Breeding 1983 54 13 0 0 2002 51 1] | 0 Non territorial* 1983 342 55 0 0 2002 620 40 0 0 Centre Territorial 1983 0 0 0 0) 2002 3 3 2 2 Breeding 1983 I | 2 3 2002 4 | 21 7 Non territorial 1983 79 44 0 (0) 2002 146 i 43 3 South Territorial 1983 3 7 0 0 2002 6 5 | | Breeding 1983 3 3 | l 2002 19 8 | 3 Non territorial 1983 92 33 0) 0 2002 43 5 4 0 Habitat Still waters River Total Number % Number % Number % 61 58 20 19 105 = 100 275 94 8 3 294 100 329 81 22 5 405 99 395 85 18 4 465 100 249 40 35 6 626 =6101 818 53 101 y 1539 ~=—-:100 16 84 3 16 19 100 59 66 23 28 89 99 60 83 9 13 72 ~ 100 257 82 3] 10 313 ~=100 80 44 21 12 180 100 1105 80 88 6 1382 100 20 49 18 44 41 100 61 54 44 39 112 99 81 73 26 23 111 100 75 74 34 14 230 99 94 33 97 34 283 100 371 a7 489 57 857 99 *Excludes Outer Hebrides, counts not disaggregated. most frequently occupied habitat amongst breeding swans. An increase occurred in the number of non territorial swans on marine shores from 79 to 146 and whilst non territorial swans were not recorded on canals in 1983 they were present in 2002. Increases were also recorded on still waters and rivers with still waters the most frequently occupied. The greater number of swans recorded on canals in the centre than in the north or south reflects a greater availability of canal habitat in the centre of Scotland. Numbers of territorial swans increased in each type of habitat in the south but a substantial increase occurred on still waters. Similarly the number of swans which bred on each type of habitat increased but the preference for still waters was again evident. Results indicated a change in distribution of non territorial birds from marine shores to rivers and still waters between 1983 and 2002. Scottish Birds (2005) Table 7 Altitude of breeding and territorial pairs of Mute Swans in Scotland in 1983 and 2002. Altitude Year Mean Range Territorial Pairs 1983 58 3-299 2002 61 3-335 Breeding Pairs 1983 45* 3-274 2002 64% 3-366 * significant difference The principal habitat for Mute Swans in Scotland was still waters with numbers of territorial, breeding and non territorial birds all increasing on that type of habitat since 1983. Additionally, large numbers of non territorial birds occurred on marine shores in the north and centre but on rivers in the south. Canal habitat increased in importance for breeding and non territorial birds in the centre. Altitude Analysis of altitude data was restricted to territorial and breeding pairs on still waters. This was due to the inherently consistent altitude of marine shore habitat, the inherent requirement for minimal altitudinal variability of canals while the altitudinal range of river territories was small. The mean altitude of still waters territories occupied by pairs which held a territory but did not breed and pairs which did breed increased between 1983 and 2002 (Table 7). In 1983 the mean altitude for breeding pairs was lower than the mean for territorial pairs which did not breed, however, results showed that by 2002 the converse was true. As the breeding population increased in number there was a Significant vertical expansion in its distri- bution (t test, df = 1014, p <0.001). This increase The 2002 census of the Mute Swan in Scotland I] Table 8 Altitude of breeding pairs of Mute Swans in the north, centre and south of Scotland in 1983 and 2002. Altitude Area Year Mean Range North 1983 27° 3-229 2002 35% 3-287 Centre 1983 77 3-244 2002 82 3-274 South 1983 90 3-274 2002 106 3-366 * significant difference was significant in the north (t test, df = 682, p 0.013), although it was not significant in the centre (t test, df = 98, p 0.547) or south (t test, df = 167, p 0.152) of the country (Table 8). The change nationally is apparent from Figure 6 which shows that in 1983 63% (n = 437) of pairs nested at an altitude up to 30m compared with only 47% (n = 829) in 2002. At sequentially higher altitudes the 2002 percentages were generally greater than those for 1983. Of the pairs which nested up to 30m in 2002, 70% bred at or below 15m (Figure 7) which indicated a preference for low altitude still water territories. That compared with 88% in 1983 which suggested that when the breeding population was low in number, and there were more vacant territories, breeding pairs had greater choice of territory and preferred to nest at a low altitude. Although the north of Scotland contains some of the highest land masses and associated water bodies in Scotland the altitude of the highest pairs of swans were recorded in the south. Requirements in terms of vegetation for adequate shelter and a supply of food are determined by water chemistry and ultimately by the underlying geology which varies from the north to the south of the country. The highest breeding pairs during 12 2002 were recorded at 335m at Acremore Loch, Selkirkshire and at 366m at Kingside Loch, Roxburghshire. Swans recorded at an altitude of over 300m comprised 2 pairs which held a territory only, 4 pairs which bred and 4 non territorial birds, all of which were located in Peeblesshire, Roxburghshire and Selkirkshire. Discussion The Mute Swan population in Scotland increased by 41% between 1990 and 2002 to number 7,028 individuals, but as a percentage it was less than 30 60 90 120 150 180 SB 25 the 70% increase which occurred between 1983 and 1990. Similarly, the breeding population increased by 47% between 1983 and 1990 but by only 22% between 1990 and 2002. These findings suggest that, while the population may continue to increase in the future, the rate of increase may be slowing and numbers may be beginning to level off. Whilst large increases were found in the centre and south of Scotland those in the north, where 45% of the population occurred, showed relatively little change and may be closer to reaching an upper plateau. Construction of reservoirs for agricultural 210 240 270 300 330 #8 360 Altitude (m) 1983 m 2002 Figure 6 The percentage of breeding pairs of Mute Swans on still waters, relative to altitude, in Scotland in 1983 and 2002. an 30 10 ; i i. _ [ee a _ 6 ) 12 15 18 21 24 27 30 Altitude (m) Figure 7 The percentage of breeding pairs of Mute Swans on still waters up to an altitude of 30m in Scotland in 2002. Scottish Birds (2005) irrigation, as has occurred in the Lothians (pers obs), has provided additional still water territories for swans. A long run of mild winters with no recurrence of severe winter weather and high mortality which occurred for example in 1962-63 (Boyd & Ogilvie 1964) has also undoubtedly benefited the species. Autumn sown cereals and oilseed rape frequently provide sources of food during the winter in lowland agricultural areas. Changes at a more local scale may be related to cyclical fluctuations in vegetation growth as occurred with Canadian Pondweed in Orkney (Meek 1993). In addition, and in particular in the centre, the increasing swan population has seen flocks developing in urban areas where supplementary feeding by humans probably helps to sustain swans during spells of adverse weather (pers obs). A decrease in availability of food may be the outcome of increased processing of sewage waste water with discharges of relatively cleaner water at discharge points on marine shores. As the population increased in number from 1983 changes occurred in the demography of the population. The percentage of the total population which was territorial decreased from 49% in 1983 to 39% in 2002 and conversely the percentage of the total population which was non territorial increased from 51% to 61%. A similar trend was identified in the Lothians during the same period (Brown & Brown 2002). The percentage of the total population which held a territory but did not breed remained little changed at 10% during the past 2 decades and was lower than the figure of 14% recorded in the Lothians with no significant change occurring there also (Brown & Brown 2002). The percentage of the total population which bred decreased from 38% in 1983 to 29% in 2002 which was similar to that in 1955-6. Six flocks of 30 or more swans were recorded in 1983 (Outer Hebrides not counted), increasing The 2002 census of the Mute Swan in Scotland 13 to 20 in 1990 and 27 in 2002. Additionally, flocks of over 100 swans increased from one or 2 in 1983 and 1990, to 7 in 2002. Such a substantial reserve of swans suggests that an increase in the breeding population may be being constrained by a lack of breeding territories causing birds with the potential to hold a territory to remain in non territorial flocks. It may also be indicative of the species potential to sustain a large breeding population at least in the immediate future. Increased numbers of non territorial swans were found on canal habitats especially in the centre of Scotland. Moreover, large numbers were found on rivers and marine shores which demonstrated the species’ ability to occupy a range of habitats. Mute Swans are able to frequent and feed on marine habitats due to the presence of salt glands situated above the eyes which extract excess salt from the blood and excrete salt through the nostrils. However, still waters were found to be the principal habitat for territorial, breeding and non breeding swans across the country. That preference appeared to have been sustained by occupation of territories at a higher altitude than in 1983, and of the 1012 pairs which bred in Scotland 4 pairs bred at an altitude in excess of 305m, all in the Borders. Rawcliffe (1958) noted that of the 463 nests recorded in 1955-56, 3 nests were above 305m, all situated in the Borders. As the percentage of pairs which nested above 305m did not increase in line with the increase in the breeding population the vertical expansion of the population may in the future be constrained by lack of suitable territories. Acknowledgements We are grateful to everyone who contributed to this census. In particular we thank all of the regional organisers for their help and enthusiasm, as well as their additional comments on the status of swans in their areas. Their support enabled a full census to be achieved 14 once again in Scotland and they were: A Stevenson, P Harvey, R Swann, D Butterfield, R Youngman, C Corse, R Proctor, J Dye, R Ormond, H Insley, C M Reynolds, A Seagrave (assisted by J Howie), J & V Wilson, N Bielby, S Laybourne, I Hopkins, A Bramhall, K Duncan, J Thomson, A Walters, M Ogilvie, I Hutchison and A Duncan. E Meek is thanked for providing additional data for Orkney and_ further information was supplied by local SOC recorders D Carmichael, C Henty, A McNee, R D Murray and A Thorpe. References Boyd H & Ogilvie M 1964. Losses of Mute Swans in England in the winter of 1962-63. The Wildfowl Trust Fifteenth Annual Report 1962-63. The Wildfowl Trust, Bristol. Brown A W & Brown L M 1985. The Scottish Mute Swan Census 1983. Scottish Birds. 13 140-148. Brown A W & Brown L M 1993. The Scottish Mute Swan census 1990. Scottish Birds. 17, 93-102. Brown A W & Brown L M 2002. Prefledging survival of Mute Swan Cygnus olor cygnets in the Lothians, UK. Bird Study. 49, 97-104. Campbell B 1960. The Mute Swan Census in England and Wales 1955-56. Bird Study 7, 208-223 Delany S, Greenwood J D & Kirby J 1992. National Mute Swan Survey — 1990. Unpublished report to the Joint Nature Conservation Committee. The Wildfowl and Wetlands Trust, Slimbridge. Dytham C 1999. Choosing and Using Statistics: A Biologist’s Guide. Blakewell, Oxford pp84-85. Meek E R 1993. Population fluctuations and mortality of Mute Swans on an Orkney loch system in relation to a Canadian Pondweed growth cycle. Scottish Birds 17, 85-92. Murray R D, Bramhall A T & Coleman J C 1996. Breeding success of Mute Swan in the Scottish Borders. Borders Bird Report 16, 70-72. SB 25 Ogilvie M A 1981. The Mute Swan in Britain, 1978. Bird Study 28, 87-106. Ogilvie M A 1986. The Mute Swan Cygnus olor in Britain 1983. Bird Study 33, 121-137. Pollitt M, Hall C, Holloway S, Hearn R, Marshall P, Musgrove A, Robinson J & Cranswick P 2003. The Wetland Bird Survey 2000-01: Wildfowl and Wader BTO/WWT/RSPB/JNCC. Rawcliffe C P 1958. The Scottish Mute Swan census 1955-56. Bird Study 5, 45-55. Counts. Allan W Brown & Lyndesay M Brown, Lothians and Fife Swan Study Group, 61 Watt’s Gardens, Cupar, Fife KY15 4UG Revised manuscript accepted August 2004 Appendix Assessment of coverage in each county The Great Britain survey methodology was based on a sampling approach including a random selection of 10 km squares. The aim of this was to ensure that results were not biased by observers only visiting those squares known to hold swans. However, in order to avoid asking too many observers to visit negative squares rather than squares known to hold birds, the selection of random squares was concentrated on all of those squares which held 50 or more birds in 1990 and those considered more likely to hold birds. Coverage of these squares had to be guaranteed for analysis of the Great Britain results. In Scotland, however, as in 1983 and 1990, regional organisers were encouraged to ensure as full coverage as possible of their area. This entailed coverage of all of the randomly selected squares which required guaranteed coverage, all other squares which recorded swans in the 1990 census and coverage of all other squares known to Scottish Birds (2005) hold or with the potential to hold swans. On this basis it was considered that coverage would be thorough and all organisers agreed to undertake this process. The results as presented in this paper are based entirely on the actual data submitted and no estimates have been made of possible under recording. The regional organisers were asked to give an indication of the extent of coverage obtained and to provide their view on any trends they were aware of for the Mute Swan population in their area since the 1990 census, including breeding numbers and changes in or establishment of non breeding flocks. Where provided these comments have been referred to in the following county summaries, otherwise the summaries are based on an assessment of the level of coverage obtained and comparisons with previous surveys. To facilitate comparison with the 1983 and 1990 surveys coverage has been assessed as good, moderate or poor. Shetland: Good - Full coverage was obtained. Breeding first occurred in 1992 since when the population has continued to show a slow expansion with 3 to 5 pairs present. The 2002 results were the best ever count of breeding and non breeding birds. The origins of these birds may be from Orkney. Orkney: Good - Full coverage was obtained. The non territorial count for Loch Harray/Stenness (172) was mid May rather than mid April and this contrasted with a count of 491 birds there in March 2002. Allowing for breeding and territorial birds this suggests that ca 140-150 birds were missing. However, over 30 dead birds were recorded on the May count suggesting that there had been some mortality for whatever reason, which may have forced many birds to disperse elsewhere (Eric Meek, pers comm). The decline in the population at Loch of Harray & Stenness in 1992 (Meek, 1993) owing to the loss of the Canadian Pondweed (Elodea canadensis) The 2002 census of the Mute Swan in Scotland TS food source was followed by further decline and then stabilisation at ca 175 birds until numbers started to recover by 1998 when ca. 420 were present (all counts in May) (Eric Meek, pers comm). Outer Hebrides: Moderate to Good - Nil return for Harris and Lewis where Mute Swans are rare vagrants (C Reynolds, pers comm). The local organiser regarded 2002 as a poor season in the Uists and Barra due to cold and wet weather which resulted in some birds not nesting at all due to high water levels. In addition stormy winter conditions are thought to have affected feeding and thus birds attaining breeding condition, as well as resulting in higher than usual mortality of young birds. Indeed it was felt that 2002 “was the worst and most unpredictable for Mute Swans in the Uists in 20 years”. The appearance of late broods in July suggested that some pairs may have been overlooked in the remoter parts of North Uist and non breeding counts, other than for Loch Bee, were fewer than would have been expected. The overall impression was that the population was relatively stable with distri- bution similar to previous surveys. Caithness: Good Sutherland: Good - Full coverage was obtained. Ross & Cromarty: Good - Full coverage was obtained. The local organiser considered overall numbers of breeding pairs had remained stable or even increased slightly. Inverness shire: Good - Full coverage was obtained of all known squares containing breeding birds. 16 Nairn and Moray: Good - All known breeding sites checked including vacant sites east of Moray which still appeared to be suitable. Aberdeenshire & Banffshire: Moderate to Good Angus & Kincardineshire: Moderate to Good - The regional organiser considered coverage was not as good as they would have liked. Additional data were obtained through the local recorder, Dan Carmichael, and this is thought to have filled in most of the gaps. Perthshire: Good - Full coverage was obtained. Stirlingshire & Clackmannanshire: Good - Full coverage was obtained. Fife & Kinross: Good - Full coverage was obtained with that in Fife being part of an annual census commenced in 199] and which has shown a rapid increase in the territorial population (Brown & Brown, unpublished reports). West, Mid & East Lothian: Good - Full coverage was obtained as part of an annual census commenced in 1978. Both breeding and non breeding populations have increased steadily and reached their highest recorded levels in 2002. (Brown & Brown, Lothian Bird Reports 1982 to 2002). Ayrshire: Good - The Regional Organiser considered that the wet weather in the spring resulted in birds either not nesting or nests being abandoned due to high water levels, and sites checked in May which usually had pairs had been abandoned. SB 25 Arran & Bute & Cumbrae: Good - Full coverage was obtained. The Regional Organiser for Arran commented that ‘all our swans are coastal and definitely on the increase. From one pair which bred successfully about 5 years ago, after a gap going back to war years, we now have 7 breeding pairs. The increase is due mainly to the establishment of a feeding station (at a hotel) at Whiting Bay”. The Bute organiser stated that “there has nearly always been a pair of swans on most of the lochs on Bute, with a slight increase since 1990”. On Cumbrae the organiser advised that “the number of breeding pairs on this island has fluctuated around the present figure for some time’. Peeblesshire, Berwickshire, Roxburghshire & Selkirkshire: Good - Some additional data were received from the local recorder, Ray Murray. Argyll and Islands: Good Dunbartonshire, Renfrewshire & Lanarkshire: Good - Full coverage was obtained. Dumfriesshire: Good - Full coverage was obtained. Kirkcudbrightshire: Good - Full coverage was obtained. Wigtownshire: Good - Full coverage was obtained. Scottish Birds (2005) 25:17-23 The eradication of Brown Rats from Handa Island, Sutherland 17 The eradication of Brown Rats from Handa Island, Sutherland J STONEMAN & B ZONFRILLO Burrow nesting seabirds declined after Brown Rats arrived on Handa Island some time after 1848. The eradication of Brown Rats in 1997 led to a rapid rise in numbers of some breeding seabirds while others increased their breeding areas or bred for the first time in living memory. The methods are based on modern commercial poisons and make the complete eradication of Brown Rats possible from the Scottish islands they have colonised, usually through the agency of man. Introduction Handa Island is located on the north west coast of Sutherland, just south of Cape Wrath, about 0.5km from the nearest mainland and is a privately owned nature reserve managed by the Scottish Wildlife Trust. It is internationally important for seabirds, holding the largest Common Guillemot Uria aalge and Razorbill Alca torda colonies in Britain and Ireland (9.4% and 8.6% of the British and Irish population, respectively), as well as having nationally important numbers of Great Stercorarius skua and Arctic Skuas Stercorarius parasiticus (1.9% and 1.0% respectively) and Black-legged Kittiwakes Rissa tridactyla (1.4%) (Handa Management Group, 1999). Seabirds thrive on predator free islands but when predators such as American Mink Mustela vison or rats Rattus sp get ashore, breeding seabirds decline or desert completely and burrow nesting species in particular tend to rapidly vanish. (Craik, 1995; Zonfrillo 2002a & b) The history of Brown Rats on Handa Both the Brown Rat Rattus norvegicus and the Black Rat Rattus rattus are introduced species to most of Europe. Their origins are fairly well documented. In the British Isles the Black Rat Great Stack 1999, \s000 at rockfall Na Geodaichean Dubha HANDA ISLAND Bothy ~y Traigh Shourie / \ Flagpole y Peninsula id Port an Eilean Traigh na b-Airigh Otter Point Glas Léac Map 1 Map of Handa showing place names and spread of Atlantic Puffins on the main island with dates of colonisation following eradication of rats in 1997. arrived via ancient trade routes from China and South East Asia, probably around the eleventh century; the Brown Rat arrived much later, around the year 1728, via shipping from present day Russia (Corbet and Southern, 1977). While the Black Rat has largely died out, the Brown Rat has continued to spread. In Scotland, the Brown Rat spread at first rather slowly, and by 1855 even some remote areas had been colonised (Matheson, 1962). However, with Chapel Bay 18 SB 25 farming practices and shipping contributing to its spread, it is now easier to name areas and islands that have no rats than those that have. It is not clear when Brown Rats first arrived on Handa, but they were certainly well established by the 1880’s. There is no mention of rats by Charles St John (1849), who visited the island just a few weeks after the last permanent inhabitants left in 1848. He wrote of the tameness of the Atlantic Puffins Fratercula arctica that alighted so close to him, at the cliff tops, that he could have ‘knocked them down with a walking stick’. The earliest reference to the presence of rats comes from Harvie-Brown and Buckley (1887) who blamed rats for reducing numbers of Black Guillemots Cepphus grille and for driving Atlantic Puffins ‘off the tops at Handa into more secure crevices in the face and slopes’. In 1904 Harvie-Brown and MacPherson stated that rats “simply swarm over the best ground’ and again attribute rats for displacing Atlantic Puffins from the tops of the cliffs, compared to their visit in 1867. They also reported that Duncan McIver, a local to the area who accompanied them on their trip, commented that about 20 years earlier ie in the early 1880s, there were so many rats on the island that he was under the impression that they were ‘migrating’. It therefore can be deduced that rats colonised Handa sometime between 1848 and 1867, a period when the island was farmed for sheep. Rats may have _ been introduced during transportation of animals and their feed from the mainland. In 1962 Handa Island became a nature reserve, and from 1972 a warden has resided on the island each year from April to September. Wildlife observations from wardens’ reports from 1974 to 1996 show that a population of rats persisted, particularly around the coast and at the Bothy, the only habitable building on the island. (See map) Evidence for rats causing declines to seabirds From circumstantial evidence Harvie-Brown and MacPherson (1904) suggested that Brown Rats were the cause for declines in Atlantic Puffins and Black Guillemots some time after 1867. In recent times, seabird monitoring since 1962 has shown that populations of most cliff nesting seabird species were either stable or increasing (Stoneman & Willcox 1995). A situation had been reached where Brown Rats were not causing further declines in seabird numbers, but were still impacting on the populations by inhibiting expansion, in the case of Atlantic Puffins, or recolonisation, for Black Guillemots and other burrow or hole nesting species. The absence of other ground or burrow nesting seabirds such as the European Storm-petrel Hydrobates pelagicus and Manx Shearwater, Puffinus puffinus in what appears to be suitable habitat, may also have been due to the presence of rats. In winter, with few birds present, it was highly likely that rats were sustained by feeding on the introduced Rabbits Oryctolagus cuniculus that were also confined to the coastal fringes of the island. To make a scientific assessment of the possible impact that rats could be having on seabird numbers on the island, a 3 year research programme was initiated in 1994 (Aragundi 1994, Wardens’ Reports 1994, 1995, 1996). A brief summary of the results of the research 1s listed below: I Rat distribution The rat population was almost exclusively based around the island’s rim apart from at the only habitable house on the island, the Bothy. This was established by noting locations of latrines, runs etc and monitoring with chewstick"! — stations throughout the island. The incisor chewing marks on the sticks can thus establish the presence or absence of rats. 1. Chewsticks are wooden spatulas impregnated with melted lard or margarine and firmly fixed at ground level. Scottish Birds (2005) 2 Effect on burrow or hole nesting birds Rats could find and decimate nests of ground nesting seabirds in any suitable breeding habitat. This was demonstrated by setting up a number of ‘false nests’, utilising chicken eggs, some waxed, some on a layer of fine, wetted peat, set up in apparent Atlantic Puffin, Black Guillemot and European Storm-petrel habitat. The predation of the eggs, indicated by tooth marks on the waxed eggs and footprints in the peat, revealed rat activity. 3 Other species There was no firm case that rats were affecting Northern Fulmar Fulmaris glacialis breeding success on the cliffs. This was deduced by observing the survival of chicks in different habitats. However there were no inland or ground nesting Northern Fulmars on Handa, as is common in Shetland and Orkney. Both species of skua appeared not to suffer from rat predation. There were few breeding waders and no large gull colonies. 4 Atlantic Puffins Puffins nested largely in rat free areas. Counting individuals above and below the cliff tops showed this. On offshore stacks, 94% of Puffins counted were found on the top, compared to just 6% of those recorded on the vertical main island cliffs. Eradication A campaign to eradicate rats from Handa was instigated and the anti coagulant poison, Warfarin, was chosen to eliminate the rats. Since rat numbers were calculated to be at their lowest from late winter to early spring, and is the time before many birds return to breed, this period became a window of opportunity to execute the programme. Eradication of the Handa Brown Rat population took place from 23 to 29 March 1997. The baiting team comprised 12 people; the authors, 2 members of Scottish Wildlife Trust staff and a team of 8 volunteers, including 2 skilled climbers. JS was also a member of the SWT staff at the time of the project. The eradication of Brown Rats from Handa Island, Sutherland 19 Two tonnes of 0.05% Warfarin on a whole wheat base were brought to the island by boat wrapped in PVC and stored in a shelter on pallets ready for use. Baiters worked in groups of 2 to 3 and distributed the bait around the coast and at the Bothy ie where any rat activity had been confirmed. Maps were used to monitor bait distribution and to avoid duplication of effort. Bait was put down burrows and under rocks thus avoiding any chance of secondary or collateral poisoning to birds. Two climbers delivered and distributed bait to areas otherwise inaccessible without ropes, especially at the base of gullies above the high tide line, where rats were likely to forage. Where there were no crevices or burrows available, or in areas used by Otters Lutra lutra for ‘lie up’ sites and holts, bait was placed under weighted plastic fish boxes, with the handles sawn out to allow rats to enter. One hundred kgs of bait was stored on the island for back up baiting. Assessing impact Following baiting, several aspects of monitoring were continued or initiated to assess the impact and success of rat eradication. 1. Presence of rats was monitored through chew stick stations set up around the coast and at the Bothy, a few weeks after baiting was completed. 2.Monitoring of Atlantic Puffins continued according to Walsh et al 1995. 3.New areas, if any, colonised by Atlantic Puffins were mapped. 4. Common Sterna hirundo and Arctic Tern Sterna paradisea populations and breeding success were monitored according to Walsh et al (1995). 5. Annual searches were made for breeding Black Guillemots. 6. A search for breeding European Storm-petrels according to Walsh et al (1995) was carried out. 20 7. A study of chick survival of Northern Fulmars was initiated. 8. Observations were made of other wildlife and changes that might be attributable to rats and noted in wardens’ reports. Results I Monitoring rats post 1997 Since 1997 no live rat has been seen on the island, though in 1998 the check in mid April showed strong evidence of ‘rat like’ activity at Port an Eilean and Chapel Bay, where chewing and ‘rat like’ prints were found. Bait was distributed in the area, after which there were no further signs. Some activity at chewsticks after this period was thought to be due to young Rabbits. The absence of rat droppings and presence of Rabbit droppings alongside the chewsticks appeared to confirm this. If by chance rats were getting ashore from the mainland their spread to the best areas of feeding might go unnoticed, at least in the short term. To address this, bait boxes were set up around the coast in 2001 to provide a permanent source of bait all around the island during summer. This complemented the monitoring system already in place. The warden regularly replaces bait during the summer months. 2 Atlantic Puffin monitoring and Puffin distribution Atlantic Puffins had been monitored since 1977 by counting peak numbers of individuals in late July, both on the Great Stack and for the whole island. This method involved counting wandering non breeding birds as well as potential breeders and was changed when Walsh et al 1995 was published, which recommended that counts should be made in late April when only breeding birds will be at the colony. A steady increase in numbers since monitoring began was noted, and this has continued after rat SB 25 eradication. However there was no monitoring of breeding success since burrows were generally inaccessible. Presumably many of the birds visiting in July, that were deterred from breeding by rats, can now be absorbed into the colony. Atlantic Puffin breeding numbers will now depend, for example, on food supply in future years and not in avoiding rat predation. The expansion of the Atlantic Puffin colony was first recorded in 1999, when between 20 and 50 occupied burrows were counted on the main island immediately west of the Great Stack. This area remains occupied by Atlantic Puffins and further colonisations have been noted (see map). One new area was above a recent rockfall (1996) at the top of the west cliffs. With Atlantic Puffins already breeding on the island, their expansion to new areas was rapid, in contrast to Ailsa Craig, Ayrshire where Atlantic Puffins took 10 years to recolonise (Zonfrillo 2002a). 3 Tern monitoring Terns nest on Handa on the skerries at Port an Eilean, islets that are cut off from the main island at high tide. This 1s a relatively undisturbed area that is not generally frequented by visitors. Despite this, numbers of nesting terns had been generally low up to rat eradication with only 8 pairs of Common Terns, and 5 pairs of Arctics nesting. In 1988, 30 pairs of terns were reported to be breeding at Glas Leac, an adjacent small island but farther offshore. The maximum number of tern chicks known to have fledged since records started was 8 in 1990. In 1998, the year after eradication, there was a marked increase in the nesting tern population with 58 pairs recorded, of which at least 25 fledged young. Since then the success of the tern colony has been variable (see Figure 1), but high numbers have been recorded in 1999 and 2001, when a new colony of Arctic Terns was Scottish Birds (2005) established at the west end of Traigh Shourie, a bay near Otter Point. The figures probably indicate that rat eradication has enabled the colony to thrive when conditions are good, but other external factors, such as food supply or weather, may be having an adverse impact on the colony in some years. Abandoned eggs show that they have not been predated. Longer term monitoring should clarify the situation. 4 Searches for Black Guillemots Since rat eradication there has been no evidence of Black Guillemots breeding on the island. There were some indications that they were at least prospecting from 1999 to 2001, when birds were heard calling either nearby or in a cave at Na Geodaichean Dubha, a little visited area in the north east of the island. However, from 2002 to 2004 such behaviour has not been recorded. 5 Searches for European Storm-petrels A European Storm-petrel survey in 1999 by AR Mainwood, conducted by playing recorded calls into burrows and crevices in apparently suitable habitat, failed to locate birds. However, European Storm-petrels have been mist netted in 2001 (BZ, JS and T P Daniels) and 2003 and 2004 (Highland Ringing Group). In 2004, 20% of the 96 birds netted had brood patches - one bird was a recapture from 2003. This suggests that European Storm-petrels might be breeding locally. In addition, European Storm-petrel remains were found in 4 Great Skua pellets in the same year (E Williams, pers comm). Further investigation is needed to confirm whether the birds could be breeding on Handa. 6 Northern Fulmar chick survival In 1996 a simple Northern Fulmar chick survival monitoring plot was established at the cliffs to see if chick survival would improve following eradication. This involved counting chicks at the end of August, when the chicks were near to fledging. Chick survival appears to differ from year to year, for reasons as yet not understood. In The eradication of Brown Rats from Handa Island, Sutherland 21 the past, rats may not have found these chicks easy prey on their cliff sites; hence so far there is no apparent increase that can be attributed directly to the eradication of rats. Other wildlife observations following the eradication of rats @ A pair of Common Shelduck Tadorna tadorna bred in 1997 and 1998; this burrow nesting species has never been recorded breeding on the island prior to this. @ Eurasian Oystercatchers Haematopus ostralegus and Ringed Plovers Charadrius hiaticula have bred successfully every year since 1998 - this was rarely reported before 1997. @ Sightings of Pygmy Shrew Sorex minutus have increased eg during 1998, 10 were counted; in the 10 year period prior to that, there were only 8 sightings in total. There is now a shrew family resident in or under the warden’s accommodation. @ A pair of Common Redshanks Tringa totanus colonised Otter Point in 2001 and continues to breed there. M@ Rock Pigeon Columia_ livia probably colonised the north east coast of Handa in 2001, although the breeding location has not yet been confirmed. @ A second colony of Mew Gulls Larus canus was established on the west side of Port an Eilean in 2002 and has thus doubled the island’s Mew Gull population. (See Figure 2) Discussion Some species have increased in numbers or extended their breeding range on the island. Other species have arrived and bred successfully for the first time. Indigenous mammals such as the Pygmy Shrew, whose population was always low, have now become more common. The poison used to target the rats appears to have had no detrimental effects on breeding birds or any other indigenous wildlife. The techniques here used for eliminating and monitoring rats can also be applied to many a SB 25 - Common Rat eradication > 50 teem Arctic 40 30 + No. nesting pairs 20 innit -tatietsteiaiatatatate iatatananenaiaieiataie naneneiaiaReREREROR RE 10 Year Figure 1 Fluctuation in breeding terns from 1974 to 2003. other small islands in Scotland. Other creatures and plants that may have been suppressed by the activities of rats will be monitored for the foreseeable future on Handa Island. Acknowledgements We would like to thank the following for their help in carrying out the rat eradication project, and in the production of this paper: Nigel Harding, Handa Warden 1994, who initiated the idea for the project; Neil Willcox, former SWT Reserves Manager, who made the project possible; Charles Thomson, Handa boatman, for ferrying materials to the island; the workers who distributed the bait: David Leather, Martin Robbins, Sam Davies, Jackie MacDonald, Sandy MacKay, Janet Maxwell, Ben Moseley, Colin Coventry, and Hans Buff. The Scottish Wildlife Trust provided the records required to write this paper and Elizabeth Williams, SWT 2004 Warden, provided data from the 2004 field season. On behalf of the Scottish Wildlife Trust, we would like to acknowledge financial contri- butions from Scottish Natural Heritage, Caithness and Sutherland Enterprise Company, the Seabird Group, East Sutherland Bird Club and SWT Glasgow and Clydeside Support Group as well as many members of the public. We trust their contributions will be rewarded by flourishing wildlife populations on Handa Island. References Aragundi S 1994. The effects of the predation by the brown rat on Fulmars on Handa Island Unpublished MSc thesis, Aberdeen University. Corbet G B & Southern H N 1977. The Handbook of British Mammals - 2nd Edition. Blackwell, Oxford. Craik JC A 1995. Effects of North American mink on the breeding success of terns and smaller gulls in the west of Scotland. Seabird 17: 3-11. Harvey-Brown J A & Buckley T E 1887. A Vertebrate Fauna of Sutherland, Caithness and Scottish Birds (2005) The eradication of Brown Rats from Handa Island, Sutherland vA) ¢ Rat eradication a1 T — ~~ fo) oo seep : . Lie a =i SO So © S oe) CS i) - 8 0 C6 6 96 Year Figure 2 Fluctuation in breeding Mew Gulls 1970 to 2003. West Cromarty pp 38-40. David Douglas, Edinburgh. Harvey-Brown J A & MacPherson H A 1904. A Fauna of the North West Highlands and Skye pp xlin - lit and 356-7. David Douglas, Edinburgh. Handa Management Group 1999. Handa Island Wildlife Reserve Management Plan 1999-2004. Unpublished - Scottish Wildlife Trust, Inverness. Matheson C 1962. Brown Rats. London, Sunday Times Publications. (Animals of Britain, No 16). St John C 1849. A Tour in Sutherlandshire. John Murray, London. Stoneman J G & Willcox N A 1995. Seabirds of Handa Island. Scottish Birds 18: 1978-1987. Walsh P M, Halley D J, Harris M P, del Nevo A, Sim I M W & Tasker M L 1995. Seabird Monitoring Handbook for Britain and Ireland. JNCC, RSPB, ITE & the Seabird Group. Zonfrillo B 2002a Atlantic Puffins return to Ailsa Craig. Scottish Bird News No.66 pp 1-2. Scottish Ornithologists Club. Zonfrillo B 2002b Wildlife Conservation on Ailsa Craig pp 22. Thomas Duncan Memorial Lectures. FMI Girvan. The following unpublished Wardens Reports from 1974-2003 were consulted: E Evans 1974, David Mower 1975, C Wells 1976, Mike Trubridge 1977, Keith Fairclough 1978, R Thorpe 1979, J Leece 1980, M Allison 1981 & 1982, M Walker 1983, C Hurlford 1984, P Fisher 1985, R Ashcroft 1986, G Macdonald 1987, Charlie Self 1988, I Lawrie 1989, Matt Self 1990, Alison Aitken 1991, Janice Breckenridge 1992, Ian Gordon 1993, Nigel Harding 1994, Julie Stoneman 1995, 1996 & 2001, Mike Thornton 1997, Uwe Riek 1998, Uwe Stoneman 1999 & 2000, Julie Stoneman & Lizzie Williams 2002, Lizzie Williams 2003. Scottish Wildlife Trust, Inverness. Julie Stoneman, The Flat, Riemore Lodge, Butterstone, Dunkeld, Perthshire PH8 OHP. Bernard Zonfrillo, Graham Kerr (Zoology) Building, University of Glasgow, Glasgow, G12 8Q@. Revised manuscript accepted January 2005 24 SB 25 Resident Golden Eagle ranging behaviour before and after construction of a windfarm in Argyll D WALKER, M MCGRADY, A MCCLUSKIE, M MADDERS & D R A MCLEOD Resident Golden Eagle ranging behaviour was monitored over 776 observation hours before and after construction of a windfarm in Argyll, western Scotland between 1997 and 2004. Overall size of the eagle range that was potentially affected by the windfarm (for male, female and both eagles) was similar before and after construction. Eagles appeared to change their ranging to avoid the windfarm site. Once built the windfarm was over flown mostly when other eagles intruded on the territory. An area of plantation forestry was felled with the aim of mitigating the potential loss of foraging habitat to the windfarm, and drawing eagles away from the windfarm thereby reducing collision risk. Eagles were seen in the tree cleared area 3 times more often after felling than before felling, and the shift in ranging was away from the windfarm and in the direction of the felled area. These findings are from a single pair and should be used cautiously when applied to other, similar, situations. However, they are an important first step in understanding the likely effects of windfarms on eagles. Introduction In the UK in 2004, 253 MW of new, wind generated electricity was added to the national grid, 5 times the annual amount in the 1990s and double the 2003 figure. In Scotland, 11 schemes are under construction and due to come on line by the end of 2005. Many more developments are being planned in Scotland, and 70% of onshore schemes being considered for planning approval in the UK are located there (British Wind Energy Association 2004). Prospecting for new, commercially viable sites continues. Scotland holds virtually all breeding pairs of Golden Eagles Aquila chrysaetos in the United Kingdom. Windfarms located within the range of Golden Eagles can cause eagle deaths due to collisions (Hunt 2002), and it has been thought that eagles may alter their ranging behaviour to avoid turbines, thus rendering the habitat within the windfarm area unavailable to foraging eagles. In Scotland these possible impacts have led to the adoption of a cautious approach to the siting of windfarms with regards to the location of territorial eagles. A 46 turbine windfarm, the Beinn an Tuirc windfarm, was constructed during 2001 within an occupied eagle territory in Argyll. In addition, another windfarm, the Deucheran Hills windfarm, was built in 2001 (9 turbines) about 6.4 km to the north of the Beinn an Tuirc site, and is more peripheral to the home range of the eagles. To mitigate the potential habitat loss resulting from the Beinn an Turic windfarm, a habitat management plan was implemented that included forest clearance and management of existing Heather (Calluna vulgaris) moorland to increase the abundance of potential eagle prey (eg Willow Ptarmigan Lagopus lagopus scoticus and Black Grouse Tetrao tetrix). The creation of new areas of foraging habitat away from the windfarm was also thought likely to reduce the risk of eagle collisions with the turbines. An on going programme of eagle monitoring was Scottish Birds (2005) 25: 24-40 Eagle behaviour before and after construction of a windfarm 25 Figure 1 Study area. Grid lines are Skm x 5 km. ¢ Turbines ~. Windfarm footprint Figure 2 Kernel analysis of resident eagle movement (n=154) 1997-2004. Legend ® Turbines ® Randomised eagle locations Eagle movement 1997-2004 [_] 50% kernel aa 95% kernel 26 undertaken from 1997 to assess effects of the Beinn an Tuirc windfarm and the habitat management plan on Golden Eagle ranging and breeding performance. The Golden Eagle is a species of medium conservation concern in Britain (Gibbons et al 1996). In Argyll habitat changes that adversely influence foraging potential (eg upland afforestation and overgrazing of Heather areas) have affected territories adjacent to the one studied by us (Watson et al 1987). In spite of the similar loss of much land to plantation forest within the estimated eagle home range that includes the Beinn an Tuirc windfarm, there remains an extensive area of open land with modest populations of important prey species such as Willow Ptarmigan. Because of this the home range continues to be potentially viable for breeding eagles. Study area The Beinn an Tuirc windfarm (255 ha) and eagle monitoring area (ca 57 km?) straddle the main ridge (Figure 1), which is generally below 300m above sea level, though there are peaks of ca 450m. The eastern slopes of this ridge, to a distance of about 3 km, are characterized by deeply cut valleys, with rock outcrops that provide a number of suitable eagle nest sites. To the west of the main ridge for a distance of about 8 km the terrain is gentler, characterized by wide, rounded ridges and shallow incised stream courses that run to the sea. This east west pattern extends both north and south of the study area. Landcover within the monitoring area includes commercial forestry blocks, mostly Sitka Spruce Picea sitchensis of varying age, and open hill, dominated by grass and Heather; open areas include both grazed and ungrazed habitats, which are mostly acidic grasslands with some areas of shrub heath and areas of blanket bog on SB 25 the higher slopes. Between October 1999 and June 2001 an area of forest (ca 280 ha) was felled to the north east of the main open area as part of the habitat management plan. Eagle monitoring focused on an area of ca 34 km? of open hill, which is bounded on the north and south by forest, but also includes ca 7 km? of open ridges within forest blocks to the north. The diversity of natural fauna is limited, and a number of species, such as Mountain Hare Lepus timidus and Golden Plover Pluvialis apricaria, no longer occur locally as breeders. Mammals include small numbers of Rabbits Oryctolagus cuniculus around the fringe of the monitoring area, occasional Brown Hares Lepus europaeus towards its western edge, Sika Cervus nippon and Roe Capreolus capreolus Deer in the plantations and Foxes Vulpes vulpes. The birds are typical of upland areas in western Scotland (Ratcliffe 1990). Birds breeding on or using the area include diurnal and nocturnal raptors, Red-throated Divers Gavia stellata, small numbers of Mallard Anas platyrhynchos, Eurasian Teal A. crecca and Mew Gulls Larus canus. The forest avifauna is dominated by passerines such as European Robin Erithacus rubecula and Chaffinch Fringella coelebs, and corvids Corvus spp. Black Grouse are present in 3 to 4 areas of the younger plantations, but also occur on the open hill. The open hill holds a scattered population of Willow Ptarmigan, which are mostly associated with areas of Heather moorland. Small numbers of Common Snipe Gallinago gallinago and Eurasian Curlew Numenius arquata occur in grass dominated wet flushes. The Beinn an Tuirc windfarm contains 46 — 660kW turbines that are divided evenly into 2 groups (north and south); within these groups the turbines are > 150 m apart. At its narrowest point the gap between the north and south areas is about 670 m. The Beinn an Tuirc windfarm itself is located in the central southern section of the Scottish Birds (2005) main block of open area with plantation forestry bordering its southern edge. Some plantation forestry (ca 50 ha) was removed to accommodate the southern section of the windfarm. Human activity in the study area prior to windfarm construction mostly comprised shepherding on the open hill, deer stalking within the forests and ecological project survey work throughout the area. Forest operations, eg felling and planting, are ongoing, but the location, timing and extent of these are controlled, especially during the breeding season, to lessen potential impact on the eagles. Since construction, regular maintenance of the wind turbines has been added to the list of human activities in the area. Human visitor pressure on the open hill by hill walkers, both before and after construction, was very limited and mostly associated with accessing the highest summit. Methods Observations of eagle movements were made from 4 vantage points (VP). From these we monitored range occupancy, habitat use and foraging effort by the individual eagles, and collected information on eagle behaviour. Two VPs have been in use since 1997, a third was added in 1998 and a fourth in 1999. The Beinn an Tuirc windfarm area and main open area have been monitored since 1997; the addition of the last 2 VPs allowed us a better view of an area of forestry felled in mitigation of the windfarm. Collectively, the area viewed from the VPs comprises the eagle monitoring area, and VPs are located around the perimeter of this area so that the greatest continuous panorama is under observation, while reducing any potential influence of observer presence on eagle behaviour. Observations were made 8 times per year (twice per quarter) from each VP between November 1997 and April 2004 except during March to Eagle behaviour before and after construction of a windfarm 27, December 2001, when fieldwork was curtailed by Foot and Mouth Disease access restrictions. Within each quarter all 4 VPs were visited; the order of visits was arbitrary. Weather could affect the area viewed from any particular VP and the duration of any particular watch period. Observation periods were chosen to avoid periods of continuous heavy rain, snow or dense fog, and ideally were 4 hours in length. Where possible, watches affected by poor weather conditions were extended to achieve 4 hours of observation time. While weather conditions could affect VP visibility they did not influence choice of VP, and all VPs were visited in a variety of conditions. While most watches tended to cover the middle of the daylight period, observations occurred at all times of the day. A total of 392 hrs of observation were made before construction, 68 during construction and 316 hrs after construction. A single, experienced observer (DW) made all observations. The viewing area was kept under continuous observation for the full watch period by above skyline scanning without optical aids, binocular scanning of all areas and regular telescopic checks of known and _ potential perches. In so doing bias in observer effort towards specific locations within the viewing field was minimized. When an eagle was seen, the time of first contact was recorded to the nearest second, and the bird’s flight path was plotted on a paper map. Simple flights were synchronously plotted in the field, prolonged flights were plotted in sections that were drawn synchronously or nearly so, and fast or short flights were plotted immediately after they occurred. Final plotting of more complex flight lines was completed as soon as was possible after the watch period. In this way a complete activity log of eagle behaviour and location was kept for each VP session. An estimation of altitude above the ground (in range 28 SB 25 Figure 3 Flight lines (left, n=811) of resident Golden Eagles (male and female). Grid (1 km?) colour shows relative use by eagles (dark red=heavy use, light pink=light use). Legend * Vantage points e Turbines ——— Flight lines Ps Forestry Metres of flight lines per km square q { | { ! i] 0 - 5643 5644 - 20095 WO 20096 - 50445 BR 50446 - 101082 BR 101083 - 164853 Scottish Birds (2005) Eagle behaviour before and after construction of a windfarm 29 Figure 4 Kernel analysis of ranging of resident eagles (male and female) showing their ranging before (left, n=57) and after (right, n=83) windfarm construction. Legend e Turbines ® Randomised eagle locations Eagle movement 1997-2004 oe 50% kernel Rececee 95% kernel 30 bands of <5m, 5-20m, 21-60m & >60m) and activity (hunting, transitional flights, species interaction, display, height gain and directional flights) were noted to the nearest second, as was the time when the bird either landed or flew from view. Factors that might influence eagle behaviour (eg human activity, presence of intruding eagles) were also noted. Even when more than one eagle was visible, all flights were followed, timed and plotted. No flights were excluded from the recording process and no assumptions were made about the route or activity of birds when they were intermittently lost from view. Analyses of eagle ranging data Two analytical approaches were taken, one based on generating a representative set of eagle locations and one that used a grid overlaid on eagle flight lines to calculate an index of use of km? areas by eagles. These were used to create maps that show location, extent and concentration of use by eagles. Data on eagle ranging and habitat were entered into a Geographical Information System (GIS, ArcView 3.3 and ArcGIS, ESRI, Redlands, CA, USA), where analyses and map making were undertaken using the Animal Movement (ver 2.0) extension (Hooge and Eichenlaub 1997). Point analysis. We framed the area in which eagles were observed by mapping the maximum extent convex polygon, the vertices of which were the most outlying of observations of eagles. The maximum extent convex polygon probably overestimates the actual range, so we also used a randomised selection of points along mapped flight lines to generate a ‘representative’ set of eagle locations that could be analysed. Points along plotted flight lines were selected in a way that promoted randomness and independence, while enhancing sample size. To do this we randomly selected a single point along the flight lines for each 4 hour observation bout, then selected the sequence of points before and after SB 25 that random point that were separated from that point and from each other by at least 45 minutes. Observations of radiotagged, territory holding eagles in western Scotland suggested that they can fly from one end of their range to the other in < 15 minutes (McGrady unpublished data), so the 45 minute limit we set is a conservative estimate of the time needed to achieve independence between points. These randomly selected eagle locations were then used to produce maps of area use for the resident male eagle, for the resident female eagle, and for the eagles as a pair. Two representations of eagle range use were employed that used randomised point data: the minimum convex polygon (MCP) (Mohr 1947) and an adaptive kernel analysis set at 95 and 50% levels (Worton 1989). The MCP maps extent of the random location’s distribution and kernel analyses map likely use of areas by eagles based on the distribution of eagle locations over time. The 50% kernel predicts the centrally located area where eagles concentrate 50% of their time, and is used by us as a nominal “core area’. One to 6 observations of intruding eagles were made per year. These are not included in our analyses, but provide useful context for interpreting behaviour of the resident eagles. Grid analysis. The study area was overlaid with a grid that corresponded to the Ordnance Survey one km grid. We then measured the total length of flight lines recorded from our direct observations that occurred in each square. Total length of flight lines per grid square was then mapped and used as a measure of eagle use. We made comparisons of ranging before (prior to August 2000) and after (after January 2002) windfarm construction for the male, the female and the pair using the kernel analyses and the flight line information. By way of these comparisons we assessed the effect of the Beinn an Tuirc windfarm and the effects of the associated Scottish Birds (2005) tree felling and habitat management. Because data are from eagles within a single range, and likely to be the same individuals, robust statistical analyses could not be undertaken. Results A total of 776 observation hours were logged over 194 watches. Prior to construction 98 watches were made, during construction 17 watches, and after construction 79 watches. No eagles were seen during 60 of the watches. Golden eagle occupancy and breeding The home range was occupied throughout the study period, apparently by the same 2 adult eagles. The eagles used a different nest in each year until 2003 when that of 1998 was reused. The eagles laid 2 eggs each year except 2003, when a single egg was laid. A single juvenile was fledged in 1997. During the study period, produc- tivity was 0.125 young per breeding attempt. Golden eagle ranging The maximum extent convex polygon in which eagles ranged covered 49.2 km’; the MCP covered 32.9 km? (n= 154). Thirty two percent of the Beinn an Tuirc windfarm was overlaid by Eagle behaviour before and after construction of a windfarm 3] maximum extent convex polygon and 28 % was overlaid by the MCP. The 95% kernel of eagle ranging covered 20.5 km’, and had 2 core areas (50% kernel) that were both outside the Beinn an Tuirc windfarm area and covered a combined area of 2.9 km? (Fig 2). The windfarm area was only overlapped by the 50-95% isopleth of kernel analyses of eagle ranging ie it was not included in the core area. Table 1 summarizes the areas of 95% and 50% kernels of eagle home ranging before and after construction and the amount of overlap between eagle ranging maps and _ the footprint of the Beinn an Tuirc windfarm. Eagle ranging kernels are illustrated in Figures 2-4. Three randomised locations of eagles (2.56% of all locations) were over the windfarm footprint, two (1.7%) were over turbines, and all of these were prior to construction. Additionally, 3 locations were within 500 m of the windfarm and 2 of these were prior to construction. Kernel areas for males were similar to those of females (Table 1). Also, for both sexes kernel areas were similar before and after windfarm construction, though the shape and spatial location of the ranges shifted, mostly east and north (Figures 5 and 6) after construction. Table 1 Areas (km?) within 50% and 95% kernels for eagles during the whole study period and before and after windfarm construction. Values in () are % of eagle range that overlap the windfarm. N 50% area kernel 50-95% kernel Total 95% kernel Male 97-04 66 3.0 (0) 17.8 (4.4) 20.8 (3.8) Male pre construction ps 6.1 (0) 19 3467) 25.4 (5:1) Male post construction 37 25.0) 15.0 (0.03) 17.3 (0.03) Female 97-04 88 4.9 (0) 20.8 (3.7) 25.1 G.0) Female pre construction 30 4.7 (0) 20.6 (8.9) ZOU) Female post construction 46 3.8 (0) 197 QA) 23.5 (2.0) All birds 97-04 154 3.2 (0) 20.9 (2.7) 24.1 (2.4) All birds pre construction a7 532 (0) 20.7 (9.0) 25:9°C7.2) All birds post construction 83 6.9 (0) 33.6 (0.5) 40.5 (0.4) 32 SB 25 Figure 5 Kernel analysis of ranging of resident male eagle before (left, n=27) and after (right, n=37) windfarm construction. Legend e = Turbines ® Randomised eagle locations Male eagle movement 1997-2004 eas 50% Kernel |__| 95% Kernel Scottish Birds (2005) Eagle behaviour before and after construction of a windfarm a3 Figure 6 Kernel analysis of ranging of resident female eagles before (left, n=57) and after (right, n=83) windfarm construction. Legend e Turbines @ Randomised eagle locations Female eagle movement 1997-2004 pesciced 50% Kernel ee 95% Kernel 34 A total of 811 flight paths were mapped. Only one eagle flight line was recorded at low to medium altitude (21-60 m) within the Beinn an Tuirc windfarm after construction and _ this passed between the 2 discrete clusters that comprise the windfarm. In that instance the nearby presence of an intruding eagle was almost certainly a contributing factor. No eagles have been seen within the turbine clusters. Two of 3 instances of eagles over flying the windfarm were when intruding eagles were in the area. Seventy seven percent of randomised locations were over open landcover types. The percentages of locations over different landcovers suggest the following ‘preference’ by the eagles: heather moor>treefell>grass hill>forest. Eighty percent of pre construction randomised locations were over open landcover types; the value was 79% for the post construction period. Regarding the area of forestry that was felled, 21.6% of random locations prior to felling (n=37), 3.1 % of random locations during felling (n=32), and 18.8% of random locations after felling (n=85) were within this area. Eagles flew 0.095 km over the forest area prior to felling per hour of observation and 0.285 km/hr of observation after felling, a three-fold increase in use. Figure 8 utilizes flight line data and shows relative use of different areas overall and proportion of use of each habitat polygon before and after tree felling. Over 70% of total eagle flight line length was over the central open area. Figure 8 illustrates that eagles shifted their ranging to the northeast after trees were felled. Discussion Impacts of windfarms on birds can include collisions (See Hunt et al 1999 and Hunt 2002) or loss of habitat (eg Leddy et al 1999). In this study, resident Golden Eagles appeared to avoid the windfarm within their home range except SB 25 when responding to intruders south and west of the centre of the territory. Studies exist that show that birds (eg Osborn et al 1998) including raptors (Curry and Kerlinger 1998) will try to avoid moving turbines. Physical accessibility does not seem to be what hinders eagle use of the windfarm. Turbines were separated by relatively large distances, larger than tree spacing in forested areas used by Golden Eagles (Tjernberg 1983), and the eagles we studied were seen hunting Willow Ptarmigan in open patches and rides within forestry smaller than those available within the windfarm (D Walker unpublished data). In combination with the fact that resident eagles continue to forage in areas comparatively close to the windfarm especially toward the centre of the range this suggests that eagles avoid the windfarm as a unit rather than individual turbines. While food densities are comparatively low within the windfarm footprint, current potential prey populations of Willow Ptarmigan, Common Snipe and sheep carrion (S Sheridan and D Walker, unpublished data) and previous use suggest that the eagles would still forage within the windfarm area if turbines were not in place. In particular, eagle foraging might be expected here at times of relatively high grouse availability, July-October, but this has not been recorded since construction. Also, the regular presence within the windfarm of corvids, upon which eagles prey, suggests that eagles may be excluded from the windfarm. Hooded Crows Corvus corone cornix are a comparatively common and easily taken prey species but appear to be safe from predation while within the farm. Rotor noise and movement or prey distri- bution, or any combination of these factors, may be influencing eagle movement. However, we had no impression that the windfarm was avoided less during periods when the turbines were not rotating (D Walker, unpublished data). Scottish Birds (2005) The kernel map of eagle ranging suggests that the windfarm may act as a barrier to some areas of the range for the eagles, however VP watches prior to construction did not suggest that the windfarm footprint was along any major transit route for the eagles. The management plan for this windfarm included activities that potentially would reduce risk of collision by reducing prey availability within the windfarm. In addition, the enhancement of other areas for eagle prey was seen as providing new feeding opportunities for eagles. According to the grid based analysis eagles did appear to more frequently use an area where trees were felled to improve foraging potential. The random point analysis did not show this, though low sample size in the pre felling period could have caused this. Willow Ptarmigan numbers have increased here (S Sheridan unpublished data) since felling, and use of the area by eagles may increase further as prey numbers recover from being limited by blanket forest and their availability increases. This may further reduce the relative attractiveness of the land within and around the windfarm to eagles. The relative use of different habitats by the eagles to some extent reflects their foraging potential. However, even within particular habitat types there can be variations in quality and prey carrying capacity. Still, so far the findings point to the Golden Eagles at Beinn an Turic being similar to eagles elsewhere and preferring open habitats to closed ones (McGrady 1997, McGrady et al 1997). In contrast, eagle use has increased in areas where managed tree felling occurred. The area where trees have been felled in mitigation of open ground lost to the windfarm notwith- standing, tree growth to canopy closure in other areas will restrict use by the eagles. McGrady et al (1997) show that eagles avoid areas of closed canopy forestry, probably because prey becomes less available. Eagle behaviour before and after construction of a windfarm 3) Our impression from direct observations of eagles and cursory examination of pellets suggest that the eagles’ most important food source is sheep carrion. It also appears that carrion availability varies spatially and temporally. Carrion hot spots are located in wet flushes on the eastern sloping open ground and the windfarm area, but there was no evidence of use of carrion within the windfarm area by eagles since construction. Most sheep carcasses are removed from the windfarm area when they are found, but some are not found and these have not been used by eagles (D Walker unpublished data). Carrion availability within the windfarm area has probably declined since construction. Rabbits, Willow Ptarmigan and Hooded Crows are the main live prey species we have recorded. This prey list is similar to that recorded for eagles elsewhere in western Scotland (Watson et al 1993). Increased human activity can influence eagle behaviour (including breeding and foraging behaviours) and productivity (Watson 1997), and in general, eagles tend to avoid human activity. We have no data to suggest that increased visitor pressure has caused the eagles to change. their ranging behaviour. Indeed, eagles did not go into the windfarm even when no people were there. However, we were unable to monitor eagle ranging at the site during construction when human activity was greatest because of access restrictions due to Foot and Mouth Disease. The windfarm is regularly visited by turbine technicians, shepherds and eagle project and other fieldworkers. None of these activities seem likely to cause reduced eagle use because they tend to be localised and relatively infrequent. It is possible that eagles are influenced more by human activity in artificial habitats (eg windfarms or newly felled forestry) than in natural habitats, but we know of no data to support this. 36 SB 25 Figure 7a Flight paths (left), and grid of relative use of km squares (right) by Golden Eagles (male and female) before windfarm construction at Beinn an Tuirc. Legend * Vantage points e Turbines Flight lines Mi Forestry Metres of flight line per km square | 0- 5000 5001 - 10000 WH 10001 - 20000 WH 20001 - 40000 He > 40001 See a si Scottish Birds (2005) Eagle behaviour before and after construction of a windfarm 37 Figure 7b Flight paths (left), and grid of relative use of km squares (right) by Golden Eagles (male and female) after windfarm construction at Beinn an Tuirc. Legend * Vantage points e = Turbines Flight lines ' i t i “t t ' i He i f t i i i fea, fi 38 Intruding eagles were mostly recorded outside the breeding season over the main area of open ground and the tree felled area (D Walker unpublished data). When detected, the resident pair routinely intercepted intruding birds, even when they were towards the fringe of their range, with interactions usually consisting of the resident pursuing the intruder, sometimes with apparently aggressive approaches. In general locations away from the territory centre were associated with territorial defence behaviour, especially by the male (eg Figure 5, western edge of left map), and these added greatly to the size of the range that we mapped. Different methods used to map = animal movements have different advantages and shortcomings (Kenward 1987). We present different mapped representations of the same data to partially overcome this problem. Also, although these data are from a single pair, the number of observations (811 flight lines) 1s large, is spread over different seasons over 7 years, and this lessens the impact of the shortcomings of the range mapping methods. Golden Eagle occupancy has not changed during the study period. Overall productivity of this range is 0.44 young per attempt (n=28, M Gregory, unpublished data), compared to an Argyll mean of 0.66 (1992, 96, 99-2004, Argyll Raptor Study Group, unpublished annual report 2004) and a Scottish mean of 0.52 (Watson 1997). Although productivity during the project was only 0.14 young per attempt, there is no evidence that links this low reproductive rate to windfarm construction or operation activities. Declines of this magnitude have been recorded in other ranges in Scotland where no windfarm, or indeed other change, has occurred, though we know of no published information that illustrates this. Rather, it seems that this home range has been relatively unproductive in recent years (only one chick since 1988), and this may be a result of the range viability already being SB 25 challenged by the expansion of forest (Watson et al 1987) and the impoverishment of the flora and fauna that has occurred (Thompson et al 1995). We have verified the presence of the adult territorial eagles every 2 weeks, and no eagles, territorial or non territorial, are known to have been killed by colliding with the turbines. There is no indication that the resident eagles have become accustomed to the windfarm area and are more likely to use it as time passes. It remains likely that any fledglings reared at the site, intruders, or new ‘naive’ replacement breeders are at greatest risk of collision. Because tree clearance roughly coincided with the construction of the windfarm, it is difficult to say to what extent eagles responded to the clearance rather than the windfarm. However, the avoidance of the windfarm — since construction suggests that the existence of relatively open areas within the windfarm is not sufficient motivation to attract eagles for foraging. Further, if the shift to the north east is a result of windfarm avoidance, then it suggests the eagles, at least at Beinn an Tuirc, ‘prefer’ recently felled forest areas to the windfarm. Interestingly, though there was an overall shift to the northeast, there was no real shift in the location of the core areas. These remained in the open area that has never been under forestry to the northeast of the windfarm between blocks of forestry. This result is likely influenced by the location of the nest sites, but supports the idea that these areas are particularly important. If this relative inflexibility in location of the core area is a feature of eagles elsewhere identifying the core area and protecting it may be particularly important. Guidance by Watson et al (1987) and modelling of eagle ranging (McGrady et al 1997, McLeod et al 2003a, 2003b) have established nominal core areas for eagles, but these are criticised as being too simplistic, and are a point of contention between developers, conservation organizations and government agencies. More Scottish Birds (2005) Eagle behaviour before and after construction of a windfarm 39 data are needed to clarify the impact of windfarms on eagles, and it would be useful if data collected at windfarm sites elsewhere in Scotland were made available for collective analyses. Acknowledgements The authors would like to thank Mike Gregory and the Argyll Raptor Study Group for use of data on eagle productivity. Scottish Power funded data collection. Saya Sheridan provided information on habitat and prey within the windfarm management area. Steph Carey-Miller of Natural Research (Projects) provided GIS support. The Central Kintyre Management Group has provided a forum for interesting discussions on windfarms and their effects on eagles, and encouraged us to publish the results. References British Wind Energy Association 2004. www. britishwindenergy.co.uk. Visited 3 Dec 2004. Curry R C & Kerlinger P 1998. Avian mitigation plan: Kenetech model wind turbines, Altamont Pass WRA, California. Proceedings of National Avian-Wind Power Planning Meeting Ill. San Diego, CA, USA. Gibbons D W, Reid J B & Chapman R A (eds.) 1993. The new atlas of breeding birds in Britain and Ireland: 1988-1991. T & AD Poyser. London. Hooge P N & Eichenlaub B 1997. Animal movement extension to Arcview. ver. 1.1. Alaska Science Center - Biological Science Office, U.S. Geological Survey, Anchorage, AK, USA. Hunt W G, Jackman R E, Hunt T L, Driscoll D E & Culp L 1999. A population study of Golden Eagles in the Altamont Pass wind resource area 1994-1997. Report to National Renewable Energy Laboratory, Subcontract XAT-6-16459-01. Predatory Bird Research Group, University of California, Santa Cruz. Hunt G 2002. Golden Eagles in a perilous landscape: predicting the effects of mitigation for wind turbine blade- strike mortality. California Energy Commission. Predatory Bird Research Group, University of California, Santa Cruz. Contract No. 500-97-4033. Kenward R 1987. Wildlife radio tagging. Academic Press. London. Leddy K L, Higgins K F & Naugle D E 1999. Effects of wind turbines on upland nesting birds in conservation reserve program grasslands. Wilson Bulletin 11: 100-104. McGrady M J 1997. Golden eagle. BWP Update 1(2): 99-114. McGrady M J, McLeod D M, Petty S M, Grant JR & Bainbridge I P 1997. Eagles and forestry. Forestry Commission Research Information Note. No. 292. HMSO. London. McLeod D R A, Whitfield D P, Fielding A H, Haworth P & McGrady M J 2003a. Predicting home range use by Golden Eagles Aquila chrysaetos in western Scotland. Avian Science 2: 183-198. McLeod D R A, Whitfield D P & McGrady M J 2003b. Improving prediction of Golden Eagle (Aquila chrysaetos) ranging in western Scotland using GIS and terrain modelling. Journal of Raptor Research 36(1 Supplement): 70-77. Mohr C O 1947. Table of equivalent populations of North American small mammals. American Midland Naturalist 37: 223-249. Newton I. 1979. Population ecology of raptors. T & A D Poyser, Berkhamstead. Ratcliffe D 1990. Birdlife of mountain and upland. Cambridge University Press. Cambridge. Thompson D B A, Hester A J & Usher M B 1995. Heaths and Moorland: Cultural Landscapes. HMSO, Edinburgh. Tjernberg M 1983. Habitat and nest site features of Golden Eagle Aguila chrysaetos (L.) in Sweden. Viltrevy 12(5): 131-163. Watson J 1997. The Golden Eagle. T & A D Poyser, London. Watson J, Langslow D R & Rae S R 1987. The impact of land-use changes on Golden Eagles in the Scottish Highlands. CSD Report No. 720. Nature Conservancy Council, Peterborough. 40 SB 25 Figure 8 Use of habitat by Golden Eagles at Beinn an Tuirc. Size of pie chart shows relative use of habitat polygons for the whole study period, dark portion is percentage use before tree felling, and hatched portion is percentage use after tree felling. Relative habitat use before and after tree felling ZA Post-felling MB Pre-felling @ Turbines HMB closed canopy - recent felling, including mitigation felling Watson J, Leitch A F & Rae S R 1993. The diet of Golden Eagles Aquila chrysaetos in Scotland. /bis 135:387-393. Worton B J 1987. Kernel method for estimating the utilization distribution in home range studies. Ecology 70: 164-168. D Walker ', M McGrady ', A McCluskie ', M Madders ', DRA McLeod ” ' Natural Research Ltd. Carnduncan, Bridgend, Isle of Islay PA44 7AS ? 14 Crailinghall Cottages, Jedburgh TP8 6LU M McGrady is the corresponding author Revised manuscript accepted February 2005 Scottish Birds (2005) 25: 41-49 The biology of a population of Willow Warblers in East Fife 4] The biology of a population of Willow Warblers in East Fife J LS COBB The breeding biology of a population of Willow Warblers in East Fife was investigated using colour ringing. Post fledging dispersal and return in following years is described. Nestlings fledge at 14 days old and are looked after by the parents for another 10 days. They then disperse locally for another 3 weeks before beginning a gradual migration south. The national ringing data is used to corroborate the findings in this study as well as to speculate on the further migration of this population. Nearly 200 nestlings have been found the following or later breeding seasons and the distribution of these returns is described. Introduction The population of Willow Warblers Phylloscopus trochilus in a wood in north east Fife has been studied from 1987 to 2004. The main aim of the study was to describe in detail the numbers, behaviour and the use of the habitat by this species throughout the breeding season in the entire wood. This data set would then allow the effectiveness of a Constant Effort Project carried out in about one fifth of the wood to be assessed (Cobb in preparation). A Constant Effort Project is a BTO scheme to use ringing to monitor populations. Study area and methods Kippo is an area of about 40 hectares of mixed woodland and moorland. There are significant gaps in the mixed wooded areas and hundreds of metres of wide rides in the conifers. Two artificial ponds have been created. Kippo Wood is entirely surrounded by arable land with little in the way of shrubs and trees for at least 1.5 km in all directions. Birds were also ringed in Redwells Wood 2 km south of Kippo and near Crail, 6 km SE. At the population peak, there were about 85 Willow Warbler territories packed into this area. The Willow Warblers in the Kippo population are almost all of the brown and white morph with little yellow except under wing. In autumn a higher proportion of the passage birds have much yellow colouration. In all 8,750 Willow Warblers have been handled during this period and all caught before the first week in August have been individually colour ringed. Observations and ringing were carried out daily for much of the duration of the study. About 80% of Willow Warbler territories over a 5 km radius were checked for singing males during May and 680 Willow Warblers were ringed at 2 sites — Redwell, 2 km south and Crail, 5 km south east - during the period 1994 to 2003. As many broods as possible were ringed after 1994, a total of 2,200 nestlings. Most nests were on the ground but a small number nested in Spruce trees, one 4m above the ground. Between 1994 and 1998 each brood carried a unique combination of colour rings but in 1999 - 2003 nestlings only received a single year colour but were individually colour ringed if subsequently retrapped after fledging. Birds were sexed on wing length (Svennson 1984) and adults, if appropriate, by cloacal protuberance or brood patch and in some cases by behaviour or song with targeted individuals. There has been an undoubted male with a wing length of 64 mm and at least 10 females with a 42 wing of 67 mm. Birds with wings of 65 and 66 were left unsexed unless other evidence was available. There is a discrepancy of the sex ratio in juveniles with apparently up to 30% more females. It seems most likely that changes in wing length in a proportion of recently independent young account for this apparent imbalance in the number of females caught and suggests that care needs to be exercised in sexing juvenile Willow Warblers until post juvenile moult is virtually complete. The national ringing recovery data up to 1996 was made available by the BTO and this was used in conjunction with the local study, in particular to try and assess onward movement of juvenile Willow Warblers during autumn migration. Results In all 727 nestlings were retrapped after fledging, some of them up to 4 times and 187 have been found on territory in subsequent years. Nestlings usually stay close to the nest for at least a week to 10 days after fledging normally at 14 days old. In rare cases, they may be moved by the parents into thick cover up to 200m from nests in exposed or vulnerable sites. They sit in thick cover and are usually fed by both parents. Some young leave the nest as early as 12 days old and can disperse on the ground where they are fed. At about 10 days after fledging the parents leave the young to fend for themselves and at this time they start post juvenile moult. Young birds begin a gradual dispersal away from the nest site (Figures | and 2). It may be that the direction of dispersal is partly dictated by the local topography but it appears largely random and in all directions. In Kippo, many of the catching sites are at the south end and this produces a strong bias in direction. Both nestlings and recently fledged young were also ringed at Redwells, 2 km south of Kippo. Some of these birds were found in Kippo during SB 25 dispersal indicating that at least some birds do briefly move north. In mid Scotland this phase of local dispersal lasts for about 3 weeks when post juvenile moult is about half completed. They may during this time form flocks of various sizes and at times be mixed with other species - partic- ularly tits (Parus sp) and Goldcrests (Regulus proregulus). Figure 3 plots the movements of juvenile males and females separately. The rates and pattern of dispersal are very similar for both sexes. There is also a tendency for birds from the same brood to keep together during this period (Figure 4). Two sets of 2 birds from the same broods were still together 3 weeks after fledging 6 km away. Analysis of the national ringing data showed that there are few records of young Willow Warblers moving more than 10 km to the north anywhere in the UK in July. Between 45 and 50 days from hatching (31 to 36 days from fledging) local Willow Warblers begin a gradual movement southwards starting in the last week of July. Figure 5 plots the hatching dates for all Willow Warblers. The peak is about 6 June and these start to leave the area on average 45 days later - about 21 July. There are differences between years of up to 10 days in average hatching date and replacement broods up fledge up to 10 July (Figure 6). At Kippo 2 females deserted their broods and nested again leaving the initial rearing of the fledglings to the male alone. Only 3 other second broods have been found, these fledged in the last week of July. Observation and ringing have revealed Willow Warblers moving in late afternoon across country on a broad front at a relatively low level rather than at heights associated with long distance migration. Each year in the last week of July and for at least the first half of August this movement is particularly obvious in open arable farm land in this part of Fife when birds arrive at isolated patches of vegetation. At nearby Fife Ness where daily observations were made Scottish Birds (2005) pullus dispersal n=556 OOOO werner oer in 8000 TOO0 6000 S000 4000 3000 distance in kilometers 2000 1000 0 days from fledging Figure 1 Distance of dispersing of fledglings from the nest. Up to 31 days (45 days from hatching) nearly all records are less than 1500m which means they were captured within Kippo wood. After about 35 days (49 from hatching) captures increase outside Kippo (more than 1.5 km) and the number of records in Kippo drops away. male data n=293 av daily distance=29.7 m 40000 8000 o¢ ¢ » 6000 + 2 4000 2000 0 a) 20 40 60 time from fledging The biology of a population of Willow Warblers in East Fife 43 numbers of pullus n=525 40 number of individuals Hl ae i | f f in WU og Bole ello a 24 28 31 34 37 40 43 46 49 52 45 days from fledging Figure 2 Captures of fledglings during dispersal. In the first 10 days after fledging nestlings remain fairly sedentary. At about 31 days after fledging (45 after hatching) the numbers again start to drop away as the birds leave Kippo to begin migration. female data n=314 av daily distance=30.3 m 8000 6000 distance a Lone) [em] © time from fledging Figure 3 a-b Rates of dispersal compared between young male and young female Willow Warblers. Both the rate of dispersal in metres per day and the pattern of dispersal are very similar. Records over 1500m are outside Kippo. during August there was no evidence of ‘coasting’ and Willow Warblers are scarce. The Isle of May, 7 km off the coast, does record larger numbers of Willow Warblers than Fife Ness during August. These are also likely to be Scottish birds as they are not associated with weather conditions that produce drifted continental birds. The regular occurrence of these birds suggests a daily southwards short distance movement of Scottish Willow Warblers. Nearly all birds caught during these movements have been juveniles. By the beginning of August most locally bred Willow Warblers have left Fife. By then most of the adults have completed a full moult and are also leaving (Figure 7). Very few adults known to have bred locally are caught during August. 44 sibling dispersal numbers ine caught together/ total pullus caught 6 9 12 15 #18 24 24 27 30 3 day periods from hatching 33. 363942 Figure 4 Significant numbers of siblings are caught together during the post fledging dispersal period. The proportion and pattern of siblings caught together mirrors the accompanying plot of all pullus caught. Willow Warblers in Kippo during August are likely to be largely from farther north; though national recoveries are few they confirm the more northerly origins of these birds. Birds move in the late afternoon and in some years form very large feeding flocks with other species during the morning. Large numbers of juveniles in these flocks were dye marked. However, these birds were never resighted nor were ringed birds caught on subsequent days. No birds trapped at Kippo in August or September showed any significant accumulation of fat. Onward movement evaluated from the national data This study has produced only 2 long distance recoveries of birds in autumn. The national data was used to investigate the early part of autumn migration. All juveniles and nestlings ringed in the UK up to 1996 recovered in the same year were used. There is good evidence from this study that juvenile Willow Warblers in Scotland do not migrate until about 45 days from hatching. The hatching date in the Kippo study averaged 6 June which gives a departure date of 21 July. This formula has been applied to all birds ringed north of a line from Cheshire to the SB 25 numbers ”. 5 go a & © 2 © Pee Ws -) " “NQ = ag oth ae Pe SM SSM iE RE Re DANI ANE EONS OP OP i et lr on “S fe oy vy vf c 3 day dates Figure 5 Average hatching dates over the years of the study. The peak date is 6 June and the earliest recorded is 25 May. Replacement broods occur throughout the rest of June and into the first week in July. Not shown are 5 second broods at the end of July. Wash but south of the Highlands. For birds ringed south of this a departure date of 14 July has been used and for birds ringed in the Highlands of Scotland a date of 28 July. Nestlings were assumed to be 8 days old at ringing. Using these approximate departure dates instead of the date of ringing, rates of movement were found of 17 km/day for the southern birds, 20 km/day for south of Scotland - northern England birds and 27 km/day for Highland birds. Two nestlings from Kippo were controlled in the same autumn from the south coast; applying the formula for middle UK to these Fife birds gives a daily movement of 19.5 and 20 km a day - which fits very nicely! Return in spring There were 187 recoveries of birds ringed as nestlings in Kippo returning the following year (106 males and 81 females). A further 91 juveniles ringed in Kippo before the end of July, and thus likely to be local birds, were also found in subsequent years breeding in Kippo. In all 170 of the nestlings returned to Kippo itself and 17 to sites up to 9 km away. Over 6 years territories were examined over a radius of up to 9 km and 830 territorial males were located, of these 9 Scottish Birds (2005) The biology of a population of Willow Warblers in East Fife 45 60 RN RR os = > oe S vw wy! a Ss > eo can RU Rm gm i QA AB AB” AB” y oe dates Figure 6 The hatching dates for 3 successive years to show variability. The peak in 1997 was exceptionally early at the end of May and 1996 was a year when the females arrived really late and the peak was not until the 9 June. 1995 was a more typical year. Presumably dispersal in any year is related to fledging date. primary wing moult average moult score Sth June 11th June ¥ 17th June 23rd June Figure 7 Moult in adult Willow Warblers. Moult is scored 1-5 for each of the 9 large primary feathers. gate I is in pin with no feather showing and 5 is complete with no further growth likely. 45 thus represents complete moult. Males start moult before females and most have virtually finished by the beginning of August when they leave. Female moult is later and many females replacing lost broods delay the moult as do the females that produce rare second broods. 29th June Sth July 1ith July 17th July 2ord July 23th July 4th Aug 9 10th Aug 16th Aug 22nd Aug Sth Aug 31 46 were Kippo nestlings; 189 adult females were caught over the same area during the breeding season with active brood patches and 6 of these were nestlings from Kippo. A juvenile ringed in Kippo when recently fledged was recovered the following year as a breeding female 9 km away. A juvenile male in Redwells on 22 July was controlled the following July at Loch Leven, presumably breeding, and until 2004 was the only record of a bird that might be breeding well away from its natal site. However it is possible from the date first caught that this bird had dispersed relatively early from the Loch Leven area when caught. The national ringing data In an analysis of the national ringing data it was assumed that any adult bird is on territory between | June and 15 July in any year, the 6 week period is deliberately conservative. It is true that many caught in May and the second half of July were also on breeding territories. Adults that have been ringed in this 6 week period and found in the same period in subsequent years show almost no evidence that they move sites in later years. Birds initially ringed in the nest or juveniles ringed before they start migration (birds ringed before 15 July south of a line from Cheshire to Wash, before 21 July for regions between this line and south of the Grampian Mountains and 28 July for further north) and found in the 6 week period in subsequent years, show virtually no evidence of a return to anywhere other than a few kilometres of their natal site in subsequent breeding seasons. Return of Kippo nestlings In all 68 of the nestlings that returned to Kippo the following year to breed had also been caught after fledging as free flying juveniles. It is therefore possible to compare the distance between the nest and the juvenile capture site SB 25 and the nest and the adult territory. This shows that there is a statistically significant correlation between juvenile capture site and its adult territory; this is the subject of a separate paper (Cobb and Gil in preparation.) A plot of return sites within Kippo for both males and females shows adult territories are evenly distributed between 100m and the maximum possible within the wood of about 1,500m from the natal site, (Figure 8). There are 10 males and 8 females (out of 170 nestlings returning) that came back to hold territory within 200m or less of their birth site. There have been no recorded cases of parent sibling incest. With the exception of a single female in this study all adult females and adult males have always returned to their previous territory. The same pairs have been recorded breeding in successive years. There are also a number of multiple returns of broods from the same nest. There have been 2 brother sister pairings; both successfully fledged young. Two sets of 3 brothers returned to exactly the same area; in one set the 3 arrived back to exactly the same song post over 3 successive days; they were all immediately displaced by up to 500m by the old male in the area they had adopted. A sister of this trio bred 200m away. The other trio of males adopted an area in the wood that was only just becoming suitable for Willow Warblers and successfully raised broods in adjacent territories that all fitted into an area of 200 square metres. These 3 males in adjoining territories was initially very confusing since they all carried the same colour ring combination. All 3 broods were exactly synchronized and in a newly colonized area of the wood. Four females from another brood returned and were spaced out over a linear distance of 1 km. Discussion There have been a number of previous studies of Willow Warblers in the UK. There was an early Scottish Birds (2005) The biology of a population of Willow Warblers in East Fife 47 distance from nest of pullus returning 12 male n=82 number of individuals female n=63 oOo o co Oo &2o & © oo ©. 2. & ©C © ff rn Oo =—- “| (TO ase WG distance in metres Figure 8 All nestlings returning to breed in subsequent seasons in Kippo up to 2001. The distance is metres between nest and adult territory. The maximum distance possible within Kippo is 1500m. Note the distribution of males and females is similar. study in Midlothian by Brock (1910) and later also in a similar area by da Prato (1982, 84) as well as a long term study by Lawn in Surrey (Lawn 1982, 1984, 1994 and 1998) and another by Norman in Cleveland (1981, 1983, 1990). Much of the basic biology described in these studies was also found in Fife, however they show the importance of basing current understanding of the biology of a particular species on a number of studies. There are significant differences in detail in these studies relating to habitat, geographical location and particularly to population levels. There is an estimate of 5% to 20% double broods based on BTO nest record cards (Cramp 1955) and the work of Brock (1910) in Lothian and May 1949 in southeast England. da Prato (1982) concluded that these studies may have misinterpreted the evidence and that double broods are rare; my own observations of just 5 over the whole study support this and it requires unequivoval evidence from marked females to prove double brooding. This will be discussed more fully in a later paper (Cobb in preparation). Once abandoned the young then gradually disperse. This may be in any direction but the national data (Cobb in preparation) suggests that it is never more than 10-15 km to the north. It has been suggested that post fledging dispersal is a time when the young scout out territories for the following year (see Berthold 1999). The data presented here suggests that at least initially broods tend to do 48 this dispersal together. This implies that they would return the following year to this same area with a risk of inbreeding between siblings. The fact that 2 such matings have been found may confirm this. However DNA fingerprinting studies (Gil 1998) in Kippo have shown substantial infidelity between some pairs and this may reduce the genetic significance of apparent incest. Norman showed in his studies of dispersal (1981) that females moved farther and faster then males and implied that this reduced the risk of inbreeding and that females were less inclined to return to the area of natal territories. This study has not shown such difference (Figure 8), but it also showed that, like the Cleveland and Midlothian populations, juveniles start their movement away from the natal site before the post fledging moult is complete. Lawn (1984) in the south of Britain found that this movement was not made until moult was completed. There is however at least a 10 day difference between mean fledging dates between the north of Britain and the south. Large flocks may form in August, feeding with other species, mainly tits but Goldcrests, Treecreepers Certhia familiaris and other warblers as well has been well documented, see da Prato (1981) and Simms (1985). Birds caught from these flocks are not resighted or recaptured and this conforms with the idea that there is a gradual daily movements south once they have achieved sufficient food for a short flight and an overnight roost. National ringing data shows reverse migration to be very rare in Willow Warblers so birds during August in Kippo are not likely to be birds from farther south. They are likely to be from farther north in the United Kingdom since there is almost no evidence for a normal migration route for Scandinavian Willow Warblers through Scotland. The high return rate found at Kippo may be influenced by the fact that it is an island of suitable habitat surrounded by open ground. SB 25 Studies in Norway have shown that in continuous large tracts of habitat returning birds may be less faithful to their immediate natal site (Cuadrado and Hasselquist 1994). However even in Kippo 2 factors suggest that a proportion of birds do not return to their immediate natal area. Every year a number of unringed male birds were found holding territory or breeding females found, as most Kippo nestlings are ringed these must be outsiders. Just 10% of ringed nestling birds have been recovered returning to Kippo. This figure is not compatible with a self sustaining population and implies a significant percentage must return to other sites. In 2004 a nestling from Kippo was trapped as an adult male in the breeding season north of Inverness. Such a movement is almost unprecedented within the UK ringing scheme for a Willow Warbler as most records show that birds return to within a few kilometres of their natal sites. When using ringing data to study migration one cannot be certain that a bird leaves immediately it has been ringed or that it has just arrived where it is recovered. Within Kippo, however, non resident juveniles are very rarely retrapped the following day or subsequently. In case birds had learnt to avoid capture after being initially caught they were dye marked but none of these birds was ever resighted. This implies that migration in the northern UK is continual from the end of July to mid August and in this region at least seems to occur late every afternoon. Over the last 6 years the weights and interclav- icular regions of a small percentage of birds caught in the reedbeds at another site in Fife (Kilconquhar Loch) and at Loch Leven in Kinross (Alan Lauder, personal communication) have shown evidence of substantial fat accumu- lation in late August and early September. It is possible therefore that either a small sub population of Scottish birds fatten up and then Scottish Birds (2005) make (long - 1000 km plus) night time flights to the south or that these heavy birds are Scandinavian migrants. Much more data is required before any sort of conclusion can be drawn about these possibilities, or indeed others. The occurrence of 2 different colour morphs in Scotland may be natural variation but the brown and white morph is more similar to some Scandinavian populations than that of the southern United Kingdom and this may be linked to migration strategies. Norman (1987) showed that normal weights allowed for nocturnal flights of a few hundred kilometres. Migration in Willow Warblers leaving the UK may consist initially of movements of a few tens of kilometres, particularly in Scotland, each afternoon, then possibly nocturnal flights of a hundred or so kilometres before they accumulate a fat loading to make very long flights into Africa (see Norman 1987). Migration north in spring seems to be rapid and nocturnal with new birds often on territory soon after dawn. Acknowledgements The author is indebted to the Cheape family of Strathtyrum for unrestricted access to Kippo wood and also to Patrick Bowden-Smith for access to Kippo moor. References Berthold P 1999. Bird Migration. Oxford University Press, Oxford. Brock S E 1910. The Willow Wrens of a Lothian wood. Zoologist 4(14):401-417. Cuadrado M & Hasselquist D 1994. Age related return of Willow Warblers Phylloscopus trochilus at two breeding sites in Sweden Ornis Fennica 71: 137-143. da Prato S R D 1981. Warblers in mixed passerine flocks in summer. British Birds 74: 9132515, da Prato S R D 1982. Polygamy in willow warblers. British Birds 75:406-411. The biology of a population of Willow Warblers in East Fife 49 da Prato S R D 1984. Polygamy by willow warblers. British Birds 77:622-623. Gil D 1998. Ph.D Thesis University of St Andrews. Lawn M R 1982. Pairing systems and site tenacity of the Willow Warbler Phylloscopus trochilus in southern’ England. Ornis Scandinavica 13: 193-199. Lawn M R 1984. Pre-migratory dispersal of juvenile Willow Warblers Phylloscopus trochilus in Southern England. Ringing and Migration 5:125-131. Lawn M R 1994. Site fidelity and annual survival of territorial male Willow Warblers Phylloscopus trochilus at four adjacent sites in Surrey. Ringing and Migration 15: 1-7. Lawn M R 1998. Arrival patterns of male Willow Warblers Phylloscopus trochilus in relation to age, territory density and pairing success. Ringing and Migration 19:31-38. Norman S C 1981. A study of post juvenile moult in willow warblers. Ringing and Migration 3:165-172. Norman S C 1983. Variations in wing-lengths in Willow Warblers in relation to Age, Sex and Season. Ringing and Migration 4: 269-274. Norman S C 1987. Body weights of Willow Warblers during autumn migration within Britain. Ringing and Migration 8:73-82. Norman S C 1990. Factors influencing the onset of post nuptial moult in Willow Warblers Phylloscopus trochilus. Ringing and Migration 11: 90-100. Simms E 1985. British Warblers. Collins. Svennson L 1984. Identification guide to European passerines BTO, Tring. J LS Cobb, 3 Station Road, Kingsbarns, Fife, KY16 8TB Revised manuscript accepted February 2005 50 SB 25 SHORT NOTES Bird numbers in an Aberdeenshire glen (1987-2004) We described counts of birds in upper Feughside in 1987-99. (1999, Scottish Birds 20:81-93). This note summarises counts from 5 more years (Table 1) and tests previous conclusions with a longer run of data on bird numbers and local weather. The year 2004 was the final year of the study in its present form. As before, DJ did the fieldwork and AW analysed the data. The winter of 1986-87, preceding the start of observations, was very cold, followed by low numbers of several species in the first year of study. Over 1987-99, 10 species increased signif- icantly, 4 species declined significantly, and snow or frost accounted well for year to year proportionate change in numbers of 5 species. For 1987-2004, correlations between the numbers of birds and the year show statistically significant increases with the year in 15 species and significant decreases in 5 species now including Eurasian Curlew (Table 2). Proportionate change in the numbers of Eurasian Oystercatcher, Winter Wren, European Robin, Stonechat and Mistle Thrush, as before, was negatively correlated with winter weather and the list now also includes Grey Wagtail, Goldcrest and Great Tit (Table 3). Goosanders are now scarce, associated with regular cooperative shooting in autumn and winter on much of the Dee catchment. In 2003, numbers of lekking Blackcocks were similar to the previous mean, but were lower in 2004 (Table 1). Blackcocks now use 2 main lek sites instead of one, with only the original lek site on the study area, and one-2 additional cocks often display solitarily elsewhere. Numbers of Eurasian Oystercatchers and Northern Lapwings remain high, though Eurasian Oystercatchers rear few young, and perhaps at least one third of the adult Eurasian Oystercatchers may not attempt to breed or lose their eggs as soon as they are laid. Northern Lapwings have bred well and July flocks numbered over 70 in 2002 and 2003. In 2003 and 2004, Black-headed Gulls failed to breed, in 2003 associated with spring drought, and their nesting area was deserted in 2004. Mew Gulls stopped breeding on the study area in 1995, though they still nest on nearby hilltops. Sand Martins dug nestholes (7 holes) in a roadside bank in the mid glen in the late 1990s and then moved to a quarry, reaching a maximum of about 40 pairs in 2000. Subsequently, the colony declined (Table 1) and no birds nested in 2004 though a few were seen nearby. Numbers of European Robins and Hedge Accentors increased in 2000-2003 above their former means but decreased in 2003-2004. Yellowhammers are now absent from Ballochan farm although land use there has stayed much the same, but Common Linnets have successfully colonized and European Goldfinches, formerly almost unknown, probably bred in 2004. An analysis of Breeding Bird Surveys in the UK has been published (MJ Raven, DG Noble & SR Baillie, (2004), BTO News 254, 10-13) covering 1994-2003. We compared this with Feughside data for the same period. The 2 sets of statistical analyses are not directly comparable because of far bigger sampling for the UK. With this proviso, 7 species (Pied Wagtail, Grey Wagtail, European Robin, Song Thrush, Common Blackbird, Chaffinch and Yellowhammer) showed significant changes in the same direction in Feughside and the UK, and only one different (Sand Martin). Acknowledgements. We thank Mr A Dykes of Glen Dye Estate for the recent weather data, and DJ is grateful to the Demigryphon and Birse Community Trusts for grants towards his expenses. Scottish Birds (2005) Short Notes ay Table i Annual total number of territories, based on number of singing passerines, lekking Blackcock, duck and wader pairs, and nest sites in 2 gull colonies to nearest 5. Mean for 1987-99 2000 01 02 03 04 Mallard 5 3 l 0 l 0 Goosander 5 0) 0 3 0 2 Black Grouse 23 16 12+ 20 23 18 Eurasian Oystercatcher 11 10 20 Nal 19 14 Northern Lapwing 14 17 pA 22 19 19 Eurasian Curlew 14 12 11 14 12 10 Common Sandpiper 1 2 3 2 bs Black-headed Gull 0 80 30 2) 15 0 Mew Gull iw 0 0 0 0 0 Common Cuckoo ©) l l 3 l 3 Sand Martin 0 40 20 20 10 0 Tree Pipit Z 0 0 4 l 0 Meadow Pipit 10* pa 28 32 32 25 Grey Wagtail 8 13 13 16 16 13 Pied Wagtail 8 16 ibe 15 iS 13 White-throated Dipper 6 2) 4 5 6 6 Winter Wren 64 99 25 53 71 fal Hedge Accentor 14 28 17 23 pp) i) European Robin 43 71 60 53 76 Si Common Redstart 6 4 5 0 0 1 Whinchat 12 18 13 16 pI 16 Stonechat 2) 5 l 0 2(0)** 4 Common Blackbird 4 4 6 6 g 8 Song Thrush 16 20 22 20 25 19 Mistle Thrush 10 8 6 12 12 7 Willow Warbler 66 87 97 84 74 oy Goldcrest “ 4 10 7 5 11 8 Spotted Flycatcher 8 10 13 5 7 | Long-tailed Tit 3 3 l 0 Z Coal Tit 4 9 17 p48 | 18 25 16 Blue Tit 13 15 19 18 17 12 Great Tit 13 13 10 17 14 18 Eurasian Treecreeper 10 15 il 14 16 o Chaffinch 7) 92 92 96 94 91 European Greenfinch “ Zz 3 2 é 2 3 Common Linnet l 12 7 5 15 5 Yellowhammer 5 0 0 0 0 0) * counts in only 7 years, + birds counted at only one lek (see text), ** Two Stonechats singing early in 2003 disappeared when their breeding sites were burned, “ See Jenkins & Watson 1999 Table 1, p.84; This paper gives scientific names omitted from this update to save space. 52 SB 25 Table 2 Spearman rank correlations between year to year proportionate changes in the numbers of territorial birds counted and the year, over 1987 — 2004 (- indicates decrease). Species Correlation Significance Mew Gull - 0.88 sera Goosander - 0.86 ee Yellowhammer - 0.82 ea Common Cuckoo - 0.78 cam Eurasian Curlew - 0.52 * Common Blackbird 0.42 ns Song Thrush 0.43 ns Grey Wagtail 0.43 ns Hedge Accentor 0.53 ‘i Great Tit 0.54 ss European Robin 0.60 = Whinchat 0.62 et Sand Martin 0.64 = Eurasian Treecreeper 0.67 ae Blue Tit 0.71 se Goldcrest O72 = Chaffinch 0.74 oe Common Linnet 0.76 oe Willow Warbler 0.76 ee Coal Tit 0.79 = European Greenfinch 0.80 cae Meadow Pipit 0.82 - Pied Wagtail 0.90 ae n = 18 except for Meadow Pipit where n = 12, * P<0.05, ** P<0.01, *** P<0.001 Table 3 Spearman rank correlation coefficients between proportionate year to year change in numbers and weather in the intervening winter, November to March inclusive. Total cumulative Number of Largest number of | Cumulative (cm) snow depth mornings with successive mornings day degrees C per winter snow lying with snow lying below 0°C Eurasian Oystercatcher = 0.52% Grey Wagtail 20.90" Winter Wren - 0.60* a0 Fie =0.72** European Robin =. 59" Stonechat si, 59° =0.70"* - 0.67* Mistle Thrush =101922 Goldcrest - 0.60* - 0.56* Great Tit - 0.49% - 0.52" n = 17 except for Goldcrest where n = 13 and Stonechat where n = 14, * P<0.05, ** P<0.01 Scottish Birds (2005) Short Notes 5 David Jenkins, Department of Zoology, University of Aberdeen AB24 3TZ Adam Watson, c/o Centre for Ecology and Hydrology, Banchory AB31 4BY Unusually large roost of Common Sandpipers on the Solway In 1961, David Bannerman wrote of Common Sandpipers Actitis hypoleucos (The Birds of the British Isles Vol 10 p 21), “As we in Scotland are so used to meeting with the bird singly or in pairs, the idea of it being found in flocks like other small waders had not occurred to me.” Valerie Thom (Birds in Scotland 1986, p 202) described the species as “non gregarious in its habits at all seasons” and this would be confirmed by the experience of most birders. It was, therefore, a surprise when we found a roost of 83 Common Sandpipers at the mouth of the Annan River on the Inner Solway on 25 July 2003. We had gone down to the shore at dusk and it happened to be high tide. Suddenly, from the shingle beach at the water’s edge, sandpipers noisily took to the wing in conveniently small groups so that counting them in the gloaming was not difficult. We returned on 5 August 2003 and Barrow’s Goldeneye — a Hastings Rarity in Scotland Ford-Lindsay (1913; British Birds 7: 21-22) records 2 male Barrow’s Goldeneyes Bucephala islandica said to have been obtained at Scalloway, Shetland on 18 March 1913. Ford-Lindsay did not shoot these birds and the name of the person who did shoot them is not recorded. Ford-Lindsay wrote that they were sent from Scalloway to the taxidermist, George Bristow, of St Leonards-on- Sea, as Common Goldeneyes B clangula. Bristow allowed Ford-Lindsay to examine them; they Revised manuscript accepted November 2004 counted 28 at the same site, but it was earlier in the evening and the tide was lower — the birds were not in a flock, but scattered along the river bank and the beach. The local bird reports contain counts of up to 36 Common Sandpipers elsewhere on the Solway in July, particularly the River Nith, but the observers we have contacted report that their tallies were of scattered feeding birds, not roosts. Roosting in flocks has been recorded on the wintering grounds in Congo (Bannerman, op cit p 20), but according to The Birds of the Western Palearctic Vol 3, pp 598-599, migration in flocks of more than 20 is exceptional. It may be that the habit of forming a flock while on passage in Scotland is simply under recorded. Not many birders go in search of roosting waders at dusk in July. Richard and Barbara Mearns, Connansknowe, Kirkton, Dumfries DG1 1SX Revised manuscript accepted May 2004 were reidentified as Barrow’s Goldeneyes. Photographs were forwarded to the editors of British Birds and the record was published. However, the editors of British Birds effectively rejected the record: ‘We think that confirmatory records should be awaited before the species is fully admitted to the British List.’ The record is interesting because both Bristow and Ford-Lindsay were at the centre of the Hastings Rarities scandal. In 1962, as a result of an investigation led by the then editors of British Birds, several hundred records of rare birds, 54 including several British firsts, that were claimed from the Hastings area of East Sussex and nearby Kent between 1892 and 1930, were deleted from the record because of the high probability of fraud (Nelder, 1962; British Birds 55: 283-298. Nicholson and Ferguson-Lees, 1962; British Birds 55: 299-384). It appears that dead birds were being imported and then passed off as locally shot rarities. Many, probably upwards of 400 of these rarities, went through Bristow’s hands and Ford-Lindsay examined and recorded nearly 100 of the specimens, especially during the critical 1910-1916 period when the Hastings Rarities were at their peak. The disputed Hastings Rarities were nearly all said to have been seen and/or collected within 20 miles of Hastings itself but, apart from the location, the Scalloway Barrow’s Goldeneyes are typical in every way of the general pattern: it was a spring record of an extreme rarity; more than one bird was involved; the identity of the supposed collector was withheld; they were supposedly sent to the taxidermist, Bristow, as a common bird; finally he ‘reidentified’ them and got confirmation from one of a small band of local authorities (Ford-Lindsay, in this case). This is a Female Merlin hunting in her nest area In raptor species females that hunt near their nests depends on the availability of prey in the nest vicinity (Newton I 1979, Population Ecology of Raptors, Berkhamsted) but specific data on female Merlins Falco columbarius hunting in the immediate vicinity of the nest area is lacking. I have previously noticed that some female Merlins hunt within their nest area going back to 1973 but no systematic observations were made. An opportunity occurred in 2004 to study the hunting behaviour of a female for 100 hours from 22 June to 16 July at a nesting site in Galloway. All hunts SB 25 classic Hastings Rarity, albeit perhaps the only Scottish one, and should clearly be rejected. What is less clear is why the record was rejected at the time, when so many other records from the same people were accepted uncritically for a further 3 years, until Witherby called time on them in 1916 (British Birds 55: 299-384). The 1915 edition of the BOU British list (BOU, 1915; A List of British Birds. British Ornithologists’ Union. London) mentions the record in an Appendix, saying that the birds were probably imported from Iceland, but no reason for this supposition is given, nor whether the BOU List Committee thought this was a deliberate importation for purposes of fraud. Nevertheless, it is tempting to speculate that even by 1913, the Hastings team was attracting suspicion. For whatever reason, BOU and British Birds were unable to act against the established pattern of Sussex and Kent occurrences, but this unusual Scottish record was perhaps easier to reject. J Martin Collinson, 31 St Michael’s Road, Newtonhill, Stonehaven AB39 3RW Revised manuscript accepted November 2004 were carried out within a radius of only 100-200m from the nest site which was in deep heather. The female always used a low flight attack from fence posts. In June and July broods of Meadow Pipits Anthus pratensis had fledged so there was a surfeit of available prey. The young were about 14-15 days old on 22 June 2004 when I first saw the female hunting in the nest area. The female chased Meadow Pipits 7 times, finally capturing one by dropping into the heather and returning to the nest with it. On 28 June 2004 the male arrived with 6 food items in 3 hours for the young and the female cached 2 items. The next time the female was seen Scottish Birds (2005) Short Notes be) hunting was on 30 June 2004, when she hunted in the nest area 9 times and caught a Meadow Pipit which she fed to her young. On 8 other occasions she caught moths or butterflies (Lepidoptera/Papilionina) which she consumed herself. On 4 July 2004 the 2 young were 26-27 days old; they had fledged and were scattered in the heather. The female hunted over the nest area capturing a Meadow Pipit which she fed to the young and continued to hunt insects. On 6 July 2004 the female fed on cached prey. On 10 July 2004 the male and female hunted outwith the nest area and both delivered one prey item each for the young. After the female brought in prey, she hunted insects and ate them herself. On 13 July 2004 both female and male were still hunting away from the nest area and on 15 July only the male delivered 3 prey items for the young in 2 hours. On 16 July 2004 the young had moved about | km from their natal area. Twin embryos in the egg of a Northern Goshawk In 2002, I found evidence that a Northern Goshawk Accipiter gentilis laid one egg with twin embryos in a nest in southern Scotland. This pair formed part of a larger, long term study of this species in southern Scotland and northern England. The first visit was on 5 March 2002, when the nesting area was found to be occupied and fresh nesting material had been added to an old goshawk nest. On 22 April, the nest tree was climbed and the female was found to have been incubating 4 eggs. On 24 May, there were 2 recently hatched chicks and 2 eggs in the nest. On 11 June, the nest contained 5 chicks; 3 females and 2 males. One of the males was very small and on the last visit, on 24 June, it was found to be missing. The remaining 4 chicks were well feathered and later fledged successfully. Thus, the only explanation for the difference in clutch and brood sizes is that one egg Female Merlins do not make important contn- butions to food gathering (Palmer 1988, Handbook of North American Birds, vol 5, London) and in most studies the male supplies food for the young and for the female. The advantage of the female Merlin hunting in the vicinity of the nest area was not readily apparent in these observations, as the male appeared to be a good food provider. During these observations the female only caught 3 prey items in the nest area and ate cached prey that the male had provided and moths and butterflies that she had caught. Perhaps when day flying insects are available they may supplement the female’s diet (cf Watson 1979, Bird Study 26:253-258; Dickson 1983, Scottish Birds 12:194). R C Dickson, Lismore, New Luce, Newton Stewart, Dumfries & Galloway DG8& 0AJ Revised manuscript accepted September 2004 must have had twin embryos. This nesting area has been used for about 10 years and is one of the most productive sites in the whole study area, regularly producing 4 or 5 chicks annually. Twin embryos are rarely reported in wild birds’ eggs. Iam aware of only 2 published accounts; from a Peregrine Falcon Falco peregrinus in Greenland (Pattee et al 1984. Twin embryos in a Peregrine Falcon egg, Condor 86: 352-353) and from a Northern Goshawk in Britain (Petty & Anderson 1989. Egg measurements from a Northern Goshawk Accipiter gentilis gentilis including one abnormally large egg with twin embryos, Journal of Raptor Research 23:113-115). The latter report was from the same study area as the 2002 record above. Malcolm Henderson, Burnhaven, Dunshill Drive, Jedburgh, Roxburghshire TD8 6QW Revised manuscript accepted July 2004 56 Little Egret attacked by a Eurasian Sparrowhawk In the autumn and winter of 2004-05, 2 Little Egrets Egretta garzetta were present at Luce Bay, Galloway from August 2004 to February 2005. On 11 September 2004 at 1400 hrs I watched a Little Egret feeding in a tidal creek. At the same time a female Eurasian Sparrowhawk Accipiter nisus was hunting the saltmarsh. On reaching the creek the Sparrowhawk swooped directly on the Egret narrowly missing its head. The Egret flew up, circled and landed in another creek. It looked like an opportunistic attack by the Sparrowhawk considering its weight of 280+grams against the Little Egret’s weight of 400+grams, although hen Sparrowhawks regularly take birds more than twice their weight in Britain (Newton 1973, Studies of Sparrowhawks, British Birds 66:271-278). On 18 October 2004 a Common Buzzard Buteo Ospreys killing a Grey Heron On 6 July 2004 I was watching a female Osprey, Pandion haliaetus, on its nest with 2 chicks above a loch in West Perthshire. At 8.50 pm she took off, and flew directly across the loch, where she was joined in flight above the loch by another Osprey, presumably her mate, and then a third large bird, which appeared ponderous in flight and less mobile than the Ospreys. The Ospreys were swooping and climbing without entering the water. Suddenly the third bird hit the water. I trained my fieldscope on the floating bird, which proved to be a Grey Heron, Ardea cinerea. The Ospreys continued to dive at its head, until it keeled over, and lay on the surface, Observations on a Red-backed Shrike nest in Shetland Early June 2004 saw a notable influx of Red- backed Shrikes Lanius collurio into Shetland. SB 25 buteo soared directly above the 2 Little Egrets at their high tide roost on an islet but did not attack them. On 9 January 2005 a Common Raven Corvus corax landed beside one of the Egrets at a high tide roost. The Raven pecked the Egret’s flank with its bill enough to disturb the Egret which flew low across the bay. Despite weekly visits to the area no more attacks were seen. No mention was made in BWP that Little Egrets may experience attacks by raptors while Musgrove (2002, The non breeding status of the Little Egret in Britain, British Birds 95:62-80) considered that “there is no reason to suspect that Little Egrets may suffer significant predation in autumn and winter in Britain’. R C Dickson, Lismore, New Luce, Newton Stewart, Dumfries & Galloway DG8 OAJ. Revised manuscript accepted February 2005 wings flapping. Increasingly feeble wingbeats by the heron failed to raise it, and after a few minutes there was no further movement, only its grey back being visible, floating without a sign of life. It seems that the Ospreys had made a lethal strike, perhaps fearing the presence of a larger bird in their territory. BWP vol 2 p272 describes Ospreys defending their territory by flying at intruders and records an instance where another Osprey was killed. I have not found any reference to a heron being killed by Ospreys. John Allen, Tay Cottage, Morenish, Killin, Perthshire FK21 8TY Revised manuscript accepted August 2004 Several were heard singing or appeared to be paired. On 15 June, Martyn Jamieson and Robin Sutton, who were leading a tour to Shetland, informed us that they had seen a pair of Red- backed Shrikes carrying nest material. The pair Scottish Birds (2005) was quickly located and, when the male caught some prey, the female immediately began begging by wing shivering and squealing. She was then fed by the male. A little later the female flew off out of sight into a nearby field. She soon returned, carrying a large ball of wool, and disappeared into a nearby clump of bushes. When the male did the same thing a few minutes later, it was clear that they were attempting to breed. The necessary licences were obtained from SNH and a watch was kept on the nest which was, fortunately, easily viewed from distance. The pair bred in the garden of an uninhabited house. The garden, with an area of about 0.1 ha, is becoming overgrown with the rose Rosa rugosa, a few Fuchsias and Sycamores Acer pseudo- platanus. An occupied house was no more than 50 m away but otherwise the site was surrounded by largely unimproved pasture. The nest was found after the juveniles had fledged. It was in the middle of the largest patch of rose, about 20 cm below the top of the bush and just over one metre above the ground. It was a rather untidy structure and quite large, almost as big as that of a Common Blackbird Turdus merula. It was lined with dry grass; the wool was incorporated into the general structure and not used for lining. It is assumed that incubation began on 20 June, as the female was very rarely seen after this date. The male was seen on most days, usually within 50 m of the garden, but he occasionally wandered further afield. The area of the feeding territory was estimated as no more than 12 ha. If the male caught any small prey items they were usually eaten immediately, but larger items were taken to the nest and fed to the female. Initially, the male was very cautious and would perch near the nest before dropping into the bushes, but by the second week he would fly straight in. There were never more than 2 visits to the nest during a watch, which lasted up to 90 minutes. The female was seen 3 times during incubation. On 2 occasions she came off the nest after being fed by the male, Short Notes oy but she was back at the nest within a minute. On 6 July, both adults were alarm calling and very agitated; a visit to the garden flushed a Eurasian Sparrowhawk Accipter nisus which had been sat in a tree next to the nest! Otherwise, the adults kept very quiet and were never heard alarm calling on any other occasion during incubation. Even after the eggs hatched, the adults could be very elusive. If the female began incubation on 20 June then the eggs should have hatched on about 4 July. A few days after this date, the male began visiting the nest much more often, on one occasion visiting 8 times in 15 minutes, although the female was never seen during this period, not surprisingly, as she should brood the chicks for about 6 days after hatching (BWP). On 19 July, the male was taking food into a site a few metres from the nest, suggesting that fledglings had moved into nearby bushes. On 23 July, a short tailed juvenile was seen flying between the bushes and at least 3 juveniles were seen on 25 July. These dates suggest an estimated incubation time of 14-16 days and an estimated fledging period of 15-19 days. BWP gives figures of 14 (12-16) days for incubation and 14-15 (11-20) days for fledging. The young remained in the garden until 11 August, almost 3 weeks after they first left the nest. Four juveniles fledged, but one died after about 2 weeks and the corpse has been sent to the National Museums of Scotland. The juveniles could be very elusive and, although the adults would alarm call when a visit was made to the garden, they would stop after a short while. Both adults fed the juveniles after they left the nest, but towards the end of their stay only the female hunted while the male patrolled around the garden or sat with the juveniles on one of their favourite branches. Towards the end of the stay he was often singing as he sat with the juveniles. On 8 August, both adults fed the juveniles, but this was also the last day the female was seen. Just one juvenile remained in the garden on 12 August; the 58 others were seen with the male patrolling fences about 800 m away. On 13 August, all 3 juveniles were with the male at an area of trees about 500 m from the nest. This was the last time the family was seen together but one juvenile remained in this area until 29 August. Prey in Shetland is rather limited and most items caught were too small to identify. The commonest were bumblebees, almost all of these being queens of the local Shetland Bumblebee Bombus muscorum agricolae, but a few were White-tailed Bumblebees Bombus lucorum. Other prey items included bluebottles Calliphora, especially later as they became commoner, ground beetles Carabus, large hoverflies of the genera Syrphus and Eristalis, and an Angle Shades Phlogophora meticulosa caterpillar. There was no suggestion that a larder was ever built up but, given the weather and food supply, it is quite surprising that the pair managed to fledge as many as 3 juveniles. Ruff breeding in Shetland in 2003 On 20 June 2003, Pete Ellis was on the Mainland of Shetland observing an area of pools and marshland as part of a breeding wader survey. A wader flew in from the south and began to run about on an area of short, wet turf. It was a female Ruff Philomachus pugnax, a rather dark individual with many blackish crescents on its upperparts, breast and flanks. The bird was running around very quickly and feeding rapidly. This behaviour suggested that it could have come off a nest and was trying to feed as quickly as possible. He watched the bird for about half an hour, after which it flew back south about 50 metres, landed in an area of cotton grass Eriophorum angustifolium, crept into a tussock and disappeared. The observer approached the area and when he was about a metre from the site, the female fluttered from the tussock and flew off. There were 4 eggs, slightly larger and more rounded than a Redshank’s Tringa totanus, Olive green with numerous dark brown SB 25 There have been 13 confirmed breeding records in Scotland, all since 1977 (A Thorpe pers comm), with no breeding anywhere in Britain since 1999 (Ogilvie and the Rare Birds Breeding Panel 2004 Brit. Birds 492-536). There is only one other confirmed breeding record for Shetland involving a female and one fledged juvenile at Pool of Virkie in south Mainland from July until mid August 1990, but Henry Saxby recorded a female feeding 3 newly fledged young at Burrafirth on Unst on the extraordinarily early date of 9 June 1870; either the date or the identification must be in error (Pennington et al 2004. The Birds of Shetland Helm, London). M G Pennington, M Maher & M Smith 9 Daisy Park, Baltasound, Unst, Shetland ZE2 9EA Revised manuscript accepted November 2004 blotches. He quickly left the area and observed at a distance, from a car. The Ruff returned to its nest within 10 minutes. The incubation period of Ruff is given as 20-23 days (Cramp, S and Simmons, K E L 1983. The Birds of the Western Palearctic, Vol 3. Oxford) so, having obtained the appropriate licence from Scottish Natural Heritage, PME visited the nest again on 12 July 2003. Unfortunately the nest had failed, as it contained only one egg, with a second punctured egg lying outside; there was no sign of the Reeve. On 22 July 2003, over 60km from the first site, Dave Okill and his son Antony were in a remote area of Mainland Shetland monitoring the breeding success of Red-throated Divers Gavia stellata. As they walked past a loch, JDO noticed a wader drop in behind a clump of cotton grass. Adjacent to the loch was an area of low mounds covered in short heather and grass, with the Scottish Birds (2005) damp hollows between the mounds, either of short grass or thick beds of cotton grass. They walked over to the area and a female Ruff took off from behind the cotton grass and flew away. When it was about 100m away, the bird turned and flew straight back towards the observers. It passed very closely in front of them and called very softly. JDO could not recall hearing a Ruff call before. The bird landed at the rear of one of the small mounds, ran up the mound and watched from over the top, occasionally bobbing its head. As they backed away, the Reeve flew around them softly calling again before landing at the rear of another mound and again running to the top to watch them. This behaviour strongly suggested that the Reeve had chicks. They backed away further and the bird ran down the mound and started to feed in a damp hollow. Scanning the area, JDO saw a small downy wader chick with very long legs run across an open area and into a large dense area of cotton grass. They soon saw a second chick wander across the side of one of the mounds. AO found it, crouched a few centimetres down a Rabbit burrow. The chick was at least 5 days old and very active. It was like a pale Redshank chick but with very long, olive grey legs. The under parts were plain, very pale buff. The crown was buff with darker (almost black) stripes. It was plain buff around the eye with a darker streak on the lores. The pattern on the back and wings was buff with darker markings, these dark marks were finely specked with buff. Its measurements were: wing = 25mm, bill = 17mm, tarsus = 34mm and weight = 29g. Yellow-billed Diver in Chelmsford Museum - the earliest record for Scotland and fourth for Britain Short Notes 59 There could have been other chicks out of view. The Reeve did not appear to be concerned at their presence, in contrast to the reaction of many other species of waders when they have chicks. At no time was a male bird seen at either site. It is indicated that males often abandon the nesting areas whilst the female is sitting and it is suggested that mating may even take place on the northerly migration in the spring (Cramp S & Simmons K E L 1983. The Birds of the Western Palearctic, Vol 3. Oxford). These are the first records of Ruff breeding in Shetland (Pennington M G, Osborn K, Harvey P V, Riddington R, Okill J D, Ellis P M, & Heubeck M 2004. The Birds of Shetland. Helm, London). The Ruff is a common passage migrant through Shetland in the autumn, but is scarce in spring, so these breeding records were completely unexpected. Previous records of Ruff breeding or apparently breeding in Scotland include an agitated female seen in the inner Hebrides on 17 July 1977, a nest with 4 eggs found in Sutherland in 1980 (although no male was seen, the eggs did hatch) and breeding was also suspected in the outer Hebrides in the early 1980s (Thom, V M 1986. Birds in Scotland. T & A D Poyser). PM Ellis, Seaview, Sandwick, Shetland ZE2 9HP J D Okill, Heilinabretta, Trondra, Shetland ZEI OXL Revised manuscript accepted May 2004 In preparation for our new book on the Birds of Essex, Simon Wood, Russell Neave, Graham Ekins and I arranged a visit to Chelmsford Museum with Dr Tony Walentowicz, Keeper of Natural Sciences, in July 2000. I was particularly interested to see the divers there, having seen a 60 sketch of the bill of one of the labelled Great Northern Divers Gavia immer drawn in 1961 by the late Stan Hudgell (1935-2002). Stan had always thought the bird to be a Yellow-billed G adamsii and, on investigation, we confirmed the identification. The specimen (now in storage, catalogue number E13541) came from the collection of G P Hope, whose notes stated the specimen was taken at Aberdeen on 17 December 1891. Although mounted and cased individually, the specimen had no label. The collection details were, however, listed in the Museum’s catalogue. Newspaper taken from the base of the display proved to be from The Times of 5 February 1892, just a few weeks after the bird had been shot. Together with other circumstantial evidence, this convinced TW that the bird was the listed Aberdeen specimen. Two Great Northern Divers at one time in the Chelmsford collections had been collected in 1881 (11 years before the newspaper) and 1920 (28 years after the newspaper), and were thus eliminated. Fraud was ruled out as the specimen had been identified as a Great Northern Diver since the outset. Details were forwarded to the Scottish Birds Records Committee. As well as confirming the identification, the Committee also considered the bird’s authenticity, documentation and the possibility of fraud. The record was accepted in September 2003. I would encourage all authors of county avifaunas to visit and maintain good relations with their local natural history museums. Other Chelmsford Museum highlights include an immature male Baltimore Oriole Icterus galbula caught alive on Unst, Shetland, on 26 September 1890, long considered an escape, and only recently accepted as the first for Britain by the British Ornithologists’ Union’s Records Committee (British Ornithologists’ Union. 2003. British Ornithologists’ Union Records Committee: 29th Report October 2002. Ibis 145: 178-183.). SB 25 Nick Green, 40 Gloucester Crescent, Chelmsford, Essex CM1 4NG The British Birds Rarities Committee’s remit generally does not extend to pre 1950 records and both they and BOURC who have ultimate responsibility for pre 1950 rarities, amongst their other duties confirmed SBRC should adjudicate the record. We were supplied with a series of photographs of the specimen and, on first circulation, identifi- cation was agreed by SBRC as an adult Yellow- billed Diver, moulting from summer into winter plumage. However, as the bird was no longer directly labelled, we were unwilling at that stage to accept the record. It was clear that more information would be required to enable the Committee to make a judgement about the bird’s origin. Bob McGowan (SBRC Museum Consultant) and Dr Alan Knox (University of Aberdeen) looked into the case with Dr Walentowicz, and reported back to the Committee. Their investigation confirmed: ™ There was no indication of fraud as the bird had, until very recently, been labelled as a Great Northern Diver. ™@ The specimen was part of a significant and reputable collection of birds assembled by George Palmer Hope (Christy R M. 1890. The Birds of Essex. Durrant, Chelmsford.). Hope prepared and cased many or most of his own birds. Although Hope _ purchased specimens, he appears to have set the diver up himself, as the taxidermy and groundwork of the mount are consistent with Hope’s style. ™ Hope collected birds mainly in Essex and Suffolk, but he also obtained a number of specimens from Scotland. @ Other diver specimens in the Hope collection, or listed in the Museum’s documentation, were Satisfactorily accounted for. Scottish Birds (2005) ™ The date of the newspaper in the base of the mount is particularly supportive. @ Although Great Northern Divers were not common off Aberdeen in 1891, and naturalists often published details of the occurrence of individual birds, there is no record of one having been reported at that time. This is not of great concern, as there were many shooters in the area in the late 1800s, and few recorded anything that they killed. @ Whilst the locality for the diver is given as ‘Aberdeen’, the locality for some of Hope’s other Scottish specimens 1s a little imprecise. It is probable that ‘Aberdeen’ refers to the county of Aberdeen, rather than the city. On the basis of this information, SBRC recirculated the record and it was accepted Hybridisation between a Whooper and Mute Swan in Shetland Swans did not breed regularly in Shetland until towards the end of the last century (Pennington MG, Osborn K, Harvey PV, Riddington R, Okill JD, Ellis PM & Heubeck M 2004. The Birds of Shetland. Helm, London). In 1992, a pair of Mute Swans Cygnus olor nested on Tingwall Loch and since this pairing Mute Swans have bred at a number of sites on Mainland or nearby islands. Up to 7 pairs have bred in some years. Whooper Swans Cygnus cygnus began nesting in 1994 and up to 3 pairs have bred in any year since, again in different areas of Mainland. In 2003, a pair of Whoopers nested on a hill loch in the east Mainland and had a brood of 4 well grown chicks when the site was last checked. In 2004, this site was first checked on 29 April when a pair of Mute Swans and Greylag Geese Anser anser were present. The next check was on 24 May when a female Mute Swan was found sitting on a nest whilst a male Whooper was nearby standing Short Notes 61 unanimously. The first Scottish record of Yellow-billed Diver was hitherto regarded as one picked up dead at the head of Whiteness Voe, Shetland on 21 January 1946 (Baxter E V and Rintoul L J 1953. The Birds of Scotland. Oliver & Boyd, Edinburgh). The Aberdeen specimen was killed over half a century before this and thus becomes the first Scottish (and fourth British) record (Burn D M and Mather J R. 1974 The White-billed Diver in Britain. British Birds 67: 257-296.). We are grateful to Alan Knox and Tony Walentowicz for their assistance with this record. Ronald W Forrester, Secretary, SBRC The Gables, Eastlands Road, Rothesay, Isle of Bute PA20 9JZ Revised manuscript accepted July 2004 guard. On 12 June the 2 adult swans were closely accompanied by 2 small chicks no older than a week. It was not possible to approach the chicks but they were very pale in colour and significantly lighter than Mute Swan chicks. The site was next checked on 18 June after a period of cold, windy and wet weather and although the 2 adults were still present there was no sign of the chicks. Hybridisation in wildfowl is fairly common in captivity and not uncommon in the wild. However the ranges of Mute and Whooper Swans do not overlap in many places and mixed pairings between these 2 species are relatively uncommon (Cramp S & Simmons KEL 1977. The Birds of the Western Palearctic. Vol 1. Oxford). In some areas feral or injured Whooper Swans have paired with Mutes when the former injured birds have been unable to migrate back to their normal breeding areas. Both of the Shetland birds were able to fly and there is no suggestion that either was injured. M Brazil (2003. The Whooper Swan. Poyser, London) notes a number of mixed pairings in scattered sites through Europe and cites a pairing 62 SB 25 a “a ae eed : as ce me Oe me 2 ee ek Male Whooper Swan paired with female Mute Swan, 2004, East Mainland, Shetland. between a male Whooper and a female Mute Swan on Loch Corrib, County Galway, Ireland in 1972 which produced hybrid chicks; 20 years later, a mixed pair in France also produced chicks. Swans have not been introduced into Shetland and it is believed that the small breeding populations are of wild birds. This is the first time that these species have hybridised in Shetland. In the Western Isles there has been a single hybrid pair recorded, again the cob was the Whooper, the pen the Mute Swan and although a nest was built they did not produce chicks (C Spray pers comm). A captive Whooper cob paired with a Mute Swan pen in the English Midlands and produced 9 chicks over a period of Merlin hunting in a wood in winter In winter the distribution of hunting Merlins Falco columbarius in Britain is mainly confined to open habitats such as farmland, moorland and the coast, generally avoiding forests and woods (Cramp & Simmons 1980, The Birds of the Western Palearctic, vol 2, Oxford; Dickson 1988, Habitat preferences and prey of Merlins in winter. British Birds 81:269-274). Larry Dalziel 4 years, the chicks have subsequently escaped or been released and are now in the wild (B Coleman pers comm). It should be noted that in all these instances the male has been the Whooper and the female the Mute. It is likely that this is the first time that hybrid swans have been hatched in the wild in Scotland. J D Okill, Heilinabretta, Trondra, Shetland. ZE1 OXL, P R Fisher, Wildland Consultants Ltd, PO Box 7137, 99 Sala Street, Rotorua, New Zealand Revised manuscript accepted February 2005 On 5 January 1985 at 1410 hrs I saw a female or juvenile Merlin actively pursuing a Chaffinch Fringilla coelebs above pasture on low ground in Galloway. The birds made their way into a conifer plantation adjoining a deciduous wood and on entering the wood, the Merlin continued to chase the Chaffinch amongst the tree branches like a Eurasian Sparrowhawk Accipiter nisus. The falcon got near enough to clutch at its prey 26 times without success. The Chaffinch managed to escape by flying deeper into the wood and the Merlin eventually landed on the Scottish Birds (2005) top of a beech Fagus sylvatica. The Merlin flew away out of sight across pasture at 1415 hrs. It is uncommon for Merlins to follow prey into large woods in winter but they are known to penetrate willow scrub Salix near their roosts and on one occasion a Merlin emerged clutching a small bird in its talons (Dickson 1970, Bird predators at passerine roosts. British Birds 63:85-86). In Germany a male Merlin was seen chasing and following a Yellowhammer Embireza citrinella into a wood and a passing female Merlin then chased it through the trees OBITUARIES Roger Frank Durman (1944-2003) With his wife Penny, Roger came to Edinburgh in 1971 moving north from London to work as a chartered surveyor with the firm of Montagu Evans. He specialised in the field of retail planning and, as a partner in Montagu Evans, built a huge reputation as an expert witness particularly in relation to planning enquiries; he took part in over 100. Not only was he very highly regarded in professional circles he was also an excellent field ornithologist. As a ringer he was initially very involved with Bardsey Bird Observatory. He had a wide range of anecdotes from Bardsey usually involving ringing during “attractions” at the lighthouse. Following his involvement with Bardsey, Roger chaired the Bird Observatories Council from 1970-1974 and edited the Poyser book entitled Bird Observatories in Britain and Ireland published in 1976. During the 1970s Roger carried out a number of studies in the Pentland Hills particularly focusing on Ring Ouzels, work that was subsequently carried forward by other Lothian ringers. Roger had a great affection for the East Lothian coast and he and Penny spent many years at Peffermill not only in studying the resident breeding birds but also the migrants Short Notes/Obituary 63 (Brauning & Lichtner 1970, Gemeinsame Jagd zweier Merline. Vogelwelt 91:32). In Montana, USA a Merlin followed Bohemian Waxwings Bombycilla garrulus to some conifers and then chased the Waxwings through the trees like a Sharpshinned Hawk Accipiter stiatus (Servheen 1985, Notes on wintering Merlins in Western Montana. Raptor Research 19:97-99). R C Dickson, Lismore, New Luce, Newton Stewart, Dumfries & Galloway DG8 0AJ Revised manuscript accepted September 2004 which arrived on the coast during spring and autumn. I am sure he would have liked to have spent longer at Peffermill but work pressures usually impinged on spare time at weekends. Roger and Penny eventually moved to Drylawhill, East Linton. Barn Swallows nested in _ the outbuildings and he and Penny took delight in following the fortunes of different broods over the years. The last significant travel which Roger was able to undertake was to visit Spain with Penny to view a site where one of their Swallows had been controlled on its migration south from Scotland. Roger was a very modest and self effacing individual whose skills and talents were not instantly on show. His moderate manner did not, however, conceal his underlying authority and professionalism whether in the context of work or the natural world. The large turnout at his memorial service in East Linton, testified to the high regard he was held in by so many. And it was entirely typical that the funeral notice asked for colourful ties to be worn! Not only his family, his partners and his many friends but also the professional and ornithological world of Scotland give thanks for a remarkable man, who will live long in our memories. Tan Darling 64 Advice to contributors Authors should bear in mind that only a small proportion of the Scottish Birds readership are scientists and should aim to present their material concisely, interestingly and clearly. Unfamiliar technical terms and symbols should be avoided wherever possible and, if deemed essential, should be explained. Supporting statistics should be kept to a minimum. All papers and short notes are accepted on the understanding that they have not been offered for publication elsewhere and that they will be subject to editing. Papers will be acknowledged on receipt and are normally reviewed by at least 2 members of the editorial panel and, in most cases, also by an independent referee. They will normally be published in order of acceptance of fully revised manuscripts. The editor will be happy to advise authors on the preparation of papers. Reference should be made to the most recent issues Of Scottish Birds for guidance on style of presentation, use of capitals, form of references, etc. Papers should be typed on one side of the paper only, double spaced and with wide margins and of good quality; 2 copies are required and the author should also retain one. We are also happy to accept papers on disk or by email at: mail @the-soc.org.uk, stating the type of word processing package used. If at all possible please use Microsoft Word . Contact the Admin Officer on 01875 871330 for further information. SB 25 Headings should not be underlined, nor typed entirely in capitals. Scientific names in italics should normally follow the first text reference to each species unless all can be incorporated into a table. Names of birds should follow the official Scottish List (Scottish Birds 2001 Vol 22:33-49). Only single quotation marks should be used throughout. Numbers should be written as numerals except for one and the start of sentences. Avoid hyphens except where essential eg in bird names. Dates should be written: ...on 5 August 1991]...but not ...on the Sth... (if the name of the month does not follow). Please do not use headers, footers and page numbers. Please note that papers shorter than c700 words will normally be treated as short notes, where all references should be incorporated into the text, and not listed at the end, as in full papers. Tables, maps and diagrams should be designed to fit either a single column or the full page width. Tables should be self explanatory and headings should be kept as simple as possible, with footnotes used to provide extra details where necessary. Each table, graph or map should be on a separate sheet, and if on disc each table, graph, map etc should be on a separate document. Please do not insert tables, graphs and maps in the same document as the text. Maps and diagrams should be either good quality computer print out and in black and white (please do not use greyscale shading) or drawn in black ink , but suitable for reduction from their original size. Contact the Admin Officer on 01875 871330 for further details of how best to lay out tables, graphs, maps etc. The Scottish Ornithologists’ Club was formed in 1936 to encourage all aspects of ornithology in Scotland. It has local branches which meet in Aberdeen, Ayr, the Borders, Dumfries, Dundee, Edinburgh, Glasgow, Inverness, New Galloway, Orkney, St Andrews, Stirling, Stranraer and Thurso, each with its own programme of field meetings and winter lectures. The Waterston Library at the Club’s headquarters at The Scottish Birdwatching Resource Centre, Waterston House, Aberlady, East Lothian EH32 6PY is the most comprehensive ornithological library in Scotland and is available for reference 7 days a week. A selection of Scottish local bird reports is held at headquarters and may be purchased by mail order. Check out our website for more information about the SOC and other bird related organisations: www.the-soc.org.uk An annual issue of Scottish Birds is published each year, in June. Scottish Bird News is published quarterly in March, June, September and December and the Scottish Bird Report is annual. The Scottish Ornithologists’ Club publishes the annual Raptor Round Up on behalf of the Scottish Raptor Study Groups with grant aid from Scottish Natural Heritage. It is sent to all members. Bird names used in publications follow those used in the Scottish List (Scottish Birds Vol 22(1): 33-49). Copies of these reports may be purchased by non members on application to the SOC. SOC annual membership subscription rates Adult £ 25.00 Family (2 adults and any juniors living at the same address) £ 35.00 Junior (aged 17 or under) £ 10.00 Student (aged 18 to 24 and in full-time education) £ 10.00 Concession (Pensioner) £ 16.00 Concession (Unwaged) £ 16.00 Joint Concession £ 2100 Life £500.00 Life Family £700.00 Overseas supplement £ 10.00 All subscriptions may be paid by Direct Debit and Gift Aided. Subscriptions paid by Direct Debit greatly assist the Club. Please ask for an application form by telephoning the Admin Officer at the above address or visiting the club’s website at www.the-soc.org.uk SMITHSONIAN INSTITUTION LIBRARIES Scottish Birds MiNi Volume 25 June 2005 Contents Main Papers The 2002 census of the Mute Swan in Scotland - A W Brown & L M Brown 1 The eradication of Brown Rats from Handa Island, Sutherland - J Stoneman & B Zonfrillo M7 Resident Golden Eagle ranging behaviour before and after construction of a windfarm in Argyll - D Walker, M McGrady, A McCluskie, M Madders & DRA McLeod 24 The biology of a population of Willow Warblers in East Fife - J L S Cobb 4l Short Notes Bird numbers in an Aberdeenshire glen (1987-2004) - D Jenkins & A Watson 50 Unusually large roost of Common Sandpipers on the Solway - R & B Mearns 53 Barrow’s Goldeneye — a Hastings Rarity in Scotland - J M Collinson op) Female Merlin hunting in her nest area - R C Dickson 54 Twin embryos in the egg of a Northern Goshawk - M Henderson op) Little Egret attacked by a Eurasian Sparrowhawk - R C Dickson D6. Ospreys killing a Grey Heron - J Allen 56 Observations on a Red-backed Shrike nest in Shetland - M G Pennington, M Maher & M Smith 56 Ruff breeding in Shetland in 2003 - P M Ellis & J D Okill 58 Yellow-billed Diver in Chelmsford Museum - the earliest record for Scotland and fourth for Britain - N Green a9 Hybridisation between a Whooper and Mute Swan in Shetland - J D Okill & PR Fisher 61 Merlin hunting in a wood in winter - R C Dickson 62 Obituaries Roger Frank Durman (1944—2003) - J Darling 63 Advice to contributors 64 Front Cover Mute Swans at St. Cyrus, N E Scotland by Harry Scott Published by the Scottish Ornithologists’ Club, The Scottish Birdwatching Resource Centre, Waterston House, Aberlady, East Lothian EH32 OPY. © 2005 2G, 640 4 S43 Birds THE JOURNAL OF THE S¢ Vol 26 June 2006 Changes in numbers and distribution of waders in the Moray Firth, 1988-2003 The status of the Northern Gannet in Scotland in 2003-04 Waterfowl counts on the Tay Estuary Scottish Birds — The Journal of the SOC Editor: Dr Stan da Prato. y Assisted by: Dr I Bainbridge, Professor D Jenkins, Dr M Marquiss, Dr J B Nelson and R Swann. Business Editor: Kate Walshaw, Admin Officer, SOC, The Scottish Birdwatching Resource Centre, Waterston House, Aberlady, East Lothian EH32 OPY. Scottish Birds, the official journal of the SOC, publishes original material relating to ornithology in Scotland. Papers and notes should be sent to The Editor, Scottish Birds, SOC, The Scottish Birdwatching Resource Centre, Waterston House, Aberlady, East Lothian EH32 OPY. Scottish Birds is published each year in June. Scottish Birds is issued free to members of the SOC, who also receive the quarterly newsletter Scottish Bird News, the annual Scottish Bird Report and the annual Scottish Raptor Monitoring Report. Published by: The Scottish Ornithologists’ Club, The Scottish Birdwatching Resource Centre, Waterston House, Aberlady, East Lothian EH32 OPY. Design by: Pica Design, 51 Charlton Crescent, Aboyne, Aberdeenshire AB34 5GN. Printed by: Meigle Colour Printers Ltd, Block 1, Tweedbank Industrial Estate, Galashiels TD1 3RS Scottish Birds (2006) 26:1-16 Wader changes in the Moray Firth, Scotland, 1988-2003 I Changes in numbers and distribution of waders in the Moray Firth, 1988-2003 BOZENA KALEJTA-SUMMERS Synchronised counts by about 45 counters were made at high tide in October, December, January and February each winter between 1988/89 and 2002/03. The average winter (December-February) numbers of all waders exceeded 40,000 in all years. Dunlin and Eurasian Oystercatcher were the most abundant waders, with average winter numbers exceeding 10,000 for each species. Sanderling and Red Knot increased significantly over the 15 year period and Purple Sandpiper and Ruddy Turnstone decreased. For Purple Sandpiper, Ruddy Turnstone and Red Knot the trends reflected patterns observed throughout the UK. However, the national index for Sanderling was stable. The seasonal variations in wader abundance were linked to different migration patterns. The Moray Firth supported internationally important numbers of Eurasian Oystercatcher, Ringed Plover, Dunlin, Red Knot, Bar-tailed Godwit, Eurasian Curlew and Common Redshank, whilst numbers of Sanderling, Purple Sandpiper and Ruddy Turnstone were nationally important. Introduction The importance of the Moray Firth, northeast Scotland, for wintering and migrating waders has been well documented (Symonds & Langslow 1986, Swann & Mudge 1989). The area is the most northerly group of estuaries in Britain to hold internationally important numbers of waders, with peak midwinter numbers exceeding 36,000 (Symonds & Langslow 1986). It is also an important staging area in autumn and spring (Swann & Mudge 1989, Swann & Etheridge 1996). The value of the Moray Firth has long been recognized and 4 areas have been designated as Special Protection Areas (SPAs), Ramsar sites and Sites of Special Scientific Interest (SSSIs): Dornoch Firth and Loch Fleet, Cromarty Firth, Inner Moray Firth (Munlochy Bay, Beauly Firth, Longman Bay, Castle Stuart Bay and Whiteness Head) and the Nairn coast (Fig 1). Counts of waders and wildfowl at high tide have been conducted in the Moray Firth since the early 1960s. However, they were initially irregular and rarely covered all the main areas. Annual and complete surveys started only in 1985. These counts were coordinated by the Royal Society for the Protection of Birds (RSBP) and have continued to the present date. The counts are part of the Wetland Bird Survey (WeBS), a joint scheme run by the British Trust for Ornithology, the Wildfowl & Wetland Trust, the RSPB and the Joint Nature Conservation Committee (on behalf of Scottish Natural Heritage, Countryside Council for Wales, English Nature and Department of the Environment Northern’ Ireland), which coordinate waterbird counts across the UK. Based on high tide counts, Swann and Mudge (1989) described wader populations in the Moray Firth between 1984 and 1988 and compared them with the counts during 1972-1975. This present paper reviews the current status of waders within the Moray Firth and describes their numbers and distribution between 1988-89 and 2002-03. The SB 26 “Broras™ © $Rockfield MORAY FIRTH East Ross coast “7 Hilton Shandwick Culbin noe Bar Whiteness LAs Head . sce Nairn ies Inverness Firth | Gu ah i astle Stuart Bay Bea uly Firth) Longman Bay Figure 1. The study area in the Moray Firth. international and national importance of the Moray Firth for waders is reassessed and comparisons made between the numbers of waders during the last 5 years and the earlier periods (1972-1975 and 1984-1988) as described by Swann and Mudge (1989). Study area and methods The study area comprised the firths and adjoining beaches and rocky shores between Brora in the north and Buckie in the east (Fig 1). The area was divided into 8 sections, mainly along natural geographical boundaries: Loch Fleet/Brora coast, Dornoch Firth, East Ross coast (Tarbat Ness to Rockfield and Hilton to Shandwick), Cromarty Firth, Beauly Firth, Inverness Firth, Culbin/Findhorn Bay and Moray rocky coast (Burghead to Hopeman, Covesea to Lossiemouth and Buckie (Fig 1). Hopeman Covesea | ossiemouth Bucki _ Spey Bay ndhor = Synchronised counts by about 45 counters were made at high tide on selected dates in October, December, January and February each year between 1988 and 2003. The majority of the sections were divided and each counter surveyed one or more subsections. The counts for subsections were summed to provide a total count for the section. Average winter numbers were based on counts in December, January and February. The East Ross coast section (Fig 1) was included in the surveys only from 1990/91. Therefore, there were no data for the first 2 winters (1988/89 and 1989/90) for this section. To test for trends in winter, Spearman correlation coefficients were derived for the average winter numbers in the whole of the Moray Firth. To identify trends in specific Scottish Birds (2006) areas, correlation coefficients were derived for the average winter numbers for each section. Finally, to test for seasonal trends, correlation coefficients were derived for the total numbers of birds in each month. To establish the national and international importance of a site for a particular species, the highest count for each winter in the most recent 5 years (1998/99-2002/03) was averaged to give the average maximum count. A site was considered to be of international importance if it regularly supported at least 1% of the western European population or if it regularly supported 20,000 or more of all waterbirds (Wetland International 2002). Similarly, a site was considered to be of national importance if it regularly supported at least 1% of the British population (Rehfisch et al 2003a). A comparison was made between the average winter number of birds in the last 5 years (1998/99-2002/03) and the 3 year periods of the 1972/73-1974/75 and 1985/86-1987/88 as described by Swann & Mudge (1989). In the earlier 2 periods, counts were obtained from 8 sites: the coast between Golspie to Brora, Loch Fleet, Udale Bay, Alness Bay, Munlochy Bay, north Beauly, Longman Bay and Findhorn Bay (Fig 1). To make the results comparable, the same sites were selected for 1998/99 — 2002/03. Results Annual and seasonal variations in numbers Twenty seven species of waders were recorded. Of those, 15 species (Eurasian Dotterel Charadrius morinellus, Grey Plover Pluvialis squatarola, Curlew Sandpiper Calidris ferruginea, Little Stint Calidris minuta, Ruff Philomachus pugnax, Long- billed Dowitcher Limnodromus scolopaceus, Common Snipe Gallinago gallinago, Jack Snipe Lymnocryptes minimus, Eurasian Woodcock Scolopax rusticola, Black-tailed Godwit Limosa Wader changes in the Moray Firth, Scotland, 1988-2003 3 limosa, Whimbrel Numenius phaeopus, Spotted Redshank Tringa erythropus, Common Greenshank Tringa nebularia, Green Sandpiper Tringa ochropus and Common Sandpiper Actitis hypoleucos) contributed, on average, less than 0.1% to the total number and were therefore excluded from the analysis. On average, over 40,000 waders were recorded during winter over the 15-year period. The winter average ranged between 31,600 (in 1989/90) and 48,100 (in 1996-97) (Table 1), with a peak count of 55,200 in February 1997. Dunlin Calidris alpina and_ Eurasian Oystercatcher Haematopus ostralegus were the most abundant species, with average winter numbers exceeding 10,000 and representing, on average, 27% and 26% respectively of the total number of birds (Table 1). They were followed by Common Redshank Tringa totanus, Red Knot Calidris canutus, Eurasian Curlew Numenius arquata and Bar-tailed Godwit Limosa lapponica which contributed on average 12%, 11%, 9% and 8% respectively, to the total number of birds. Statistical analysis of the average winter numbers between 1988-89 and 2002-03 revealed some trends. Sanderling Calidris alba and Red Knot increased significantly, whereas Purple Sandpiper Calidris maritima and Ruddy Turnstone Arenaria interpres declined (Table 1). When their winter averages for the 5 year periods 1988/89-1992/93 and 1998/99-2002/03 were compared, Sanderling and Red Knot increased by 72% and 67% respectively, whereas Purple Sandpiper and Ruddy Turnstone declined by 52% and 35% respectively. Some species showed very little seasonal variation with Eurasian Oystercatcher, Sanderling, Common Redshank and Ruddy Turnstone maintaining stable numbers between October and February (Fig 2). By contrast, SB 26 (sere) (Ur) (ElLt%) (ees) (7771) (rl) (81) (Grrl) (ZL) (8e) (M67) (er) (SED Soke = SOL °*CC6S CLLR S66C £81 CC POLE VEL OT S6LL OLI ree ILTOI 66/r661 (CSL) (rs) (7rL) (L771) CLS) (66) (OL) (690°7) (S867) (SEI) (ZL) (OO) (M7Z9'T) OSI'se 60L O66€ IL6E FIST LE8Z 6S 668r v8L'OI 80Fr 69 SSE —s- SSL6 76/€661 (tETS) (891) (OTT) (89L'T) (78) (EL) (91) (LTD @8rD (e9L) (8D (Li) ASO) Ole9e 88 996b LOSE TSE ESE 6¢ SILT 8806 668 Ol 81S 9Fr 01 ¢6/C66I (90L'7) (StI) (OS) (S9L'I) (Orb) (G8) (Sg) (Ov8) (S67) (Hse) (Or) (BI) (ZSL) Trlip Srl 1699 TL9r O67E I6E 19 eelT eH LBL 6br €OS SEcol 76/1661 (tcS'€) (791) (OIL) (EOS) (Or) (ET) (9E) (hI6) (898) (90S) (LIT) (817) (O€Z) 90T8E S86 ILTS 6877 SSTE 9S 8L LSvl c6Lcl 109 SCC 66S SLOOI 16/0661 (OPT 1) (OET) (9SZ) (LEST) (68E) (ES) (Og) (EOE) (OSTZ (TOS) (L9E) (89) (HSL) 6o9 IE CEB COS 887 169C vr6r Of pov tl S9CL LOO 1O¢ VV L606 06/6861 (OE€9'€) (901) (S69) (79E) (7st) COI) (SI) (tre (L8e¢e) (979) (OL) CL) (r79) TI96E C86 SoCs o6cre L8LT 6Lt CC pOLT OSScl s89Tl LSI 6Lb 3 =9E8°6 68/8861 v7 i i A Oh, i S/S MA, % on 2, Vey, % y 2 4 a; % y %, &) 0) ry) Yy Yy ®) ? Se. (og (og @ % Y G ~@ a e, Y ? % % ch % % % % %y * “~ %, eg % @ my % @& mm 8 %e& G9 v co o ov Vv ‘ly Se, a, % , ? “SJIYIDAG U1 UIAIS AAD SUOIDIAIP PADPUVIY °YJA1 ADAOY 9Y} U1 £0/Z00T — 68/886] Uaamjag (Aavn.Aga J-4aquiazaq) 1aJuiM U1 Ssapom fo saizads Juasaffip fO SAaqUINU ABDAIAY *{ I[QVL Wader changes in the Moray Firth, Scotland, 1988-2003 Scottish Birds (2006) (ceo) 868 CV (1Sv'7) SIO LE (9879) 79 CP (pr8'7) 9€L OV (6LT' 1) 8IS Ir (COS €) SSE bY (LSV' OL) Ivl8r (LIT‘S) GE GY [00 0> 198°0- (L9) 91S (SOL) €6S (O91) c69 (€8) vs9 (tr) L89 (36) cO9 (L8) C19 (Ce) cL9 ‘SU tvc O- (TE€) Tees (Z6L) S60'S (17S) 60'S (COC) ero r (699) c8lr (ple) S697 (ZET) 9E9'E (610 1) 96€ € ‘SU vScO c8l0O- 9P8'0- (618) Ipl€ (90L) E8V'e (17S) 619'€ (C19) 086 € (LOL) L9T VP (6LL) SOV (€76) 100°€ (v8) LO6 7 ‘SU (618) CLO E (OO 1) LOVC (rrr) 9ST (98h) COE C (178) 860° € (T87 1) (SOT) Orr (6LL'1) S8er (VEL) 67 V 100°;0> S00'0> S00 0> (67) trl (OS) OIC (79) col (CT) SOC (€L) [8¢ SCC (98) SVC (T€) 88C 9VL 0 (€6) IIc (62) vOC (6p) l6l (€Z) 91 (€) v9 (€€) 801 (1) Oe (c) LC ScLO (8oC' 1) 6L19 (886) 960° (O8L'7) LLLO (066) CLV (96S' 1) ocs's (1b) COTY (LOI'9) LO€'6 (919'T) 906°8 CU vso 0 (Z19) S06 OI (LEL) S88 (CIS 1) CLS Ol (LOV'7) ceo ll (18h) 6cT OL (168) Iv6 TI (L69'Z) Ore v1 (LOL) 107 TT ‘SU ITO (r6r) LVL (ZI) Loot (60) 19S (710) VC8 (S9€) SVs (LT) 0Z0' I (101) CSC (Z9¢€) 66S “SU “SU 9810 89€0 (Tpz) = (LS) 99€ LOS (671) (S¢€) ZOT OLS (10Z) (99T) SEE O8s (SIT) (SZI) ZI ELS ($6) (91) LVE TrV (6) (€Z) LS es (79) =(LT) (6 90S (pL) (SST) AS lIp ‘Su vO (SLIT) ZIT OI (Z6€) 8L1°6 (rrr) L90 II (78) Tes ol (€0S) 68S OI (C8TT) 06L°TI (78h) C80 II (L9L) 866 6 JUDTOTJJIO9 UOT}PTALION ueulteads €0/CO0C CO/LOOC [0/000¢ 00 /6661 66/8661 86/L661 L6/9661 96/S661 numbers of Dunlin, Red Knot, Purple Sandpiper and Bar-tailed Godwit were low in October and peaked in January and/or February, whilst Northern Lapwing Vanellus vanellus and Ringed Plover Charadrius hiaticula reached peaks in abundance in October and the majority were gone by February. Spatial distribution Most species occurred in almost all sections of the Moray Firth but their distributions were uneven (Fig 3). Sanderling, for instance, were more numerous in Culbin/Findhorn Bay, Purple Sandpiper and Ruddy Turnstone preferred the Moray rocky coast, whereas’ Eurasian Oystercatcher and Red Knot occurred in highest numbers in the Cromarty’ Firth and Culbin/Findhorn Bay. Lapwing preferred the Cromarty Firth, Ringed Plover the Loch Fleet/Brora coast whereas Eurasian Curlew, Common Redshank, Dunlin and Bar-tailed Godwit occurred in high numbers at several localities including the Cromarty Firth, Dornoch Firth, Culbin/Findhorn Bay and/or Inverness Firth (Fig 3). For species which showed significant trends in average winter numbers in the whole Moray Firth, further significant patterns emerged at particular sections. Sanderling increased at Loch Fleet/Brora coast and Culbin/Findhorn Bay by 63% and 79% respectively, whereas numbers of Red Knot increased in the Dornoch Firth (by 74%), Inverness Firth (by 99%) and Moray rocky coast (by 55%) (Table 2). Ruddy Turnstone and Purple Sandpiper declined in the Dornoch Firth (by 44% and 53%) and along the Moray rocky coast (by 34% and 66%). In addition, Purple Sandpiper also declined by 60% on the East Ross coast and Ruddy Turnstone by 37% in the Inverness Firth. For species that showed no significant trends in the mean winter numbers for the Moray Firth as SB 26 a whole, some significant changes occurred at certain sections (Fig 3, Table 2). Numbers of Eurasian Oystercatcher, for instance, increased between 18% and 51% in the Cromarty Firth, Beauly Firth and East Ross coast, but at the same time declined by 25% at Loch Fleet/Brora coast. Similar shifts in distribution were observed for Common Redshank. Their numbers significantly increased by 54% and 71% on the Moray rocky coast and in the Beauly Firth but declined by 24% and 35% in Culbin/Findhorn Bay and the Inverness Firth. Dunlin, on the other hand, declined by 54% in the Beauly Firth as did Bar- tailed Godwit in the Moray and East Ross rocky coasts and Culbin/Findhorn Bay. Ringed Plover decreased by 36% on the Moray rocky coast but, at the same time, increased by 45% in the Dornoch Firth. International and national importance The average maximum numbers of 7 species, Eurasian Oystercatcher, Ringed Plover, Dunlin, Red Knot, Bar-tailed Godwit, Eurasian Curlew and Common Redshank, in the Moray Firth exceeded 1% of the international total for those species in the most recent 5 years (1988/89- 2002/03) (Table 3). In addition, the Moray Firth supported nationally important numbers of Sanderling, Purple Sandpiper and Ruddy Turnstone. Furthermore, the Moray Firth as a whole supported over 40,000 waders in all years over the 15 year period (Table 1). When individual sections were considered, internationally important numbers of Common Redshank were found in the Cromarty and Inverness Firths. The Cromarty Firth also supported internationally important numbers of Bar-tailed Godwit and nationally important numbers of Red Knot. Culbin/Findhorn Bay supported nationally important numbers of Eurasian Oystercatcher and Bar-tailed Godwit whereas the Dornoch Firth was _ nationally important for Bar-tailed Godwit. Scottish Birds (2006) Wader changes in the Moray Firth, Scotland, 1988-2003 7 16000 Ringed Plover 12000 8000 Eurasian 4000 Oystercatcher O . 2 Sage sal bowen G T | ee) eae I T T T lege 1 1988/89 1992/93 1996/97 2000/01 1988/89 1992/93 1996/97 2000/01 1500 European 4000 Northern Lapwing 4000 Golden Plover 3000 2000 - rie 1000 e) ) : — 1988/89 1992/93 1996/97 2000/01 1988/89 1992/93 1996/97 2000/01 20000 Dunlin 15000 10000 5000 5000 O Leaf T T ola ale plissalescaleplinaeal O 1988/89 1992/93 1996/97 2000/01 1988/89 1992/93 1996/97 2000/01 500 : 800 -; 2 Sanderling R Purple Sandpiper ® 600 + : Q = S = 8 Ks. 1988/89 1992/93 1996/97 2000/01 1988/89 1992/93 1996/97 2000/01 Bar-tailed Godwit 9000 ; 6000 Eurasian Curlew 4000 2000 O eeais i fo ke ee ee oe ed O eS a a ea tae eg eee r- 4 1988/89 1992/93 1996/97 2000/01 1988/89 1992/93 1996/97 2000/01 10000 Common Redshank 0 aa let aah tat a Oe ee 1988/89 1992/93 1996/97 2000/01 1988/89 1992/93 1996/97 2000/01 ——®— October --*#--December -—t— January -— ©-- February Figure 2. Numbers of waders in October, December, January and February in the Moray Firth between 1988/89 and 2002/03. SB 26 Average numbers 1988/89 1992/93 1996/97 2000/01 300 5 European Ks 200 - Golden Plover ; [@) . 100 - IN a 0 = ) 1988/89 1992/93 1996/97 2000/01 8000 - 6000 Qa Dunlin lad 4000 - : 2000 - : Oo . i me) O z cr ee Ooooh BR yt 1988/89 1992/93 1996/97 2000/01 200 5 etna anderling 150 wo | ££ a 100 - A 7 50 : a ee ee Se O xX —--. 1988/89 1992/93 1996/97 2000/01 2500 Bar-tailed Godwit o 2000 - A i 1500 - eA. il 1000 {~& ha 500 - = O Cr ~ a. . i. ok Se i. T “OO 1988/89 1992/93 1996/97 2000/01 2000 7 Common Redshank 1500 -& Moy ep 1000 -— my 4 1988/89 1992/93 1996/97 2000/01 —~@— Loch Fleet/Brora coast 22> => Cromarnry Firth — - Inverness Firth — ~ - East Ross coast 8 6000 Eurasian 300 + Ringed Plover 4000 Oystercatcher a ro Ob Ot oO Ok ORR Ory. 2000 /§ OSes pooh S-0 O i (anos T i T T T at aaa ar or T T T T 1 1988/89 1992/93 1996/97 2000/01 700 Northern Lapwing oO Oo A Pay Se a 350 | ra o-O : oe \ oO O - ane mas > i ae 1988/89 1992/93 1996/97 2000/01 6000 - Red Knot eo} 4000 - ° : Fy 2000 - : - O oi re ioe a-2 1988/89 1992/93 1996/97 2000/01 S00 > Purple Sandpiper 400 - 300 -+ 200 - 100 +e. 1988/8 1992/93 1996/97 2000/01 1500 - Eurasian Curlew 1000 - 500 - Oo a T T a. 9 T =o a a T i T aie fob aa Tie 1 1988/89 1992/93 1996/97 2000/01 600 - 300 - O + : 1988/89 1992/93 1996/97 2000/01 —tla— Dornoch Firth —*«z— Beauly Firth — #@— Culbin/Findhorn Bay —x— Moray rocky coast Figure 3. Average winter numbers (December-February) of waders in different sections of the Moray Firth between 1988/89 and 2002/03. Only 4 to 5 of the numerically most important sites were Selected for each species. 9 Wader changes in the Moray Firth, Scotland, 1988-2003 Scottish Birds (2006) 100°0>d 100°0>d I0'0=d 100°0>d SO 0>d CO 0>d JSBO9 LS80- vl60 vS90O- LS80- CoS 0 96S 0- AYSOI Keo SO 0>d SO 0>d 100° 0>d CO 0>d Avg wioypul{ 6c 0- 6LS 0- 98L'0 909'0- /urq|n) COO 0>d S00 0>d 100° 0>d yur LSLO- 9tLO- C160 SSOUTOAUT 100 0>d C00 0>d CO 0>d qu evs 0 VIL O- LSS O A[neag SO 0>d Yur S7S'0 Ayrewosy L100 0>d S00 0>d SOO 0>d JSBOL) IL8°0 Sol Ur 9rL'0 SSOeY SPY 10 0>d SO 0>d 100° 0>d SO00>d cO00>d Yul L69'0- evs O- vI8 0 6890 S790 yooulod S0'0>d S0'0>d }SvOO VIOIG 88° 0 Tes O- /199|4 YOO'T o % q, 4 % & 40 Y & 4. “ a) On “ey, con w, Gry %, % “My Gu, “tg Ny, %y %, ey o) ry) Yy &y a) ? es ~ G @ % y Gp, 2 e ( ev ? % a, “Q, % en mn ul “ye %, “¢ Gs % Mn, 8 On. EC = ee <4 9 “4 ? Sd 4 “My py Ce “D > ? "(€ 81 99S) €0/ZOOT PUY 68/886] UIAMIJAg SAIQUINU 19JUIM ASDAIAD AY] U1 Spuas JuDIL/IUSIS pamoys YI1YM Sa1zads asoy sof 4]UO Uad1s AAD SAN]DA-q “YJ ADAOPY AY) U1YJIM SvasD Juasaffip 40f Sanjva-q ay) pun (54) syuainif{ao09 uoyvjas1109 uDULADAdS *Z 3{QUI, 10 Comparisons with previous records Comparisons between mean winter numbers revealed significant differences between the 1970s, 1980s and 1998 — 2003 for 4 species (Table 4). Between the 1970s and 1980s, numbers of Red Knot and Dunlin declined by 82% and 56%, respectively, (Swann & Mudge 1989). However, since then, they have increased by 80% and 38% (Table 4). Numbers of Ringed Plover increased significantly by 44% in the last 15 years and by 55% between 1970s and 1998 —2003 (Table 4) (Swann & Mudge 1989). Eurasian Oystercatcher, on the other hand, increased significantly by 44% between the 1970s and 1980s and although their numbers dropped by 22% between 1980s and 1998-2003, the increase by 28% between 1970s and 1998-2003 was significant. Numbers of Northern Lapwing decreased significantly by 77% between 1970s and 1980s, but their increase between 1980s and 1998-2003 was not significant. No other species showed significant long term changes in abundance over the last 30 years. Ruddy Turnstone and Purple Sandpiper were not counted in the earlier years so it is not possible to comment on their long term changes in numbers. Discussion Annual variations in numbers This study has revealed significant trends for 4 wader species in the Moray Firth over the 15 year study period. The decrease in numbers of Purple Sandpiper (by 53%) and Ruddy Turnstone (by 35%) (Table 1) reflected trends observed elsewhere in the UK. A survey conducted over 38% of the UK’s non estuarine coastline between 1984/85 and 1997/98 showed a decline of 16% for Ruddy Turnstone and 21% for Purple Sandpiper (Rehfisch et al 2003c). However, declines of both species in East Lothian, southeast Scotland, between 1970s and 1990s were much higher: 73% for Ruddy Turnstone and 88% for Purple Sandpiper (Dott 1997), indicating SB 26 spatial variations in these declines. The scale of the decline in the Moray Firth was greater for Purple Sandpiper than for Ruddy Turnstone in line with the other studies. The declines are probably also slightly underestimated because the East Ross rocky shore was not surveyed during the first 2 years of this study. The increase in numbers of Sanderling in the Moray Firth over the 15 year period did not reflect their national index which is relatively stable (Pollit et al 2003). This also contrasts with the estimated UK wintering population on the non estuarine coastline between 1984/85 and 1997/98 which showed a decrease of 20% (Rehfisch et al 2003c). For Red Knot, on the other hand, the national index indicates a decline by 15% in the past 30 years but an increase by 4% over the last 10 years (Pollitt et al 2003). The comparison of the counts in the 8 selected areas in the Moray Firth between the 1970s and 1998-2003 indicates that Ringed Plover has continued to increase over the last 30 years (Table 4). This contrasts with the national trend which indicates a decline by more than 25% over the last 25 years (Pollitt et al 2003). Similarly, a decrease by 15% has been reported for Ringed Plover on non estuarine coasts between 1984/85 and 1997/98 (Rehfisch et al 2003c). Although there was no significant trend for Dunlin in the Moray Firth over the last 15 years (Table 1), the average winter number in the 8 selected areas over the last 5 years was signifi- cantly higher than during 1985-1988 (Table 4). This contrasts with the UK annual index which in the late 1990s was low and similar to that of the mid 1980s (Pollitt et al 2003). The average winter numbers of Eurasian Oystercatchers, on the other hand, was higher over the last 5 years than during 1972-1975 (Table 4). This pattern mirrors the national index which increased by 45% over the last 30 years (Pollitt et al 2003). I] Wader changes in the Moray Firth, Scotland, 1988-2003 Scottish Birds (2006) 9L8 Sols LOLS 6tlP 6te rats O86 OSTPE 100€ 8971 StL = TLOTT yur Aero Avg wsloypuly 9¢ £66 866 sss e 991 96¢ I Sor LV 6S LSC 66 600° /utq[n’) yu OTT scr I OS8 V8C C 0 9ST I O10 T LO9 6L C6 699 | SSOUISAUT yun 6V TLE 6 6SOG TIT: 6ULSET lz 0 Lwee 617 L68 COL OLT Esle AYWRWOID yt c6 TIO T vel L66 Ce 9 Orit try VSI G9 IvlI vos | yooulod proysezyy OOS OOTI OOS 0c9 O81 OIC 00sz 009S ODDO O00S7 Of 007 € yeuonen proysazyy 000 T 00¢ T 00Tr O07 T OSL 007 I OOSr OOF EI OODOZ 0008 OcLl 007 OI [euorjeusojuy 9 % q 2 ‘@ % 40 % 7 ws * % “ap, ty Xp Gy, kon Ge, % “My %y “9 Ng %y Cok. en De ee & ae Se te. % 0, @ “@ y S) 40, Ka ip) ©, %, “ : Ks % ) Y % %, ee “e “% “Wo %, “, 9) 7 Ka a yp “Se . Q Xy % rs ‘UMOYS adv Sp.1ig fo SAaqunu junjiodun KyouoYyvU 40 KouoYyousaqul pjay Joy] SuOYIaS asoyj 4]UQC “PjOg ui Kjuo asp aounjsodun youoynu fo says svasaym pauysapun pup pjog ul asp aoupjiodun jouoynusajus fo suoysag "sa12adS asoyj 4of jupjsodun (vgggz 1» 12 yasifyay) Kjouoypu 40 (Z7QQT JOUOYDUAaJUT puvjjayy) K]JOUOYoUAIJUI SD SUONIAS Afyjonb 0] spjoysasyy %l PUY “EO/ZO0T PUP 66/8661 Uaamjag YI Koso ay fo suoysas juasaffip ui Ssapom fo ssaquinu wnunxpu aspsaay *¢ QPL, 12 Several factors may be responsible for annual fluctuations and long term trends in numbers. These includes breeding success. Ruddy Turnstones and most Purple Sandpipers wintering in the Moray Firth come from the same breeding grounds, Canada (Nicoll et al 1988, Swann & Etheridge 1996), and _ their declines in the Moray Firth may be due to poor breeding success. For Purple Sandpipers the decline in numbers can be accounted for by the low recruitment in the late 1980s and early 1990s (Summers et al 2005). Climatic changes and increasingly mild winters may cause geographical shifts in the distribution of birds, resulting in local population changes. The increasing mean minimum winter temperature in the UK may partially explain a northward and eastward shift of 6 species of waders between 1984/85 and 1997/98 in their non estuarine coastal distribution, and an eastward shift of 5 species over the last 15 years in their distribution among estuaries (Austin ef al 2000). The Ringed Plover distribution, for instance, has moved partially from the west to east coast following milder winters (Austin et al 2000). It has also been suggested that some waders that used to winter in the UK might now be wintering elsewhere as milder conditions make previously unsuitable areas hospitable (Rehfisch er al 2003c). The decline in Ruddy Turnstone and Purple Sandpiper might simply be a result of more birds spending winter closer to the breeding grounds eg Iceland, instead of migrating to Britain. It is also possible that some waders might decide to stay in the Moray Firth for the winter instead of moving further south. More evidence, however, 1s required to support these ideas. Severe weather can also affect wader distri- butions. Prys-Jones et al (1994) suggested that the widely fluctuating numbers of Bar-tailed Godwit in Britain between the 1970s and 1990s were due to immigration of birds from the SB 26 Wadden Sea during severe weather. However, ringing has shown that not many birds come to the Moray Firth from continental Europe in response to severe weather in comparison with wetlands further south in Britain (Swann & Etheridge 1996). Food supply is another important factor regulating bird numbers. An invertebrate survey in the 1960s in Nigg and Udale Bays indicated that the food supply for waders, although rich in species, was not particularly abundant (Anderson 1970). Lack of large expanses of uniform sediments with optimum amounts of organic nutrients were thought to be responsible for the low density of invertebrates. A later survey, in the early 1980s, showed an increase in species richness (Raffaelli & Boyle 1986). It is unknown, however, if the change was part of a long term natural cycle of invertebrates or were induced by onshore development in the area (Raffaelli & Boyle 1986). The present status of invertebrates in the Moray Firth is unknown. It has been demonstrated that organic input can have a positive impact on the presence of waders in estuaries and non estuarine coasts (Rehfisch & Austin in press). The discharges from sewage and trade effluents provide organic nutrients for intertidal invertebrates, resulting in an increase in invertebrate abundance and diversity (Pearson & Rosenberg 1978, Kropp et al 2000). This is likely to affect the number of waders feeding in the vicinity of outfalls. Although there are 5 distilleries and several sewage outfalls around the Moray Firth, their organic input is now believed to be low because of improvements to sewage treatment in the 1980s (SEPA, pers comm). The most recent (1999) changes involved moving the discharge outfalls from 2 distilleries in the Cromarty and Dornoch Firths into deeper water and away from the shore (SEPA, pers comm). Similarly, the sewage output at Inverness was upgraded in 2000, 13 Wader changes in the Moray Firth, Scotland, 1988-2003 Scottish Birds (2006) (OST) (ESS) (O€¢) (Trl 1 €8S-) O8 (O8S ©) TLT-) PSI LSS'I Cee | COLT yueyspoy uowuo0D (6€1) (TEE) (261) (TEV 1 P8E-) HZ (6€T 1 60E-) SE- 9+6 776 186 MoIND uUeIseINg (9¢€¢) (C8¢) (pT7¢) (79S 01 96S-) LI- (8ZE 01 OEL-) LOT- vOs ae SOL WMPOD pejie}-1eg (66S) (09¢) (SI9'T) (PE9T 91 OS) THS (€CO' IT 0 [L8°T-) 616- SITT OLE I LOVE uljyund (PLL) (P67) (VET 1) (ZL6 1 91 96) PEOT (TpS I 91 6SL'I-) 60I- 687 I CoCo 86¢ I JOU poy (€Z1) (SOI) (OTP) (L8Z 91 6L-) VOI (9ZI 01 998-) OLE- CVT Ivl C19 SulMmde’y] WsyyIoN (60) (Or) (S€) (PST 93 OL) ZS (OLT 93 €€) ZOT C8I cOl €8 IJAO[d Posury (S€Z) (L711) (€Tr) Tayoed191skO (CLE 01 967'T-) 766- (POST°} LOP) 9S6 oor'¢ TSP OS‘ uviseing COOT-866T1 PUP 88-C86l COOC-8661 Pur C/6I-CLOI €00C- 8661 8861-S86l SL6I-CLOl ([BAIOJUL DOUIPTJFUOD 2%CG) DOUSIOJJIP UII] Z [ ‘pjog ul asp saouasaffip juvoifius8ig “(7 AS+z IS)AXO6" JF wolf pauinjgo som aouasaffip ay fo Jvasajul aouapifuod aspvjuarsad aalf KJOUIN (686) A8pnW x» UUuDMG WOL UayD] alam Spolsad sausof Z ay] A0f SAaquinu uvap *SQO7Z-S661 PUP S8-SS86I ‘SZ-7ZZ6I U! YJAL] ADIOW AY] Ul (SPOYyjJaul aas) Sajis pajIajas 9 JD SAaPYM AOL (SJAYIDAG U1) SUOIIYIAap PADPUDIS PUD SAaquiNU 4AAJUIM UDA “py IQVL 14 reducing the amount of nutrients, and moved further east into deeper water. It has been demonstrated that improvements to sewage treatment may result in a decrease in invertebrate abundance as communities return to those more typical of an _ unenriched environment (Kropp et al 2000). On the Clyde estuary, for instance, improvements to 4 sewage treatment works have been linked to a decline in the abundance of Corophium volutator and Nereis diversicolor (Curtis & Smyth 1982, Thompson et al 1986). It is unknown, however, whether the long term changes in wader populations at the Moray Firth have been affected by any changes in invertebrates associated with sewage outfalls. In the case of Purple Sandpiper and Ruddy Turnstone, both species are rocky shore specialists and are site faithful within and between winters (Atkinson er al 1981, Burton & Evans 1997; Rehfisch et al 2003b). Their food is less affected by cold weather than that of waders on estuaries but could have been influenced by the changes in sewage outfalls (Eaton 1999). Seasonal variations in numbers The seasonal variations in numbers of waders in the Moray Firth can be explained by different migration patterns. The birds that winter in the Moray Firth can be grouped according to their origins. The northwestern group includes Red Knot, Ruddy Turnstone, Purple Sandpiper and Sanderling, which breed in Arctic Canada and northern Greenland. The northeastern group includes Bar-tailed Godwit, Grey Plover and Dunlin that breed in Arctic Russia (Swann & Etheridge 1996, Swann & Insley 1997). The western European group comprises Eurasian Oystercatcher, Common Redshank, Golden Plover, Dunlin, Eurasian Curlew, Purple Sandpiper and Ringed Plover whose breeding grounds include Iceland, Faroe Islands, northern SB 26 Scotland and Scandinavia. Finally, there is a group of locally breeding birds: Eurasian Oystercatcher, Northern Lapwing, Ringed Plover and Eurasian Curlew (Swann & Etheridge 1996). As a result of the diverse origins of the waders (some of which originate from more than one breeding population), arrivals and departures to and from the Moray Firth occur at different times. Adult birds of species such as Eurasian Oystercatcher, Common Redshank and Ruddy Turnstone start to return to the Moray Firth in July. Peak numbers for these species occur in late October, when young birds arrive and adults move through on passage (Swann & Mudge 1989). Numbers then remain high and stable throughout the winter (Fig 2). For species such as Dunlin, Red Knot, Purple Sandpiper and Bar-tailed Godwit, the first influx of adults usually occurs in late October and November, and peak numbers are not recorded until the midwinter months (December-January) (Fig 2; Swann & Mudge 1989). For Ringed Plover and Northern Lapwing numbers peak in October, though, the latter species is present in large numbers also in August (McNee 2003). During midwinter, numbers of Ringed Plover decrease (Fig 2) as the larger flocks disperse (Swann & Mudge 1989). Fluctuations in numbers of Golden Plover, Northern Lapwing and Eurasian Curlew throughout the winter (Fig 2) are probably due to severe weather as many of these birds feed locally in the nearby fields and rely more on to the coast when fields become frozen (Swann & Mudge 1989). The results of this study confirmed the interna- tional and national importance of the Moray Firth as a wintering site for waders. However, in comparison with the 1980s, the area no longer supports internationally important numbers of Ruddy Turnstone but it became internationally important for Ringed Plover, Dunlin, Red Knot Scottish Birds (2006) and Bar-tailed Godwit (Table 3; Swann & Mudge 1989). In addition, the Moray Firth now supports nationally important numbers of Sanderling and Purple Sandpiper. The total number of waders in the Moray Firth regularly exceeded 30,000, with 6 species accounting for 93% of the total. This places the Moray Firth as second only to the Solway Firth in importance in Scotland, and twelfth of all the estuaries in Britain (Pollitt et al 2003). Acknowledgements Enthusiasm and dedication of many WeBs volunteers over the years made this study possible. It is impossible to acknowledge all of them but special thanks go to F Symonds, D Patterson, R L Swann, C Crooke, B Etheridge and M Cook. I also thank D Butterfield, K Chisholm, M Hancock, Dr R Summers and R L Swann for their comments on the earlier draft of this manuscript. M Hancock’s assistance with computing and T Burns’s help in producing Fig 1 are gratefully appreciated. Financial support was provided by the Talisman Energy (UK) Ltd. References Anderson S S 1970. Nigg and Udale Bays: survey of intertidal invertebrates. Unpubl. Report. Coastal Ecology Research Station, Norwich. Atkinson N K, Summers R W, Nicoll M & Greenwood J D 1981. Population, movements and biometrics of the Purple Sandpiper Calidris maritima in eastern Scotland. Ornis Scandinavica 12: 18-27. Austin G E, Peachel I & Rehfisch M M 2000. Regional trends in coastal wintering waders in Britain. Bird Study 47: 352-371. Burton N H K & Evans P R 1997. Survival and winter site-fidelity of Ruddy Turnstones Arenaria interpres and Purple Sandpiper Calidris maritima in north-east England. Bird Study 44: 35-44. Wader changes in the Moray Firth, Scotland, 1988-2003 15 Curtis D J & Smyth J C 1982. Variations in densities of invertebrate benthos of the Clyde Estuary tidal flats. Chemical Ecology 1: 57-60. Dott H E M 1997. Declines in Ruddy Turnstones and Purple Sandpipers wintering in south-east Scotland. Scottish Birds 19: 101-104. Eaton M 2000. Studies on Purple Sandpipers and Ruddy Turnstones at Hartlepool, 1999-2000: Have recent Changes in the Treatment and Discharge of Sewage had an Impact? Report to Northumbrian Water plc. University of Durhan. Kropp R K, Diaz R J, Dahlen D, Shull G H, Boyle J D & Gallagher E D 2000. 1998 Harbour Monithoring Report. Report ENQUAD 00-06, Massachusetts Water Resources Authority, Boston. McNee A F 2003. Highland Bird Report 2002. Scottish Ornithologist’s Club, Inverness. Nicoll M, Summers R W, Underhill L G, Brokie K & Rae R 1988. Regional, seasonal and annual variations in the structure of Purple Sandpiper Calidris maritima populations in Britain. [bis 130: 221-233. Pearson T H & Rosenberg R 1978. Macrobenthic succession in relation to organic enrichment and pollution of the marine’ environment. Oceanography and Marine Biological Annual Review 16: 229-311. Pollitt M, Hall C, Holloway S, Hearn R, Marshall P, Musgrove A, Robinson J & Cranswick P 2003. The Wetland Bird Survey 2000-01: Wildfowl and Wader Counts. BTO/WWT/RSPB/JNCC, Slimbridge. Prys-Jones R P, Underhill L G & Waters R J 1994. Index numbers for waterbird populations. II. Coastal wintering waders in the United Kingdom 1970/71-1990/91. Journal of Applied Ecology 31: 481-492. Raffaelli G & Boyle P R 1986. The intertidal macrofauna of Nigg Bay. Proceedings of the Royal Society of Edinburgh 91B: 113-141. Rehfisch M M & Austin G E in press. The effect of water quality on overwintering waterfowl population. Maine Forum for Environmental Issues. 16 Rehfisch M M, Austin G E, Armitage M, Atkinson P, Holloway S J, Musgrove A J & Pollit M S 2003a. Numbers of wintering waterbirds in Britain and the Isle of Man (1994/95-1998/99: IT. Coastal waders (Charadrii). Biological Conservation 112, 329-341. Rehfisch M M, Insley H & Swann R L 2003b. Fidelity of overwintering shorebirds to roosts on the Moray Basin, Scotland: implications for predicting impact of habitat loss. Ardea 91: 53-70. Rehfisch M M, Holloway S J & Austin G E 2003c. Population estimates of waders on_ the nonestuarine coast of the UK and the Isle of Man during the winter 1997-98. Bird Study 50: 22-32. Summers R W, Butterfield D P, Swann RL & Insley H 2005. The decline of Purple Sandpiper Calidris maritima in the Moray Firth, Scotland — an effect of recruitment. Wader Study Group Bulletin 106: 34-38. Swann R L & Etheridge B 1996. Movement of waders to and from the Moray Firth. Ringing & Migration 17: 111-121. Swann R L & Insley H 1997. Survey of movement, age structure and biometrics of waders residing in the Moray Firth. Unpubl. Rep., Highland Ringing Group & Scottish Natural Herritage. SB 26 Swann R L & Mudge G P 1989. Moray Firth wader populations. Scottish Birds 15: 97-105. Symonds F L & Langslow D R 1986. The distri- bution and local movements of shorebirds within the Moray Firth. Proceedings of the Royal Society of Edinburgh 90B: 143-167. Thompson D B A, Curtis D J & Smyth J C 1986. Patterns of associations between birds and invertebrates in the Clyde estuary. Proceedings of the Royal Society of Edinburgh 90B: 185-201. Wetland International 2002. Waterbird Population Estimates - Third Edition. Wetlands International Global Serious No.12. Wangeningen, The Netherlands. Bozena Kalejta-Summers, Royal Society for the Protection of Birds, Etive House, Beechwood Park, Inverness IV2 3BW. E-mail: Bozena@waders .fsworld.co.uk Revised manuscript accepted July 2005 Scottish Birds (2006) 26: 17-29 The status of the Northern Gannet in Scotland in 2003-04 17 The status of the Northern Gannet in Scotland in 2003-04 S MURRAY, S WANLESS & M P HARRIS A census of the Northern Gannet colonies in Britain and Ireland was carried out in 2003- 04. Scotland had 14 colonies with a total of 182,511 apparently occupied sites, representing 58% of the east Atlantic population. Numbers were divided unevenly between the colonies with St Kilda, the Bass Rock and Ailsa Craig together holding 74% of the Scottish population and 43% of the east Atlantic population. Two new colonies have been founded since the last major census in 1994-95, at Sule Skerry (2003) and Westray (2003). Comparison of colony totals with previous estimates indicate that numbers at St Kilda and Sule Stack were stable whereas Ailsa Craig and Sula Sgeir had decreased. Rates of increase varied considerably, but in general were highest at the most recently founded and smallest colonies. Although overall numbers continued to increase, the sustained period of population growth throughout the 20th century has slowed over the last 10 years. Introduction At the time of the last comprehensive census of the Northern Gannet Morus bassanus in 1994- 95, there were 12 gannetries in Scotland. The Scottish population was estimated at 167,407 occupied sites which represented 61% of the total for the east Atlantic (Murray & Wanless 1997). Numbers at most colonies were increasing and the Scottish population had increased at an average rate of 2.4% per annum (pa) between 1984-85 and 1994-95. We co- ordinated another survey of British and Irish Gannetries in 2004 and this paper details the counts from all the Scottish colonies in 2003 or 2004 and compares them with the results of the 1994-95 census. Methods The locations of the 14 colonies are shown in Figure |. We carried out aerial surveys of the Bass Rock, Sule Stack, Sula Sgeir, the Flannan Isles, St Kilda, Ailsa Craig and the Scar Rocks in 2004, and counted Sule Skerry from a land photograph. Other observers counted Westray, Foula, Fair Isle and Troup Head in 2004 and Noss and Hermaness in 2003. The 2003-04 survey was carried out using similar methods to those adopted in 1984-85 and refined in 1994-95, with colonies counted either from field counts made from the land or sea, or from photographs (mainly transparencies) taken from the land, sea or air. The methods and the associated problems are described in Murray & Wanless (1986, 1997). In brief, for counts made from photographs the only practical counting unit was the apparently occupied site (AOS, a site occupied by one or 2 Gannets irrespective of whether nest material was present). In most field counts the unit was the apparently occupied nest (AON, one or 2 birds at a site with nest material present). Sites with a chick but no obvious nest were included in this category. Neither count unit provides an estimate of the number of breeding pairs, nor is it strictly correct to equate occupied sites with pairs, as some sites may be held by a single bird for at least a year (Nelson 1978). The unavoidable lack of standardisation of count units across colonies makes it impossible to 1S calculate a grand total for Scotland in terms of a common unit. Our estimate of the Scottish population, and the east Atlantic total, is therefore, a combination of totals of apparently occupied sites at the majority of colonies and a few counts of nests. No correction factors were applied to either unit, and for convenience the grand total is expressed in terms of apparently occupied sites. Counts were made between 26 May and 26 July, which is within the time period recommended by Nelson (1978). The counts for most colonies are presented in terms of their constituent sections, to facilitate future comparisons of population changes. These boundaries are shown in Murray & Wanless (1997) and Wanless et al (2005a). Average rates of change for each colony were calculated using the equation: P2/P1l=(1+r)t where r is the rate of population change, P| is the nest or site count in 1994 or 1995, P2 is the nest or site count in 2003 or 2004 and t is the number of years between the 2 counts. It is inevitable that some of the observed changes in numbers will be Hermaness f [150km Foula@abNoss Sule Stack ir | Sige eee oe sle fe stray q , Troup Head Sula Sgeir Fiannan Isies (Boreray, Stac Lee, , Stac an Armin) =~ @ Bass Rock JJ / Figure 1. The locations of Northern Gannet colonies in Scotland 2004. SB 26 due to counting or sampling errors rather than actual changes in abundance. Variations in count methods, dates or times when counts are made and the quality of photographic images, all contribute to counting errors. Even when photographs are of a high standard, there is variation among counts made by each observer and among different observers (Harris & Lloyd 1977). The 2004 results of replicate counts made between 2 observers, indicated that changes associated with observer error were generally between 5-10%, similar to those recorded in the 1984-85 and 1994-95 surveys (Murray & Wanless 1986 & 1997, Wanless 1987, Wanless et al 2005a). Therefore, differences in count totals between 1994-95 and 2003-04 of <10%, were interpreted as indicating no overall change and approximately stable numbers. However, for all colonies careful inspections were made to pinpoint any major changes in colony extent or nesting density. Unless otherwise stated, counts from 2003-4 and 1994-5 are directly comparable. Results There were 14 active gannetries in Scotland in 2003-04 (Fig 1). Since the 1994-95 survey new gannetries have become established in Orkney, on Sule Skerry and Westray. There has been no recorded nest building on the Shiant Isles since 1986 and no birds or nests were present in May 2004 (pers obs). There have been no reports from Rockall since 1992, when a nest with an egg was found (Belaoussoff 1993). The rock was not visited in 2004. The 2003-04 counts and population trends are summarised below for each of the Scottish gannetries. Scar (Scare) Rocks An aerial survey was made on 6 June 2004. Three counts made by 2 observers gave an average total of 2,394 AOS, representing an overall increase of 22.6% since 1995 at an Scottish Birds (2006) average rate of increase of 2.3% pa. There was no change in overall density, but small increases in colony extent were apparent at the lower edge of the Plateau section. Table 1 Counts (AOS) of the Scar Rocks Gannetry in 1995 and 2004. Count Section 5 August 6 June 1995 2004 West Slope Zi 226 West Stack 111 121 North Cliff 28 38 Plateau 1604 2009 Total 1952 2394 Ailsa Craig An aerial survey was made on 6 June 2004. Three counts made by 2 observers averaged 27,130 AOS, an overall decrease of 16.4% since 1995 at 2% pa. Comparison with the 1995 photographs showed small increases along the upper colony limit of Section 9/10/11, but a decrease in breeding density, leaving a near empty swath through this section (Table 2). In contrast, an increase in breeding density has taken place in Section 21, but without noticeable changes in the colony extent. Elsewhere there appeared to be no changes to colony boundaries, and the small gaps separating Section 6 from Section 7, and also Section 13/14 from adjacent sections, apparent in 1995, were still obvious. The decrease in the population seemed to be due to an overall decline in breeding density, reflected in both observers results for 2004 (Wanless et al 2005a, b & c). Observer 1 counted 25,017 and 26,905 AOS respectively and Observer 2, 28,298 AOS, giving a mean for the 3 counts of 27,130 AOS. The late date of the 1995 count may have contributed to the higher count that year, but that does not explain the The status of the Northern Gannet in Scotland in 2003-04 19 large drop in numbers, in for example Section 8 (Table 2). There may have been an element of over counting in 1995, and repeat counts made of some sections support this view. However, even if the 1995 total was too high and is adjusted downward to 30,844 AOS, the lower of the 2 counts, rather than the published average of 32,456 AOS, then the colony has still shown an overall decrease of 13.4%. Table 2 Counts (AOS) of the Ailsa Craig Gannetry in 1995 and 2004. Count Section 5 August 6 June 1995 2004 1 946 1009 2 187 85 ) 44] 245 4&5 1877 1222 6 2097 1333 7 954 809 8 4300 DIMA 9,10 & 11 11959 10338 12 140 157 13 & 14 1994 1411 15 & 16 2263 2279 17 Zl 38 18 908 vig 19 3410 3482 20 531 270 21 250 127, 22 0 0 23 177 236 Total 32455 27130 St Kilda (Boreray, Stac Lee and Stac an Armin) The aerial survey made on 21 June 2004 gave 100% photographic coverage of the stacks and 97.3% coverage of Boreray, the same percentages as in 1994. Two complete counts were done by one observer and sample counts 20 totalling 38% of the group total by a second observer. The average group total was 59,622 AOS (Table 3), a decrease of 1.3% since 1994 at 0.1% pa. No changes were found in nest densities and colony extent remained largely unchanged, except for small increases in one section on Boreray and 2 on Stac an Armin. The slow but consistent increase that had occurred between 1959 and 1994 (at an average of 0.9% pa Murray & Wanless 1996) had ceased and numbers are probably stable. Boreray Because of the complexity of the cliff structure, small areas in some sections are virtually impossible to photograph from the air. These have been identified from past land visits and were counted from the land on 18 August 2003. The aerial survey achieved almost complete coverage of the colony, but the quality of the images was too poor to give a count of sections 3 to 12. Here, 582 AOS were added to the section total, based on these earlier land counts. The figures were included in the final section totals (Table 3). The combined air and land counts gave an average total of 32,333 AOS (Table 3), an overall decrease of 1.5% since 1994, at 0.1% pa. The land visit in 2003 and comparison with the 1994 photographs failed to detect any changes in colony extent, except for a small increase on the upper east side of section 83 to 88. Given the available information, numbers on Boreray are probably stable. Stac Lee The average total was 13,369 AOS (Table 3), a decrease of 8.8% since 1994, at 0.8% pa. Comparison with the 1994 photographs failed to find any obvious changes in the densely packed Top Table or on narrow ledges that comprise breeding sites on other areas of the stack. Areas SB 26 unoccupied in 1994 that were apparently suitable breeding habitat remained unoccupied, and had not changed in extent. The apparent decrease is within the limits of intra and inter observer variation and suggests that numbers are stable. Stac an Armin The average total was 13,920 AOS (Table 3), an increase of 7.5% since 1994, at 0.7% pa. Comparison with the 1994 photographs clearly shows areas of increase, at the foot of the prominent centre gully of the South Summit / East Face section and the lowest corner of the South Centre section. Increases in both these areas were first noted in 1994. Table 3 Counts (AOS) of the St Kilda Gannetry in 1994 and 2004. Sub Count 15 May 21 June colony Section 1994 2004 Boreray | 21 30 2 50 58 3 to 12 1892 1721 13 to 15 1756 1558 16 to 41 5815 5506 42 & 43 1316 1410 44 89 65 45 500 593 46 576 610 47 784 684 48 835 854 49 1063 1044 50 34 D5 51 to 57 4068 4247 58 to 63 2447 2408 64 549 506 65 455 386 66 463 450 67 to 72 1828 1795 We 426 322 74 to 78 2622 235d Scottish Birds (2006) 79 381 542 80 382 332 81 & 82 1341 1587 83 to 88 2831 2875 89 294 362 Boreray Total 32818 32333 Stac Lee Top Table 7141 7247 Bothy Face =1525 [273 Casting Point 2877 2242 North Face i 2607 Stac Lee Total 14660 13369 Stac an South Summit / East Face Armin 9089 9830 Lower East Face 1487 1436 West Face 1204 1155 South Centre / Lower South Ledge 1259 1500 Stac an Armin Total 12950 13920 St Kilda Total 60428 59622 Flannan Isles The aerial survey was made on the 26 May 2004. The average total of 3 counts by one observer was 2,760 AOS (Table 4), an increase of 91.9% since 1994 at 6.7% pa. There were increases in all sections where breeding had occurred in 1994, most notably in the South colony of Roareim, and the offshore islet Sgeir an Eoin. In 1994 breeding was suspected in the East section, but in 2004 this appeared to be used entirely by non breeders, which left the ground on the approach of the aircraft. On the adjacent island, Eilean a Gobha, where displaying pairs were seen on 24 May 1992, there were no birds present in 1994, 1998 or 2004. A count by M Tasker, G Leaper and S O’Brien, from both land and sea on 3 June 1998 reported The status of the Northern Gannet in Scotland in 2003-04 21 1244 AOS (Wanless & Harris 2004). This was unexpectedly low compared to the aerial survey results obtained in 1994 (1436 AOS) and 2004 (2760 AOS). However, despite the small scale of the colony it is difficult to find a clear viewpoint that adequately covers all the breeding groups, either from the land or the sea, whereas the low lying nature of the islets make them ideal for aerial photography. The most likely explanation for the count fluctuations are the different count methodologies used, and the more complete colony coverage obtained by aerial survey. Table 4 Counts (AOS) of the Flannan Isles Gannetry in 1994 and 2004. Count Section 15 July 26 May 1994 2004 West Stac 8 51 Arch Stac 107 145 Sgeir an Eoin 248 625 Main 960 1302 South 13 637 East 0 0 Total 1436 2760 Sula Sgeir The aerial survey was made on 26 May 2004 and achieved 100% photographic coverage, with good to excellent image quality. Four counts made by 2 observers gave an average total of 9,225 AOS (Table 5), a decline of 11.6% since 1994 at 1.2% pa. Comparison with the 1994 photographs, partic- ularly those of the summit plateau and around the helipad, revealed no changes in colony extent, either increases or decreases, nor any differences in the limits of sub colony boundaries, however birds appear to be nesting at relatively low densities throughout the colony. a2 SB 26 Table 5 Counts (AOS) of the Sula Sgeir Table 6 Counts (AOS) of the Sule Stack Gannetry in 1994 and 2004. Gannetry in 1994 and 2004. Count Section 15 July 26 May —_ Count Section 15 July 26 May 1994 2004 1994 2004 I 879 546 = North 0 0 2 384 Sof North West qd 748 3 2954 2350 North East 632 615 4a 2038 1944 East 357 351 4b 1154 1121 South East | 1296 ] 569 5 0 0 Centre | ] 633 6 h353 1339 ~— Top 1755 1581 7 1680 1589 South 73 121 helipad 0 O Total 4888 4618 Total 10440 9225 Sule Skerry Sule Stack The first successful aerial survey was made in July 1969 for Operation Seafarer and found 4018 pairs occupying nests (Cramp et al 1974). The second aerial survey, on 15 July 1985, was too distant from the rock and the total of 5880 AOS is undoubtedly an overestimate (Murray & Wanless 1997). Subsequent surveys in 1994 and 26 May 2004 (Table 6) have delivered photographs of outstanding coverage and quality. Two whole colony counts made by one observer gave an average total of 4618 AOS, an overall decrease of 5.5% since 1994, at an average rate of 0.6% pa. However, comparisons with the 1994 photographs found no changes in the largest sections, but a small increase in the number of AOS in the South section. This population appears stable, and most probably has been since at least 1969. Figure 2. The site of the Gannetry on Sule Skerry in 2004. Between 1992 and 2002 Gannets were period- ically reported ashore on the west side of the island, (Fig 2) and on the 13 July 2003 15 pairs had sites with either eggs or chicks among nesting Common Guillemots Uria aalge (Budworth & Blackburn 2004). To prevent disturbance to this newly established colony, no aerial survey was made. A land count, made by D Budworth on 8 July 2004 gave c40 occupied nests, at least 10 of which had chicks. Counts by the authors from a photograph taken on land that day all suggested 57 AOS. However, there were Helipad Lighthouse Gannetry Scottish Birds (2006) slight differences in interpretation of the photograph, so the total was probably between 55-60 AOS (Wanless et al 2005a). Westray Gannets were first noticed on The Noup, Westray in 2003, when 5 nests were found among breeding Common Guillemots, but only 2 eggs were laid and probably only one chick fledged. A count on 21-22 June 2004 by K Judd (RSPB) found numbers had increased to 14 nests in 3 distinct groups (Wanless et al 2005b). Fair Isle The 2004 count of 1875 AON was made on 13 June 2004 by C Bailley and D Shaw (Fair Isle Bird Observatory Trust) (Table 7). Counts were made from the land except for Sheep Rock and the outer face of the Inner Stack of Scroo. There had been a 127% overall increase since 1994, at an average rate of 8.6% pa. All the established sub colonies increased, and in 1997 a pair colonised Sheep Rock; only a single bird was present in 1998, but since then an increasing number of pairs has bred each year. Table 7 Counts (AON) of the Fair Isle Gannetry in 1994 and 2004. Count Section 13 June 13 June 1994 2004 Outer Stack 219 584 Inner Stack of Scroo 66 205 Yellow Head 34 48 Dronger TW 7 North Felsigeo 285 436 Toor ‘o’ da Ward Hill 82 209 Matchi Stack 30 65 Kame ‘o’ Guidicum 32 226 Sheep Rock 0 23 Total 825 1875 The status of the Northern Gannet in Scotland in 2003-04 23 Foula A total of 879 AOS was counted from the land and sea by S Gear, and a further 40 on the summit of Da Stab stack estimated from the air, giving a total of 919 AOS on 3 July 2004. The 1994 survey on the 2 and 16 July found c600 AOS, thus giving an increase of 130% at an average rate of 8.7% pa over the decade. Hermaness The majority of nests were counted from the land on 22 and 23 July 2003 by S E Duffield, J Swale and M Pennington (Duffield 2003), with areas hidden from the land counted from a boat on the 26 June 2003. The total count was 15,633 AON, an increase of 30% since 1994 at 3.0% pa (Table 8). Colony extent also expanded between 1999 and 2003, with birds colonising both the Greing and Flodda Stac. The 2003 total would undoubtedly have been higher had a landslide not destroyed several hundred nests on Saito prior to the survey. Table 8 Counts (AON) of the Hermaness Gannetry in 1994 and 2003. 12 to 18 July 26 June & & 22 August 22 & 23 July Count Section 1994 2003 Rumblings East 736 869 Vesta Skerry East 1381 1745 Vesta Skerry West 623 485 Humla Stack North Day 265 Humla Stack West ZO 101 Humla Stack Southwest 0 214 Humla Houl North 133 15] Humla Houl South 435 478 Burra Stack East 335 377 Burra Stack West 238 264 Clingra Stack 207 317 Neap North Face 492 1190 24 Neap - Soorie 3096 3873 Soorie - Saito 1799 2606 Saito 1509 1941 Soorie Stacks 54 60 Neapna Stack 453 550 Flodda Stack 0) 38 Greing West 0 109 Total 11993 15633 Noss Two whole colony counts of nests were made by S Smith, the first between 8-14 June 2003 and another between 9-30 July (Marshall & Thomas 2003). About 90% were counted from the land and those in hidden areas from the sea. The average colony total was 8652 AON, an increase of 18% since 1994 at 1.9% pa (Table 9). Table 9 Counts (AON) of the Noss Gannetry in 1994 and 2003. Count Section 18-20 June 8-14 June & & 17 July 1994 9-30 July 2003 Cradleholm l 4 Holmoless 108 93 Holmoless to Geordies Hole 1569 2008 Geordies Hole 49 70 Rumblewick Zao 355 Rumblewick Face 162 320 Cuddacks Geo 0 3 Noup South 2659 2850 Noup East a7 709 Noup North 910 965 The Rump aa 885 Rump North 339 387 Geo Heogatoug 0 3 Total 7310 8652 Figure 3. The main features of the Bass Rock and the boundaries of count sections in 2004. SB 26 Troup Head The only gannetry on the Scottish mainland, Troup Head continues to expand, from 530 nests on 31 May 1995 (Wanless et al 1996) to 1547 AON on | July 2004 (Mavor et al 2005). This is the largest percentage increase of any Scottish colony, 200% since 1995 at 20.5% pa. Bass Rock The aerial survey was made on 21 June 2004 and covered the entire colony. The rock is now completely sub divided into 12 sections, using natural and man made features as boundaries (Fig 3).Two counts were made by one observer and 24% of the colony total was counted by a second observer, giving a mean count total of 48,065 AOS (Table 10). Although an increase of 20.9% since 1994, at an average rate of 1.9% pa this was a much reduced rate compared with the 5.3% pa increase for the previous decade, although this had been calculated using totals of AONs (Murray & Wanless 1997). Comparison with the 1994 photographs clearly shows large increases in colony extent, the most obvious being in Section 9 (Table 10). Here, breeding birds have become established up to the summit of the rock, but gauging the extent of Footpath from lighthouse to foghorn Scottish Birds (2006) this expansion is complicated by the large numbers of non breeding and club birds around the section edge, which are reluctant to leave the colony at the approach of the aircraft. A land visit had to be made on 17 July 2004, to define both the limits of breeding in this section and in Section 12. Although most of the birds in Section 12 were non breeders, pockets of nesting occurred throughout the section, with partic- ularly high numbers adjacent to Sections 7 and 8. Section 11 had been colonised since 1994 and both here and in Section 12 there is space for future increases. Sections 1,3,7 and 10 all show clear increases in colony extent. Section 8 has no room for expansion and appeared to be full in 1994. Numbers in cliff sections 2, 5, and 6 appeared to have declined, compared with both 1985 and 1994, but the reason for this is unclear. Overall, the population has more than doubled in size since 1985, and despite the reduced growth rate, the large numbers of non breeders attending the colony suggest that the numbers of breeders will continue to increase. The status of the Northern Gannet in Scotland in 2003-04 Za) Total numbers in Scotland The counts documented above were combined to provide an estimate of the total Scottish population in 2003-04. The overall total for colonies counted using AOS was 154,790, the total for those counted using AON was 27,721 (Table 11). Combining these figures, and for convenience expressing the total as the number of AOS, gave a grand total of 182,511 AOS. Numbers were divided very unequally between the 14 colonies, with St Kilda, the Bass Rock and Ailsa Craig together holding 74% of the total. In 1994-95 the Scottish population was estimated at 167,407 AOS (Murray & Wanless 1997). Comparing the 2003-04 total with this value indicates that numbers increased by 6% at an average rate of 0.6% pa, compared with the previous decadal increase of 27% at an average rate of 2.4% pa. Table 10 Counts (AOS) of the Bass Rock Gannetry in 1985, 1994 & 2004. Count Section 1 & 11 June 1985 1 1227 Ds 320 3 3436 4 505 5) 1909 6 1560 7 4185 8 2707 9 5371 10 369 1] 0 12 0 Total 21589 11 July 21 June 1994 2004 273) 4087 645 240 6867 7226 613 349 1026 442 2119 1066 7885 8191 3630 3530 12651 18855 1578 2925 0 157 0 1000 39751 48065 26 SB 26 Table 11 Summary of counts of Scottish Gannetries in 2003-04 and changes since the 1994-95 survey. Colony Scar Rocks Ailsa Craig St Kilda Flannan Isles Sula Sgeir Sule Stack Sule Skerry Foula Bass Rock Westray Hermaness Noss Fair Isle Troup Head Total Grand total Count AON AOS 2394 27130 59622 2760 9225 4618 57 919 48065 14 15633 8652 1875 1547 2hi2k 154790 182511 % Scottish total 1.3 14.9 cyae 1.5 el 2 <0.1 0.5 26.3 <0.1 8.6 4.7 1.0 0.8 % change overall pa 22.6 23 - 16.4 -2.0 -13 -0.1 91.9 6.7 - 11.6 -1.2 - 5.5 -0.6 colonised 130.0 8.7 20.9 1.9 colonised 23:3 3.0 15.6 1.9 127.0 8.6 200.0 20.5 6.0 Table 12 Summary of results from the 2003-04 east Atlantic Northern Gannet survey 0.6 (Wanless et al 2005a & b) Country Number of Year Total % total colonies counted AOS/AON east Atlantic Scotland 14 2003-04 182511 58.0 Ireland 5 2004 36111 11.5 Wales pi: 2004 32095 10.2 Channel Islands ! 2 2005 7409 23 England | 2004 3940 1.3 Iceland ” 5 1999 28536 9.0 Norway 5 2002-03 4500 1.4 Faeroe Islands l 1995 2340 OF Russia 1 1998 35 0.0 Germany ] 2004 190 0.0 France 1 2004 17000 5.4 Total 38 314667 100.0 Note: ! Sanders & Atkinson 2006 7 Gardarsson 2005 Scottish Birds (2006) The status of the Northern Gannet in Scotland in 2003-04 27 Discussion Although the Northern Gannet population continues to increase in Scotland and 2 new colonies have been founded in Orkney, this steady rise in numbers could be coming to an end, as the overall rate of increase appears to be slowing down. St Kilda, although still Scotland’s largest gannetry, has seen negligible growth since 1994 and is at present stable. By contrast, the Bass Rock, the second largest colony, continued to increase, currently at 1.9% pa, but this is a marked reduction on the 5.3% pa increase recorded up to 1994. Although the Bass Rock could theoretically overtake St Kilda within a decade, it possibly does not have the space to accommodate the 12,000 nests needed to do so. In contrast, Ailsa Craig the third ranking gannetry has experienced a decline of 16.4% since 1995. The losses were not uniform across the colony, and they have not been offset by the small increases in extent and density in some sections. Sula Sgeir is unique among British and Irish Gannetries in that licensed killing of young birds, ‘gugas’, by the men of the parish of Ness in Lewis (Beatty 1992) takes place annually and here the colony has declined by 11.6% since 1994. The current licence, issued by the Scottish Executive with advice from SNH, is for 2000 young to be taken. In every year since 1985, except 1999 (1690) and 2004 (1984) the licence returns have been of exactly 2000 birds (Scottish Executive returns). Without more data on colony demography it is not possible to determine the true effect of the guga hunt, but it would appear to be the most likely cause for the lack of growth of the colony (Wanless et al 2005a). A clearer picture of the total number of young taken, and the effects of disturbance on non target young would go some way to clarifying the situation. The gannetry on Sule Stack has also shown little change during the last 20-35 years, but here there appears to be a shortage of space, and that may have resulted in the founding of a new colony on nearby Sule Skerry in 2003. Coincidentally, also in 2003, Gannets bred for the first time on Westray, also in Orkney. The 2 larger Shetland colonies of Hermaness and Noss have also increased since the mid 1990s, and all the 5 smaller colonies, Scar Rocks, Flannan Isles, Foula, Fair Isle and Troup Head, that were extant in 1994-95, showed substantial increases. In 2003-4 the east Atlantic and total world population of Northern Gannets was estimated at c314,000 and c418,800 AOS (Wanless eft al 2005b). Scotland remains the species stronghold with 58% and 44% of these populations respec- tively. In 1995 we wrote ‘that all the evidence suggests that conditions are extremely favourable for Gannets’. At that time all the Scottish colonies were expanding, and there seemed no reason to suppose there would be limits to this growth in the near future, as most colonies were not constrained by a shortage of breeding sites or food resources. However, the evidence from this survey shows that the overall growth rate has halved and this suggests that conditions are becoming less favourable. Why this might be happening is unclear, but there is increasing concern that climate change is beginning to affect marine life in the east Atlantic, and the North Sea in particular. How this might affect Gannets is not yet obvious, and while they were one of the few breeding seabirds not to suffer major breeding failure in Scotland in 2004 (Mavor et al 2005), it may be only a matter of time before they do so. An analysis of Gannet ringing recoveries for the UK and Ireland has found evidence for an increase in adult mortality over the last 10 years (Wanless et al 2006). The reason for this is unknown, but if adult mortality remains high and breeding success falls, we can expect to see a further 28 Northern Gannets at Sula Sgeir decline in the rate of increase, or perhaps even in absolute numbers of the Northern Gannet in Scotland by the time of the next survey. Acknowledgements We sincerely thank the many people who took part in this survey, particularly those whose efforts are acknowledged in the text. Hebridean Air Services pilot George Cormack made the aerial surveys possible, for which we are extremely grateful. Jill Harden, Matt Parsons, Stuart Benn and Tony Mainwood provided photographic support for the flights. Sharon Evans meticulously carried out the replicate counts, and with Malcolm Collie and Denise Wright created the database and CD archive. We are also grateful for the assistance of Natalie MacColl on St Kilda and in various ways to Andy Stronach, John Davies, John Chardine, Jez Blackburn and Roddy Mavor. The survey was made under contract for Scottish Natural Heritage and we thank Rhys Bullman for facili- tating this work. SB 26 Arthur Grosset References Beatty J 1992. Sula:The Seabird Hunters of Lewis. Joseph, London. Belaoussoff S 1993. Northern Gannet and Common Guillemot nesting on Rockall. British Birds 86: 16. Blackburn A & Budworth D 2004. A new gannetry for the UK. Seabird Group Newsletter 96: 10. Cramp S, Bourne W R P & Saunders D 1974. The Seabirds of Britain and Ireland. Collins, London. Duffield S E 2003. Hermaness and Keen of Hamer National Nature Reserves annual report 2003. Unpublished report, Scottish National Heritage, Lerwick. Gardarsson A 2005. Numbers of Northern Gannet Morus bassanus in Iceland 1999. Bliki 26: 17-20. Harris M P & Lloyd C 1977. Variations in counts of seabirds from photographs. British Birds 70: 200-205. Scottish Birds (2006) Marshall L & Thomas A 2003. Noss National Nature Reserve annual report 2003. Unpublished report, Scottish National Heritage, Lerwick. Mavor R A, Parsons M, Heubeck M & Schmitt S 2005. Seabird numbers and breeding success in Britain and Ireland, 2004. Joint Nature Conservation Committee (UK Nature Conservation, No, 29), Peterborough. Murray S & Wanless S 1986. The status of the Gannet in Scotland 1984-85. Scottish Birds 14: 74-85. Murray S & Wanless S 1996. A census of the St Kilda gannetry in May 1994. Scottish Birds 18: 152-158. Murray S & Wanless S 1997. The status of the Gannet in Scotland in 1994-95. Scottish Birds 19: 10-27. Nelson J B 1978. The Gannet. T & A D Poyser, Berkhamsted. Sanders J & Atkinson M 2006. Alderney Nesting Gannets. The Seabird Group Newsletter 101:16-17. Wanless S 1987 A survey of the numbers and breeding distribution of the North Atlantic gannet (Sula bassana) and an assessment of the changes which have occurred since Operation Seafarer 1969/70. Research and Survey in Nature Conservation 4: 1-100. Wanless S, Matthews J & Bourne W R P 1996. The Troup Head Gannetry. Scottish Birds 18: 214-221. Wanless S, Frederiksen M, Harris M P & Freeman S N 2006. Survival of Northern Gannets Morus bassanus in Britain and Ireland, 1959-2002. Bird Study 53: 79-85. Wanless S & Harris M P 2004. Northern Gannet Morus bassanus. \n Seabird populations in Britain and Ireland. (eds) P I Mitchell, S F Newton, N Ratcliffe & T E Dunn), pp. 115- 127.T & AD Poyser, London. Wanless S, Murray S, Evans S & Harris M P 2005a. Scottish Gannet Survey 2004. Scottish Natural Heritage Commissioned Report FO4AC301. The status of the Northern Gannet in Scotland in 2003-04 29 Wanless S, Murray S & Harris M P 2005b. The Status of the Northern Gannet in Britain and Ireland in 2003-04. British Birds 98: 280-294. Wanless S, Murray S & Harris M P 2005c. Reply to Zonfrillo. British Birds 98: 552-553. Zonfrillo B 2005. The Ailsa Craig Gannetry in 2004. British Birds 98: 551-552. Stuart Murray, Easter Craigie Dhu, Cardney, Dunkeld, Perthshire PH8 OEY. e-mail murraysurvey @yahoo.co.uk Sarah Wanless & Mike Harris, NERC Centre for Ecology and Hydrology, Banchory Research Station, Hill of Brathens, Banchory AB31 4BW. e-mail swanl@ceh.ac.uk Revised manuscript accepted March 2006 30 SB 26 Waterfowl counts on the Tay Estuary N ELKINS Counts of waterfowl populations on the Firth of Tay between 1995 and 2005 are compared to the results from earlier decades, especially the decade 1985-95. Changes have led to a decline in the importance of the estuary to some species and the total number of waders using the estuary has declined significantly. Eurasian Oystercatcher, Dunlin and Common Redshank are the most notable species to have been affected. Disturbance, climate change and roost site configuration are suggested to be possible reasons for the declines. Introduction The Firth of Tay is the third largest tidal estuary in eastern Scotland and is a key site for both migrating and wintering waterfowl. Together with the nearby Eden estuary to the south, it is subject to several conservation designations (FERN 2005) on account of internationally important populations of waterfowl. The whole area was declared a Special Area of Conservation (SAC) in 2002, with parts having been a Special Protection Area (SPA) and a Ramsar site since 2000. The area also includes Sites of Special Scientific Interest (SSSI), and both Local and National Nature Reserves (LNR, NNR). An earlier paper (Elkins & Lynch 1997) described the estuary and its waterfowl, giving a summary of the status of most species resulting from Wetland Bird Survey (WeBS) counts during the period September 1985 to August 1995. During the 10 years since then, ie from September 1995 to August 2005, several changes have occurred, some of them significant. Methods Monthly high tide roost counts of Great Cormorants, wildfowl and waders are made throughout the year as part of the Wetlands Bird Survey (WeBS) organized by the British Trust for Ornithology (BTO) in partnership with the RSPB, WWT and JNCC. During winter 1996- 97, a series of counts of birds was also made at low tide under contract to Scottish Natural Heritage (SNH), the data from which contributed to the WeBS Low Tide Count scheme (see Musgrove et al 2003). High tide roost counts were made on a specified date each month by a team of observers, usually at spring tides near the middle of the month, although alternative dates were chosen during short winter days when high tide occurred during the hours of darkness. The 4 roost counts are undertaken in the outer Tay at Tentsmuir Point, Tayport and Monifieth to Broughty Ferry and in the inner Tay at Invergowrie and Kingoodie. There were, understandably, some months when one or more of the roosts remained uncounted, mainly due to observer shortage. These missing counts were chiefly in the summer months, with a high degree of coordi- nation otherwise maintained. During the 20 years from 1985-86 to 2004-05, 92% of counts during the priority months between September and March were complete. 66% of all counts were fully coordinated, with the remainder counted within 1-5 days. During April to August, 49% of counts were complete, with 33% fully Scottish Birds (2006) 26: 30-38 Waterfowl counts on the Tay Estuary 3] coordinated. Wildfowl were less comprehen- sively counted in the early years and remain poorly covered in the upper reaches of the firth in the vicinity of Mugdrum Island. Counts made before 1985 (some published in Bird of Estuaries Enquiry (BoEE) annual reports and in Prater (1981)) were very intermittent but a few have been used in comparisons with later counts. Data from the Eden estuary are included where relevant. Winter counts described below are attributed to the year at the start of the winter. All counts used in calculating means are those resulting from WeBS Core Counts ie counts made on the specified monthly dates. Casual counts on other dates can often be significantly different and, if used here, are only quoted for illustrative purposes. Winter means are calculated using the highest count in the months from December to February. Results The mean maximum number of waders (i e excluding wildfowl) roosting on the Tay estuary in midwinter was 7,723 for the period 1995- 2004, a decrease of 24% from the previous 10 year mean of 10,130. These figures were derived from summing the highest monthly counts of each species between December and February. They varied from 12,484 in the winter of 1985- 86 to 5,918 in 2004-05, coincidentally the first and last winters of the 20 year period (see Fig 1). Monthly totals at other seasons illustrated changes in the number of passage migrants. In September 1996, the total number of waders roosting on the estuary was 13,400, the highest number ever counted in any month during the 20 years. This contrasts with 3,535 in September 2002. Fewest waders were present in June, with a mean of 706, although few comprehensive counts were made during this month (see Fig 2). There are few published figures of total wader counts before 1985-86. However, those available compare well with the early years of this study, being in the order of 13,000-14,000, although 20,300 were reported in the winter of 1973-74 (BoEE annual reports). If wildfowl are included in the totals for the whole estuary, a decline is still evident. The 5 year mean had dropped from 26,844 in the late 1980s to 22,332 for the period 1999-2003, a fall of 17%. Species Accounts Eurasian Oystercatcher Haematopus ostralegus The 5 year winter means show a decrease over the 20 year period (see Fig 3). The midwinter population fell from 2167 in 1985-89 to 1645 in 2000-04, a fall of 24%. Most of this decrease was due to fewer birds roosting in the inner Tay and occurred during the last 5 years. Despite this, the species remains the second most abundant species in midwinter. However, numbers on autumn passage have not changed significantly and have shown a steady increase at Tentsmuir Point at the expense of the Monifieth roost site, mainly in October. This contrasts with the Eden estuary where a marked autumn decrease has occurred from the mid 1990s. The species is now the most abundant autumn wader at Tentsmuir Point in contrast to winter, when only a small percentage of the estuary’s total is now found there. There was an increase in spring at Tentsmuir Point during 2001-2005, mainly in March. This suggests a growth in the use of the outer estuary by migrants. Midsummer flocks of immature birds and non breeding adults have remained steady at around 400 birds, by far the commonest roosting wader at this season when the highest proportion occurs in the inner estuary. As in the earlier years, low tide counts in 1996-97 showed that the winter roosting population was lower than that using the estuary to feed. This has been attributed to movements outwith the estuary. Total Mean count 5 yr mean 32 ay 10000 - Y ; AV Z Y —YOAZZ my Y G [7 6000 Y WAZ Y YY WN y YOO wo AAGIAAAAZVinV7 WO 0! 7 ZAAAAA AAA Us Z| ” 7 7 ” Winter stave : 7 ° . Figure 1. Peak wader totals roosting in winter on the Firth of Tay \ \ININ Feb Mar Apr May Jun Jul Aug Month Oct Nov Figure 2. Monthly mean number of waders roosting on the Tay, 1985-2005. Tay Estuary - midwinter @ 1985-89 & 1990-94 WN 1995-99 ® 2000-04 u Y, tn MAAS ; \ y \ ZN. Yr. 1500 AN Ve GY 1000 ZN \ Z = GN GR Ge DN BA Species (OC = Eurasian Oystercatcher, DN = Dunlin, BA = Bar-tailed Godwit, RK = Common Redshank, CU = Eurasian Curlew) Figure 3.5 year means for 5 species of waders wintering on the Tay, 1985-2005. SB 26 Numbers in the early years of this study were similar to those published for 1969-74 (Prater 1981) but the latest data for Great Britain show a decrease from a peak in 1990-91 (Collier et al 2005). Ringed Plover Charadrius hiaticula Minor fluctuations have occurred during winter but with little overall change and a mean of 113 birds; this showed an increase from data published for 1969-74 by Prater (1981). The main roosts remain in the outermost Tay, where a few pairs also breed. There was an increase here during autumn passage, especially at Monifieth in October, with a mean of 273 in the last 5 years representing a sevenfold increase since 1985-89. Estuary totals as a whole in August have increased threefold during the period 1992 to 2005, with a mean of 306 during 2001-2005. Late spring totals, presumed to involve Greenland breeders, reached in peak in 1999, but have declined since. However, the transitory nature of these spring migrants could mean that WeBS count dates may miss the largest flocks on both the Eden and the Tay. For example, the Eden estuary recorded its highest ever casual count of 1030 in May 2002, not reflected in WeBS counts. Over Great Britain as a whole, the wintering population has declined steadily from a peak in 1988-89 (Collier et al 2005). European Golden Plover Pluvialis apricaria The European Golden Plover remains a very mobile species, with its occurrence on the estuary mainly confined to low tide. Wintering flocks roosting at high tide, often on coastal fields, have declined markedly since the mid 1990s, although autumn migrants show an increase in October. This contrasts with a decrease in November, and may represent earlier migration. Combined totals on the Tay and Eden also show this change. On the nearby Eden Scottish Birds (2006) estuary, large flocks also roost and feed at low tide. Here, winter high tide roosts remain substantial with a mean of 200 birds, although a recent decrease has been noted. Grey Plover Pluvialis squatarola After a peak in the mid 1990s, the wintering population has shown only a slight decline from a mean of 259 in 1995-1999 to 236 in 2000- 2005. The species roosts almost entirely at Tentsmuir Point, where the recent decline has been most noticeable in January and February in both 2004 and 2005. On the nearby Eden estuary totals halved during the same months, although a mean of 280 was still present during 2004-05. The Grey Plover population increased markedly during the past few decades but reached a peak in Great Britain in 1996-97 after which a steady decline has been recorded (Collier et al 2005; Maclean et al 2005). Both spring and autumn passage totals have declined on the Eden and the Tay since 2002. Northern Lapwing Vanellus vanellus Like the European Golden Plover, Northern Lapwing numbers fluctuate widely, but they are more frequently seen at high tide roosts. They occur earlier than the European Golden Plover on autumn migration, peaking in August and September. The inner estuary holds the largest proportion of Northern Lapwings throughout the year, but there has been a fourfold decline in numbers of roosting birds throughout the winter, perhaps reflecting the high incidence of mild winters when birds are able to feed inland. The Eden estuary has also experienced a decline in wintering birds. Red Knot Calidris canutus This species, never abundant on the Tay, has decreased in winter. This decline has also occurred on the Eden, although there is still a mean winter population of 400 birds and casual counts still exceed 1000 at times. Red Knot roost Waterfowl counts on the Tay Estuary 33 mainly at the outer Tay sites, where a decline at Tentsmuir Point has been partially offset by an increase at Monifieth. Marked fluctuations occurred during the early years but the wintering population has stabilised at around 250 birds since the mid 1990s. Sanderling Calidris alba Sanderling decreased in the earlier decade but the wintering population has stabilised at around 100 birds since 1998. The species remains more abundant during passage periods, especially in autumn when around 150 are present. However, in common with other species, the main passage roost has shifted from Tentsmuir Point to Monifieth, although this is partly dependent on the amount of disturbance. At low tide, feeding birds remain confined to the outer estuary, chiefly on the Buddon beach. There is some evidence that Sanderling roost on uncounted adjacent sandy shores, with increases at counted sites during disturbance elsewhere. This was highlighted in December 2005 when a flock of 560 Sanderling arrived at Tentsmuir Point on high tide. This was the highest count for 18 years. Dunlin Calidris alpina A major decrease of Dunlin took place after 1999 with a mean midwinter population of just over 1600 birds in the period 2000-04. The roost at Invergowrie contributed to this decline to a large degree, falling significantly between 1990-94 and 2000-04. Some of these birds may have moved to Monifieth, which showed an increase in the winter roost size, but it is believed that the majority of birds displaced from Invergowrie may have moved to the Eden estuary where the midwinter population doubled between 1997 and 2004. However, the proportion of Dunlin to the whole wintering wader population remains at about 30%, accentuating the decline in other species. The large Dunlin passage roosts in the past have also declined, again mostly at Invergowrie with a 34 SB 26 small increase at Monifieth partially offsetting this. Overall, autumn passage has declined on both the Tay and Eden. The low tide counts in the winter of 1996-97 showed a similar distribution to those in 1993- 94, although birds were more evenly distributed along the north shore mudflats upriver from Invergowrie. Bar-tailed Godwit Limosa lapponica Totals show only minor fluctuations over the period; the species has thus increased its proportion of the total wintering wader population from 15% to 21% during the last 2 decades. The main roosts remain at Tentsmuir Point and Monifieth although a significant decline at the former has been offset by a substantial rise at the latter. Indeed few birds were recorded at Tentsmuir Point in the last 2 winters of this study. The species is of interna- tional importance on the Tay (i e numbers exceed 1200 birds) with a mean wintering population of 1457 during the decade in question. The winter of 1996-97 held the highest population during the period (2305) coincident with the peak in the Great Britain index. Recent individual months have produced high counts both in autumn (e g 2664 in September 2003) and in winter (e g 1944 in January 2002). Eurasian Curlew Numenius arquata Few wintering Eurasian Curlews roost at the entrance to the Tay, favouring the more sheltered sites at Lucky Scalp and Invergowrie. The species remains one of the less common waders with a winter mean of around 340, probably due to its habit of feeding inland during high tide. On both the Tay and the Eden, Eurasian Curlews have become more abundant on passage in late summer and early autumn since the mid 1990s. For example, Tentsmuir Point is now used much more during autumn compared to the earlier decade. Common Redshank Tringa totanus Unlike the earlier decade, wintering numbers of Common Redshank on the Tay do not now reach the criteria for importance either internationally or for Great Britain. The mean midwinter roosting population is now around 700-1000. This is due almost solely to the decline in birds roosting at Invergowrie where numbers have rarely exceeded 500 in recent years. Roosting patterns have also changed at Tayport, where it is the only wader to use the saltmarsh as a roost. Disturbance here often flushes them on to newly created islets on the pond at Tayport. Autumn passage numbers have also declined, with a peak of 1979 in September 2003, compared to numbers in excess of 3000 prior to 1997. The peak was 5358 in September 1996, of which the majority were at Invergowrie. This decline has also been noted at the Eden estuary. Such decreases are all the more notable since the Common Redshank is relatively site faithful and numbers have been relatively stable in Great Britain as a whole for many years (Collier et al 2005). Ruddy Turnstone Arenaria interpres The small wintering Ruddy Turnstone population favours the rocky outcrops in the outer estuary, mainly on the north shore. Fewer than 100 birds are present and little change has occurred over the period. Rather more pass through in spring, with 170 in March 1998. Other wader species Small numbers of other species are recorded on the estuary, mainly during migration. The only other regular wintering species is Common Snipe Gallinago gallinago, which favours salt marshes in the inner estuary and at Tayport. Counts appear to have increased in recent years with a peak of 53 in November 2002, but numbers using the saltmarshes are governed by 5 yr mean Mean count Scottish Birds (2006) Tay Estuary - midwinter 1985-89 i ' 1990-94 4 9 1995-99 | & 2000-04 | 300 - 7 250 | =N yy % 2Ne = a YN Yew =NAW =| 150 - YN = S G Ww 2NY 100 Ge WG 2NY YEN YeNY 50 YG w 2NB : LN ON. RP GV KN ss Species (RP = Ringed Plover, GV = Grey Plover, KN = Red Knot, SS = Sanderling, TT = Ruddy Turnstone) Figure 4.5 year means for 5 species of waders wintering on the Tay, 1985-2005. Yaa Month Figure 5. Mean monthly count of roosting Great Cormorant on the Firth of Tay, 1995-2005. the height of the tide and many of those in the inner estuary remain hidden and uncounted. Black-tailed Godwits Limosa limosa are very scarce on the Tay although the Eden estuary wintering population regularly exceeds 200. However, in the winter of 2005-06, after the end of the study period, a small population roosted at Invergowrie, with 50 birds in December 2005. This coincides with the sharp increase in the UK Waterfowl counts on the Tay Estuary 35 winter population since 1999 (Maclean et al 2005). A record count of 615 on the Eden in December 2004 mirrors this. Great Cormorant and Wildfowl Great Cormorant Phalacrocorax carbo There are normally Great Cormorant roosts on Lucky Scalp and the Pile platform off Tayport, but the majority uses the old Tay rail bridge columns where most were also recorded during low tide counts. Overall, mean counts begin to increase in August and reach a peak of 183 in September (Fig 5). Thereafter, they decline steadily to a mean of 34 in April. However, individual counts have exceeded 220 between August and October and in December. Mute Swan Cygnus olor Mute Swans flock on the Tay in considerable numbers between April and September. In recent years, flocks have grown, with a peak of 226 in July 2003. Most of these birds favour the shore at Monifieth, where a small wintering population of around 20 birds is also found. Further parties are found upriver and are not often monitored. Common Shelduck Tadorna tadorna Common Shelduck numbers have shown little change over the years. A spring build up occurs from February with a peak reached in April when 187 were counted in 1996. Since then, there has been a slight decrease although over 100 are still present at this time. Although a few pairs breed, numbers fall to single figures by August and only a few overwinter. Common Shelduck are distributed throughout the whole estuary and significant numbers near Mugdrum Island are not often counted. For example, 77 were present in this area in February 2003 only a few days prior to a total WeBS count of 41. The Eden estuary holds double the numbers present on the Tay. 36 Eurasian Wigeon Anas penelope This winter visitor arrives on the estuary in September and departs in March. The peak is reached in early winter, and the mean midwinter population (chiefly in the Tayport area) has increased from 160 in 1985-89 to 319 in 2000- 2004. An earlier decline in the mid 20th century (Atkinson-Willes 1963) had not reversed even by 1969-74, when the mean count was only 30 (Prater 1981). The population exceeded 400 in 2 of the 5 latest winters and an exceptionally high number of 824 was present in October 2005, after the period under analysis had ended. This is another species for which flocks of several hundred are often found upriver near Mugdrum Island and remain regularly uncounted. Small numbers of Eurasian Wigeon are present intermit- tently in summer (normally in single figures) although none has been recorded on July counts. Eurasian Teal Anas crecca This species declined in the early 1990s (Elkins & Lynch 1997), but has since recovered to previous levels. The decline was first evident during 1990- 94, when the mean midwinter count fell to 24 individuals from a mean of 103 during 1985-89. The level remained at a similarly low level until 1997-98 after which numbers rose steadily to a mean of 150 in 2000-04 with a record 302 in January 2002. Although Eurasian Teal are a highly mobile species in winter, this mobility is often linked to unfrozen inland waters. Winters from 1996, however, have been milder than normal, and there is no obvious reason for this recovery in numbers. Most Eurasian Teal counted at high tide are found at Lucky Scalp. Mallard Anas platyrhynchos Mallard were very abundant in the past but numbers had fallen to a winter mean of 850 during 1969-74 (Prater 1981). Counts of wintering birds reached this number only in 5 winters between 1985 and 2004 (the last being in 1994-95), and there has been a general decline SB 26 since 1996. The most recent 5 year winter mean has been 350 and individual monthly counts now rarely exceed 400 at any time of the year. Tufted Duck Aythya fuligula Although Tufted Ducks mainly frequent the inner Tay estuary, counts have been intermittent. However, mean midwinter counts in excess of 200 prior to 1995-96 fell to a mean of only 22 during 2000-2004. Common Eider Somateria mollissima The Tay Common Eider flock is of international importance but, unfortunately, complete counts are extremely difficult to make. Tides and weather make accurate counts only possible from the south shore under ideal conditions of sunshine, calm sea and within the 2 hours before high tide. Nevertheless, intermittent counts from 1969 show that 5 year means exceeding 10,000 in early years have declined to a mean of 6737 in the period 1999-2004. However, this is probably an underestimate, as there is movement between the estuary and St Andrews Bay; a count in February 2006 within the estuary resulted in 10,000 birds. Dean et al (2003) found the largest numbers off Abertay sands and Buddon Ness. Common Goldeneye Bucephala clangula Common Goldeneye are normally scattered throughout the Tay estuary, although only one large group off Invergowrie is counted by WeBS and has declined during the 2 decades. The peak occurs between January and March and exceeded 200 regularly until 1997. Since then, the mean peak has been around 100. Two other concentrations on the Tay are found outwith the WeBS count area. One is at Newburgh, where peaks of up to 600 can occur in early winter. The decline in this group from December may be the result of some movement upriver to Perth harbour, where numbers often exceed 200 in December and January. Scottish Birds (2006) Goosander Mergus merganser The moulting flock in late summer remains nationally important with the peak occurring between July and September. Most are found on and around Lucky Scalp and exceed 200 in most summers. A few arrive in late May, building to an August peak and mostly leaving by the end of September. All are red heads, with most drakes moving to Norway to moult (Collier et al 2005). This moulting flock is a fairly recent phenomenon, not being recorded prior to the mid 1980s. Other wildfowl species Other wildfow! not monitored regularly by this study, mainly because they are not present within the count areas at high tide, include geese, mainly Pink-footed Goose Anser brachyrhynchus and Greylag Goose Anser anser, and sea duck, such as Long-tailed Duck Clangula hyemalis, scoter Melanitta spp and Red-breasted Merganser Mergus serrator. The geese normally roost in the inner estuary and the most recent 5 year means, as counted by the WWT/JNCC National Grey Goose census, are 6014 Pinkfeet and 1224 Greylags (Collier et al 2005). The sea duck enter the outer estuary from St Andrews Bay and have been subject to aerial counts by JNCC (Dean et al 2003, 2004). Other duck species are rare, a far cry from the flocks of Northern Pintail Anas acuta, Common Pochard Aythya ferina and Greater Scaup Aythya marila present in the 1920s (Berry 1930). Discussion The geography and geomorphology of the Firth of Tay in relation to its wetland bird populations was described by Elkins & Lynch (1997). Since that time, little has apparently changed, except that the extensive accreting sand dune systems at the entrance to the firth have changed the shape of the shoreline, particularly at Tentsmuir Point on the south shore. This may be the cause of the Waterfowl counts on the Tay Estuary 37 shift of roosting birds from the main high tide wader roost at Tentsmuir Point to Monifieth and Barry Buddon on the opposite shore. Declines in birds using the inner estuary roosts at Kingoodie and Invergowrie Bays are thought to be due to disturbance, particularly on the increasingly busy Dundee airport nearby. More short term fluctuations in the use of the Tay as a feeding and roosting site are occasionally due to the severity of winter weather. The 2 winters of greatest wind chill during the past decade were 1995-96 and 1999-2000, coincident with high mean wind speeds and low mean temperatures. Unlike the previous decade, when no significant local effects on wintering populations were apparent, there was a marked decrease in the number of waders on the estuary during these 2 winters. This was particularly noticeable in 1995- 96, when extensive ice covered the mudflats in the inner estuary at times. Conversely, the highest numbers of waders roosting on the estuary was in the winter of 1997-98, coincident with the lowest wind chill (see Fig 1). It is known that wader populations are higher on west coast sites in Britain during cold winters, and that recent milder winters have seen an eastwards shift in wintering populations (Austin & Rehfisch 2005). Thus a significant part of the Tay wader population may have moved to western Scotland in the coldest winters. Climate change may also be responsible for other variations in wintering populations in eastern Scotland, particularly affecting the ability of high latitude waterfowl to winter further north and east during warmer winters. Total numbers of waterfowl using the Tay estuary have probably declined for several reasons. Some have mirrored national changes, while roost disturbance and changes in roost characteristics all play a part. Considerable movements of some species occurs between the Tay and adjacent waters and shores, where some remain uncounted (cf Sanderling q v). 38 Acknowledgements A great debt is due to those whose commitment overcomes all weathers to count the birds. Apart from the author, the majority of the counts between 1995-96 and 2004-05 were carried out by Dave Ferguson, Bruce Lynch, Dan Carmichael and Donald Stewart. I would also like to thank the following for acting as reserve counters and for taking part in low tide counts: A. Davidson, A. Lees, D. Arthur, G. Adam, D. Bell, S. Cairns, P. Blackburn, G. Johnston, D. Shepherd and G. Wardrope. My apologies to anyone who has been inadvertently omitted. I am grateful to Les Hatton for allowing me access to Eden Estuary data and to the WeBS partnership (BTO, RSPB, WWT and JNCC) for the opportunity to continue these valuable counts. References Atkinson-Willes G L 1963. Wildfowl in Great Britain. HMSO, London. Austin G & Rehfisch M 2005. Shifting non- breeding distributions of migratory fauna in relation to climatic change. Global Change Biol. 11: 31-38. Berry J 1930. Notes on the movement of duck off Tentsmuir Point, Fife. Scottish Naturalist 50: 43-46. Collier M P, Banks AN, Austin G E, Girling T, Hearn R D & Musgrove A J 2005. The Wetland Bird Survey 2003/04: Wildfowl and Wader Counts. BTO/WWT/RSPB/JNCC. Thetford. Dean B J, Webb A, McSorley C A and Reid J B 2003. Aerial surveys of UK inshore areas for wintering seaduck, divers and grebes: 2000/01 and 2001/02. JNCC report, No. 333. Peterborough. Dean B J, Webb A, McSorley C A, Schofield R A & Reid J B 2004. Surveillance of wintering seaducks, divers and grebes in inshore areas: Aerial surveys and_ shore-based_ counts 2003/04. JNCC report, No. 357. Peterborough. SB 26 Elkins N & Lynch B M 1997. Waterfowl counts on the Tay Estuary, 1985-1995. Scottish Birds. 18: 36-54. Fife Environmental Recorder Network (FERN) 2005. Natural Heritage Designations in Fife: A Guide. Fife Council, Glenrothes. Maclean I M D, Austin G E, Mellan H J and Girling T 2005. WeBS Alerts 2003/2004: Changes in numbers of wintering waterbirds in the United Kingdom, its Constituent Countries, Special Protection Areas (SPAs) and Sites of Special Scientific Interest (SSSIs). BTO Research Report No 416, Thetford. Musgrove A J, Langston R H W, Baker H & Ward R M (eds) 2003. Estuarine Waterbirds at Low Tide: the WeBS Low Tide Counts 1992/93 to 1998/99. WSG/BTO/WWT/RSPB/JNCC. Thetford. Prater A J 1981. Estuary Birds of Britain and Ireland. Poyser, Calton. Norman Elkins 18 Scotstarvit View, Cupar, Fife KY15 5DX Revised manuscript accepted March 2006 Scottish Birds (2006) Short Notes 39 SHORT NOTES Secondary nest usage by a pair of Lesser Whitethroats In Ayrshire, Renfrewshire and Lanarkshire, 9-12 Lesser Whitethroat Sylvia curruca territories have been studied regularly since 1983 (T Byars et al 1991, Scottish Birds 16: 66-76). A Renfrewshire study site contained 2-3 territorial males within a 2 km square south east of Paisley at Brownside Braes (NS 4860). All territories were located in areas of mature Hawthorn Crataegus monogyna scrub with a dense mosaic understorey of Bramble Rubus sp, Gorse Ulex europaeus, Dog Rose Rosa canina and Goat Willow Salix caprea (T Byars et al 1991, Scottish Birds 16: 66-76). Breeding Lesser Whitethroats were individually colour ringed, allowing territorial behaviour, site fidelity and interchange to be studied. Sexing Lesser Whitethroats can be difficult as males develop well vascularised brood patches (S C Norman 1992, Ringing and Migration 13: 167-174; L Svensson 1992, Identification Guide to European Passerines, Stockholm). Research in NE England found that trapped females could be sexed if birds showed distended stomachs and higher body mass through egg carrying, adults can also be sexed on the basis of brood patch size and/or cloacal protuberance (M Boddy 1994, Ringing and Migration 15: 65-78). Trapped Lesser Whitethroats at the study site were sexed using a combination of these features. At Brownside Braes on 8 May 1996, a male Lesser Whitethroat was vociferously holding territory (Territory A). It was mist netted and given an orange colour ring that morning and observed singing throughout the day. The following morning, I discovered a new territory within the study site (Territory B) which was located 800 metres away from Territory A. Although the male was singing infrequently, this second territory did contain a breeding pair (Pair B) and efforts were made to mist net them. Despite various attempts, only the female from pair B was caught and given a green colour ring. Pair B were observed nest building in a gorse bush that afternoon. The female lined and shaped the nest cup, an indication that pair bond establishment had taken place (S Cramp, {ed}, 1992, Birds of the Western Palaearctic, Vol 6: 449-450. Oxford University Press, Oxford). Six days later, a new unringed male had taken over Territory A and the orange ringed male had disappeared. On 16 May, the unringed male on Territory B was singing again, suggesting that mate loss had occurred (S Cramp, {ed}, 1992, Birds of the Western Palaearctic, Vol 6: 450. Oxford University Press, Oxford). The green ringed female could not be located at Territory B and was never seen again anywhere on the study site. Mist netting at Territory B that morning revealed the orange ringed male, approxi- mately 30 metres away from the now vacant nest site. The following day, the unringed male was still singing on Territory A and the unringed male from Territory B had now moved into a new territory 86 metres away, leaving Territory B vacant. Returning to Territory B on 7 June to obtain data on the unused nest site, I heard 2 Lesser Whitethroats alarm calling close by. On inspecting the nest in Territory B, I was surprised to find it contained a clutch of 5 Lesser Whitethroat eggs. Both adults became visibly agitated at my _ presence and immediately started distraction displays which allowed for close observations of this pair. The male was instantly identified by his orange colour ring, while the female was unringed. This new pair had utilised the old nest and successfully raised 4 young, which left the nest on the 17 June. Although rare, this behaviour has been noted in other birds, eg the Wren 40) Troglodytes troglodytes (J J Sweeney pers comm). This appears to be the first documented account of a pair of Lesser Whitethroats taking on an empty nest. Great Skuas feeding on bread Using scraps of bread to entice ringed birds close enough to a car or hide so that their rings can be read is a well proven method for generating ringing returns, and one which is especially effective with gulls Laridae. On 24 June 2004 whilst trying this technique at Scrabster Harbour, Caithness, we were surprised to attract not only the usual handful of Herring Gulls Larus argentatus but also 2 Great Skuas Catharacta skua, though none of either species was ringed. The latter were quite aggressive and managed to corner most of the bread we provided, but took care not to approach our vehicle closer than about a metre, and for the most part were content to sit on the concrete quay about a metre and a half Some early records of Northern Gannet movements in the coastal waters off Caithness Movements of Northern Gannets Morus bassanus are regularly observed in coastal waters around much of Britain and Ireland. Such movements can occur at almost any time between March and October, when birds are settled at their breeding colonies and are most numerous in the seas around Britain and Ireland, and may continue for periods extending from hours to several days. Often they involve consid- erable numbers of birds and are generally orientated in one particular direction. The extent of these movements, now known to _ be associated with movements between breeding SB 26 Tom Byars 81 Sharphill Rd, Saltcoats, KA21 5QU (e-mail:tombyars @btinternet.com) Revised manuscript accepted March 2005 away. We can find no previous record of Great Skuas taking bread but some have developed the habit of scavenging on Sullom refuse tips with gulls and corvids (see Furness, Monaghan & Sheddon 1981. Exploitation of a new food resource by the Great Skua. Bird Study 28:49- 52). 2004 was a very difficult season for seabirds in the north of Scotland, and it seems likely that these 2 birds were driven to taking bread due to extreme hunger. Mark Oksien, 32 Struan Drive, Inverkeithing, Fife KY11 LAR Robin M Sellers, Crag House, Ellerslie Park, Gosforth, Cumbria CA20 IBL Revised manuscript accepted May 2005 colonies and feeding areas, first became apparent in the 1880’s in connection with the enquiry into the movements of birds at coastal lighthouses held under the auspices of the British Association for the Advancement of Science (1880-89) and published in their ‘Migration Reports’. Such movements have been described in many parts of Britain and Ireland, the culmination being a detailed study carried out in NE Scotland by Bourne (1982. Ibis, 124:81-88; which also contains references to the extensive earlier literature on these movements) who emphasised the important role of wind drift in the phenomenon. It is not generally appreciated that the occurrence of these coastal movements had been reported well before the British Association enquiry in the 1880’s. Our purpose here is to draw attention to Scottish Birds (2005) 25: 41-49 3 old records relating to movements of Northern Gannets in the waters off the east coast of Caithness, one from the 18th century, and 2 from the 19th century. The first published reference to the fact that Northern Gannets were passing through coastal waters off Caithness is contained in Thomas Pennant’s A Jour in Scotland 1769 (1771, J Monk, Chester). He travelled through Caithness en route to the John O’Groats area in the north of the county in the late summer of 1769, and as he returned south he: ‘saw multitudes of gannets, or soland geese, on their passage northwards they went in small flocks from 5 to 15 in each, and continued passing for hours; it was a stormy day: they kept low and near the shore; but never passed over the land, even when a bay with promontories intervened, but followed (preserving an equal distance from shore) the form of the bay, and then regularly doubled the capes. I saw many parties make a sort of halt for the sake of fishing; they soared to a great height, then darting down headlong into the sea made the water foam and spring up with the violence of their descent; after which they pursued their route’. Unfortunately he does not say precisely where he saw these birds but from the context it seems to have been somewhere north of Wick, probably between Freswick and Keiss. A few days later he continued his journey south from Wick and between Thrumster and Dunbeath he: ‘again saw numbers of flocks of gannets keeping due north, and the weather being very calm they flew high. It has not been observed that they ever return this way in the spring, but seem to make a circuit of the island, till they again arrive at the Bass, their only breeding-place on the east coast’. He reiterates much of this in his British Zoology (1812. British Zoology Vol IT, Class II Birds, Div IT Water. Wilkie & Robinson, London p292). Short Notes 4] Some similar observations were reported almost a century later by R I Shearer, who for much of his life lived near Sarclet Head, to the south of Wick, and where he had many opportunities to observe Northern Gannets. Between 1859 and 1867 he published a long series of articles on the birds and mammals of Caithness in the local newspaper, the John O’Groat Journal (details in Clark & Sellers 2005. Birds and Mammals of Caithness - Robert Innes Shearer’s Contributions to the Natural History of Caithness, 1859-1867, Bellfield Publications, Wick), and he mentions the movements of Northern Gannets in 3 of these. The first, published in the edition of 21 July 1859, refers to such movements twice, and begins as follows: “This bird, though frequently seen at sea, and passing northwards along our coast, is not known to breed in Caithness’. A little later in the article he returns to this point in the following words: ‘There is one thing I have noticed rather strange in regard to those birds, viz, that they are occasionally seen through the summer, when I would expect them to be at their breeding stations, flying northwards, but are never seen to return. Whether they be birds migrating, having been unable to procure stations, or from other causes not inclined to stay till the breeding season is over, or belong to fixed breeding stations, I am unable to say. If, however, they belong to a station south of this they must return far out to sea, or perform a journey round our island’. In a second article published in the issue dated 20 October 1859 and primarily concerned with the Little Auk Alle alle, he adds: ‘before closing this notice, allow me to mention that solans are still (22nd September) flying northwards in great numbers. The flocks vary in number from two to twenty or thirty. They are very shy and watchful. In the morning of this day the wind blew from the south, and these birds, following a bight or bay opposite Ulbster House, had 42 either to fly out round Ulbster Head, or to cross over the promontory. They very seldom chose the latter. ... This day the flight of some of these birds was very high, and they were evidently intent on making the most of a favourable wind in their migration.’ The third article appeared on 5 January 1860 and was primarily concerned with the air sacs of the Northern Gannet. He had evidently received some fairly strident criticism for his statement that the Northern Gannets he saw always flew northwards, and addresses the final 2 paragraphs of this piece to his critics. The key section of the article reads as follows: ‘In some former letters, when writing of the solan, I spoke of its always being seen on the north east coast of Caithness flying northwards in flocks. ... The nearest breeding station south of this is Bass Rock, and the nearest north or north west of this 1s said to be the Stack of Sule Skerry, an outlying rock, part of the Orkneys; and there are also some stations on the west coast of Scotland. Now, those birds that fly so regularly north on our coast must perform a long migration, whichever of these stations they come from. It is also very strange that this bird is found in Orkney in great abundance, especially in winter, while there are few or none of them found ai the same season in Britain to the south of this, so far as I know.’ The existence of Shearer’s articles in the John O’Groat Journal has only recently come to light. They were, however, known to Henry Osborne, a friend of Shearer’s and another very capable amateur ornithologist and who also makes reference to the movements of Northern Gannets in the waters off the east coast of Caithness. Osborne prepared a document describing the birds of Caithness, but died before it could be published. The manuscript appears now to be lost, but when Harvie-Brown & Buckley (1887) came to write their Vertebrate Fauna of Sutherland, Caithness and West Cromarty in the SB 26 1880’s they relied to a large extent on Osborne’s Manuscript Notes for information about Caithness, and quote them extensively. The section on the Northern Gannet reads: ‘A common summer visitor to the Caithness coast, but does not breed, appearing merely to follow the shoals of herrings that abound at that season. “A curious feature in their local history is that they are observed invariably to fly in a northward direction; that, while they follow the east coast on their journey northward, their return must be effected by some other route, as the bird is seldom or never seen on the east coast of Caithness to pursue a southward course” (H. Osborne in MSS).’ Recent observations in Caithness confirm what Pennant, Shearer and Osborne reported. On the east coast of the county, northward movements of birds are to be seen regularly March-October, often involving 200-1000 birds per hour, the most intensive movements usually coinciding with periods of strong on shore winds (Caithness Bird Report 1976-1999; Manson 2002. A History of Caithness Birds 1769 to 2001). Very occasionally birds can be seen moving south, but by far the majority pass north. Neither Pennant, Shearer nor Osborne make any comment about movements in the Pentland Firth, but we now know that movements here occur in both directions with approximately equal frequency (based on our analysis of records in the Caithness Bird Report 1976-99 and personal observations). In the 18th and 19th centuries the only gannetries north of Caithness were on Sule Stack, some 60 km west of Orkney, and Sula Sgeir, about the same distance north of Lewis, and the most reasonable interpretation is that the movements seen by Pennant, Shearer and Osborne were birds returning to one or other, or both, of these colonies after fishing excursions into the North Sea. The outward leg of the journey from Sule Stack and/or Sula Sgeir Scottish Birds (2006) appears to pass through the Pentland Firth, but we suspect that some birds enter the North Sea to the north of Orkney on occasion. Pennant, Shearer and Osborne accurately described most of the essential features of these movements a century and a half ago, and both Shearer and Osborne seem to have understood that for the most part they were nothing to do with migration but were much more likely to be connected with feeding movements. What neither Shearer nor Osborne seems to have understood is that they were witnessing the return journey, not the outward one, and then only after the birds had drifted or been blown close to the shore by winds from between north and east. Shearer did, however, realise that the other leg was most probably made well out to sea and so went unseen by land based observers. Despite an extensive search through the 19th century ornithological literature we can find no other mention of coastal movements by Northern Gannets published prior to 1880, and Predation of European Storm petrels by Great Black-backed Gulls on the island of Auskerry, Orkney Auskerry is approximately 1.5 km by | km and situated on the eastern side of the Orkney archipelago, 5 km south of the island of Stronsay. It has a rocky coastline with some boulder and shingle beaches and, on the west side, cliffs which reach a height of 18 metres. The vegetation is mainly Calluna heath and acidic grassland. There is also some marshy ground and standing water and, in the south of the island, a number of peat banks, the majority of which are no longer worked. The island has long been used for grazing sheep Short Notes 43 conclude that the observations reported by Pennant, Shearer and Osborne are not just the first records of this phenomenon for Caithness but probably for Britain as a whole. We are indebted to the staff at the Carnegie Library, Wick for their help in locating and copying the material published by Shearer in the John O’Groat Journal, and to staff at the Natural History Museum Library, London and Gosforth Public Library, Cumbria for their assistance in locating the other material consulted. Hugh Clark, 3 Lindsay Place, Wick, Caithness KW1 4PF (e-mail:hugh @lindsayplace fsnet.co.uk) Robin M.Sellers, Crag House, Ellerslie Park, Gosforth, Cumbria CA20 IBL (e-mail: sellers@craghouse7 freeserve.co.uk) Revised manuscript accepted July 2005 At least 10 species of seabird breed regularly on Auskerry including Great Black-backed Gull Larus marinus and European Storm petrel Hydrobates pelagicus. The number of Great Black-backed Gulls increased from 20 pairs in 1969 to 180 pairs in 1976, then declined. It has remained fairly stable since 1983 (Adam RG & Booth C J 1999. The breeding birds of Auskerry, Orkney 1969-1998. Scottish Birds 20:1-5) at about 60 pairs. The majority nest on rocks on the east side of the island with 2 to 3 pairs on the west side and a similar number in the centre of the island. In 1954 Myxomatosis reached Orkney (Booth C & Booth J 1994. The Mammals of Orkney, Kirkwall) and, by the early 1960s, the Rabbit Oryctalagus cuniculus population on Auskerry 44 Tabie 1 No of pellets found annually 1997 - 2004. 1997 1998 1999 ay 17 49 59 had been wiped out by the disease. Their burrows, however, remain and provide the main nesting sites for storm petrels. These petrels also nest in natural holes in the peat, under stone slabs, in the ruins of ancient stone buildings and amongst stones on the boulder beaches. Surveys have indicated that the number of petrels breeding on the island has declined. In 1995 there were estimated to be 3,613 apparently occupied sites (Wood D 1997. An estimate of the number of Storm petrels Hydrobates pelagicus breeding on Auskerry, Orkney, Seabird 19: 40-46) but in 2001 only 994 apparently occupied sites were found (Mitchell P I & Newton S F 2004. European Storm petrel in : Mitchell P I, Newton S F, Ratcliffe T & Dunn T E (Eds) Seabird Populations of Britain and Ireland, 81-99. T and A D Poyser. London). The greatest decline has occurred in the number of petrels nesting in the old burrows of rabbits. It 1s thought that overstocking with sheep has led to collapse of burrows and subsequent loss of nesting sites. However, the number of petrels nesting in the ancient ruins and boulder beaches was also found to have decreased. Since 1970 the authors have made a number of visits to the island to monitor some of the breeding sea birds; these visits have been annual since 1992. In 1984 a ring was noticed in a Great Black-backed Gull pellet (Adam R G & Booth C J 1999. Storm petrel Hydrobates pelagicus rings in Great Black-backed Gull Larus marinus pellets. Ringing & Migration 19, 298) and, on closer examination, it was found that the pellet contained the remains of a whole petrel. The ring was from a petrel that had been ringed on Auskerry in a previous 2000 SB 26 2001 2002 2003 2004 88 10 36 53 year. This discovery of a ring prompted a more intensive search for pellets with petrel remains, concentrating especially around gull nesting sites. Since 1997 the number of pellets found has been recorded, although effort has varied from year to year depending on time available. In the period 1997 - 2004 the extreme dates of visits have been 6 July and 17 July, and the number of pellets containing petrel remains found annually is shown in Table 1. Pellets were found mainly at 3 sites : an area of peat banks close to 2 gull nests in the centre of the island, around a nest on the edge of rocks at the east side and by a nest on the north east side. It appeared that the gulls from these nests may have specialised in catching petrels but this may also be due to the fact that they were probably very close to the main petrel nesting area. No pellets with petrel remains were found at gull nests on the outlying rocks or on the area used by non breeding birds as a daytime roost. The pellets were all fairly fresh and would have been produced within a few weeks prior to our visits. It can be seen from Table | that a total of 369 pellets was found. We could not be certain, because of variation in size, that every pellet would have contained the remains of one petrel. Even so, given the number of pellets, it was thought that at least 300 petrels would have been killed by the gulls in the 8 years between 1997 and 2004. This would be a minimum figure as it does not take into account pellets that were missed and petrels killed during the remainder of the breeding season. Scottish Birds (2006) An examination of the pellets revealed 36 storm petrel rings. Five were from birds ringed on Auskerry, 27 were from birds ringed at various localities at distances ranging from 21kms to 77 kms from the island. Two were from birds ringed in Norway and there were single birds from northern England and Portugal. All the petrels had been ringed as full grown birds hatched before the year of ringing. As the pellets were found before mid July and in the early part of the petrel breeding season, it is probable that some of the petrels killed would have been potential breeding birds. The killing over the years, by the gulls, of these birds could be an additional reason for the decline in the European Storm petrel breeding population of Auskerry. Predation by Great Skuas Stercorarius skua is considered to pose a serious threat to the Storm petrel colonies on St.Kilda (Mitchell P I & Newton S F 2004. European Storm petrel, in: Mitchell P I, Newton S F, Ratcliffe T & Dunn T E (Eds) Seabird Populations of Britain and Ireland, 81-99. T and A D Poyser. London). Peregrine Falcon catching and killing a bat in daylight At 1220 hrs on Sunday 20 March 2005, we saw a large bat hawking insects c30-50 m above the Water of Ken near the point where it enters Earlstoun Loch in the Stewartry' of Kirkcudbrightshire. It was an overcast and humid day, though the misty cloud was thin in places and it was fairly bright at times. From a distance of cl100m we estimated the bat to have a total wingspan of no less than 30 cm. It occasionally showed light brown on the body as it repeatedly turned and dipped to feed; the head appeared owl like. The bat was almost certainly a Noctule Nyctalus noctula, a species Short Notes 45 References Adam, R. G. and Booth, C. J. 1999. The Breeding birds of Auskerry, Orkney 1969- 1998. Scottish Birds (1999) 20:1-5. Adam, R.G. and Booth,C.J. 1999 Storm petrel Hydrobates pelagicus rings in Great Black- backed Gull Larus marinus pellets. Ringing & Migration 19, 298. Booth Chris and Jean 1994. The Mammals of Orkney. Kirkwall. Mitchell, P. I. and Newton, S. F. 2004 European Storm petrel. In: Mitchell, P.I., Newton, S.F., Ratcliffe, N. & Dunn, T.E. (Eds) Seabird Populations of Britain and Ireland. 2004 81- 99.T. and A.D. Poyser. London. Wood, D. 1997 An estimate of the number of Storm petrels Hydrobates pelagicus breeding on Auskerry, Orkney. Seabird 19, 40-46. C J Booth, Ronas, 34 High Street, Kirkwall, Orkney, KW15 [AZ R G Adam, Northolme, St. Ola, Orkney, KW15 ISB Revised manuscript accepted August 2005 known to be present in Galloway that behaves in this way (Stewart Pritchard and Peter Norman pers comm). An immature female Peregrine Falcon Falco peregrinus attacked the bat, which detected the falcon and evaded capture by losing height. The Peregrine turned and repeated its attack on the bat, which again evaded capture successfully. Two Carrion Crows then briefly harried the Peregrine, but it was undeterred and turned to capture the bat in level flight at the third attempt. In all, this action lasted some 10-15 seconds. Having secured the struggling prey between its talons, the falcon flew off, briefly slowing twice in mid air to bite and presumably 46 kill the bat. We lost sight of the Peregrine in the haze about 1-2 km away. We have watched Peregrines attack a variety of birds over the past thirty years. Usually, when the victim evades capture, it goes to ground or into cover, with the Peregrine characteristically ~ waiting on’ in an agitated and often highly vocal state. Only when the prey is flying high is there normally any chance of multiple attacks in mid air. However, on this occasion the Peregrine pressed home 3 attacks so swiftly that it seemed to know that the bat was unlikely to go to ground, suggesting the falcon had encountered bats before. The bat seemed aware of the danger posed by the Peregrine. In the UK, Common Kestrels Falco tinnunculus have been known to take bats at dusk (Gordon Riddle pers comm; Simms, 1977 Kestrels hunting long eared bats. British Birds 70:499- 500) and Noctule has been recorded amongst prey items at a Eurasian Hobby’s Falco subbuteo nest in Hampshire (Insley and Holland, 1975. Hobbies feeding on bats and notes on other prey. British Birds 68: 242.). The taking of bats by Peregrines in South America is well known (Albuquerque, 1978. Contribuicao ao conhecimento de Falco peregrinus Tunstall, 1771 na America do Sul. Revista Brasileira de Biologica 38:727-737; Pearson & Donohue, 1983. Peregrine Falcon feeding on bats in Suriname, South America. Am Birds 37257-259; Sick, 1961. Peregrine Falcon hunting bats while wintering in Brazil. Auk 78:646-648), and this behaviour is also reported in the USA (Sprunt, 1951. Aerial feeding of Duck Hawks, Falco peregrinus anatum. Auk 68: 372-373; Palmer, 1988. Habits [Peregrine Falcon]. in Handbook of North American birds Vol. 5:363-378). Byre (1990. A group of young peregrine falcons prey on migrating bats. Wilson Bulletin 102 [4]: 728- 730) described how a group of 5 recently released juvenile Peregrines repeatedly watched SB 26 for and attacked migrating bats flying across lake Michigan on 3 autumn mornings. These Peregrines hunted together and a total of 28 kills were seen over the 3 days, with evidence found of at least another 15. Ratcliffe (1980. The Peregrine Falcon. Poyser, Calton) cited the taking of bats in Texas by Sprunt (no reference), but noted that they had not been recorded as prey in Britain or Ireland. However, John and Geoffrey Kaczanow found Noctule bat remains in pluckings and prey debris at 2 separate Peregrine sites inland in west Devon in 1990 and 1994 respectively. They also found hair of a Pipistelle bat Pipistrellus sp in a Peregrine pellet from an eyrie at a third inland site in the same area in 2000 (John Kaczanow pers comm). However, the present report appears to be the first record of a Peregrine being seen to prey on a bat in the UK. Acknowledgements We are very grateful to Stewart Pritchard and Peter Norman for information on bats in southern Scotland, and to Gordon Riddle and Hugh Insley for information on other falcons taking bats in the UK. One evening on a visit to Galloway in 2005, just 3 weeks before he died, Derek Ratcliffe commented on this record and recalled that the Kaczanow brothers had recently recorded bat parts in Peregrine prey remains in Devon. Next morning, with typical thoroughness, he was able to produce the contact address for these fine naturalists, without whose information this note would have been so incomplete. We are grateful to John and Geoffrey Kaczanow for their information, and to the great man himself for his inspiration and example. Chris Rollie & George Christie, RSPB Scotland, Old School, Crossmichael, Stewartry of Kirkcudbrightshire, DG7 3AP. Revised manuscript accepted August 2005 Scottish Birds (2006) House Sparrows nesting in sites other than buildings in Scotland In Britain the great majority of House Sparrows Passer domesticus site their nests in holes or cavities in buildings, including some in nest boxes. A small minority construct round, untidy, roofed nests of dry grass or straw with a side entrance, in a tree, hedge or climbing plant, typically a few metres above the ground (Summers-Smith 1988. The Sparrows. Poyser, Calton; Cramp and Perrins 1994. The Birds of the Western Palearctic, Vol 8. Oxford). There is a presumption that this also applies in Scotland (Thom 1986. Birds in Scotland. Poyser, Calton) though little has been published for Scotland except for a few records in Scottish Bird Reports and regional bird reports, and some early information in Baxter and Rintoul (1953. The Birds of Scotland. Oliver & Boyd, Edinburgh). To investigate the extent to which House Sparrows nest elsewhere than in buildings in Scotland, I contacted all Scottish Local Recorders in late 2003 and placed a request for information in Scottish Bird News (No 71, March 2004). These enquiries asked for any records of House Sparrows nesting in trees, hedges, climbers, other vegetation, or in other unusual sites, with if possible information on place, year(s), how many pairs were involved, whether young were successfully reared, and whether any nesting was associated with Tree Sparrows Passer montanus or other birds. The results from this enquiry together with published Scottish records are summarised in Table 1. Most reports, at over 20 locations, were of nests in vegetation. Nesting in nests of other species and usurping other species’ nests were reported from over 12 locations, and nesting in holes in rocks or trees from 9 locations. Of nests in vegetation most were in trees, nearly as many were in hedges, fewer in climbing plants, and only one each in a Short Notes 47 bush or low plant. Evergreen trees were strongly favoured over deciduous trees. More nests were reported in deciduous than evergreen hedges and more in evergreen than deciduous climbers, though this could have related as much to availability as preference. When nesting in vegetation a single nest was usual, 2 less so, and more than 2 uncommon, though in thick climbers and hedges reporters were not always sure how many nests were involved. When nest character was mentioned, all those in vegetation were roofed and with a side entrance. The communal nest, reported by G D Joy in Ayrshire, was about 20cm wide and 75cm tall, domed, with separate side entrances for each of the 3 pairs of House Sparrows, and situated in dead interlocking branches of 2 Sitka Spruce trees growing close together and forming part of a line of spruce trees. This appears to be a fairly unique occurrence in Britain, though communal nesting by House Sparrows has been described in Canada (McGillivray 1980. Journal of Field Ornithology 51: 371-2). None of the reports in this survey were from cities or large towns (though the majority of readership of Scottish Bird News is from cities and towns); the great majority came from villages and rural locations. A similar number of reports of nests in vegetation came from the Northern and Western Islands as came from the Scottish mainland. Most people replying to the survey believed that young were successfully reared from the nests reported. Predation was reported from one site where several pairs of Sparrows nested in thick ivy on a large tree and Carrion Crows Corvus corone were seen to pull a nest apart and take young Sparrows (R N Cinderey in litt); perhaps hinting at one important reason why nesting in vegetation is much less common than in cavities in buildings. Climate may also be an influence; Summers-Smith (1988. The Sparrows. Poyser, Calton) found House Sparrows nesting away from 48 SB 26 Table 1. House Sparrows nests in Scotland in sites other than in buildings. SBR = Scottish Bird Reports, rbr = regional bird reports, repl = reply to unpublished enquiries (see text), B&R = Baxter and Rintoul 1953, Penn = Pennington et al 2004. Type of nest site Number of Number Years Scottish Source locations of nests Recording per location Area domed nest in tree (spruce) | l 2004 Shetland repl domed nest in tree | l 1999 Outer Isles SBR domed nest in tree (spruce & willow) several l several recent Outer Isles repl domed nest in tree (pine) l a 1934 Highland B&R domed nest in tree (spruce) l 2 1997-8 Highland repl domed nest in tree (Cupressus) l l 1950s Borders repl communal domed nest | | communal nest 2001 Ayrshire repl in tree (sitka spruce) with 3 pairs birds domed nest in bush (hawthorn) I l 1996 Orkney repl domed nest in hedge | 7-8 1998 Orkney rbr domed nest in hedge (hawthorn) | I 1995-6 Orkney repl domed nest in hedge (holly) | l 1990s Moray & Nairn repl domed nest in hedge (beech-holly) | | recent years Perth & Kinross repl domed nest in hedge (beech-hawthorn) l pi 2004 Clyde repl domed nest in hedge (hawthorn) l 2-3 c1990-2004 Dumfries & Galloway _ repl domed nest in climber | abandoned | 2004 Shetland repl on house (ivy) domed nest in climber on house (ivy) | Z 2004 Highland repl domed nest in climber | | 1990s Moray & Nairn repl on house (clematis) domed nest in climber on tree (ivy) | several c1992-2000 Dumfries & Galloway repl nest in rhubarb plant | | 2000 Orkney rbr in hole in tree (ash) | | c1995 Ayrshire repl in hole in tree l | 2003 Lothian rbr in clay pit l l 1998-99 Lothian rbr in crevices in rocks at coast 3 | pre-1955, 2002 Shetland Penn in crevices in cliff at coast l small colony at least to 1966 Shetland Penn in crevices in rock quarry | up to 6 1980s-90s Shetland repl in crevices in rock quarry l many early 20th cent Clyde B&R in House Martin nests on cliffs l ? several at least to 1974 Angus & Dundee repl in House Martin nests on houses 2 ? several 2001 Clyde Islands repl in Swallow nests on several - 1980s-recent Orkney repl buildings and culverts in bases of Rook nests l few 1990s Shetland repl in bases of Rook nests - - pre-1924 Dumfriesshire B&R in bases of Buzzard nests several 1-2 per Buzzard nest 1990-recent Highland repl in bases of Grey Heron nests few 1-2 perHeronnest 1990-recent Highland repl in centre of disused Magpie nest l 2004 Highland repl Scientific names of birds in table: House Martin Delichon urbica, Barn Swallow Hirundo rustica, Rook Corvus frugilegus, Common Buzzard Buteo buteo, Grey Heron Ardea cinerea, Black-billed Magpie Pica pica. Scottish Birds (2006) Short Notes 49 buildings to be commoner in southern parts of the Sparrow’s Northern Hemisphere range. The proportion of House Sparrows nesting away from buildings in Scotland seems to be extremely small, which the low number of reports received by this enquiry supports, and the proportion may be lower than in southerly parts of UK. This survey received no reports of House Sparrows nesting in association with Tree Sparrows, though hybrids between the 2 species are known in Scotland, mainly from remote islands (regional bird reports, Scottish Bird Reports). In one Ash tree in Ayrshire, 3 holes were occupied in the same year by, from highest to lowest, a pair each of Common Starlings Sturnus vulgaris, Tree Sparrows and House Sparrows (A Hogg in litt). Whether there has been any change in the nesting preferences of House Sparrows in Scotland over time is uncertain. Baxter and Rintoul (1953. The Birds of Scotland. Oliver & Boyd, Edinburgh) ‘often found’ House Sparrows’ nests in hedges, conifers and ‘other evergreens’ and cite records of nesting in Ivy covering walls and cliffs. There is no doubt that Interaction between Merlins and a Hen Harrier over a nesting area Relationships between Merlins Falco columbarius and Hen Harriers Circus cyaneus on their breeding moors are mainly antagonistic. In the spring of 2005 I watched an extension of this behaviour when a pair of Merlins and a male Hen Harrier were involved disputing a nesting area on a heather moor in Galloway. Both Merlins and Hen Harriers breed on_ this particular moor usually 200-300m apart. On 29 April a male Hen Harrier hunted near the Merlins’ nest. The male harrier returned and Sparrows can be adaptable; the same authors mention nests low in Gorse bushes in Skye, and that in the 1880s Sparrows spread over the north part of Lewis ‘preferring to nest in holes in rocks rather than in thatched houses’. Acknowledgements I thank those who responded to enquiries and made this paper possible: R G Adam, A Bramhall, D Carmichael, P Carter, R N Cinderey, P Collin, M Cook, C Henty, A Hogg, G D Joy, D Okill, A Stevenson, R L Swann, J Williams, R Youngman and B Zonfrillo. Appendix Scientific names of plants mentioned in text and table: ash Fraxinus excelsior, beech Fagus sylvatica, clematis Clematis sp, gorse Ulex sp, hawthorn Crataegus monogyna, holly Ilex aquifolium, ivy Hedera helix, pine Pinus sp, rhubarb Rheum rhaponticum, sitka spruce Picea sitchensis, willow Salix sp. Harry E M Dott, 8 Mortonhall Park Gardens, Edinburgh, EH17 8SL Revised manuscript accepted August 2005 directly above the Merlins’ nest ‘skydanced’ 11 times; at the same time the sitting female Merlin flew from her nest and landed about 30m away on a fence post. The male Merlin arrived and the female returned to her nest but was reluctant to settle and flew back to the post. The male Merlin flew up and aggressively attacked the male harrier until he flew away. The male Merlin visited his nest followed by the female which landed just above him. The male flew from the nest and landed a metre away and the female Merlin entered her nest and remained. On | May the male harrier attacked a Carrion Crow Corvus corone directly above the Merlins’ 50 nest and he landed in heather 12-15m above; the female Merlin watched the harrier from behind some heather. On 3 May a male Hen Harrier ‘skydanced’ 91 times above the Merlins’ nesting area and then landed in heather 12-15m above the Merlins’ nest. The female Merlin rose from a fence post and aggressively attacked the male harrier. The male harrier then began building a ‘cock’s nest’ 12-15m from the Merlins’ nest. He gathered nest material 3 times watched by the female Merlin on a fence post. On 9 May the male harrier landed on the moor above the Merlins’ nest whereupon a female Hen Harrier flew from the ‘cock’s nest’ and took small prey from the male on the ground. The female harrier did not remain long and flew away. One hour later the male harrier rose from the heather less than 2-3m from the Merlins’ nest but the Merlin remained, apparently incubating. On 12 May a male Mean laying dates in the Northern Gannet Since 1961 the mean laying date (MLD) of Atlantic Gannets Morus bassanus on the Bass Rock has consistently fallen within the fourth week of April (Nelson J B 2002. The Altantic Gannet second edition. Fenix Books Ltd. Great Yarmouth). This date can be reliably calculated by working back from the age distribution of chicks in July or early August, before any have fledged but after the latest eggs have hatched. The approximate age of chicks is easily assessable, by size and plumage, within the categories 0-3 weeks, 3-6, 6-9 and 9+ (see illustrations in Nelson 2002). In 2005 I calculated that the MLD was some 3 weeks later than usual. Unlike some Bass seabirds such as Common Guillemots Uria aalge, Black-legged Kittiwakes Rissa tridactyla SB 26 harrier ‘skydanced’ 50 times above the Merlins’ nest. A female harrier rose from the heather about 250m away and followed the male back to a burnt area 30m above the Merlins’ nest. The male ‘bowed’ to her on 3 occasions but the female flew away, followed by a male, outwith the area. For a male Hen Harrier to ‘skydance’, ‘bow’ toa female and build a ‘cock’s nest’ so close to a Merlins’ nest is unusual. Whether the behaviour of the male harrier was to intimidate the Merlins or to attract a female harrier is not known. The pair of Merlins abandoned their nest sometime in May. The Hen Harriers did eventually breed successfully about 250m from the Merlins’ nest. R C Dickson, Lismore, New Luce, Newton Stewart, Dumfries & Galloway DG8 0AJ Revised manuscript accepted August 2005 and European Shags Phalacrocorax aristotelis, whose MLD can vary by several weeks (pers ° obs), Gannets are largely unaffected by weather during the normal laying period (late March to early July). Furthermore, and again unlike other seabirds, Gannet breeding success (young fledged from eggs laid) is consistently high at around 75%. Whilst I have no evidence that breeding success in 2005 was lower than usual, the late laying does seem significant and could potentially be linked to a relative paucity of sand- eels Ammodytes spp within the foraging range of Bass Gannets. These fish are an important part of the Gannet’s diet prior to egg laying. A reduction in the abundance of sand eels could result from a combination of over fishing and climate change. J Bryan Nelson, Mine House, Auchencairn DG7 IRL Revised manuscript accepted October 2005 Scottish Birds (2006) Crossbill numbers in old pinewoods on upper Deeside and Speyside in 1944-75 Currently there is renewed interest in Scotland’s crossbills (Loxia spp), which justifies fuller use of previous data. Here we give a fuller analysis of A Watson’s data, which had been published only in summary form, including a brief note on breeding in upper Deeside (1955, Scottish Naturalist 67, 121). D Nethersole-Thompson (1975, Pine Crossbills, Table 14, Poyser, Berkhamsted, for brevity called DNT below) compared his estimates of numbers in upper Speyside with AW’s unpublished estimates in upper Deeside. For this, he and AW used grades such as ‘high’ and ‘peak’. This showed years when the 2 sets of records differed, such as 1950 when few birds were seen on Speyside, yet many on Deeside. He suggested that birds may move between Speyside and Deeside. Analyses using scores for grades of abundance Here we present statistical analyses from AW’s notes, supplemented by interpretations from MM’s experience. Each grade for abundance had a score, | for Very Low, 2 for Low, 3 Fair, 4 High, and 5 Peak. Because most data could not be transformed into normal distributions, non- parametric correlations were used. Scores on DNT’s area and on AW’s area in the same year showed hardly any association (n = 17 years, r, = -0.19). This also held with proportionate change (the score in year 2 divided by the score in year 1,n = 13,7, =-—0.20). Moreover, although DNT reported cases where estimates on Speyside and Deeside differed, such as 1950, this was not general, and any _ inverse correlations were extremely weak. Analyses with a precise index of numbers on upper Deeside To get a better index of numbers in old Scots pinewood Pinus sylvestris near Braemar, AW Short Notes oy divided the total of birds seen by the number of days when he was in the woods. This index was more precise than anything attainable with the 5 grades. Nonetheless the indices were strongly correlated with scores for AW’s grades in DNT’s book (n = 20, r, = 0.945, P = 0.000). The index was crude, however, because a bird calls more frequently when flying than when not flying, and perhaps more when numbers are high than when low. Also, AW’s time in the woods varied from day to day. He was not doing a study of crossbills, and merely passed through the woods on the way to or from high ground. Fluctuations must have been big, however. For instance, he saw no birds in 5 years with considerable effort (6-41 days, mean 17), yet saw many in 1950. The indices were not associated with scores for DNT’s grades on Speyside (n = 29 years, r, = 0.08), and likewise for proportionate change on Deeside and Speyside (n = 27, r, = —0.02). Hence analyses using the more precise indices for Deeside birds confirmed that Deeside and Speyside populations did not fluctuate up or down together in the same year, or inversely (one up when the other was down). In short, the evidence goes against the ideas that Deeside and Speyside populations generally fluctuate together or inversely. Does local movement affect abundance in upper Deeside? Common Crossbills Loxia curvirostra move in summer, but Parrot Crossbills Loxia pytyop- sittacus and Scottish Crossbills Loxia scotica in autumn (M Marquiss 2002, The Migration Atlas). To study movement further, AW abstracted his notes by calendar month. Most birds were seen in February-August, and in 1948 none after mid July. The 1949-50 notes are revealing. In 1949, he saw no birds in January—August, but in September 5 in 2 days at Quoich and 25 on each of 2 days at Derry/Luibeg, suggesting that some had immigrated there in September. At D2 SB 26 Table 1. Crossbill abundance (annual total of birds seen by AW, divided by the number of days when he was in the woods) in old pinewood at Mar Forest and Ballochbuie, and scores for British irruption years (big irruption 2, small 1, none 0). Year Index of abundance No of days Irruption score 1944 0.2 19 0 1945 Dd 12 0 1946 a4 38 0 1947 6.0 38 l 1948 0.6 99 l 1949 0.0 4] 0 1950 58.3 25 0 195] 0.5 29 0 1952 1.0 32 0 1953 0.0 10 ps 1954 0.034 a2 0 1955 0.3 20 0 1956 0.4 1] 2 1957 0.4 12 0 1958 2.0 5 2 1959 0.8 5 2 1960 0.0 6 0 1961 0.8 9 0 1962 a, 10 2 1963 0.7 13 2 1964 OF 15 0 1965 0.0 14 0 1966 0.0 12 1 1967 0.7 14 0 1968 0.9 12 7) 1969 0.7 12 0 1970 2.0 16 0 197] 1.0 [7 0 1972 yi) 16 2 L973 Pe 2 0 1974 32 10 0 1975 6.0 15 0 ‘ Taken to 2 places of decimals to distinguish from 0.0 (none seen). Scottish Birds (2006) Derry/Luibeg he saw 15 on 28 October and 9 January, none on 12 other days there in October—January, 20 on a February day, none on 3 other February days, and none on 29 and 30 March there or in Glen Lui and Glen Dee. On the early morning of 31 March, in contrast, many were at Luibeg and about 120 at nearby Derry, with ‘Very large numbers at Luibeg, Carn Crom and Derry, never seen so many’ and ‘singing everywhere and some displays’. On 6 April ‘dozens’ were at Derry Lodge and ‘everywhere’ in lower Glen Derry and Luibeg, and many in old pinewood at Glen Lui. Watson (1955) stated, ‘there were hundreds in the woods’. The inference is that many came at the end of March, about the time when Scots pine cones begin to open and provide accessible seed. An influx of Scottish Crossbills occurs into pinewood from February onwards, so the many birds that suddenly appeared at the end of March may have been Scottish Crossbills. Luibeg stalker Robert Scott and AW saw many flying young from 9 to 29 June, and some newly fledged young being fed on | July. Yet ‘the majority of the birds remained in_ large wandering flocks throughout the breeding season’ (Watson 1955). AW was in Norway for the rest of July, but RS made notes that fledglings were still being fed until 7 July, and that all birds had gone by mid July. In August and September 1950 he and AW saw none. Although AW stopped noting crossbills in 1975, he has gone through the woods annually and has not seen numbers as high as in 1950. Do irruptions affect abundance on upper Deeside and Speyside? Because AW seldom saw flocks closely, the flocks may have consisted of any of the 3 species. Irruptions of continental Common Crossbills occurred in 1927, 29-30, 35. 38-39, 42, and later years shown in Table | (Newton 1970, and S Short Notes 55 Cramp & K E L Simmons (eds), 1994, BWP, vol 8, Oxford University Press). Up to 1975, the only big recorded influx of Parrot Crossbills to Britain was in 1962-63 (DNT 1975). DNT’s period included 16 years with irruptions and 26 without. The scores for abundance of Speyside crossbills in the 2 sets were very alike (means 2.87 v 3.00, both medians 3.0). AW’s period included 11 years with irruptions and 21 without, Deeside indices in the former being if anything lower (means 1.98 v 4.18, medians 0.95 and 0.70). Correlations between annual scores for irruptions and for abundance of local crossbills were very weak and insignificant (Table 2). This goes against the idea that irruptions affected the abundance of crossbills seen in old pinewoods on upper Deeside and Speyside. Variation in abundance between woods in the same year In 1950, the peak year at Derry, Luibeg and Glen Lui, RS and AW saw far fewer in most parts of the Derry and Luibeg woods (ie parts at higher altitudes), none in Glen Dee, Quoich and Ballochbuie, and at Dubh Ghleann only 3 in October. In 1973, AW found more at Ballochbuie than at Mar Lodge and Derry (5 v | per day), and in 1974 vice versa (0.5 v 3.1). Abundance in different woods was not in phase. Dates of fledging The first fledglings at Derry/Luibeg appeared on 6 June in 1950 and 8 June in 1947. On 9 June 1950, some with fully grown tails were already foraging away from any adults. In 1946, 1947 and 1950 the fledgling peak came in late June, with fewer on 1-10 July. One nest of small chicks on 19 April 1946 (Watson 1955) was in a young plantation of Scots pine and larch at Knock by Birkhall, a much earlier date than the records in old pinewood at Derry and other sites with old pinewood 54 SB 26 Table 2. Correlation coefficients (r,) in comparisons of the ‘irruption score’ as in Table 1 with a) the numbers score for local crossbills, and b) the proportionate change from year t-1 to year t in the numbers score. Correlation Yearly comparisons with the irruption score n years ry l DNT score for upper Spey crossbill numbers“ 42 0.003 ps Proportionate change in 1 36 0.09 3 AW score for upper Dee crossbill numbers4 20 —0.06 4 Proportionate change in 3 18 —0.15 5 AW index of crossbill numbers on upper Dee 32 —0.05 6 Proportionate change in AW index* 31 —0.10 “ DNT’s book gave 4 grades (note below his Table 11) plus very low. In the book copy sent by DNT to AW, he corrected his 1973 grade as Fair, not Low. *After adding 0.01 to each value, so that proportionate change could be calculated where the earlier year’s value was 0. mentioned above. A few other dates are later than those in old pinewood. The latest note of a fledgling, on 24 Juiy 1945, involved one being fed in old larches at Braemar village. The record in Watson (1955) at Lion’s Face involved two young being fed in planted old larches and Norway spruces, as seen by B W Tucker and AW on 21 July 1945. A late observation that is not in Watson (1955) concerned a nest with big young in mature planted Scots pine, larch and Norway spruce at Corndavon Lodge on 15 July 1945. Another was of a fledgling being fed on 10 July 1947 in old planted larches by Invercauld Bridge (BWT & AW). Scottish or Parrot Crossbills? All adults in the last paragraph were big billed and thought to be Scottish Crossbills, and the mixed plantations of larch, Norway spruce and Scots pine would fit this (M Marquiss & R Rae, 2002, Ecological differentiation in relation to bill size amongst sympatric genetically undiffer- ntiated crossbills Loxia spp. Ibis 144: 494-508). Writing of the 1950 birds, Watson (1955) noted that any seen closely at ‘less than 12 feet’ had heavy bills unlike Common Crossbills that he had seen in Lapland. In his diary he described a pair seen feeding as close as 2.4 m, at the top of Craig Doin in Ballochbuie pinewood on 1 April 1948, where the cock had ‘a tremendous pickaxe of a bill’. Having seen Parrot Crossbills with MM and Robert Rae in recent years, he thinks that this cock was a Parrot Crossbill, and likewise cocks seen less than one metre away on a skylight window at Luibeg in April 1949, 1952, 1954 and 1955. DNT had decided from 1966 onwards that the birds in old pinewood were Parrot Crossbills, when he with AW visited the woods in upper Deeside each spring. Because of this, AW called them Parrot Crossbills in April 1969 (discussion after I Newton, 1970, Animal Populations in relation to their Food Resources, A Watson (ed), pp. 337-353, Blackwell Scientific Publications, Oxford & Edinburgh) when describing a comparison by DNT and AW of abundance on Speyside and Deeside, and when reporting that DNT ‘has done a population study of Scottish parrot crossbills’ on upper Speyside. Of old pinewood in the Cairngorms, DNT wrote (in D Nethersole-Thompson & A Watson, 1974, The Cairngorms, Collins, London) that he ‘and others consider that our resident Scottish Crossbills should be treated as pytyopsittacus’ , and (1975) called them pine or Parrot Crossbills. Scottish Birds (2006) Habitat features in the most favoured Sites seen on upper Deeside The most used sites were at lower Glen Derry and at Luibeg, and to a lesser extent near Mar Lodge and on both sides of the river Dee at Ballochbuie. Pines in these places grew at low density, with wide crowns and many cones. This typifies open woodland (H M Steven & A Carlisle 1959, The Native Pinewoods of Scotland, Oliver & Boyd, Edinburgh & London), and old conifers (C W Benkman, 1993, Conservation Biology 7: 473-479). The pines were old, the Derry/Luibeg ones with a median age of 181 years in a cut sample in 1948 (A Watson, 1983, Biological Conservation 25: 289-305). At all these sites, soils comprised freely drained alluvium with a good crumb structure, at least 20 cm thick. The pines grew in herb-rich, fine- leaved grassland, which was grazed by many Red Deer Cervus elaphus and _ Rabbits Oryctolagus cuniculus, and fertilised by their dung and urine at all seasons. The site in lower Glen Derry included a vegetable plot, and the Luibeg one a field that was partly cultivated before 1940 (Ian Grant, pers. comm., who was brought up there in the Black Grouse and Western Capercaillie on upper Speyside in 1934—42 In 1988, Desmond Nethersole-Thompson (DNT) sent his notes on Black Grouse Tetrao tetrix and Western Capercaillie Tetrao urogallus to Adam Watson, following AW’s questions. The notes are of considerable interest, because they come from years when both species were abundant, long before the large declines recorded since the 1970s. Short Notes 2) early 1900s). In the 1930s—70s, stalkers spread dung and bedding straw on both plots from cattle, horses and poultry, and uneaten hay and other leftover food given to deer in winter would have boosted soil fertility outside the plots (observations by AW after 1942, Donald Macdonald and James Beattie, pers comm for earlier years). Much of the ground in the sites at Ballochbuie and Mar Lodge lay on former arable farmland, as shown on old estate plans cited by A Watson & E Allan (1984, The Place Names of Upper Deeside, Aberdeen University Press). Moles Talpa europea and molehills of fertile topsoil on all these sites also indicated fertile soil, as did abundant earthworms, especially where stock manure and straw had been added. Hence, big cone crops per tree in the above favoured sites may be due to trees being isolated, or old, or open to sunshine, or to fertile soil caused by alluvial deposits or cultivation or additions of stock dung and straw, or to 2 or more of these together. Adam Watson & Mick Marquiss, Centre for Ecology & Hydrology, Banchory, AB31 4BW Revised manuscript accepted February 2006 In the 1930s, DNT lived at Dorback near Nethy Bridge, and at Whitewell croft on Rothiemurchus from the 1940s till 1961. AW analysed the data and used the Tables to write a summary paragraph on breeding. Habitat use In Abernethy and Rothiemurchus, DNT found that Black Grouse were essentially birds of old Scots pinewoods Pinus sylvestris and the edges of pine plantations or birchwood Betula spp, but many nested in nearby moorland. On Dorback moor he saw them only occasionally more than 56 one km from woodland, and such birds were usually broodless greyhens. Western Capercaillies seemed more plentiful in the interior of old pinewood. He saw no fighting or threat between the 2 species, but foresters and keepers in the 1930s and 1940s believed that capers were increasing in the interior pinewood, while Black Grouse were declining there. SB 26 He noticed that when adult Black Grouse were slightly disturbed, they sometimes did not rise and fly as a flock, but one by one or 2 by 2, and did not always fly in the same direction. His impression was that they did not move in flocks during years with low numbers, whereas several coveys appeared to join when numbers were high, though there were insufficient notes on this to form a sound judgement. Table 1 Features of leks of Black Grouse at Abernethy and Dorback (top 3), and Rothiemurchus (bottom 5). Type of habitat Altitude Distance from the (m) nearest woodland (m) Loch Garten bog at edge of mature 250 100 Scots Pine plantation Badenedin mature Scots Pine plantation 270 inside wood Lurg peat bog 250 1500 Whitewell* permanent grass field 350 300 from mature Scots Pine plantation Lochan Deo south burnt old Scots Pinewood, of Whitewell # short Heather and Cowberry 325 150 Allt Dhru towards Lairig Ghru4 permanent grass at disused croft oP) 50 West of felled Scots Pinewood, Loch Morlich** short Heather and Cowberry 370 <50 peat bog 325 <50 Existed since at least 1902, not always in the same part of the field. From 1960 it tended to break into several small leks, and in January 1984 was finally deserted. # Maybe overflow from Whitewell, very few birds but I once saw a single cock there mating with a hen. ‘A small lek which was in the heart of Western Capercaillie country. On both sides within 50 m were old pinewoods in which Capercaillies had leks and nested. *** Established in late 1950s, abandoned in early 1980s. Scottish Birds (2006) Table 2 Number of greyhen nests with clutches of 5-13 eggs. 1935 1936 1937 1938 1939 1940 194] 1942 Total =— Oo. 2 So oO = © oO UT NocrOCOOCOCO OF N Mean 7.3, Standard error 0.5 Short Notes D7 7 8 13 | 0 | 2 0 0 0 ] 0 ] 0 0 | 0 0 0 0 0 | 0 0 | ] 0 7 yy ] First egg dates in days after 30 April in the above years starting in 1935 were 8, 13, 16, 17, 17, 19, 21 and 22, mean 17 May. Woodland clutches were: 3 of 7 eggs, one of 8 and one of 13, and moorland clutches one of 5, 2 of 6, 4 of 7 and one of 8, difference not significant (Wilcoxon rank-sum, P > 0.1). Diet and feeding habits At the Whitewell lek (Table 1), blackcocks and greyhens fed on birch buds and on pines, and on shoots of Cotton Grass Eriophorum spp in a nearby bog. Beside the lek at Loch Garten, blackcocks from February onwards fed and displayed on Larch Larix europea treetops where they ate buds, and young shoots later in the spring. In a bog at the edge of the lek, greyhens ate shoots of Cotton Grass, and both sexes likewise in a bog near a lek at Lurg farm on Dorback. As well as their staple diet of Scots Pine needles and buds, Western Capercallies in spring were observed like Black Grouse to be taking Larch at Loch Garten, and cotton grass in bogs at Abernethy and below Whitewell. In early autumn, flocks of blackcocks with greyhens fed on berries of Cowberry Vaccinium vitis-idaea and Blaeberry V. myrtillus, and on tips of young Heather Calluna vulgaris. During late autumn they often ate loose oats in stubble fields and occasionally on oat stooks. At Whitewell and at Badenedin east of Nethy Bridge, greyhens sometimes joined poultry in spring to eat chicken food, and in 1935 2 greyhens nested in the wood beside Badenedin house. A fairly tame greyhen laid 13 eggs there after feeding frequently with the gamekeeper’s poultry on chicken food in the few weeks before and during the egg laying period, and she continued to feed with the poultry daily through her incubation period. Nesting Nests of greyhens were well sheltered in tall Heather, apart from 2 sites that were more typical of Western Capercaillies, at the foot of Scots Pines 100 m inside a wood. DNT once found 2 greyhens nesting less than 5 m apart in an old peat cutting; one nest was in a clump of tall Heather surrounded by a Cowberry carpet. Greyhens seldom nested in steep pinewoods such as at Bognacruie in Abernethy, Carn Eilrig in Rothiemurchus, and on Inshriach and Invereshie near Kincraig, in all of which the Capercaillie hens frequently nested. Capercaillies often nested in pinewoods with 58 SB 26 Table 3 Success of 13 greyhen nests with clutches of 5-13 eggs. n nests n nests robbed n eggs hatched NO; WM ONN AN | 8 13 7 Z ] 2 l Og 7 13 Out of 95 eggs, predators took 41%, with Foxes robbing 3 nests and Crows 2. Of the remaining 61 eggs in 8 nests, 92% hatched. Table 4 Brood sizes of Black Grouse at < 2 weeks old and > 2 weeks. < 2 weeks 1934 4,5,5 1935 3,4 1936 _ 1937 - 1938 2,3 1939 5 1940 4 1942 5 Mean 4.0 > 2 weeks Chicks/hen at > 2 weeks l 0.33 l 1.0 2 0.5 l 0.33 2 0.5 1,2 1.0 1.4 0.48 Chicks/hen included hens without young. Breeding success was similar in woodland and moorland. In 2 years when DNT had data for chicks < 2 weeks in both woodland and moorland in the same year, 2 woodland broods had 4 and 2 chicks, and on moorland slightly larger at 2 broods of 5 in the first of these years and one of 3 in the second year. The number of chicks per woodland or moorland hen was 2.0 or 3.3 in the first year, and 3.0 or 0.7 in the second. The proportion of hens without young when chicks were > 2 weeks old was 50% out of 8 hens in woodland, and 86% out of 14 on moorland. The proportion of broodless greyhens varied from 0% in 1936, and 67% in 1934, 1938 and 1940, to 75% in 1939, and 100% in 1942. considerable growth of Juniper Juniperus communis bushes, which provided very good cover and shelter. Greyhens laid their first eggs from 8 May to 22 May, with a mean clutch of 7.3 eggs (Table 2). Red Foxes Vulpes vulpes and Crows Corvus corone and C. cornix took 41% of the eggs, and 92% of the remainder hatched (Table 3). In the 8 nests which were not robbed, the mean number that hatched was 7. There was a mean of 4 chicks in broods less than 2 weeks old, and 1.4 at more than 2 weeks old (Table 4). When broodless hens were included, the mean number of chicks per hen at more than 2 weeks of age was 0.48. Number of birds Both species fluctuated in numbers, but showed no obvious overall decline in 193442. However, some leks of Black Grouse have been abandoned since 1960 (Table 1). Leks west of Loch Morlich were established when open Scottish Birds (2006) ground increased after tree felling at Glen More (Forestry Commission) in 1939-47. The open land was later planted and deer fenced, as was nearby moorland in east Rothiemurchus in 1970-71. When the tree canopy closed over, birds deserted both leks. The decline of the Whitewell lek coincided with a lack of short vegetation on moorland and open woodland near the lek, following declines of sheep and cattle grazing outside the farm fields, and new deer fences to exclude Red Deer Cervus elaphus. Numbers of gamekeepers and predators The number of gamekeepers remained fairly similar in 1934-42, but declined after 1950. Foxes and Crows increased during the 1939-45 Sandwich Tern egg on Common Tern raft For the last 6 years we have maintained and monitored 3 floating tern rafts on the sea at Avoch, on the Black Isle in Ross and Cromarty. An average of 120 pairs of Common Terns Sterna hirundo use the rafts and productivity is over twice that of Common Terns on mainland sites elsewhere in the Moray Firth. In 2005, a pair of Sandwich Terns Sterna sandvi- censis were seen around the rafts in May and early June. On 24th June whilst undertaking the first ringing visit, we found a single Sandwich Tern egg on the largest of the 3 rafts. It was cold, as were many of the Common Tern eggs, Short Notes 59 war, when many keepers were away in the armed services, and then declined as keepers returned and increased their killing. After 1960, the numbers of Foxes and Crows rose above wartime levels, following a decline in the number of keepers, and the many new densely planted woods made keepering more difficult. Acknowledgement Des B A Thompson made useful comments. Adam Watson, Centre for Ecology & Hydrology, Banchory AB31 4BW The late Desmond Nethersole-Thompson, Culrain, Ardgay IV24 3DW Revised and accepted March 2006 following a bout of severe weather a couple of weeks earlier. As far as we are aware only Common Terns have been recorded using floating rafts provided by man and this is the first record of Sandwich Terns attempting to breed on a raft. The nearest Sandwich Tern colony is at Morrich Mor, 35 km further north in the Dornoch Firth. Kenna Chisholm & Brian Etheridge, RSPB North Scotland, Etive House, Beechwood Park, Inverness [V2 3BW. kenna.chisholm@rspb.org.uk brian.etheridge @rspb.org.uk Revised manuscript accepted March 2006 60 Advice to contributors Authors should bear in mind that only a small proportion of the Scottish Birds readership are scientists and should aim to present their material concisely, interestingly and clearly. Unfamiliar technical terms and symbols should be avoided wherever possible and, if deemed essential, should be explained. Supporting statistics should be kept to a minimum. All papers and short notes are accepted on the understanding that they have not been offered for publication elsewhere and that they will be subject to editing. Papers will be acknowledged on receipt and are normally reviewed by at least 2 members of the editorial panel and, in most cases, also by an independent referee. They will normally be published in order of acceptance of fully revised manuscripts. The editor will be happy to advise authors on the preparation of papers. Reference should be made to the most recent issues of Scottish Birds for guidance on style of presentation, use of capitals, form of references, etc. Papers should be typed on one side of the paper only, double spaced and with wide margins and of good quality; 2 copies are required and the author should also retain one. We are also happy to accept papers on disk or by email at: mail@the-soc.org.uk, stating the type of word processing package used. If at all possible please use Microsoft Word . Contact the Admin Officer on 01875 871330 for further information. SB 26 Headings should not be underlined, nor typed entirely in capitals. Scientific names in italics should normally follow the first text reference to each species unless all can be incorporated into a table. Names of birds should follow the official Scottish List (Scottish Birds 2001 Vol 22:33—49). Only single quotation marks should be used throughout. Numbers should be written as numerals except for one and the start of sentences. Avoid hyphens except where essential eg in bird names. Dates should be written: ...on 5 August 1991...but not ...on the Sth... (if the name of the month does not follow). Please do not use headers, footers and page numbers. Please note that papers shorter than c700 words will normally be treated as short notes, where all references should be incorporated into the text, and not listed at the end, as in full papers. Tables, maps and diagrams should be designed to fit either a single column or the full page width. Tables should be self explanatory and headings should be kept as simple as possible, with footnotes used to provide extra details where necessary. Each table, graph or map should be on a separate sheet, and if on disc each table, graph, map etc should be on a separate document. Please do not insert tables, graphs and maps in the same document as the text. Maps and diagrams should be either good quality computer print out and in black and white (please do not use greyscale shading) or drawn in black ink , but suitable for reduction from their original size. Contact the Admin Officer on 01875 871330 for further details of how best to lay out tables, graphs, maps etc. The Scottish Ornithologists’ Club was formed in 1936 to encourage all aspects of ornithology in Scotland. It has local branches which meet in Aberdeen, Ayr, the Borders, Dumfries, Dundee, Edinburgh, Glasgow, Inverness, New Galloway, Orkney, St Andrews, Stirling, Stranraer and Thurso, each with its own programme of field meetings and winter lectures. The Waterston Library at the Club’s headquarters at The Scottish Birdwatching Resource Centre, Waterston House, Aberlady, East Lothian EH32 OPY is the most comprehensive ornithological library in Scotland and is available for reference 7 days a week. A selection of Scottish local bird reports is held at headquarters and may be purchased by mail order. Check out our website for more information about the SOC and other bird related organisations: www.the-soc.org.uk An annual issue of Scottish Birds is published each year, in June. Scottish Bird News is published quarterly in March, June, September and December and the Scottish Bird Report is annual. The Scottish Ornithologists’ Club publishes the annual Scottish Raptor Monitoring Report on behalf of the Scottish Raptor Monitoring Group with grant aid from Scottish Natural Heritage. It is sent to all members. Bird names used in publications follow those used in the Scottish List (Scottish Birds Vol 22(1): 33-49). Copies of these reports may be purchased by non members on application to the SOC. SOC annual membership subscription rates Adult £ 25.00 Family (2 adults and any juniors living at the same address) £ 35.00 Junior (aged 17 or under) £ 10.00 Student (aged 18 to 24 and in full-time education) £ 10.00 Concession (Pensioner) £ 16.00 Concession (Unwaged) £ 16.00 Joint Concession 2 PALAU, Life £500.00 Life Family £700.00 Overseas supplement £ 10.00 All subscriptions may be paid by Direct Debit and Gift Aided. Subscriptions paid by Direct Debit greatly assist the Club. Please ask for an application form by telephoning 01875 871330 or visiting the club’s website at www.the-soc.org.uk NSTITUTION LIBRARIES Scottish Birds - TN Volume 26 June 200 Contents Main Papers Changes in numbers and distribution of waders in the Moray Firth, 1988-2003 - Bozena Kalejta-Summers 1 The status of the Northern Gannet in Scotland in 2003-04 - § Murray, S Wanless & M P Harris 1, Waterfowl counts on the Tay Estuary - Norman Elkins 30 Short Notes Secondary nest usage by a pair of Lesser Whitethroats - Tom Byars 39 Great Skuas feeding on bread - Mark Oksien & Robin M Sellers 40 Some early records of Northern Gannet movements in the coastal waters off Caithness - Hugh Clark & Robin M Sellers 40 Predation of European Storm petrels by Great Black-backed Gulls on the island of Auskerry, Orkney - C J Booth & R G Adam 43 Peregrine Falcon catching and killing a bat in daylight - Chris Rollie & George Christie 45 House Sparrows nesting in sites other than buildings in Scotland - Harry E M Dott 47 Interaction between Merlins and a Hen Harrier over a nesting area- RC Dickson 49 Mean laying dates in the Northern Gannet - J Bryan Nelson 50 Crossbill numbers in old pinewoods on upper Deeside and Speyside in 1944-75 - Adam Watson & Mick Marquiss Sil Black Grouse and Western Capercaillie on upper Speyside in 1934-42 - Adam Watson & the late Desmond Nethersole-Thompson 55 Sandwich Tern egg on Common Tern raft - Kenna Chisholm & Brian Etheridge 59 Advice to contributors 60 Front Cover Northern Gannet, Buckie, North-East Scotland by Harry Scott Published by the Scottish Ornithologists’ Club, The Scottish Birdwatching Resource Centre, Waterston House, Aberlady, East Lothian EH32 OPY. © 2006 i } & nna” — 4 c » ( rr, gr | }} f A | i" | tes I CT A A as ae es CN Ha - a ae a Amendments to the Scottish List; Birds in a Borders valley Autumn skua migration; Black Grouse in Lochaber Goshawks in Argyll; White-tailed Eagles in Sweden Loch Leven ducks; Firth of Tay counts Scottish Birds - The Journal of the SOC Editor: Dr Stan da Prato. Assisted by: Dr I Bainbridge, Professor D Jenkins, Dr M Marquiss, Dr J B Nelson and R Swann. Business Editor: Wendy Hicks, Office Manager, SOC, The Scottish Birdwatching Resource Centre, Waterston House, Aberlady, East Lothian EH32 OPY. Scottish Birds, the official journal of the SOC, publishes original material relating to ornithology in Scotland. Papers and notes should be sent to The Editor, Scottish Birds, SOC, The Scottish Birdwatching Resource Centre, Waterston House, Aberlady, East Lothian EH32 OPY. Scottish Birds is published each year in June. Scottish Birds is issued free to members of the SOC, who also receive the quarterly newsletter Scottish Bird News, the annual Scottish Bird Report and the annual Scottish Raptor Monitoring Report. Published by: The Scottish Ornithologists’ Club, The Scottish Birdwatching Resource Centre, Waterston House, Aberlady, East Lothian EH32 OPY. Design by: Pica Design, 51 Charlton Crescent, Aboyne, Aberdeenshire AB34 5GN. Printed by: Meigle Colour Printers Ltd, Block 1, Tweedbank Industrial Estate, Galashiels TD1 3RS Scottish Birds (2007) Amendments to the Scottish List: species and subspecies ] Amendments to the Scottish List: species and subspecies THE SCOTTISH BIRDS RECORDS COMMITTEE SOC Council has delegated to SBRC responsibility for maintaining the Scottish List and SBRC has appointed a subcommittee to carry out this function. The current subcom- mittee consists of Dave Clugston, Dougie Dickson, Ron Forrester, Angus Hogg, Bob McGowan and Roger Riddington. The Scottish List was last published in full in 2001 (Clugston et al 2001), since when there have been 2 sets of amendments (Forrester 2003, 2004). This is therefore the third update. For a full explanation of the procedure for maintaining the list see Scottish Birds 22: 33-35. The taxonomic subcommittee of The British Ornithologists’ Union’s Records Committee (BOURC) has made recommendations all of which have been adopted by BOURC for the British List (Sangster et al 2004, 2005). The 7th edition of the British List was published in 2006 (Dudley et al 2006) containing a number of additional changes. These recommendations and changes automatically apply to the Scottish List: Canada Goose Branta canadensis now treated as 2 separate species: ™@ Greater Canada Goose B canadensis (polytypic, with subspecies canadensis, fulva, interior, maxima, moffitti, occidentalis and parvipes) @ Lesser Canada Goose B hutchinsii (polytypic, with subspecies hutchinsii, leucopareia, minima and taverneri) Greater Canada Goose is on Category C of the Scottish List. There are many outstanding claimed Scottish records of various subspecies of Lesser Canada Goose. A paper on Canada Goose taxonomy is in preparation which should assist with the process of determining which, if any, of these records can be accepted. Meanwhile this species does not appear on the Scottish List. Greater Scaup Aythya marila Treat as polytypic (previously monotypic), with subspecies A m marila and A m nearctica. Only nominate marila has been recorded in Scotland. Common Scoter Melanitta nigra Common Scoter was previously a polytypic species, with both subspecies occurring in Scotland. Each taxa should be treated as a separate species. Both species are now monotypic. Treat as 2 separate species: # Common Scoter M nigra (monotypic) @ Black Scoter M americana (monotypic). Add Black Scoter Melanitta americana to Category A. The status code SV is applicable. Velvet Scoter Melanitta fusca Treat as 2 separate species: @ Velvet Scoter M fusca (monotypic) @ White-winged Scoter M deglandi (polytypic, with subspecies deglandi and stejnegeri). Velvet Scoter is on Category A of the Scottish List. There are no Scottish records of White- winged Scoter. Red-throated Diver Gavia stellata Treat as monotypic. Little Shearwater Puffinus assimilis The 3 taxa breeding in the tropical and subtropical parts of the Atlantic Ocean (P / lherminieri, P a baroli, P a boydi) that were included in a_ recent study formed a monophyletic group which is not closely related to the other forms of the complex (including nominate P a assimilis). The 3 tropical and subtropical Atlantic taxa are best treated as 2 species: @ Macaronesian Shearwater P baroli (polytypic, with subspecies baroli and boydi) @ Audubon’s Shearwater P lherminieri (monotypic) Macaronesian Shearwater P baroli replaces Little Shearwater P assimilis on Category A of the Scottish List with the subspecies described as being ‘nominate baroli (presumed)’. There are no Scottish records of Audubon’s Shearwater. White-tailed Eagle Haliaeetus albicilla Previously Categories A, D. The introduced population now appears self sustaining and BOURC has changed the British List to Categories A, C. It is appropriate to make the same change for the Scottish List. Yellow-legged Gull Larus michahellis The combined evidence of morphology, vocalization and molecular phylogenetics strongly indicate that Yellow-legged Gull L michahellis should be treated as a separate species. There are 2 subspecies nominate michahellis and atlantis. Add Yellow-legged Gull to Category A. The status code SV is applicable. It is the nominate subspecies that has been recorded in Scotland. Generic limits of terns As a result of a molecular phylogeny based on mitochondrial DNA _ sequencing, which compared the relationship of tern species, the species on the Scottish List should now be listed in the following sequence: SB 27 @ Sooty Tern Onychoprion fuscata @ Bridled Tern Onychoprion anaethetus @ Little Tern Sternula albifrons # Gull-billed Tern Gelochelidon nilotica @ Caspian Tern Hydroprogne caspia @ Whiskered Tern Chlidonias hybrida @ Black Tern Chilidonias niger @ White-winged Tern Chlidonias leucopterus @ Sandwich Tern Sterna sandvicensis @ Royal Tern Sterna maxima @ Lesser Crested Tern Sterna bengalensis @ Forster’s Tern Sterna forsteri # Common Tern Sterna hirundo @ Roseate Tern Sterna dougallii @ Arctic Tern Sterna paradisaea Note that the scientific names for Sooty, Bridled, Little, Gull-billed and Caspian Terns have changed. Atlantic Puffin Fratercula arctica Treat as monotypic. There were previously 2 subspecies on the Scottish List the nominate subspecies and grabae. Snowy Owl Bubo scandiaca Change scientific name to Bubo scandiacus. Generic limits of swallows A recent molecular phylogeny improved our understanding of the relationships of the species of swallow on the British List and as a result the species on the Scottish List should be listed in the following sequence: @ Sand Martin Riparia riparia @ Tree Swallow Tachycineta bicolor @ Eurasian Crag Martin Pryonoprogne rupestris # Barn Swallow Hirundo rustica @ House Martin Delichon urbicum ® Red-rumped Swallow Cecropis daurica Note that the scientific name for Red-rumped Swallow has changed. Scottish Birds (2007) Amendments to the Scottish List: species and subspecies 3 Richard’s Pipit Anthus novaeseelandiae Anthus novaeseelandiae was treated as a polytypic species occurring in Europe, Asia, Africa and Australasia. The taxon occurring in Britain, richardi, is now treated as a monotypic species, retaining the English name Richard’s Pipit. Richard’s Pipit Anthus richardi (monotypic). Richard’s Pipit is in category A of the Scottish List. Pallas’s Leaf Warbler Phylloscopus proregulus Treat as monotypic. Generic limits of tits Molecular phylogenetic analysis of tits suggests that our previous understanding of their relationships to each other was incorrect and a new species sequence has been recommended. The species on the Scottish List should be listed in the following sequence: # Blue Tit Cyanistes caeruleus # Great Tit Parus major @ Crested Tit Lophophanes cristatus HZ Coal Tit Periparus ater B Willow Tit Poecile montana @ Marsh Tit Poecile palustris Note changes to some scientific names. As a result of records that have recently been accepted, the following species have all been added to the Scottish List: Canvasback Aythya valisineria 2000 Orkney Loch of Rummie, Sanday, f, 21-23 June, photo (I Dillon, B Ribbands, E J Williams et al) (British Birds 98: 634). The status code SV is applicable. Monotypic. Ist Scottish record. Add to Category A. Redhead Aythya americana 2003 Outer Hebrides Loch Tangasdale, Barra, first winter f, 20 September to 15 April 2004 (K Gillon, S L Rivers, C Scott et al per M S Scott) (British Birds 97: 563; Birding Scotland 7: 130- 135, Birding World 17: 59); same, Loch an Duin, Barra, female, 7-8 November 2004 (British Birds 100: 20). The status code SV is applicable. Monotypic. Ist Scottish record. Add to Category A. American Coot Fulica americana 2003 Shetland Loch of Clickimin, first winter, 30 November to 5 April 2004; presumed returning bird, Loch of Benston, Mainland, 13 November to 28 March 2005, photo (S E Duffield et al); same Loch of Benston, Mainland 24 September — 2 October 2005 (J S Lees, M J McGill et al) (British Birds 97: 575-576, 97: plates 27 and 336, 98: 647, 100: 35; Birding Scotland 7: 16-18 and plates 7-11). The status code SV is applicable. Unlikely to be other than nominate race. Ist Scottish record. Add to Category A. Lesser Sand Plover Charadrius mongolus 2004 Lothian Aberlady Bay, M C m mongolus, 8-9 July (J B Bell et al) (British Birds 98: 649). The status code SV is applicable. Attributed to the nominate subspecies. Ist Scottish record. Add to Category A. Belted Kingfisher Megaceryle alcyon (1) 2005 North-east Scotland Peterculter, first summer male, 4-8 April, photo (I D Broadbent, K Landsman et al) (British Birds 98: plates 144 & 145, 100: 60, plate 22). This bird had previously been seen in Staffordshire on | April and in East Yorkshire on 2 April. The status code SV is applicable. Monotypic. Ist Scottish record. Add to Category A. Purple Martin Progne subis 2004 Outer Hebrides Butt of Lewis, Lewis, juvenile, 5-6 September 2004 (S P Coyle, T Grant, M Witherall et al) (British Birds 97: plate 327, 98: 665-666, plates 429 & 430). BOURC have admitted Purple Martin to the British List. This is the first British record. The status code SV is applicable. Ist Scottish record. Add to Category A. Rufous-tailed Robin Luscinia sibilans 2004 Fair Isle first winter 23 October (/bis 148: 594: British Birds 100: 78). BOURC have admitted Rufous-tailed Robin to the British List. This is the first British record. The status code SV is applicable. Ist Scottish record. Add to Category A. Taiga Flycatcher Ficedula albicilla 2003 Shetland Sandgarth, Mainland, first winter 12-15 October, trapped, photo (M S Chapman, D P Hall, S J Minton er al) (British Birds 98: 682, Chapman 2003, 2004, 2005). This record closely followed the Ist British record, at Flamborough Head, East Yorkshire on 26 April 2003 and BOURC have admitted the species to Category A of the British List. The status code SV 1s applicable. Ist Scottish record. Add to Category A. Masked Shrike Lanius nubicus 2004 Fife Kilrenny, 29 October to 14 November, trapped, photo (T Glass, A W Lauder, M Oksien, K D Shaw et al) (British Birds 98: 685, plate 441; 97: plate 395). This record has been accepted by BOURC as the first British record (Ibis 148: 594). The status code SV is applicable. Ist Scottish record. Add to Category A. Chestnut-eared Bunting Emberiza fucata 2004 Fair Isle first winter, 15-20 October (D Shaw et al) (British Birds 100: 100; Birding World 17: 415-419; Birding Scotland 7: 145-151). Accepted as belonging to the nominate race. This record has been accepted by BOURC as the first British record (Press Release 5 December 2006). The status code SV is applicable. Ist Scottish record. Add to Category A. The Scottish List Subcommittee of SBRC has made the following changes to the Scottish List: SB 27 Soft-plumaged/Madeira/Cape Verde Petrel Pterodroma mollis/madeira/feae In line with names now in common usage, the English name for Prerodroma madeira is now changed from Madeira Petrel to Zino’s Petrel and that of Prerodroma feae from Cape Verde Petrel to Fea’s Petrel. There were 2 records from England in 2001, accompanied by good photographs that enabled each to be confirmed as Fea’s Petrel. As a result the British List has recently been changed to show Fea’s Petrel (BOURC 2006a, 2006b), with Soft- plumaged/Zino’s/Fea’s Petrel removed. The 1996 Scottish record was at the time accepted by BBRC as Soft-plumaged/Zino’s/Fea’s Petrel and that of the 2002 bird as Zino’s/Fea’s Petrel. BBRC has in recent years decided that Soft- plumaged Petrel is most unlikely to occur in Britain, but neither Scottish record has been specifically identified. The Scottish List is now changed to Zino’s/Fea’s Petrel. Squacco Heron Ardeola ralloides A record from the ‘Glasgow Canal’ in October 1852 (Martin 1853) was examined by SBRC in 2002 and not found to be acceptable. Although the identification of the bird was not in question, the locality where the specimen was obtained was inadequately documented and, as the reporter lived in Stockton-on-Tees, may possibly have been in northeast England. 2005 Fife Kilconquhar Loch, 22 May, photo (T Moodie et al) (British Birds 100: 27). There are now 4 Scottish records, this being the first since 1950. Removed from Category B to Category A. Red Knot Calidris canutus The nominate race of Red Knot has appeared on the Scottish List with the status code ‘PV’. Our knowledge of the migration route of this subspecies is exceedingly limited and although Scottish Birds (2007) Amendments to the Scottish List: species and subspecies ») in 1992 the nominate subspecies was described as a winter visitor and passage migrant to Britain (BOURC 1992), its occurrence in Britain and Ireland was by 2002 considered ‘exceptional’ (Migration Atlas). There is no evidence for its occurrence in Scotland, although it is likely to have occurred as a limited passage visitor or vagrant. It is removed from the Scottish List until such time as ringing evidence proves its occurrence in Scotland. C c islandica is the race found commonly in Scotland during winter. Black-tailed Godwit Limosa limosa The nominate subspecies has appeared on the Scottish List with the status code MB. There is no evidence that nominate /imosa breeds, and in fact, although it probably does occur, at least on passage, there are no formally substantiated records of this subspecies in Scotland. As a result the subspecies has been removed from the Scottish List. SBRC would welcome claimed records being submitted to it. Bridled Tern Sterna anaethetus The subspecies shown on the British List is antarctica, as a result of a specimen belonging to that subspecies having been found dead in Kent in 1931. It is clear that not all records of this species in Britain necessarily refer to that subspecies, and none of the 5 Scottish records have been racially determined, therefore the Scottish List is changed to show ‘race undetermined’. Dartford Warbler Sylvia undata The only race accepted by BOURC for Britain is dartfordiensis which is shown on the Scottish ist--There are 2 Scottish records: neither of which has been assigned to race, in fact it has been suggested that at least one of the records is more likely to have been the nominate race. This subspecies should be changed to ‘race undetermined’. Brown Shrike Lanius cristatus 2004 Shetland Skaw, Whalsay, adult L c cristatus, 19-24 September, trapped, photo (J Dunn, J L Irvine, B Marshall et al) (British Birds 97: plate 394, 98: 683-4). This is the first occasion that the subspecies of a Brown Shrike recorded in Scotland has been determined. Change from ‘race undetermined’ to Le crisiaus. Ovenbird Seiurus aurocapilla At the time subspecies were added to the Scottish List (Scottish Birds 22:46) the nominate subspecies was shown. Although this subspecies appears on the British List, the only Scottish record (Shetland 1973) was not assigned to race. Whilst the nominate subspecies is most likely, neither of the other 2 subspecies can be ruled out, therefore the race is now shown as ‘undetermined’. Category D Add Ross’s Goose Anser rossii and Yellow- headed Blackbird Xanthocephalus xantho- cephalus, both of which now appear on Category D of the British List as a result of Scottish records. Indigo Bunting Passerina cyanea Indigo Bunting was on Category D of both the Scottish and British Lists. It is on the Scottish Category D list due to a 1964 record from Fair Isle (there is also a 1974 Fair Isle record). BOURC have recently accepted Indigo Bunting on to Category A of the British List on the strength of a 1996 record from Wales. The Fair Isle records, pre date the Welsh record and therefore have not been found acceptable for Category A by BOURC. BOURC/SBRC do not have a joint Category A/D, therefore as the Fair Isle record does not meet Category A criteria it requires to be downgraded to Category E. As a result of the above changes, the Scottish List totals are now: Category A 496 Category B 7 Category C 6 TOTAL 509 Category D 13 Category C Category C had previously been subdivided into 5 sections. A 6th section has been added to the British List and should similarly be added to the Scottish List ‘C6 former naturalised species’. At present there are no species that fall into this category. BOURC conducted a review of the British Category C (Dudley 2005). As a result of _ this review the following change should be made to the Scottish List: English name Current category New category Snow Goose A ro & A population of Snow Goose originating from a waterfowl collection on the Isle of Mull (Argyll) has existed since the 1950s or 1960s and meets the Category C2 criteria. Changes to Status Codes Due to work undertaken by the team working on the new ‘Birds of Scotland’ book, it is clear several status codes are incorrect and require to be changed. Table 1 shows the agreed changes. Update to records of species and subspecies recorded in Scotland on up to 20 occasions A list of all records of species and subspecies recorded in Scotland on up to 20 occasions was published on behalf of SBRC (Andrews & Naylor 2002) since when there has been a single update (Forrester 2004). Since that time BBRC has published its Reports on rare birds in Great Britain in 2003 (British Birds 97: 558-625), 2004 SB 27 (British Birds 98: 628-694) and 2005 (British Birds 100: 16-61, 72-104). What follows is the second of what is intended to become regular updates. Red-breasted Goose Branta ruficollis 2001 Moray & Nairn Redhill, adult, 17-18 April, and Roseisle, 19 April (British Birds 96: 555), now considered not same as Perth & Kinross 2001 (British Birds 95: 484, 98:634). 2002 Perth & Kinross Findatie and other sites, adult, 15 February to 27 April (British Birds 96: 555), now presumed returning bird, same as Perth & Kinross 2001 (British Birds 95: 484). 2003 Argyll Loch Gruinart, Islay, adult, 18 October 2002 to 9 April, same 10 November to at least 28 March 2004 (A W Reid et al); presumed returning individual of 2002 (British Birds 96: 555, 97: 562; Scottish Bird Report 2001: 10). The Moray & Nairn bird is the 15th Scottish record. American Black Duck Anas rubripes 2002 Shetland Loch of Hillwell, m, 13-25 May (G J Fitchett (et al) (British Birds 97:563). 2003 Shetland Dales Voe, m, 18 January-24 February (B H Thomason et al) (British Birds 972503). 2004 Northeast Scotland New Pitsligo, m, 5 December to | January 2005 (J M Wills et al) (British Birds 98: 634). 8th-10th Scottish records. Ferruginous Duck Aythya nyroca 2003Argyll Loch Bhasapol, Tiree, m, 21-29 April (J Bowler et al) (British Birds 97: 563). 2003 Fife Loch Gelly, m, 30 July to at least 18 August (D Ogilvie, K D Shaw, J J Squire); Perth & Kinross Vane Farm, Loch Leven, m, 1-16 September (K D Shaw et al) (British Birds 97: 564). 13th and 14th Scottish records (of 15 birds). Scottish Birds (2007) Lesser Scaup Aythya affinis 2001 Outer Hebrides Loch an Eilean, South Uist, f, age uncertain, 27 January (P R Boyer, B Rabbitts, A Stevenson et al) (British Birds 97: 566). 2003 Dumfries & Galloway Castle Loch, Lochmaben, m, 29 December to 27 January 2004 (A W Reid ef al); same plus one other Milton Loch, 27 February to 10 March, one to 21st; one or other, Lochrutton Loch, | March; one or other, Caerlaverock, 16-17 March; both first noted in 2002 (British Birds 96: 558) (per P N Collin) (British Birds 97: 566, 98: 635). 2003 Perth & Kinross Vane Farm, Loch Leven, first winter m, Ist and 11th February (D Abraham, K D Shaw et al). Findatie, Loch Leven, m, 16-20 March (R Shand, K D Shaw, A Wilson et al) (British Birds 97: 566). 2004 Forth Gart Gravel pits, m, 19 June (A W Lauder, C McKeever, K A & K D Shaw) (British Birds 98: 635). 2004 Perth & Kinross Loch Leven, m, 4 December (A W Reid et al); first winter, 12-16 December (W McBay, J S Nadin, K D Shaw) (British Birds 98: 635). 2005 Perth & Kinross Blair Drummond, m, 23 March to 4 April (N Bielby et al) (British Birds 100: 22). 2005 Perth & Kinross Vane Farm, m, 3-4 July, photo (A W Lauder, J S Nadin, K D Shaw et al) (British Birds 100: 22). There are now 21 reports of 25 birds in Scotland. Now removed from the list of species recorded on 20 or fewer occasions. Harlequin Duck Histrionicus histrionicus 2004 Outer Hebrides Coll, Lewis, f, mid January to 20 May (M Hague, B Rabbitts, A Walker et al) (British Birds 98: 635; 97: plate 117). 9th Scottish record (11 birds). Black Scoter Melanitta americana 2005 Moray & Nairn Burghead Bay, male 10 October (R Proctor) (British Birds 100:23). Sth Scottish record. Amendments to the Scottish List: species and subspecies 7 Bufflehead Bucephala albeola 2004 Outer Hebrides Balranald, North Uist, m, 10 May (M Darlaston, B Rabbitts, A Rennels); same, Loch Branahuie, Lewis, 20 May, photo (C A & R P Cockbain et al) (British Birds 98: 638). 2nd Scottish record. Barrow’s Goldeneye Bucephala islandica 2005 Northeast Scotland Ythan Estuary, m, 13- 22 May, photo (P Shepherd ef al); same Loch of Strathbeg 23 May — 23 June (per Birding Scotland) (British Birds 98: plates 177 & 178, 100: 24, plate 1). 2nd Scottish record. Black-browed Albatross Thalassarche melanophris 2005 Outer Hebrides Sula Sgeir, adult, 25-31 August, photo (D Macfarlane per M S Scott) (British Birds 98: plates 392 & 393, 100: 25). 5th Scottish record. Black Stork Ciconia nigra 1998 Highland Munlochy, first year, 22 July to 3 August (D C & JA Jardine et al) (British Birds 100: 30). This is presumed to be the same bird as in North east Scotland from 3-19 July. Black Kite Milvus migrans 2003 Highland Kinbrace, Sutherland, 15 June to at least 8 July (N Dales er al); 2004 Kinbrace, Sutherland, 3 June to 3 July (per A F McNee), presumed returning individual of 2002 (British Birds 96: 560; 97: 572, plate 267; 98: 645 SBR 2001: 11); probably same Glen Glass, Easter Ross, 2 August (R H Hogg) (British Birds 97: 572). 2005 Ayrshire Pinwherry, 7 July (R H Hogg) (British Birds 100: 30). 19th Scottish record. SB 27 Pallid Harrier Circus macrourus 2003 Shetland Baltasound/Haroldswick area, Unst, juvenile, 2-17 September, photo (S E Duffield, M A Maher et al) (British Birds 97: 573). 2003 Shetland Haroldswick, Unst, first summer m, 23-25 May (S E Duffield, M A Maher ef al) (British Birds 98: 645). 2004 Shetland Trondra, juvenile, 25-26 September, photo (J D Okill et al) (British Birds 97: plate 384, 98: 645, plates 416 & 417). 2005 Shetland Sumburgh, Mainland, juvenile, 29 September (R Riddington) (British Birds 100: 31). 7th-10th Scottish records. American Coot Fulica americana 2004 Dumfries & Galloway Castle Loch, Lochmaben, 10-17 February (A A Murray, A White et al) (British Birds 98: 647). 2004 Outer Hebrides West Loch Ollay, 25 January to 7 April (A Stevenson ef al) (British Birds 98: 647). 2nd and 3rd Scottish records. Collared Pratincole Glareola pratincola 2003 Orkney Loch of Tankerness, 30 May, photo (K E Hague) (British Birds 97: 576). 8th Scottish record. Killdeer Charadrius vociferus 2004 Outer Hebrides Knockintorran, North Uist, first winter, 25 December to 8 January 2005, photo (J & S Entwhistle-Baker, B Rabbitts et al) (British Birds 98: 649, plates 56 and 421). 2005 Lothian Musselburgh, 22 January (B D & M Griffin, B A Hickman et al) (British Birds 100: 36). 10th and 11th Scottish records. Pacific Golden Plover Pluvialis fulva 2003 Northeast Scotland Ythan Estuary, juvenile, 2-16 November, photo (P A A Baxter, P S Crockett, C N Gibbins et al) (Birding Scotland 7: 19-22 and plates 12-16; British Birds 97: 578). 2003 Outer Hebrides North Boisdale/Daliburgh, South Uist, adult, 9 February to at least 5 April 2003 (A Stevenson et al), presumed returning bird from 2001 and 2002 (British Birds 96: 567,97: 378). 2004 Orkney North Ronaldsay, adult, 30 June (P A Brown, P J Donnelly) (British Birds 98: 650). 15th ad 16th Scottish records. Semipalmated Sandpiper Calidris pusilla 2003 Fair Isle adult, 13 August, photo (A J Bull, D N Shaw et al) (British Birds 97: 578-579). 2005 Shetland Grutness, Mainland, juvenile, 1- 6 November, photo (D Andrews, R Butcher, S Mitchell et al) (British Birds 99: plate 18, 100: 39, plate 8). During 1999 there were multiple observations of this species in Argyll and the Outer Hebrides, making it difficult to accurately assess the numbers involved. This has been reviewed and there are now 14 Scottish records of 21 birds. Western Sandpiper Calidris mauri The 1956 Fair Isle record (British Birds 56: 55- 58) is now considered to be unacceptable (Garner 2005). There are now 3 Scottish records. Upland Sandpiper Bartramia longicauda 2004 Shetland Foula, 4-8 May, photo (G Atherton et al) (British Birds 98: 657, plate 427). 7th Scottish record. Solitary Sandpiper Tringa solitaria 2003 Outer Hebrides Gravir, Lewis, 23-24 October, photo (M Witherall et al) (Birding Scotland 6: 183-185; British Birds 96: plate 405, 97: plate 341, 97: 582). Ath Scottish record. Terek Sandpiper Xenus cinereus 2005 Shetland Funzie, Fetlar, 10-13 June, photo, (A Grove et al) (British Birds 100: 44). 10th Scottish record. Scottish Birds (2007) Amendments to the Scottish List: species and subspecies 9 Franklin’s Gull Larus pipixcan 2004 Shetland Hametoun, Foula, adult, 10 June (S Bearhop, C J McInerny, S C Votier) (British Birds 98: 659-660). 9th Scottish record. Bonaparte’s Gull Larus philadelphia 2004 Caithness Castlehill and Thurso, adult, 12 September intermitently to 15 January 2005 (S Laybourne, S A M Manson et al) (British Birds 98: 660, plate 90.). 2004 Outer Hebrides Penrine, South Uist, adult, 31 March to 9 April, photo (D J Britton) (British Birds 98:660). 2004 Outer Hebrides Ardivachar, South Uist, adult, 10 April (A Stevenson et al) (British Birds 98:660). 2004 Outer Hebrides Penrine and South Bay, South Uist, adult, 18-22 April, North Bay, South Uist, adult, probably same 23 April (A Stevenson et al), possibly same as one or other of the above (British Birds 98:660, 100:49). 2005 North-east Scotland Peterhead, adult, 31 January (A Thiel et al) (British Birds 100: 49). 2005 Outer Hebrides Luskentyre, Harris, adult, 26 February to 5 March, photo, (T ap Rheinallt et al) (British Birds 100: 49). 2005 Outer Hebrides Balranald, North Uist, adult, 10 May, photo, (B Rabbitts et al) (British Birds 100: 49). 2005 Shetland Veensgarth, Tingwall, Mainland, adult, 17 July, photo (P Sclater et al) (British Birds 100: 49). The South Uist birds in 2004 have been treated by BBRC as 3 separate birds, making the above records the 17th-24th Scottish records. Now removed from the list of species recorded on 20 or fewer occasions. Bridled Tern Sterna anaethetus 2003 Angus & Dundee Arbroath, adult, 19 July (S R Green et al) (British Birds 98: 661). Sth Scottish record. Gull-billed Tern Sterna nilotica 2003 Outer Hebrides Rubha Ardvule, South Uist, 13-31 July (A Stevenson et al) (British Birds 98: 661). 9th Scottish record. Forster’s Tern Sterna forsteri 2001 Orkney Tingwall and Eynhallow Sound area, adult, 24-27 October (I A Dillon, E J Williams et al) (British Birds 97: 585). 2003 Argyll Oban Bay and Loch Feochan, first winter, 8-11 January, photo (W Jackson et al) (British Birds 97: 585). 3rd and 4th Scottish records. Oriental Turtle Dove Streptopelia orientalis 2002 Orkney Stromness, 20 November to 6 December (British Birds 96: 581; SBR 2001: 11). Extended last date to 20 December (British Birds 97: 587). 2003 Caithness Hill of Ratter area, first winter O s meena, 5 December to at least 24 March 2004, photo (N Money ef al), same, St John’s Brough, 23 February 2004, 6 and 24 March, photo (J Smith) (British Birds 97: 587, 97: plate 28, 983.663). Ath Scottish record. Barn Owl Tyto alba Andrews & Naylor (2002) included 23 Scottish records of the dark breasted Continental subspecies 7 a guttata, but they suggested some might require to be reassessed, thus leaving fewer than 20 records. There have been at least 2 more records in recent years to add to those included by Andrews & Naylor and until such time as records are reassessed it is best to treat this race as having in excess of 20 records. Now removed from the list of subspecies recorded on 20 or fewer occasions. 10 Pallid Swift Apus pallidus 2004 Northeast Scotland Newburgh, 31 October to | November, photo (P A A Baxter, S Morrison) (British Birds 98: 664). 2005 Shetland Skaw, Whalsay, 3 July (J L Irvine, B Marshall et al) (British Birds 100: 58). 4th and 5th Scottish records. Calandra Lark Melanocorypha calandra 2002 Orkney North Ronaldsay, 10-11 May (P J Donnelly et al) (British Birds 97: 590). Sth Scottish record. Siberian Rubythroat Luscinia calliope 2003 Fair Isle first winter f, 17-19 October, photo (A J Bull et al) (Birding Scotland 6:181- 182; British Birds 96: 97: 596, plate 406). 2005 Fair Isle Ist winter female, 23-27 October, trapped, photo. (M Culshaw, E Douglas et al)(British Birds 100: 78; Birding Scotland 8: 159-162). 3rd and 4th Scottish records. Red-flanked Bluetail Tarsiger cyanurus 2003 Fife Denburn Wood, Crail, f or first winter, 15-17 October, photo (C Andrews, R Armstrong et al) (Birding Scotland 7: 9-11, plate 4; British Birds 97: 596). 2003 Shetland Funzie, Fetlar, first winter, 28 September (P Crockett, I Gordon) (British Birds OF 296): 2003 Fair Isle f or first winter, 16 October (R J Curtis) (British Birds 97: 596). 2004 Shetland Fair Isle, M, 29 September (D N Shaw et al) (British Birds 98: 671). 12th-15th Scottish records. Isabelline Wheatear Oenanthe isabellina 2004 Shetland Sumburgh Head, Mainland, 22- 25 October (P M Ellis, P V Harvey, M Heubeck et al) (British Birds 98: 672). Sth Scottish record. SB 27 Pied Wheatear Oenanthe pleschanka 2002 Orkney North Ronaldsay, m, 13-27 October, photo (P J Donnelly et al) (British Birds 97: 597). 2003 Northeast Scotland Collieston, first winter m, 18-21 October, photo (P S Crockett et al) (British Birds 97: 597). 15th and 16th Scottish records. Desert Wheatear Oenanthe deserti 2003 Fife Kilminning, f, 10-15 November, photo (A Whitehouse et al) (Birding Scotland 7: 31-34, plates 25 and 26; British Birds 97: 599). 2003 Northeast Scotland Girdleness, f, 12-15 November, photo (P S Crockett, I Gordon et al) (British Birds 97: 599, plate 56). 2003 Shetland Gulberwick, Mainland, m, 10 November (L Dalziel) (British Birds 98: 673). 15th-17th Scottish record. Swainson’s Thrush Catharus ustulatus 2003 Shetland Sandwick, Unst, first winter, 27- 30 September, photo (M G & M J Pennington et al) (Birding Scotland 7: 35-38, plate 27; British Birds 97: 600, plate 353). 2003 Shetland Burrafirth, Unst, first winter, 15 October, photo (M A Maher, MG & M J Pennington et al) (Birding Scotland 7: 35-38, plates 28-31; British Birds 96: plate 409, 97: 600). 6th and 7th Scottish records. Grey-cheeked Thrush Catharus minimus 2003 Shetland Foula, first winter, 27-30 September, trapped 27th, photo (A R Mainwood et al) (British Birds 97: 601). 9th Scottish record. Veery Catharus fuscescens 2005 Shetland Northdale, Unst, Ist winter, 22 September, trapped, photo. (J Fairclough, R Lockwood, G Woodburn et al), later killed by cat. (British Birds 98: plate 402, 100: 81). 3rd Scottish record. Scottish Birds (2007) Amendments to the Scottish List: species and subspecies Il Savi’s Warbler Locustella luscinioides 2005 Perth & Kinross Cairnie Pier, male in song 10-16 May, (A J Leitch et al) (British Birds 100: 83). 8th Scottish record (of 9 birds). Thick-billed Warbler Acrocephalus aedon 2003 Shetland Fair Isle, adult, 16-17 May, trapped 16th, photo (A J Bull, G Tyler et al) (British Birds 96: plate 250, 97: 604, plate 357). 4th Scottish record. Sykes’s Warbler Hippolais rama 2002 Orkney North Ronaldsay, 26 August, trapped, photo; (P A Brown, M Gray, J S Lees et al); previously accepted as indeterminate Booted or Sykes’s Warbler (British Birds 97: 606) but now accepted as the latter (British Birds 98: 677). 2003 Orkney North Ronaldsay, 29 September to 1 October, trapped 29th (P A Brown, A E Duncan, N Gates et al) (British Birds 97: 606). 2003 Shetland Baltasound, Unst, 4-8 October, trapped Sth, photo (M A Maher, M G Pennington, B H Thomason et al) (British Birds 96: plate 413, 97: 606). Ath-6th Scottish records. Sardinian Warbler Sylvia melanocephala 2004 Shetland Skaw, Whalsay, first summer f, 30 May to 10 June, trapped, photo (J Dunn , J L Irvine, B Marshall et al) (British Birds 98: 677-8). 2005 Shetland Lerwick, female, 2-4 June (D Coutts, P V Harvey et al) (British Birds 100: 86). 2005 Fife Fife Ness Muir, male, 15 October to at least 3 November (A MacCormick, M Oksien et al) (British Birds 100: 86; Birding Scotland 8: 171-174). 10th-12th Scottish records. Hume’s Warbler Phylloscopus humei 2003 Fair Isle 13-14 November, trapped 13th, photo (D N Shaw et al) (British Birds 97: 611). 2003 Fair Isle 17-23 November (D N Shaw et al) (British Birds 97: 611). 2003 Fife Denburn Wood, Crail, 11 November (AM & TC Smout et al) (British Birds 98:679). 2003 Shetland East Ham, Bressay, 16-23 November, probably since 13th, photo (N Davies, S E Duffield et al) (British Birds 97: 611, plates 30 and 362). 2004 Northeast Scotland Cove, 16-22 October, trapped, photo (P AA Baxter, C N Gibbins, RA Mavor et al) (British Birds 98: 679). 6th-10th Scottish records. Western Bonelli’s Warbler Phylloscopus bonelli 1983 Shetland Seafield and Helendale, Lerwick, 27 September to 3 October; (M S Chapman, C Robson et al); previously accepted as indeterminate Western or Eastern Bonelli’s Warbler (British Birds 79: 575), but now accepted as the former (British Birds 98: 681). 2003 Shetland Scatness, Mainland, 13 May, photo (D P Hall, S J Minton et al) (British Birds 96: plate 251, 98: 681). 2004 Orkney Sanday, 12 October, photo (K E Durose, J Wright) (British Birds 98: 681). 2004 Orkney Herston, South Ronaldsay, 29 October to 15 November, trapped, photo (J A. & R McCutcheon et al) (British Birds 98: 681). 2005 Fair Isle 10 September (D, DR & J F Cooper et al) (British Birds 100: 91). 7th-11th Scottish records. Collared Flycatcher Ficedula albicollis 2004 Shetland Fair Isle, first summer m, 9-12 May (A J Perkins et al) (British Birds 97: plate 186, 98: 682). 2004 Shetland Muness, Unst, adult m, 2 June, photo (M Grey, M G Pennington ef al) (British Birds 97: plate 225, 98: 682). 15th & 16th Scottish records. 12 Brown Shrike Lanius cristatus 2004 Shetland Skaw, Whalsay, adult L c cristatus, 19-24 September, trapped, photo (J Dunn, J L Irvine, B Marshall et al) (British Birds 97: plate 394, 98: 683-4). 3rd Scottish record. Isabelline Shrike Lanius isabellinus 2003 Shetland Out Skerries, first winter m, killed by cat, 30 September, now at National Museums of Scotland (J Lidster, M J McKee, E Tait et al) (British Birds 97: 615). 2004 Shetland Vidlin, Mainland, first winter, 17 October (M S Chapman et al) (British Birds 98: 684). 2005 Angus & Dundee Montrose 22-28 October (R A Bramhall et al) (British Birds 100: 92: Birding Scotland 8:167-170). 1Sth-17th Scottish record. Long-tailed Shrike Lanius schach 2000 Outer Hebrides Howbeg and Howmore, South Uist, first winter, 3-4 November, probably since 27 October, photo (J C Brain, B Rabbitts, A Stevenson er al) (SBR 2001: 10). Ist Scottish record. Add the following references: (British Birds 97: 615; Ibis 147: 246-250). Red-eyed Vireo Vireo olivaceus 2003 Outer Hebrides Aird Mhor Plantation, Barra 5-7 October, photo (M Oksien ef al) (Birding Scotland 6: 185-187, 172 - 174; British Birds 97: 617). 7th Scottish record. Arctic Redpoll Carduelis hornemanni Northwest Greenland race C h hornemanni 2003 Shetland Fair Isle, C h hornemanni, 22-27 September (D N Shaw et al) (British Birds 98: 686). 2004 Outer Hebrides Bru, Lewis, C h hornemanni, 11 October (M S Scott) (British Birds 98:686). 2004 Outer Hebrides Bru, Lewis C h SB 27 hornemanni 23 October, photo (L & P Cunningham, M S Scott) (British Birds 98:686). 2004 Shetland Foula, C h hornemanni, 29 September to 8 October (P J Wright, R B Wynn et al) (British Birds 98: 686). 2004 Shetland Skaw, Unst, first winter C h hornemanni, 10-11 October, photo (M A Maher) (British Birds 98: 686). 20th- 24th Scottish records. Now removed from the list of subspecies recorded on 20 or fewer occasions. Yellow Warbler Dendroica petechia 2004 Outer Hebrides Breibhig, Barra, 2-7 October, photo (S L Rivers et al) (British Birds 97: plate 396, 98: 687). 2005 Shetland Garths Ness, Mainland, Ist winter male, 15-17 September (R M Mellor et al) (British Birds 98: plate 409, 100: 98, plate 36). 3rd and 4th Scottish records. Yellow-rumped Warbler Dendroica coronata 2003 Orkney Evie, 31 October to 6 November, photo (D Matson, J B Ribbands, E J Williams et al) (Birding Scotland 6: 175-177) (British Birds 96: plate 419, 97: 619). 6th Scottish record. Blackpoll Warbler Dendroica striata 2003 Outer Hebrides Snishival Plantation, South Uist, 17-20 October, photo (P MacEwan, A Stevenson et al) (British Birds 98: 687). 2005 Highland Glasnakille, Skye, Ist winter, 4 October (R D Day, R Macmillan) (British Birds 100:99; Birding Scotland 8: 177=178),; 2005 Outer Hebrides Loch Druidibeg, South Uist, Ist winter, 29 September (A Stevenson et al) (British Birds 100: 99). 5th-7th Scottish records. Common Yellowthroat Geothlypis trichas 2004 Shetland Foula, first winter m, 9-10 October, trapped, photo (J M. & T P Drew, AR Mainwood, M A Wilkinson et al) (British Birds Scottish Birds (2007) Amendments to the Scottish List: species and subspecies je) 97: plate 398, 98:689, plate 444). 3rd Scottish record. Savannah Sparrow Passerculus sandwichensis 2003 Fair Isle first winter, 14-19 October, trapped 14th, photo (D N Shaw, J G Walmsley et al) (Birding Scotland 6: 178-180; British Birds White-throated Sparrow Zonotrichia albicollis 2003 Shetland Fetlar, 10-11 May, photo (A Hughson, M Smith et al) (British Birds 97: 619-620). 2003 Fair Isle m in song, 9 June, photo (D N Shaw et al) (Birding Scotland 6: plate 119; British Birds 96: plate 279, 97: 619-620). 14th and 15th Scottish records. 96: plate 420, 97: 619, plate 367). 2nd Scottish record. Table 1. Changes to status codes English name Bean Goose (nominate race) Snow Goose (nominate race) Tufted Duck Long-tailed Duck Spotted Crake Dunlin (subspecies schinzii) Ruff Grey Phalarope Pomarine Skua Long-tailed Skua Mediterranean Gull Sabine’s Gull Lesser Black-backed Gull (graellsii) Glaucous Gull Great Auk Eurasian Collared Dove Common Kingfisher Horned Lark Meadow Pipit (nominate race) Winter Wren (nominate race) Black Redstart Redwing (coburni race) Mistle Thrush Firecrest Bearded Tit Common Linnet (nominate race) Common Redpoll (nominate race) Parrot Crossbill Current status code New status code WV WV PV SV SV IB RB WV PV RB MB WV PV CB WV WV CB PV MB PV MB PV WV MB PV CB PV CB PV WV PV PV WV PV PV WV PV CB PV PV PV WV PV PV WV MB PV MB PV WV WV WV HB Extinct FB Extinct FB FMB FWV RB RB PV RB MB RB CB WV CB WV PV MB RB PV WV MB RB PV WV PV PV PV WV CB PV WV WV PV CB? WV PV RB MB PV WV RB MB PV PV PV WV RB? PV RB PV WV CB? PV WV PV CB WV PV [RB?] SV RB SV The following additional status codes should be added: HB = hybrid breeder, FMB = former migrant breeder, FWV = former winter visitor 14 Dark-eyed Junco Junco hyemalis 2003 Shetland Out Skerries, first summer m, 1- 9 May, trapped Ist, 9th, photo (E Tait et al per R Riddington) (British Birds 96: plate 219, 97: 620; Birding Scotland 6: plate 118). 7th Scottish record. Yellowhammer Emberiza citrinella citrinella Andrews & Naylor (2002) included only 2 Scottish records for this subspecies. It is clear that most birds recorded in Shetland belong to this subspecies (Pennington et al 2003). There are therefore over 20 records. Now removed from the list of subspecies recorded on 20 or fewer occasions. Rose-breasted Grosbeak Pheuticus ludovicianus 2005 Outer Hebrides Ardmhor plantation, Barra, Ist winter, male, 8 October (K Gillon ef al) (British Birds 100: 102). 2nd Scottish record. Bobolink Dolichonyx oryzivorus 2005 Shetland Foula, 30 September to 4 October, photo. (K B Shepherd, N D & P J Wright, R D Wynn et al) (British Birds 100: 102). 6th Scottish record. References Andrews I J & Naylor K A on behalf of the Scottish Birds Records Committee. 2002. Records of species and subspecies recorded in Scotland on up to 20 occasions. Scottish Birds 23: 61-116. British Ornithologists’ Union Records Committee. 2005. Thirty-first Report (January 2005). /bis 147: 246-250. British Ornithologists’ Union Records Committee. 2006a. Thirty-second Report (October 2005). Ibis 148: 198-201. British Ornithologists’ Union Records Committee. 2006b. Thirty-third Report (April 2006). /bis 148: 594. Chapman M 2003. The Taiga Flycatcher in Shetland. Birding World 16: 429-432. SB 27 Chapman M S 2004. Scotland’s first Taiga Flycatcher, Sandgarth, Shetland, October 2003. Birding Scotland 7: 39-44. Chapman M 2005. Taiga Flycatcher in Shetland. British Birds 98: 547-550. Clugston D L, Forrester R W, McGowan R Y & Zonfrillo B on behalf of the Scottish Birds Records Committee 2001. Scottish List — species and subspecies. Scottish Birds 22:33-49. Dudley S P 2005. Changes to Category C of the British List. bis 147: 803-820. Dudley S P, Gee M, Kehoe C, Melling TM & The British Ornithologists’ Union Records Committee (BOURC) 2006. The British List: A Checklist of Birds of Britain (7th edition). Ibis 148: 526-563. Forrester R W on behalf of Scottish Birds Records Committee 2003. Amendments to the Scottish List. 2000 Scottish Bird Report: 5-7. (ed R Murray). Forrester R W on behalf of Scottish Birds Records Committee 2004. Amendments to the Scottish List. 200] Scottish Bird Report: 7-12. (ed R Murray). Garner M 2005. The Fair Isle Sandpiper: a review. British Birds 98: 356-361. Martin W 1853. Miscellaneous notices. The Naturalist 3: 61. Sangster G, Collinson J M, Helbig A J, Knox A G & Parkin D T 2004. Taxonomic recommen- dations for British birds: second report. [bis 146: 153-157. Sangster G, Collinson J M, Helbig A J, Knox A G & Parkin D T 2005. Taxonomic recommen- dations for British birds: third report. /bis 147: 821-826. Ronald W Forrester, Scottish List Subcommittee, Scottish Birds Records Committee, The Gables, Eastlands Road, Rothesay, Isle of Bute PA20 9JZ Revised manuscript accepted March 2007 Scottish Birds (2007) — Seasonal and altitudinal distribution of birds in a Scottish Borders valley 15 The seasonal and altitudinal distribution of birds in a Scottish Borders valley D G PYATT Regular monitoring of birds has been undertaken from 2000 to 2005 in a 75 km? river catchment. Records were grouped into breeding seasons and winter periods and 2 altitudinal zones. In the breeding season tetrads in the lower altitudinal zone contained up to two thirds of the 107 species found in the valley; in the upper zone tetrads contained up to 30 species. In the winter periods tetrads in the lower zone contained up to two thirds of the 85 species found in the valley; tetrads in the upper zone contained only up to 10 species. Introduction National and regional surveys leading to the publication of bird atlases at 10km square or tetrad scales are underway or in preparation in Great Britain and Ireland (see BTO or SOC websites). National and regional atlases of breeding birds (Sharrock 1976, Gibbons er al 1993, Murray et al 1998, Stott et al 2002) have been joined by a national atlas of wintering birds (Lack 1986) and regional atlases that include breeding and wintering birds (eg Elkins et al 2003). Surveys in lowland areas during the breeding season follow well established procedures but for less accessible upland areas especially in winter such procedures may be impracticable. In remote areas and where birds are sparsely distributed there is a need to make the best use of survey time given the short daylight period available. The present paper describes the number of Species present in an inland valley in south east Scotland and compares the breeding and winter seasons within 2 altitudinal zones. This information could inform the discussion about winter atlas fieldwork for the hilly areas of Scotland. Methods Survey area The results presented are from the Manor Valley in the Scottish Borders. At an earlier stage in the survey the Valley and its birds were briefly described by Pyatt (2003). The survey area comprises the complete catchment of the Manor Water, a tributary of the River Tweed, together with a minor extension into an adjoining catchment to furnish a neater boundary at the north east corner near to the town of Peebles. The area totals some 75 km’ and lies within the rectangle defined by national grid references NT1725 and NT2541. The valley is within the Southern Uplands of Scotland and is deep and steep sided. The lowest point in the valley, in Peebles, is at 160m and the highest at 817m. The 375m contour is used to divide the catchment into upper and lower altitudinal zones. Habitats in the lower zone mainly comprise improved pasture interspersed with small but mature woods of varied species. The upper zone is dominated by rough grazing and_ heather moorland with extensive immature conifer plantations. There is virtually no native woodland. The Manor Water has many tributaries supplied by springs that never dry up. As a consequence all 1 km squares have permanent running water. 16 SB 27 Survey method The data were collected during 2000-2005 and comprise some 22,000 records. An individual record consisted of one or more birds of one species at a particular location. Until May 2002 locations were recorded as 6 figure grid references from a map (Harvey 1994), since then most recording was of 8 figure grid references obtained using a Global Positioning System. Survey was undertaken mainly through ‘targeted walks’. These aimed to cover all parts of the valley and all habitats, roughly in proportion to their bird richness. In open country a band up to Ikm wide could be observed, although of course not all small birds would have been flushed within the band. Two BTO Waterways Bird No. of spp (sq) re, & sak 1-4 (1483) | 7 pea |= 8 (264) i oh r 42 (196) 1 : Pe ie eae it Ee i 13 - 16 (56) | ms & 20136) | ia Aa i 21 - 24 (12) hy | it e 75 IE) 7 at Pili ™ | ie | & 29 G) ” . ; | | rercray AS hae en | |, oS get g et = 4 : ; f | 7 = { ya | 9 1 = x = Ay ABS co ee Sia i i 2 $e deste 4 = ake a a ee | a — | H | | HARVEY lic. no 56046, 30 0406 18 19 20 2} y 23 4 25 Figure 1. Number of species/Iha square, breeding season (no of squares) Survey (WBS) plots covered the lower half of the Manor Water itself. Each plot involved 9 walks at fortnightly intervals along the river. These plots comprised about 10% of the survey effort given to the whole Valley. In winter there were fewer species present than in the breeding season and far fewer individuals. The survey effort in the winter was less than in the breeding season, perhaps by a factor of 4, but was deemed sufficient to detect all the species present (see Discussion and Appendix). Outside the WBS plots no attempt was made to count all individuals or pairs except for a small number of ‘key’ species including Ring Ouzel Turdus torquatus, Black Grouse Tetrao tetrix and White-throated Dipper Cinclus cinclus. No. of spp (sq) 1.4.07 71) 5-8 (126) 9-42.40) fd maT antl Sat Mes Re : ‘ F] f i 13 - 16 (10) : F< 209) Hy Po : mie tren inn 2i1 - 24 (1) 5-2 29 - 32 Q) 3-41 (1) Figure 2. Number of species/Iha square, winter (no of squares) Scottish Birds (2007) For The Atlas of Wintering Birds in Britain and Ireland (Lack 1986) the period chosen to represent ‘winter’ was mid November to end February. The beginning of November was avoided because some migrants, especially Fieldfare Turdus pilaris and Redwing Turdus iliacus, were still very mobile at that stage. In The New Atlas of Breeding Birds in Britain and Ireland (Gibbons ef al 1993) the period taken to represent the breeding season was April to July. For this paper the records from the 4 months November to February were combined to represent the winter and the 4 months March to June were taken to represent the breeding season. The early start to the breeding season chosen here partly reflected the earlier arrival of migrants in these years. The inclusion of November in the winter period here equalised the length of the 2 seasons and ignored the mobility of winter visitors in that month. Data were processed in a database (Microsoft Access) and records were then transferred to the DMAP mapping program (Morton 2003). In this paper extensive use is made of the DMAP utility of ‘coincidence mapping’ in which the number of species recorded in each grid square is counted. In what follows this number will be referred to as N. The number of species in each square reflects the Number of species per square, breeding season Number of species Size of square (length of side) Figure 3. The effect on N of increasing the size of the square, breeding season. Seasonal and altitudinal distribution of birds in a Scottish Borders valley 17 range and richness of the habitats in the square. The size of the square can be chosen at will, but for the current purpose the side lengths used are 100 m (1 ha), 200 m, 500 m, 1 km and 2 km (a tetrad). Squares of 1 ha and | tetrad are the practical limits for a survey area of this size. By progressively enlarging the square from a given starting location additional species are included. In uniform habitat the rate of inclusion may be small but where there is a range of habitats the rate of inclusion should be greater. Thus the number of species in the starting | ha and this rate of inclusion together provide a measure of local habitat diversity. For display on maps it was convenient to choose 9 classes for VN. DMAP also counts the number of squares in each N class; these numbers are shown in the key to each map in parentheses and are hereafter referred to as S. Although DMAP easily handles the data for breeding season or winter period separately, it cannot count squares in altitudinal zones (it is not a Geographic Information System). For the upper zone the squares within each N class were counted manually and then subtracted from the totals to get the numbers in the lower zone. Number of species per square, winter Size of square (length of side) Figure 4. The effect on N of increasing the size of the square, winter. 1S SB 27 Table 1. Number of ha squares (S) with different numbers of species (N) in 2 altitudinal zones, in breeding season and winter. Number of species per Iha square (N) Zone 1-4 5-8 9-12 13-16 17-20) = 21-24) =. 25-28). 29-32 333-41 Breeding season (S) Lower III11 250 104 56 36 IZ 10 3 2 Upper 372 12 2 Winter (S) Lower 682 125 AO 10 9 l l Upper 89 l Results Number of species per square Seasonal and altitudinal distribution Some 13,000 records were accumulated in the 6 breeding seasons, from 108 species, of which some 87 were thought to breed (see Appendix). By choosing the | ha grid square, the precise distribution of these records can be shown (Figure 1). The number of squares (ha) S decreases rapidly (actually exponentially) as class size N increases. What is not evident in Figure | is that of the 1483 ha in N class 1-4 about half (784) actually had only one species present. Over the 6 winters 4,300 records were collected from 86 species. The | ha distribution of these records 1s shown in Figure 2. Of the 771 ha in N class 1-4 more than half (421) actually had only one species present. Although the highest N class had the same number of species as in the breeding season, there was only one ha in this class. There are indeed only 3 ha spanning the 4 highest N classes. The preponderance of records at the lower altitudes in both seasons evident in Figures | and 2 is quantified in Table |. In the breeding season the overall proportion of records in the upper zone was 19%, whereas in the winter it was only 9%. For the breeding season records, as the size of square was increased to 200 m, then to 500 m the increase in N was modest. At the I km size there was an appreciable increase in N and the inverse relationship between N and S was less evident. The 3 most bird-rich 1 km squares had slightly more than half of the total number of species in the survey area. When the squares were increased to tetrads, the highest value of N was 65-79 species and there were 4 such tetrads. For the winter records, as the size of square was increased to 200 m the N classes remained the same but there were then 12 squares in the highest 4 classes. With squares of 500 m the N classes were again unchanged but there were 22 squares in the highest 4 classes. The 1 km distribution had similar S$ values to the breeding season but with much lower N classes. As in the breeding season, the highest N class (with 4 squares) had about half as many bird species as the total for the survey area. When the squares were increased to tetrads, the highest value of N was 49-60 species and there were 4 such tetrads. The effect on number of species per square (NV) of increasing the size of the square was studied Scottish Birds (2007) by taking as starting points the grid reference of each tetrad that had at least half its area within the survey area. For each size of square the value of N was found to the nearest unit by manipulating the class intervals within the DMAP coincidence mapping utility. By this means it was possible to ‘home in’ on the precise value of N. For the breeding season and winter records separately the results are shown in graphical form in Figures 3 and 4. Breeding season values of N were always larger than winter values, but the differences were small for many squares. Seasonal and altitudinal distribution of birds in a Scottish Borders valley 19 Comparison with existing atlases During the 6 years of the survey 111 species were recorded in the Manor Valley. In the first 4 columns of Table 2 these are classified in terms of number and breeding status. In the breeding season 107 species were recorded of which some 87 were thought to breed. In the winter 86 species were recorded. In Table 2 the column headed Breeding Atlas lists 86 species found breeding by Murray ef al (1998) in 31 tetrads wholly or partly lying within the Manor Valley. The Breeding Atlas included a further 6 species that were not recorded in the present survey, namely Dunlin Calidris alpina, Stock Pigeon Columba oenas, Short-eared Owl Asio flammeus, Wood Warbler Phylloscopus Table 2. Checklist of birds in the Manor Valley 2000-2005 Number Code = breeding status Blank = does not breed 0. = has bred, not regularly 1 = 1-10 pairs 2 = 11-100 pairs 3 = 101-1000 pairs 4 = over 1000 pairs Standard Name/Scientific Name Little Grebe Tachybaptus ruficollis Great Cormorant Phalacrocorax carbo Grey Heron Ardea cinerea Mute Swan Cygnus olor Whooper Swan Cygnus cygnus Greylag Goose Anser anser Eurasian Wigeon Anas penelope Eurasian Teal Anas crecca Mallard Anas playrhynchos Tufted Duck Aythya fuligula Common Goldeneye Bucephala clangula Goosander Mergus merganser Hen Harrier Circus cyaneus Eurasian Sparrowhawk Accipiter nisus Common Buzzard Buteo buteo — SO WAND MN BWN _ — — = — — Mm BW NY Letter Code = time of year when most likely to be seen B = Breeds, but not resident all year R = Resident all year W = Winters P = Mainly seen on passage S = Non-breeding summer visitor V = Vagrant, recorded less than annually Number Letter Breeding Winter Code Code Atlas Atlas W = W 2 R S H l R 5 i V S V pi RW B 2 R # “ V V u l RW A i 0 B H l R i H 2 R fn 20 16 17 18 19 20 21 29) 23 24 25 26 27 28 29 30 31 32 33 34 35 36 a7 38 39 40 4 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Golden Eagle Aquila chrysaetos Osprey Pandion haliaetus Common Kestrel Falco tinnunculus Merlin Falco columbarius Peregrine Falcon Falco peregrinus Willow Ptarmigan Lagopus lagopus Black Grouse Tetrao tetrix Red-legged Partridge Alectoris rufa Grey Partridge Perdix perdix Common Pheasant Phasianus colchicus Common Moorhen Gallinula chloropus Common Coot Fulica atra Eurasian Oystercatcher Haematopus ostralegus Ringed Plover Charadrius hiaticula Eurasian Dotterel Charadrius morinellus European Golden Plover Pluvialis apricaria Northern Lapwing Vanellus vanellus Common Snipe Gallinago gallinago Eurasian Woodcock Scolopax rusticola Eurasian Curlew Numenius arquata Common Redshank 7ringa totanus Common Sandpiper Actitis hypoleucos Black-headed Gull Larus ridibundus Mew Gull Larus canus Lesser Black-backed Gull Larus fuscus Herring Gull Larus argentatus Great Black-backed Gull Larus marinus Feral Pigeon Columba livia Common Wood Pigeon Columba palumbus Eurasian Collared Dove Streptopelia decaocto Common Cuckoo Cuculus canorus Barn Owl Tyto alba Tawny Owl Strix aluco Common Swift Apus apus Common Kingfisher Alcedo atthis Green Woodpecker Picus viridis Great Spotted Woodpecker Dendrocopus major Sky Lark A/auda arvensis Sand Martin Riparia riparia Barn Swallow Hirundo rustica House Martin Delichon urbica Meadow Pipit Anthus pratensis Grey Wagtail Motacilla cinerea Pied Wagtail Motacilla alba Bohemian Waxwing Bombycilla garrulus NO We Re NY — Ne Re RK WwW DN NNR NNN WN LY NN ABANNNNNR KWWWWDWDWDWAAANAATAAASENNSATWWAATWDVGSD SAA AAA AATAUNSK + Ht HHH +t Ht +H tre HHtHH HH -rtHHtHHtHtHt Ht HHH HHH HH + + +H +H SB 27 Scottish Birds (2007) 61 62 63 64 65 66 67 68 69 70 7A 2 le: 74 us 76 om, 78 79 80 81 82 83 84 85 86 87 88 89 90 9] 92 93 94 95 96 oF 98 99 100 101 102 103 104 105 White-throated Dipper Cinclus cinclus Winter Wren Troglodytes troglodytes Hedge Accentor Prunella modularis European Robin Erithacus rubecula Common Redstart Phoenicurus phoenicurus Whinchat Saxicola rubetra Stonechat Saxicola torquata Northern Wheatear Oenanthe oenanthe Ring Ouzel Turdus torquatus Blackbird Turdus merula Fieldfare Turdus pilaris Song Thrush Jurdus philomelos Redwing Turdus iliacus Mistle Thrush Turdus viscivorus Sedge Warbler Acrocephalus schoenobaenus Lesser Whitethroat Sylvia curruca Common Whitethroat Sy/via communis Garden Warbler Sylvia borin Blackcap Sylvia atricapilla Common Chiffchaff Phylloscopus collybita Willow Warbler Phylloscopus trochilus Goldcrest Regulus regulus Spotted Flycatcher Muscicapa striata Long-tailed Tit Aegithalos caudatus Coal Tit Parus ater Blue Tit Parus caeruleus Great Tit Parus major Wood Nuthatch Sitta europaea Eurasian Treecreeper Certhia familiaris Eurasian Jay Garrulus glandarius Black-billed Magpie Pica pica Eurasian Jackdaw Corvus monedula Rook Corvus frugilegus Hooded Crow/Hybrid Crow Corvus corone corvix Carrion Crow Corvus corone corone Common Raven Corvus corax Common Starling Sturnus vulgaris House Sparrow Passer domesticus Chaffinch Fringilla coelebs Brambling Fringilla montifringilla European Greenfinch Carduelis chloris European Goldfinch Carduelis carduelis Eurasian Siskin Carduelis spinus Common Linnet Carduelis cannabina Twite Carduelis flavirostris a) NOR NNR RK WN W LV WNN RNY RW We Re NN NWN EWHA REAR AAAAAAAVAA AAA POP WWWHWVOR SH SSW WWW WAAR Seasonal and altitudinal distribution of birds in a Scottish Borders valley tHteHtHtHtHt Ht HH HS HK te th #+t#HrHHt Ht HH HHH HH + tH +t Ht Ht +H ++ HH ++ HH +H ttt H +H tb t+ +H +h +HRHHtHtH HH HH 21 22, 106 Lesser Redpoll Carduelis cabaret 107 Common Crossbill Loxia curvirostra 108 Common Bullfinch Pyrrhula pyrrhula 109 Hawfinch Coccothraustes coccothraustes 110 Yellowhammer Emberiza citrinella 111 Reed Bunting Emberiza schoeniclus Totals sibilatrix, Pied Flycatcher Ficedula hypoleuca and Marsh Tit Parus palustris. The status of Coot Fulica atra and Fieldfare Turdus pilaris differed between the Breeding Atlas and the present survey. Four 10 km squares intersect within the Manor Valley, of which square NT23 provides the best match of habitats with the valley as a whole. In Table 2 the column headed Winter Atlas lists the 66 species found wintering in square NT23 by Lack (1986). Discussion Seasonal and altitudinal distribution The apparent concentration of birds near to the watercourses in both seasons is_ striking. Riverside and streamside habitats are diverse, relatively sheltered and presumably tend to be productive of food for many bird species but this explanation is more convincing for the winter distribution than for the breeding season. Furthermore, Figure 2 is likely to be a more accurate representation of the winter distribution of birds than Figure | is of the breeding distri- bution. For example, in spring and summer it would be possible to find Meadow Pipits in many of the apparently blank areas of the upland zone. The ‘string of beads’ records that can be seen in various places such as along the watershed boundary show the birds encountered along a particular walking route. Detailed surveys by Pearce-Higgins and Grant (2006) have shown that Meadow Pipits occupy habitat typical of the upland zone at a density of about 1 pair per ha. At the tetrad scale the concentration of breeding species along watercourses was also SB 27 R Hf #t R # 2 R # # V Ht 0 B Ht Ht l B # 111 86 66 found by Murray ef al (1998) in south east Scotland (see their map on p27). The | ha square with the highest number of species, 42 in the breeding season and 41 in winter, surrounded the surveyor’s garden and although a bird feeding station may have had some influence in winter, this was inevitably the most observed part of the survey area. On the other hand all the species recorded in this square were found elsewhere in the survey area, indicating that concentrated survey effort increased the number of species locally but not for the survey area as a whole. Number of species per square The 1 km size of square is used in national surveys such as the British Trust for Ornithology (BTO) Breeding Bird Survey. In the Manor Valley there are more species per 1 km square in the breeding season than in winter (Figures 3 and 4). In each season the highest N value is just over half of the total number of species for the survey area. The upland zone (see Figure 2 for the boundary) is almost entirely composed of the lowest N value, 1-7 species in the breeding season and 1-5 in the winter. It should be emphasized however that these are numbers in individual squares and not the total number of species in the upland zone. The tetrad is used in most regional atlases, such as that for South east Scotland (Murray et al.1998) that includes the Manor Valley. In the breeding season there are 4 tetrads with the highest N value of 65-79, which represents Scottish Birds (2007) Seasonal and altitudinal distribution of birds in a Scottish Borders valley | about two thirds of the total number of species for the survey area. In the winter there are also 4 tetrads with the highest N value of 49-60, again about two thirds of the number of species for the survey area. Comparison with existing atlases The comparison of species lists between the present survey and existing breeding and winter atlases shows remarkably good agreement for both seasons. This adds further weight to the view that it is not necessary to spend anything like the 1500 hours fieldwork of the present survey to obtain an accurate list of the species present in an area the size of the Manor Valley. Where there are omissions of species in the atlases, these are species that are known to have increased in the last decade eg wintering Little Grebe Tachybaptus ruficollis, wintering Common Buzzard Buteo buteo. Ringed Plover Charadrius hiaticula recently extended its breeding range to the Manor Valley. It is a minor criticism of the winter atlas that it underestimated the status of the Twite Carduelis flavirostris or that the same may be said of the breeding atlas about the Reed Bunting Emberiza schoeniclus. The numbers of species per | km square or per tetrad recorded in the Manor Valley are probably typical of the western and southern parts of the Scottish Borders. Murray et al (1998) found 43 and 34 breeding species per tetrad in altitudinal bands 200-299 m and 300-399 m respectively, equivalent to the lower zone of the Manor Valley. They found 24, 17 and 9 breeding species per tetrad in altitudinal bands 400-499 m, 500- 599 m and over 600 m respectively, equivalent to the upper zone of the Manor Valley (see their p27). Moreover it should be noted that the Valley lacks a substantial loch and therefore is deficient in waterfowl, particularly in the breeding season, compared to many parts of the Scottish Borders. References BTO 2006. Why another atlas? Atlas 2007- 2011. British Trust for Ornithology internet website. www.bto.org. Gibbons D W, Reid, J B and Chapman R A 1993. The New Atlas of Breeding Birds in Britain and Ireland: 1988-1991. T & A D Poyser, London. Harvey 1994. Peebles, Manor Hills & St Mary’s Loch. Map at 1:40,000 scale, Harvey Map Services. Lack P 1986. The Atlas of Wintering Birds in Britain and Ireland. T & A D Poyser, Calton. Morton A J 2003. DMAP for Windows (2003). Software for Distribution Mapping, Version 7.2a. Internet Web Site: http://www.dmap.co.uk/ Murray R D, Holling M, Dott H E M and Vandome, P 1998. The Breeding Birds of South-east Scotland: a tetrad atlas 1988-1994. The Scottish Ornithologists’ Club, Edinburgh. Pearce-Higgins J W and Grant M C 2006. Relationships between bird abundance and the composition and structure of moorland vegetation. Bird Study 53, 112-125. Pyatt D G 2003. Birds of the Manor Valley. Scottish Bird News 67 :12-13. Scottish Ornithologists’ Club. D G Pyatt, The Schoolhouse, Manor, Peebles EH45 9JN Revised manuscript accepted August 2006 24 SB 27 Autumn skua migration at Hound Point, Lothian C J MCINERNY & M GRIFFIN The autumn migration of 4 species of skuas was studied at Hound Point, Lothian. Systematic observations over a 20 year period from 1986-2005 revealed regular passage of a large number of birds at this site. Arctic Skua was the most frequently observed species, largely in September and October, usually seen in small groups, with day totals rarely exceeding 100. Long-tailed Skua was seen more irregularly in smaller numbers during August and September. In most years only a few immatures and adults were recorded; very rarely, day totals of 100-200 adults in flocks of up to 20 were recorded. A similar, irregular pattern was observed for Pomarine Skua, although with larger numbers and flocks of up to 50, and a broader seasonal occurrence from August to November. Great Skua was seen every year, usually as singles or in small groups of up to 10, mostly in September and October. Many birds of all 4 species were observed passing west, apparently moving overland to the west coast of Scotland and the Atlantic Ocean. Introduction The 4 species of skua breeding in the northern hemisphere undergo long migrations, wintering from equatorial coastal waters to deep waters and rich currents in the southern oceans (Furness 1987, Olsen & Larsson 1997). For 2 species, Arctic Skua Stercorarius parasiticus and Great Skua Stercorarius skua, their breeding range includes northern and western Scotland (Loyd et al 1991, Mitchell et al 2004). These, along with Long-tailed Skua Stercorarius longicaudus and Pomarine Skua Stercorarius pomarinus, are observed migrating through Scottish waters each spring and autumn (Scottish Bird Reports, Critchley 1972, Davenport 1979, 1987, 1992, Fox & Aspinall 1987, Dunn & Hirschfield 1991, Harrop ef al 1993). In autumn most migrating skuas seen in Scotland are recorded in the North Sea, observed from land based sites along the east coast, particularly at exposed headlands after strong onshore winds (Wallace & Bourne 1981, Steele 1987, Innes 1993). Large numbers have also been recorded at coastal sites in the Firth of Forth and Moray Firth at geographic bottlenecks, many kilometres from the open North Sea. The 2 best known of these are Hound Point, Lothian (Griffin & McInerny 2001), and Chanonry Point on the Black Isle, Ross & Cromarty (McInerny 2004). In the Firth of Forth skuas have been recorded in its upper stretches since the late 18th Century (Pennant 1771, Gray 1871, Godfrey 1899, Harvie-Brown 1906, Rintoul & Baxter 1935), with a regular autumn passage first noted at Hound Point in the late 1970s (Andrews 1986). Occasionally, exceptional influxes have occurred resulting in very large numbers being observed (Dalgleish 1880, Rintoul & Baxter 1935, Brown & Andrews 1986, Andrews 1986, Andrews & Griffin 1989, McGarry & Speak 1992, Griffin & McInerny 2001). Since the mid 1980s this migration has been monitored on a regular basis, with birds seen from July to early December. In this paper we describe our observations from Hound Point during the period 1986-2005, and discuss the autumn movements of skuas at this site. Scottish Birds (2007) Autumn skua migration at Hound Point, Lothian 25 Study area and methods Hound Point is a small promontory on the south side of Firth of Forth, 8 km west of Edinburgh and 4 km east of South Queensferry, where the firth becomes constricted. In this area the waters narrow to 3 km, with the Forth Railway Bridge artificially ‘sealing’ the firth 2 km to the west. These geographic features result in the concentration of migrating seabirds entering from the east, with Northern Gannet Sula bassana, Black-legged Kittiwake Rissa tridactyla, Northern Fulmar Fulmaris glacialis and skuas the most commonly seen. Hound Point, although only at elevation of 10 m, has the advantage of giving an unrestricted view to the north and east out of the firth. Regular visits to the site began in 1986, with between 5 and 47 per year (Table 1). Each visit lasted for at least 2 hours, resulting in the total viewing time each year of 15.5 to 81.5 hours. Binoculars and viewing telescopes were used to scan for skuas, which could pass both low over the water and high in the sky. The species, number, and direction of movement were recorded, with age class determined for Arctic, Pomarine and Long-tailed Skuas. Duplication of records was avoided wherever possible, with recognisable individuals and flocks only counted once. This was important as some birds were observed passing the point on a number of occasions, either entering and leaving the firth, or re entering a number of times. Results Pomarine Skua This species was observed on 53% of visits with 2,635 recorded. The annual total numbers varied considerably, from 3 to 790; 61% were observed moving west, 14% east, and the remainder indeterminate (Figure 1). Most passed in October (57%) and September (23%), with fewer in August (15%) and November (6%); in July just 4 in total were noted (Figure 2). The majority were adults (73%) and 27% immatures. Most adults passed in October (56%) and September (26%), with fewer in August (18%), November (2%) and July (<1%). Immatures passed later, as more were noted in October (67%) and November (11%), with fewer in September (16%) and August (7%). The number seen on each visit varied enormously. On many days singles and small groups of up to 6 were observed, but a number of larger flocks of up to 20, and day counts of 50-100 were recorded. The largest day count was of c 420, including 350 adults and 70 Table 1. Observer coverage at Hound Point, Lothian, 1986-2005. Year Visits Viewing time (hr) 1986 2) 15.5 1987 22. 60.0 1988 24 59.0 1989 3] TED 1990 47 rsh Ue) 199] 34 67.5 1992 p16) 56.0 1993 25 52.0 1994 Si) (hi 1995 25 50.0 Year Visits Viewing time (hr) 1996 14 PH | 1997 23 53 1998 19 34 1999 14 29 2000 15 33 2001 20 31 2002 20 54 2003 13 P| 2004 13 26 2005 9 26 26 A oO @ jo) N vt wo fee) jo) N vt ice) @ ron) fon) toa) on) a j=) j=] j=) on) on) on a a fon) a oOo oO jo} o aoa al ae aaa al 4 N N N Figure 1. Annual occurrence and direction of migration of Pomarine Skuas at Hound Point, Lothian, 1986-2005. Grey, total numbers; black, numbers moving east; white, numbers moving west. immatures, on 16 October 1992, the highest day tally for any of the 4 skua species. This influx was followed 4 days later by 145 passing on 20 October 1992, which included 130 adults and 15 immatures. Weather conditions on _ both occasions were force 4-5 north to north easterly winds, after a period of westerly and north westerly airflows. Another exceptional passage occurred on | October 1994 during a period of moderate easterly winds and low cloud: 186, including 173 adults and 13 immatures, were watched flying west past the point. This movement contained the largest flock of Pomarine Skuas seen at the site, 48 birds. Adult Pomarine Skua exists in 2 colour phases, the light phase being the most common in the North Atlantic (Olsen & Larsson 1997). We observed only 70 dark phase birds, 4.6% of the total number of adults. Arctic Skua This species was the most regularly seen passing Hound Point and was observed on almost all visits (98%), with numbers reasonably constant between years, although showing a range of 50 to 541; 50% were observed moving west, 9% east, and the remainder indeterminate (Figure 3). Out a total of 5,501 recorded the majority passed SB 27 a wn Oo Oo o oO wi 2 500 x) “5 400 3 300 3 3 E 200 2 100 0 © ree) ° N st ito) 00 ° N st ica) oo a a a aD roa ro) ° ° a a a a a ro.) roy 3} to) is) mr oa re a we baal Poel N N N 900 800 3 700 ‘5 600 ‘6 500 @ 400 2 £ 300 3 2 200 100 0 o ms cod i M im bn & s 4 g a 2 rs 5 ox oH £ fe! E (S Zz o QO © ® a 9 z a 2 2 a Figure 2. Annual and seasonal occurrence of adult (black) and immature (white) Pomarine Skuas at Hound Point, Lothian, 1986-2005. in September (63%), with fewer in August (21%) and October (15%); less than 1% were seen both in July and November (Figure 4). Of these c 60% were immatures. The number seen on each visit was the most constant amongst the 4 species of skua: between 10 and 20 were seen, with counts above 50 being unusual. The largest day counts were 105 on 12 September 1996 in a northerly force 4-5 wind, with 84 moving west; and 102 on 19 September 1999 in an easterly force 5-6 wind, including 78 migrating west. Arctic Skua only rarely formed larger flocks of more than 10 birds. Most were seen migrating either as singles or in small loose groups of up to 5. The largest flock recorded was of 21 on 11 September 1989. Scottish Birds (2007) Number of birds w So o TES SE BEI AIT EE NET 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 Figure 3. Annual occurrence and direction of migration of Arctic Skuas at Hound Point, Lothian, 1986-2005. Grey, total numbers; black, numbers moving east; white, numbers moving west. | aus June July August September L October November December Figure 4. Seasonal occurrence of Arctic Skuas at Hound Point, Lothian, 1986-2005. Arctic Skua exists in 2 colour phases, pale birds predominating in the northern part of the breeding range and dark birds in southern latitudes, including Scotland (Furness 1987). Numbers of each phase varied between years, but of the adults, 60-70% were dark and 30-40% were pale. Long-tailed Skua This was the rarest species with 920 recorded, observed on 27% of visits, and annual total numbers varying considerably, from 0 to 415; 79% were observed moving west, 4% east, and the remainder indeterminate (Figure 5). Most passed in September (57%) and August (35%), with fewer in October (15%); just one, an adult, Autumn skua migration at Hound Point, Lothian a7, was recorded in July (Figure 6). Most recorded were adult (85%) and 15% immature, although the latter were likely under recorded due to the difficulty of separating individuals of this age class from Arctic Skua. The majority of adults passed in September (55%), and August (38%), with few in October (8%). Immatures appeared to pass later as more were noted in October (21%), although most were seen in September (64%), and a significant number, 14%, were recorded in August. Typically, birds passed as singles or in small groups of up to 4. However, a number of exceptional passage days were recorded, when over 100 were seen. The first of these was on 7 September 1991 when 180 adults and 10 immatures were counted, of which 177 moved west, during a westerly force 2-3 wind followed by north easterly force 3-4. This movement continued over the next 3 days, with 35, 24 and 60 recorded. The largest passage was on 7 August 1995, when 217 migrated west past the point, including 214 adults and 3 immatures, during a force 4 north easterly wind. This day had the largest group, of 22 adults, recorded at Hound Point. Just 4 were noted the following day. Number of birds 2002 2003 2004 2005 ama i=) oO N Figure 5. Annual occurrence and direction of migration of Long-tailed Skuas at Hound Point, Lothian, 1986-2005. Grey, total numbers; black, numbers moving east; white, numbers moving west. 28 400 300 Number of birds N [=} oO 100 2001 2002 2003 2004 2005 a Number of birds June July August September October November December Figure 6. Annual and seasonal occurrence of adult (black) and immature (white) Long-tailed Skuas at Hound Point, Lothian, 1986-2005. Two adults showing characteristics of the Nearctic subspecies pallescens were observed. One was on 30 September 1988 and the other on 27 August 1998. While the status of this subspecies on the British List is currently under review (Sangster et al 2004), its appearance in Britain seems plausible, considering the migration pattern of east North American and Greenland breeding birds through the mid Atlantic, and their likely displacement westwards after strong anticyclonic winds. Both adults and immatures were sometimes noted in mixed flocks, especially with Arctic Skuas. Lone, immature Long-tailed Skuas were also rarely noted migrating in flocks of Black- legged Kittiwakes. SB 27 Great Skua This species was the second most regularly seen passing Hound Point, being observed on 67% of visits, with some fluctuation in numbers between years, showing a range of 17 to 248; 55% were observed moving west, 30% east, and the remainder indeterminate (Figure 7). Out a total of 2,099 recorded, the vast majority passed in September (49%) and October (41%); just 8% were seen in August and less than 1% and 2%, respectively, were noted in July and November (Figure 8). Separation of adults from immatures is extremely difficult under distant field conditions, and so we did not differentiate age classes. Most were seen migrating as singles or in small groups of up to 6; larger groups of up to 34 were rare. The largest flock recorded was on 30 September 1998 when a group of up to 73 was seen during a north easterly force 4-6 storm. These birds did not move west in the low cloud and rain, but instead formed a raft on the sea off the point, with individuals continuously entering and leaving the Firth of Forth from the east. Great Skua was occasionally noted migrating with Northern Gannet, typically singles within various group sizes. Such migration by skuas along side their host species has been noted elsewhere (Olsen & Larsson 1997). Mixed flocks Mixed flocks of various skua species were observed on many occasions migrating together past the point. The most common combinations were Pomarine and Arctic Skuas, Arctic and Great Skuas, and Arctic and Long-tailed Skuas. Pomarine and Great Skuas were also more rarely seen together, and on a few occasions flocks containing all 4 species were noted. Scottish Birds (2007) Autumn skua migration at Hound Point, Lothian 29 Number of birds 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 Figure 7. Annual occurrence and direction of migration of Great Skuas at Hound Point, Lothian, 1986-2005. Grey, total numbers; black, numbers moving east; white, numbers moving west. Number of birds June July om a =| Da 3 x September October November December Figure 8. Seasonal occurrence of Great Skuas at Hound Point, Lothian, 1986-2005. Discussion The North Sea is an established regular autumn passage route for skuas (Webb ef al 1990, Stone et al 1995). The annual appearance of Arctic and Great Skuas in these waters indicates that they intentionally use this sea area each year on their journey to southerly wintering areas. At least for Great Skua, satellite tracking of Shetland bred birds has revealed a number of individuals moving south along the east side of Britain (Robert Furness, personal communication). These 2 species probably pass through the North Sea because of the large number of other seabird Species present at the time of the year, which they kleptoparasitise for food (Wourinen 1992). The Firth of Forth attracts very large flocks of post breeding Sterna terns in August and September, particularly Sandwich Tern Sterna sandvicensis. In contrast, Pomarine and Long-tailed Skuas are seen very much more irregularly in the North Sea, with only small numbers noted in most years, and occasional larger influxes. Both species are thought to prefer to migrate over deeper waters through the mid Atlantic (Stone et al 1995, Pollock et al 2000), those seen in the North Sea brought there only after unusual weather conditions such as strong westerly winds. Another crucial factor influencing their occurrence is breeding success. Both skua species show dramatic fluctuations in annual breeding productivity due to changes in numbers of lemmings (Lemmus, Myopus and Dicrostonyx) in the Arctic breeding grounds (Bell 1965). At least 2 of the years when large numbers of Long-tailed and Pomarine Skuas were seen off Hound Point, 1991 and 1994, coincided with ‘lemming years’ (Stenseth et al 1998). However, it must be emphasised over 90% of birds seen in 1991 and 1994 were adults, and large numbers of the 2 species, including in 1992 when 790 Pomarine Skuas were counted, were present in years when lemming numbers were low. Furthermore, small counts of both Pomarine and Long-tailed Skuas were recorded at Hound Point since 1995, although a number of lemming years occurred during this time. Most of the skuas seen at Hound Point were observed entering the Firth of Forth from the east, moving in a westerly direction toward the Forth Railway Bridge. On many occasions, flocks were seen to gain height in front of the bridge, before continuing to fly in a south westerly direction overland. We are convinced that many skuas seen at Hound Point use an overland route between the Firth of Forth and the west coast of Scotland to 30 SB 27 return to the Atlantic Ocean. Such overland passage by Great Skuas has been proved in England on the Wash in Cambridgeshire (Easy 1990, 1993), and overland movements by skuas have been observed elsewhere in both North America and the Palearctic region (Olsen & Larsson 1997). Similar westward overland passage by Black-legged Kittiwakes has also been observed in the Firth of Forth at South Queensferry and Skinflats (Sandeman 1974, 1975, Griffin & McInerny 2001). We also observed a smaller number of skuas flying east past the point out of the Firth of Forth. Some of these were recognisable as individuals or groups that had previously arrived from the east. We believe that all birds seen moving east in autumn past Hound Point were leaving the Firth of Forth, having entered earlier from the North Sea and, instead of attempting overland passage, had opted to return to the North Sea. Numbers of skuas are seen flying north past east coast headlands elsewhere in Scotland and England in autumn, with some believed to re enter the North Atlantic via a route around the north of Scotland (Wallace & Bourne 1981, Innes 1993). For all species where age classes could be identified, adults predominated earlier in autumn than immatures. However, immature Pomarine and Long-tailed Skuas were both recorded in August, suggesting in some cases very rapid dispersal from the breeding areas. These early immatures were invariably migrating with adults, possibly their parents. In contrast, many later immatures were seen passing by themselves. The fact that 30-40% of adult Arctic Skuas were pale phase indicates that significant numbers of this species breeding in more northerly latitudes, where this colour phase predominates, pass through Scottish coastal waters in autumn. Acknowledgements We thank Rosebery Estates for allowing access to Hound Point and Ian Andrews for help with the graphics. References Andrews I J 1986. The Birds of the Lothians. SOC, Edinburgh. Andrews I J & Griffin M 1989. Influx of Long- tailed Skuas in Lothian, autumn 1988. Lothian Bird Report 1988: 79-81. Bell D G 1965. Studies of less familiar birds. 133. Long-tailed Skua. British Birds 58: 139-145. Brown A & Andrews I J 1986. Influx of Pomarine Skuas in autumn 1985. Lothian Bird Report 1985: 104-108. Critchley G 1972. A large movement of skuas in the Outer Hebrides. Scottish Birds 7: 205-206. Davenport D L 1979. Spring passage of skuas at Balranald, North Uist. Scottish Birds 10: 216- 220. Davenport D L 1987. Large passage of skuas off Balranald, North Uist in May 1986. Scottish Birds 14: 180-181. Davenport D L 1992. The spring passage of Long-tailed and Pomarine Skuas in Britain and Ireland. Birding World 5: 92-95. Dalgleish J J 1880. Pomarine Skuas in the Firth of Forth in the autumn of 1879. Proceedings Royal Physical Society of Edinburgh April 27, 4: 274. Dunn P J & Hirschfield E 1991. Long-tailed Skuas in Britain and Ireland in autumn 1988. British Birds 84: 121-136. Easy G M S 1990. Skua migration over Cambridgeshire. Cambridge Bird Club Report 41: 41-42. Easy GM S 1993. An update on skua movements inland from the Wash. Cambridge Bird Club Report 67: 85-87. Fox AD & Aspinall S J 1987. Pomarine Skuas in Britain and Ireland in autumn 1985. British Birds 80: 404-421. Scottish Birds (2007) Autumn skua migration at Hound Point, Lothian Sf Furness R W 1987. The Skuas. T & A D Poyser, London. Godfrey R 1899. Pomatorhine Skua in West Lothian. Annals of Scottish Natural History £8992 52, Gray R 1871. The Birds of the West of Scotland. Thomas Murray, Glasgow. Griffin M & McInerny C J 2001. Skua watching at Hound Point, Lothian. Birding Scotland 4: 2-7. Harrop H, Mellor M & Suddaby D 1993. Spring passage of Pomarine Skuas off Shetland in May 1992. Scottish Birds 17: 50-55. Harvie-Brown J A 1906. Richardson’s Skua at Kincardine-on-Forth. Annals of Scottish Natural History 1906: 118. Innes M 1993. The movement of Skuas off Peterhead, Grampian. North-east Scotland Bird Report 1992: 10-14. Lloyd C S, Tasker M L & K Partridge 1991. The Status of Seabirds in Britain and Ireland. T & A D Poyser, Calton. McGarry O & Speak P W 1992. Long-tailed Skua passage, autumn 1991. Lothian Bird Report 1991: 113-115. McInerny C J 2004. Sea-watching at Chanonry Point, Highland. Birding Scotland 7: 49-55. Mitchell PI, Newton S F, Ratcliffe N & Dunn T E 2004. Seabird Populations of Britain and Ireland. T & A D Poyser, London. Olsen K M & Larsson H 1997. Skuas and Jaegers. Pica Press, Sussex. Pennant, T. 1771. A Tour of Scotland in 1769. John Monk, Chester. Pollock C M, Mavor R, Weir C R, Reid A, White R W, Tasker M L, Webb A & Reid J B 2000. The Distribution of Seabirds and Marine Mammals in the Atlantic Frontier, North and West of Scotland. JNCC, Aberdeen. Rintoul L J & Baxter E V 1935. A Vertebrate Fauna of Forth. Oliver and Boyd, Edinburgh. Sandeman G L 1974. A large movement of Kittiwakes in the Forth. Scottish Birds 8: 77-78. Sandeman G L 1975. Another Kittiwake movement in the Firth of Forth. Scottish Birds 8: 324-325. Sangster G, Collinson J M, Helbig AJ, Knox AG & Parkin D T 2004. Taxonomic recommen- dations for British birds: second report. [bis 146: 153-157. Steele R D 1987. Large movement of skuas off Rattray Head. North-east Scotland Bird Report 1986: 50. Stenseth N, Chan K-S, Framstad E & Tong H 1998. Phase-and density-dependent population dynamics in Norwegian lemmings: interaction between deterministic and _— stochastic processes. Proceedings of the Royal Society of London B 265: 1957-1968. Stone C J, Webb A, Barton C, Ratcliffe N, Reed T C, Tasker M L, Camphuysen C J & Pienkowski M W 1995. An Atlas of Seabird Distribution in North-western European Waters. JNCC, Peterborough. Wallace DI M & Bourne W R P 1981. Seabird movements along the east coast of England. British Birds 74: 417-426. Webb A, Harrison N M, Leaper G M, Steele RM, Tasker M L & Pienkowski M W 1990. Seabird Distribution West of Britain. JNCC Report. Wourinen J D 1992. Do Arctic Skuas Stercocarius parasiticus exploit and follow terns during autumn migration? Ornis Fennica 69: 198-200. Christopher J McInerny, 10 Athole Gardens, Glasgow G12 9AZ C.MclInerny @bio.gla.ac.uk Mervyn Griffin, 9 Baberton Mains Lea, Edinburgh EH14 3HB mbgeriffin@ dunedin14.wanadoo.co.uk Revised manuscript accepted August 2006 32 SB 27 Numbers, distribution and habitats of Black Grouse leks in Lochaber AM MaCLENNAN & R W SUMMERS In spring 2000, a survey of lekking Black Grouse in 3 study areas in Lochaber (Glen Garry, Glen Spean and South Lochaber, comprising 933 km’) recorded 133 males at 43 lek sites. Thirteen of the leks had only single males, and over half of these were in the Glen Garry area. The average number of males per lek was highest in Glen Spean (4.4) and lowest in Glen Garry (1.6) with an overall average for the 3 study areas of 3.1males per lek. The average distance between nearest neighbour leks was 1.42 km. Most of the leks in the Glen Garry area were associated with Heather dominated clearings in young conifer plantations which were reaching thicket stage. Black Grouse in Glen Spean were found lekking on a mosaic of hill farmland, forest and moorland habitats on the north side of the glen. In South Lochaber, there were 2 Black Grouse groups isolated by almost 10 km of largely unsuitable habitat. Both groups were associated with Birch and Ash woodland. The risk of Black Grouse being killed on forest fences was high in all areas except around Kinlochleven, where there were no deer fences within 1.5 km of any of the lek sites. The average length of deer fence within a 1.5 km radius around the lek sites was 6.0 km (range 0 — 13.2 km). It is recommended that a Lochaber Black Grouse Group forms and that the landowners take steps to reduce possible collisions with fences. Introduction The Black Grouse Tetrao tetrix is a species of high and growing conservation concern (Anon 1999). Black Grouse have been in decline in many parts of Britain for over a century, but in the last 25 years, the rate of decline has accelerated (Hancock et al 1999). Despite their widespread distribution in Europe, their conser- vation status on the continent is equally unfavourable, apart from Sweden (Tucker & Heath 1994). Although the overall downward trends are clear, our knowledge and understanding of the status and dynamics of some local populations is poor. Systematic counts of Black Grouse males at leks in Britain were undertaken in spring 1995 and 1996 (Hancock ef al 1999). Although this survey gave a general picture of the status of the population nation wide, it did not provide sufficient local information on which to base conservation objectives and actions. For some areas, basic knowledge of Black Grouse numbers and distribution exists (Robinson et al 1993). However, Lochaber has not had a comprehensive survey. The first aim of this study was to describe the numbers and distribution of males at leks, and to describe associated habitat. Studies have highlighted the effect of fencing associated with woodland on a variety of birds (Catt et al 1994, Andrew & Baines 1997, Baines Scottish Birds (2007) & Summers 1997). Owls, passerines, pigeons and woodpeckers have all been found dead against fence lines, but grouse appear to be particularly vulnerable. Therefore, a secondary aim was to describe the amount of fencing close to lek sites, as a measure this potential hazard. Information on numbers, distribution, habitat and potential collision risk with fences can then be used as a baseline for future monitoring and targeting conservation action. Study areas The 3 study areas are shown in Figure |. The Glen Garry study area extended over 308 km° between Glen Moriston in_ the — north, Inchnacardoch Forest and the Great Glen in the east, Kingie in the west and the watershed of the Glen Garry deer forest in the south. Much of this area is moorland and conifer plantation, along with substantial areas of clear fell around the ancient pinewoods on the south shore of Loch Garry. The mountainous terrain along the southern margin and the northwest corners of the . Loch Laggan Glen Spean yt a ead Y Glen Nevis South Lochaber — Gr” Blackwater Reservoir z Figure 1. The study area in Lochaber showing the different sections that were surveyed. oa km Numbers, distribution and habitats of Black Grouse leks in Lochaber 33 study area were largely unsuitable for Black Grouse and were excluded from the survey. The 337 km? Glen Spean study area extended in a 5-10 km wide band from Loch Laggan in the east to Loch Eil in the west. Land use and habitats contrasted on either side of Glen Spean; extensive dense maturing conifer plantations occurred on the lower ground on the south side of the glen, while the south facing slopes comprised a mosaic of hill farmland, Heather Calluna_ vulgaris moorland and _ small commercial plantations. The 288 km” South Lochaber study area abutted the Glen Spean area at Fort William and extended south almost as far as Argyll. It was bounded by Loch Linnhe in the west, and Glen Nevis and the Blackwater Reservoir in the east. With the exception of the Mamore deer forest and the western end of the Aonach Eagach ridge, the terrain was made up of a mosaic of mature conifer plantations, Birch Betula spp woodland, hill farmland and grassy moorland. Field methods The survey methods were based on those used in the national survey in 1995-96 (Hancock ef al 1999). The 3 study areas, totalling 933 km’, were split into 5-km squares. The survey was carried out between late March and early May 2000. Surveyors walked to within 500 m of all suitable habitat during the early morning or evening, to locate display sites (leks). Areas deemed unsuitable for Black Grouse were ground above 550 m, dense conifer and broadleaved woodland, enclosed arable farmland and built-up areas (Hancock et al 1999). In addition, poor weather was avoided (Koivisto 1965, Baines 1996). Once located, surveyors revisited the leks within one hour either side of dawn to obtain peak counts of attending males (Cayford & Walker 1991). Where possible, counts were made from 34 vantage points to avoid disturbing the birds. However, there were a few sites in woodland where there was no suitable vantage point, and the birds were counted by flushing them. All locations where males were found displaying were defined as lek sites. This is contrary to the guidelines followed in other Black Grouse surveys, where single lekking birds were not classed as leks (Robinson et al 1993, Hancock er al 1999). Displaying males separated by 300 m or less were counted as belonging to the same lek site. Habitat assessments were made within a 200 m radius of the leks. The following categories were recognised: grass moor, Heather moor, field/rough grazing/improved grass, conifer plantation and Birch or Birch/Ash Fraxinus excelsior native woodland. Categories were assigned as primary or secondary depending on their relative abundance within the 200 m radius. Deer fences within a radius of 1.5 km from the lek sites were noted in the field and measured from 1:25.000 maps. It is the radius recommended by Cayford (1993), within which management Table 1. Black Grouse lek statistics in Lochaber. SB 27 improvements for Black Grouse should be targeted. Therefore, birds within this radius were assumed to be at risk from collisions. Results Glen Garry Sixteen lek sites were identified and 26 males counted (Table 1, Figure. 1). One lek site (GG4) extended over a broad area of hillside, with displaying birds about 200 m apart. At GG7, birds appeared to use an alternative site on different days. The average number of males at a lek was 1.6, with half the sites holding only two males. One site had three males and the remaining leks had only one male each. Ignoring leks which were nearer the boundary of the study area than their nearest neighbour, the average distance between neighbouring leks was 1,260 m and the closest leks were separated by only 400 m. The length of deer fence within a 1.5 km radius of leks varied from 2.94 km (GG4) to 13.2 km (GG14), and the average for the 16 leks was 7.5 km (Table 1). Proximity of deer fences to leks ranged from 50 m (GGS) to 575 m (GG12), and the average was 290 m. Glen Garry Glen Spean South Lochaber All sites Number of leks 16 17 10 43 Maximum number of males per lek 3 12 10 12 Total number of males 26 74 33 [33 Maximum number of females per lek 4 2 + Total number of females > 15 4 24 Average number of males 1.6 44 333 3.1 Average distance (m) between leks * 1260 2130 870 1420 Total length (km) deer fence within 1195 107.9 Sie 258.5 1.5 km radius of the leks Average length (km) deer fence in VR) 6.3 3.1 6.0 1.5 km radius * These figures exclude leks close to edge of the survey area that may have biased results. Scottish Birds (2007) Numbers, distribution and habitats of Black Grouse leks in Lochaber 3) Fifteen out of the 16 leks were associated with varying amounts of Heather moorland and conifer plantation (Table 2). Although Heather moorland was the most common primary habitat in the immediate vicinity of the leks, most of this occurred in clearings or rides within extensive conifer plantations. Glen Spean Seventy four males were recorded at 17 leks (Table 1, Figure. 1). The number of males counted on a lek ranged from one (at 3 sites) to 12 (one site), with an average of 4.4 males per lek. The nearest neighbouring leks were separated by 500 m, and the average distance between neighbouring leks within the study area was 2,130 m. This calculation ignored 5 lek sites (GS1, GS2, GS3, GS6 and GS10), which were closer to the edge of the study block than the nearest neighbours within it. The length of deer fence within a 1.5 km radius of lek sites ranged from 0.96 km to 9.75 km and the average was 6.3 km (Table 1). The closest deer fence to a lek site was 30 m (GS13), although lek GS2 was only 10 m away from a stock fence. Thirteen of the 17 lek sites recorded were located in Glen Spean itself and 10 of those were on the north side of the glen or the valley bottom. This distribution pattern reflects the availability of suitable habitat: a mosaic of conifer plantation, Heather moor, grass moor and small fields or rough grazing (Table 2). There was a tendency for the grassy habitats to dominate the vegetation on the lower slopes and on the valley bottom where the main concentrations of sheep were found. Most of the lek sites were located on grass moors, with conifer plantations or Heather making up the secondary habitat. The young, open plantations, such as the one close to GS7 supported a lush dwarf shrub layer, and appeared to be used by birds, particularly before and after visiting the adjacent lek site (J Gordon pers comm). In contrast, the extensive dense conifer plantations on the south side of Glen Spean occupied all the lower, more fertile and sheltered slopes. Apart from birds attending leks GS13 (one male) and GS17 (2 males), there were few Black Grouse associated with this habitat. Most of the open ground available on the south side of the glen was steep, exposed deer forest and was probably unattractive to Black Grouse. South Lochaber Thirty three males occurred at 10 leks in 2 distinct geographical areas (Table 1, Fig. 1). A cluster of 7 leks was recorded between Blarmafoldeach and Cow Hill, south of Fort William. The remaining 3 were between the Table 2. The number of Black Grouse leks associated with different primary and secondary habitat categories in Lochaber. Glen Garry Glen Spean Prim Second Prim Second Wood 0 0 0 0 Field 0 0 3 | Heather 10 5) 4 5 Grass ] 5 10 0 Plant 5 8 0 11 South Lochaber Prim Second All areas Prim Second 3 6 S) 6 | 0 + l 5) | 19 11 l 3 12 6 0 0 5 19 Key to habitat categories: Wood - Birch/Ash and Ash native woodland; Field - Field/rough grazing/improved grassland; Heather - Heather moor; Grass - Grass moor; Plant - Conifer plantation. 36 SB 27 shore of Loch Leven and the Leven Valley. Black Grouse at SL4 appeared to use alternative fields on different days. The frequency distri- bution of numbers of birds at leks was skewed, with 10 birds at one site but only | - 4 birds at the remaining 9. Therefore, the average of 3.3 males per lek is somewhat misleading. Three lek sites held only single males. Although the SL6 site appeared to be favoured by displaying birds, there was considerable interchange of males within the cluster of sites immediately south of Fort William. Due to the grouping of lek sites, the closest were only 500 m apart and the average distance between sites was 870 m. This calculation excludes SL1, which was over 6 km from the nearest lek site. Three of the sites (those around Kinlochleven) did not have any deer fences within a 1.5 km radius. With these sites included in the calculation of the average length of deer fence within a 1.5 km radius of leks, the result (3.1 km) is less than half of that recorded in the Glen Garry (7.5 km) and Glen Spean (6.3 km) study areas (Table 1). The Kinlochleven sites were predominantly associated with Birch woodland. Grass moor and Heather moor habitats were of secondary importance (Table 2). In contrast, the primary habitat in the northern group of sites was largely Heather moorland, with Birch and Birch/Ash woodlands of secondary importance. Although the risk from fence collision may be less in South Lochaber compared to the other study sites, the risk of collision from overhead cables was more of an issue here. Pylons carrying high voltage cables run through a gully separating lek SL6 from leks SLY and SL10. One fresh Black Grouse corpse was found beneath these cables during the survey. Black Grouse collisions with electricity transmission wires have also been noted in Norway (Bevanger 1995). One additional factor influencing Black Grouse displays at SL9, SL10 and to a lesser degree at SL6, was disturbance from early morning joggers and dog walkers from Fort William. During the periods of observation, birds were regularly flushed from the leks (T Millard pers com). Disturbance may also compound the risk of collision with the overhead cables. Discussion Numbers and distribution The Glen Garry Black Grouse population was fragmented and lek sites were almost entirely associated with patches of Heather in rides and clearings within conifer plantations. Several of these plantations were young, pre thicket woodlands, but much of the area was at the thicket stage and becoming increasingly unattractive for Black Grouse. It is now recognised that fragmented groups of Black Grouse tend to have lower genetic diversity due to inbreeding (Héglund et al 2003), so steps should be taken to connect neighbouring groups through habitat management. By contrast, the Black Grouse population in Glen Spean was less fragmented than in Glen Garry and the birds appeared to favour the mosaic of habitats on the north side of the glen. The high number of birds at GS16 in the west of this area was likely to reflect the favourable conditions for Black Grouse on the Creag Meagaidh reserve to the east of the study area. The South Lochaber Black Grouse population was localised and split into 2 groups, isolated by a large area of unsuitable habitat. Disturbance is an issue regarding sites close to Fort William and may have contributed to the apparent decline in the number of birds attending the leks close to houses (J Rowbottom pers com). With limited movement of birds and dwindling numbers around Kinlochleven (only 7 males recorded) the risk of losing this group is acute. Scottish Birds (2007) This was the first comprehensive study of Black Grouse in Lochaber. It provides baseline data in the three study areas that comprise the bulk of the Lochaber population. In order to retain this population and guide ongoing conservation management, it is important to continue monitoring numbers at regular intervals. It would be useful to initiate formation of a Lochaber Black Grouse Study Group, similar to that in Tayside (Robinson ef al 1993), to continue monitoring. Habitat The lek sites were predominantly located on Heather moorland with conifer plantations forming the most common secondary habitat (Table 2). This association was most clearly seen in the Glen Garry study area. In Glen Spean, grass moors formed the dominant primary habitat in association with secondary habitats of conifer plantations or Heather. The 2 groups of leks in south Lochaber showed a stronger association with semi natural native woodland and moorland. Although there are clear differences in the proportions of the various habitats being used by, and available to the Black Grouse in each of the study areas, it was beyond the scope of this project to examine the quality of these habitats. A condition assessment of the various habitats of importance to the Black Grouse and examination of the changes in habitats in the various areas over the past 25 years might shed more light on the apparent changes in numbers. Habitat management Landowners and forestry agencies should make better provision for Black Grouse in new and existing woodlands by adopting the design and management recommendations that are known to benefit this species (Cayford 1993). These include maintaining significant amounts of open Space with ericaceous vegetation and damp, boggy ground, linked by wide rides and riparian zones in new conifer plantations, maximising Numbers, distribution and habitats of Black Grouse leks in Lochaber S7. forest edges and providing a broad (at least 200 m) graded interface adjacent to moorland habitats. These can be achieved by leaving areas unplanted, using variable spacing of trees and using species such as Scots Pine Pinus sylvestris, Larch Larix sp and Birch (Cayford 1993). Management measures for Black Grouse can be adopted at the design, restructuring or clear felling stages of forestry management or in a continuous cover forestry system. The average length of deer fence within a 1.5 km radius of surveyed lek sites was 6.0 km (Table 1), although the figures varied enormously from 0 in the case of 3 South Lochaber sites to 13.2 km at one site in the Glen Garry study area. The shortest distance from a deer fence to a display site was 10 m and may on occasions have separated displaying birds. Twenty three percent of surveyed sites were located 100 m or less from a deer fence, several of which were recently constructed. Only one of those fences in Glen Garry was reported to have been colour marked to enhance its visibility to woodland grouse. Baines & Summers (1997) demonstrated that collision rates for different grouse species vary seasonally, with 2/,; of Black Grouse collisions occurring between February and May. This coincides with the period when birds are most active on lek sites. It is likely that certain lengths of fence claim more lives than others, according to, for example, their visibility, type of wire mesh and/or the length of time they have been in place. However, with possible collision rates of 0.7 Black Grouse per kilometre of fence per year (Baines & Summers 1997), and the lengths of fences in the vicinity of lek sites, it is likely that fences will account for a significant proportion of the annual mortality. Although fence strikes may be contributing to the decline of this species, they are not solely responsible for the changes in numbers. The 3 lek 38 sites around Kinlochleven did not have any fences within a 1.5 km radius and were associated with a higher proportion of semi natural habitats. These leks were attended by few Black Grouse in 2000, but anecdotal information suggests that numbers were much higher only a few years previously (T Piggot pers com). There are now a number of reports that make recommendations with regard to reducing fence collisions (Goddard et al 2001, Summers & Dugan 2001). These include recommending that the Deer Commission for Scotland (DCS) should adopt a more strategic approach to deer fencing, particularly in consultations regarding woodland grant schemes and in the development of local deer management plans through the individual Deer Management Groups. In addition, the Forestry Commission Scotland should continue to rationalise use of deer fences in Black Grouse areas, particularly where they merely mark ownership boundaries in woodland. Posts without wire could be as effective in boundary marking. Landowners should ensure that where Black Grouse occur new fences associated with woodland schemes or existing fences should be marked to reduce the number of bird collisions. Further, removal of fencing after it has served its useful purpose should be made a condition of Woodland Grants where its purpose is to temporarily ensure woodland establishment. Where it is erected for other purposes, it should become the responsibility of individual landowners to remove deer fences after they have served their purpose. In addition, steps should be made to mark the electric cables that pass through the cluster of leks south of Fort William. SB 27 Disturbance Given the potential human disturbance at the leks in the Fort William area, the community at Achintore should be requested to use alternative routes for walking and jogging in the early morning during the main time for lekking by Black Grouse (April - early May). Distribution of leaflets to occupants in the adjacent housing estate explaining the situation may be sufficient to alleviate the problem. Acknowledgements This project was funded by RSPB and Scottish Natural Heritage, with financial contributions from Forest Enterprise and Alcan Smelting and Power Ltd. Thanks go to James Gordon, Duncan MacNicol and Tony Millard for their dedicated work in the field. We would also like to take this opportunity to thank the numerous landowners, land managers, estate staff, farmers, crofters, foresters and rangers who permitted access to their land and provided local information. Chris Donald and Murray Grant commented on the draft. References Andrew M & Baines D 1997. The impact of deer fences on woodland grouse and other birds. Game Conservancy Report. Anon 1999. UK Biodiversity Group. Tranche 2 Action Plans. Vol. VI — terrestrial and freshwater species and habitats. English Nature, Peterborough. Baines D 1996. The implications of grazing and predator management on the habitats and breeding success of Black Grouse Tetrao tetrix. Journal Applied Ecology 33: 54 - 62. Baines D & Summers RW 1997. Assessment of bird collisions with deer fences in Scottish forests. Journal Applied Ecology. 34: 941 - 948. Bevanger K 1995. Estimates and population consequences of tetraonid mortality caused by collisions with high tension power lines in Norway. Journal Applied Ecology 32: 745 - 753. Scottish Birds (2007) Catt DC, Dugan D, Green RE, Moncrieff R, Moss R, Picozzi N, Summers RW & Tyler GA 1994. Collisions against fences by woodland grouse in Scotland. Forestry 67: 105 - 118. Cayford JT 1993. Black Grouse and forestry: Habitat requirements and management. Forestry Commission Technical Paper 1. Forestry Commission, Edinburgh. Cayford JT & Walker F 1991. Counts of male Black Grouse Tetrao tetrix in North Wales. Bird Study 38: 80-86. Goddard PJ, Summers RW, Macdonald AJ, Murray C & Fawcett AR 2001. Behavioural responses of Red Deer to fences of five different designs. Applied Animal Behaviour Science 73: 289-298. Hancock M, Baines D, Gibbons D, Etheridge B & Shepherd M 1999. Status of male Black Grouse Tetrao tetrix in Britain in 1995-96. Bird Study 46: 1-15. Hoéglund J, Baines D, Larsson JK & Segelbacher G 2003. Population fragmentation and genetic variability in European Black Grouse — a progress report. Sylvia 39 (suppl): 17-23. Numbers, distribution and habitats of Black Grouse leks in Lochaber 2) Koivisto I 1965. Behaviour of the Black Grouse, Lyrurus tetrix (L.), during the spring display. Finnish Game Research 26: | - 60. Robinson MC, Baines D & Mattingley W 1993. A survey of Black Grouse leks in Perthshire. Scottish Birds 17: 20-26. Summers RW & Dugan, D 2001. An assessment of methods used to mark fences to reduce bird collisions in pinewoods. Scottish Forestry 55: 23-29. Tucker GM & Heath MF 1994. Birds in Europe: their conservation BirdLife International, Cambridge. Status. Alison MacLennan, Royal Society for the Protection of Birds, Sutherlands, Broadford, Isle of Skye, 1V49 9AB Ron Summers, Royal Society for the Protection of Birds, Etive House, Beechwood Park, Invervess, 1V2 3BW Revised manuscript accepted May 2006 40) SB 27 An unsuccessful attempt to establish Northern Goshawks in Argyll STEVE J PETTY Details are provided of an unsuccessful attempt to establish a breeding population of Northern Goshawks in Argyll with 11 birds imported from northern Europe. Birds were released in 2 batches, during 1969-1973 (8 birds) and in July-August 1980 (3 birds). One successful breeding attempt was reported following each batch of releases, in 1973 and 1982-83, and although apparently genuine, with the passing of time the validity of these records cannot be confirmed. Reasons why Goshawks failed to establish a self sustaining population in Argyll are discussed. A lack of suitable food is considered to be the most likely cause, unlike elsewhere in Scotland where persecution is the main factor restricting population growth. Introduction Northern Goshawks Accipiter gentilis are expanding their range in Britain following unregulated releases of birds by hawk keepers in the latter half of the 20th century (Marquiss & Newton 1982, Petty, 1996a). These releases started in the 1950s, but were few in number, with birds coming mainly from central Europe. By the 1970s and 1980s, more were being imported into Britain, especially from Scandinavia, where many birds were taken as nestlings or trapped annually around Pheasant Phasianus colchicus release pens. On their arrival in Britain, some birds were intentionally released into the wild to establish breeding populations and others escaped while being flown by falconers. Kenward (1981) noted that from 1970 to 1975, 549 Goshawks were imported under licence into the UK, and during 1970-78, 25 birds a year were estimated to have entered the wild as the result of deliberate releases or escapes. The success of this clandestine attempt at reestablishment was due largely to the method employed. Small groups of birds were released in numerous, widely scattered locations over a short period of time, which largely overcame the problem of extinction due to random events. However, some released birds failed to establish viable populations, but such failures have received little attention in the literature. Even where establishment was successful, there was no site specific information provided about numbers, ages, sexes and dates when birds were released, although early breeding records can be used to estimate the general area where birds were liberated. The present population in Scotland appears to have originated mainly from birds released in the north east (NES) and in the Scottish/English Borders (SEB) (Petty in press). The first recorded breeding attempts in Scotland occurred in SEB in 1972 and in NES in 1973, indicating birds had been released a year or 2 before these events. Subsequently, populations in both areas have increased in numbers and range, although population growth rates in the 2 areas differ, probably due to different levels of persecution (Marquiss et al 2003). This investigation started during preparation of the Goshawk account for BS3. In this paper I review the status of Goshawks in Argyll, drawing on previously unpublished information, to document Scottish Birds (2007) An unsuccessful attempt to establish Northern Goshawks in Argyll 4] an apparently unsuccessful release of birds at around the same time that the SEB and NES population were established. Although successful breeding was reported on 2 occasions, these events did not generate a self sustaining population in Argyll as occurred elsewhere in Scotland. The reasons for this failure are discussed. Methods The data used to piece together the recent history of Goshawks in Argyll were obtained from a number of sources; unpublished records from friends and contacts, literature searches/references and records from the Argyll Bird Database, managed by the Argyll Bird Recorder (see acknowledgements). Results Release of birds The first batch, consisting of 5 males and 3 females, was released near the head of Loch Sween in Knapdale Forest, Mid Argyll between December 1969 and 1973 (Table 1). The 1972 birds were imported from Finland after being taken from the nest as large chicks. Only 2 were successfully released, although 12 birds had been imported. Ten of these died possibly due to lack of food during transit from Finland and adverse weather in Argyll (D Kent pers comm). They were placed on man made platforms in trees at various locations in the forest and fed daily. They were able to pull food apart themselves, but there were several days of heavy rain during which the birds chilled. They were taken back into captivity, but only 2 survived. Later, these were returned to one artificial nest and fed. They subsequently fledged, learned to hunt and were noted to be in very good shape. They were seen well into August 1972 and possibly September, by which time they had become very unapproachable. All the liberated first winter and adults birds (Table 1) were in pristine condition and were fed well for a period before release. This project was conducted in tandem with similar numbers being released on Speyside through Doug Weir (cross fostering with Common Buzzards Buteo buteo) and in South Scotland (cross fostering with Eurasian Sparrowhawks Accipiter nisus, David Kent pers comm, Kenward 2006). A further batch of 3 birds was released in the same area in July/August 1980 (Table 1). David Kent was not aware of any other releases of Goshawks in Argyll before or after those above (Table |), and was doubtful that any would have been released without his knowledge, at least in the 1970s and 1980s. Lost falconers’ birds from the Glasgow area could conceivably have found their way into Argyll and account for some earlier sightings in Argyll. Possible breeding records The first recorded breeding attempt was in 1973 (Rainier 1975). It is worth quoting this account in full: ‘A nest with young was found on 7th August 1973 in a Forestry Commission Plantation within the area (Mid Argyll); the nest was watched daily until 12th August when the young birds were seen in flight. Although the site has remained undisturbed the birds have not returned. Since then Goshawks have been seen as follows: One on 23rd March and 13th April 1974; 2 on 9th March 1975 and single birds on 27th March, 12th April, 23rd August and 21st November 1975. The possibility that the nesting record and subsequent sightings had their origins in falconers’ escapes must be born in mind.’ In a further paper, Rainier (1983) mentioned some additional records of Goshawks in Mid Argyll: ‘No further breeding records known, but several additional sightings of single birds, as follows: Achnamara 18th January 1976 and 15th October 1976, Lochgilphead 19th October 1977, 42 Achnamara 12th February 1978, Minard 13th May 1979, and Tarbert 20th October 1979.’ Unfortunately, the author of these papers died a few years ago, and it has not been possible to trace the original source of her records. Neither is the actual location of the breeding attempt given, but all sightings in the second paper were within 25km radius of the original release site, suggesting that the breeding attempt was in this area too. The second reported breeding attempt was in early 1980s, possibly in 1982-83. The nest was in a larch tree, near Lochgilphead in Mid Argyll and contained at least 3 large nestlings. The nest was approximately 8km east of the release site. Matthew Wilson, who reported this event, was familiar with Goshawks, having flown falconry birds of this species, and also Sparrowhawks. SB 27 Neither this nor the 1973 breeding record have been submitted to the Argyll Bird Records Committee (ABRC). Discussion Prior to 1970, there had been few records of Goshawks in Argyll. There were no historical breeding records for Argyll (ap Rheinallt in press), although they may have bred prior to widespread forest clearances in the Middle Ages. The only apparently genuine record in the 19th century was at Glenforsa on Mull on 9 November 1892, although no description of the bird was given (Buckley 1893). Three reports, of 4 birds, from the late 1850s into the 1860s in Gray (1871) were not mentioned by Harvie- Brown & Buckley (1892), and were thus probably unreliable, as were some of the other Goshawk records mentioned by Gray. Table 1. Details of goshawks released in Knapdale Forest’, Argyll during 1969-1980. (David Kent pers comm) Date Sex Age Origin’ Other information Dec 1969 male first year” Norway Jan or Feb 1970 male first year” Norway March 1970 male adult? Germany Shot in Achnamara in 1970 after killing a chicken! May 1970 female first year” Norway June 1972 male large nestling —- Finland Food provided for a period after release June 1972 female large nestling = Finland Food provided for a period after release 1O72773 male not known not known 1O72/73 female 5-6 years old? Sweden Imported in 1967 as first year bird July/August 1980 male juvenile Norway July/August 1980 female juvenile Norway July/August 1980 female juvenile Norway ' All birds were liberated either by the Fairy Isles (NR7688) or close to Loch Coille-Bharr (NR7890). > Trapped at game rearing establishments, 2 birds had spent brief periods in the care of falconers. Scottish Birds (2007) The first record and description of the 20th century was not until 1959 when an adult was seen at Loch Ballygrant, Islay on 28 April (Greenwood 1960). This was followed by a first year bird on 15 September 1963, near the Mull of Kintyre lighthouse (Gordon & Merrie 1964). Both latter records occurred before the release of birds in Mid Argyll (Table 1), but were likely to be of imported origin. At this time, there was no evidence of Goshawks breeding anywhere in Scotland and true migrants from northern Europe were very scarce. For example, Petty (in press) showed that over a 143 year period, only 34 Goshawks had been recorded on Fair Isle, Outer Hebrides, Orkney and Shetland, and some of these might not have been genuine migrants; or even Goshawks, as unfortunately, much confusion still exists over identification. The 2 claimed breeding attempts (1973 and 1982-83) occurred soon after each batch of birds (1969-73 and 1980) had been released (Figure 1). Goshawks can breed as yearlings, but are more likely to breed for the first time at 2-3 years of age (S J Petty unpublished). Thus, both breeding attempts were at a time when they would have been expected. In Goshawks, adult males appear to stay within their home range for most of the year in contrast to females, which female @ male w Number of birds dho4: 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 Year Figure 1. Number of goshawks released in Argyll and years when breeding was reported (each arrow is one record) An unsuccessful attempt to establish Northern Goshawks in Argyll 43 tend to be more mobile outside the breeding season (Petty 2002). In the first batch of releases, males had been liberated a few years before the first female, thus allowing them time to establish home ranges before females were released. In fact, a 5-6 year old female had been released just before the first breeding attempt in 1973. The second batch, of 3 fledged juveniles, was released in 1980 followed by a breeding attempt 2-3 years later, again at a time when it would have been expected, even if no birds had survived from the initial breeding attempt and releases in the early 1970s. More information is available about the first breeding attempt than about the second, although the brood size was not recorded (Rainier 1975). The chicks fledged sometime during 7-12 August, which meant the first egg must have been laid in mid May. In the ESB population this date falls towards the end of the recorded laying period (SJ Petty unpublished), but bearing in mind the generally poorer habitat in Argyll and the likelihood of a pair breeding for the first time, this date is certainly feasible. It is too late for Sparrowhawk chicks to be fledging (Newton 1986). There is less information available on the second breeding attempt, apart from at least 3 large chicks being recorded in the nest. The modal successful brood in the SEB population was 3 chicks (Petty & Anderson 1996). This second breeding attempt was observed by a person with experience in flying falconry Goshawks, who was unlikely to misidentify the birds. Why did these apparently successful breeding attempts fail to lead to Goshawks becoming established in Argyll? In both NES and SEB, the number of breeding pairs started to increase within a year to 2 of the first broods being successfully reared (Marquiss et al 2003, Petty in press). Persecution is widely perceived to be the main factor limiting population growth in Scotland (Marquiss et al 2003), but this was 44 unlikely to be the main cause of failure in Argyll, as persecution levels are much lower than in either SEB or NES. Most illegal killing of Goshawks occurs around Pheasant release pens, and while Pheasants are reared for shooting in Argyll, this is on a far smaller scale than in eastern and southern Scotland. Food shortage is the most likely reason why Goshawks failed to colonise the forests of Argyll. Peregrines Falco peregrinus have a considerable dietary overlap with Goshawks (Petty 1996a). Thus, the breeding performance of Peregrines may reflect to some extent the suitability of habitats for Goshawks, although Goshawk are more versatile than Peregrines, being able to take medium sized mammals, such as Rabbits Oryctolagus cuniculus and Red Squirrels Sciurus vulgaris, as well as birds (Marquiss & Newton 1982; Petty et al 2003). In the national Peregrine survey of 1991, mean brood size per territory holding pair in Argyll was 0.92 for occupied inland territories compared to 1.64 for occupied coastal territories (Crick & Ratcliffe 1995). Inland territories would be closer to the forested parts of Argyll that Goshawks would most likely occupy. Furthermore, the food supply for Peregrines may be worsening, as the latest national survey in 2002 revealed a 38% decline in the number of breeding pairs in north west Scotland including Argyll since 1991; a greater decline than anywhere else in the UK (Crick er al 2003). It is hard to imagine any factor, other than food, being linked to such a widespread deterioration in the performance of Peregrine. Moreover, winter food supply may be even more critical. At this time of the year, little suitable bird prey remains in and around upland forests in mainland Argyll. Thus, Goshawks would have to move to survive the winter, either to the coast or to prey rich locations on the larger islands, such as Islay and Mull. From 1970 until 1996 there were 77 records ‘of Goshawks submitted to the Argyll Bird Recorder. Inevitably, these will include an unknown proportion of misidentified Sparrowhawks, but it is SB 27 interesting to note that 59% of records came from Islay and Mull and just 27% from Mid Argyll. Interestingly, the latest accepted record of an adult male, in March 2006, was within 2km of where birds were released (Table 1). What are the prospects for the future? A few pairs breed to the north of Argyll, in Highland Region, but numbers are small, breeding success is low and the population may have declined over recent years, again apparently reflecting prey scarcity. Therefore, few if any recruits are likely to enter Argyll from the north. The largest breeding population is in southern Scotland with a few pairs extending into south Ayrshire (Petty in press), but still a substantial distance from Argyll. In addition, a few pairs now breed in Central Scotland. Recent reports of birds breeding on Arran (Arran and Clyde Islands Bird Reports), just a few kilometres from the Argyll mainland, are incorrect (SJ Petty unpublished). Thus, recent sightings of Goshawks in Argyll are most likely to originate from south/central Scotland, unless some progeny survive from the original breeding attempts. The other possibility is that some birds have continued to breed in Argyll since the initial reports of breeding, but this is unlikely. Goshawks often use the same restricted area of forest for breeding, and over a period of years are unlikely to avoid detection, particularly by forestry workers (Petty 1996b). Thus, the chance of recruiting enough birds to establish a viable population in Argyll from population elsewhere in Scotland would, at present, appear to be slim. Acknowledgements I thank David Kent for details of Goshawks he released in Argyll, Matthew Wilson for information about one breeding attempt and Paul Daw, Argyll Bird Recorder, for records of Goshawk sightings in Argyll. Helpful comments on a first draft were received from Paul Daw, Jim Dickson David Kent, Peter Kirk and Stan da Prato. Scottish Birds (2007) References ap Rheinallt T in press. Northern Goshawk Accipiter gentilis. In Birds of Argyll (eds) T ap Rheinallt, J C A Craik, PC Daw, R W Furness, S J Petty. & D Wood). Argyll Bird Club, Lochgilphead. Buckley T E 1893. Goshawk in Mull. Annals of Scottish Natural History 5: 45. Crick H, Banks A & Coombes R 2003. Findings of the national Peregrine survey 2002. BTO News 248: 8-9. Crick H Q P & Ratcliffe D A 1995. The Peregrine (Falco peregrinus) breeding population of the United Kingdom in 1991. Bird Study 42: 1-19. Gordon, AG & Merrie, TD H 1964. Goshawk in north and south Argyll. Scottish Birds 3: 27. Gray R 1871. Birds of the West of Scotland, including the Outer Hebrides. Thomas Murray & Son, Scotland. Greenwood S 1960. Goshawks in Islay. Scottish Birds 1: 186. Harvie-Brown J A & Buckley T E 1892. A Vertebrate Fauna of Argyll and the Inner Hebrides. David Douglas, Edinburgh. Kenward R 2006. The Goshawk. T & AD Poyser, London. Kenward R E 1981. Goshawk re-establishment in Britain: causes and implications. Falconer 7: 304-310. Marquiss M & Newton I 1982. The Goshawk in Britain. British Birds 75: 243-260. Marquiss M, Petty S J, Anderson DI K & Legge G 2003. Contrasting population trends of the Northern Goshawk (Accipiter gentilis) in the Scottish/English Borders and_ north-east Scotland. In Birds of prey in a changing environment. (eds) D B A Thompson, S M Redpath, A H Fielding, M Marquiss & C A Galbraith, pp. 143-148. The Stationery Office, Edinburgh. Newton I 1986. The Sparrowhawk. T & A D Poyser, Berkhamsted. Petty S J 1996a. History of the Northern An unsuccessful attempt to establish Northern Goshawks in Argyll 45 Goshawk ! 85,7 Op” Gp? 796, 796,78550 SOF OOP 95 SO 8580 O57 O55 SO; ES S52 892 95 950 7 Og FT) DOSS SoH EYY PG PPG O10) 01 YOY 054 0G, ILI Figure 6. Mean number of nestlings in white- tailed eagle nests on the Swedish Baltic coast over time (no of broods in brackets). The reference line and shaded area represent mean brood size up to 1950, with 95% confidence limits (Helander 2003b). The sea eagle in environmental monitoring The national Environment Monitoring Programme (EMP), under the Swedish Environment Protection Agency, includes the monitoring of selected organisms and parameters in a range of environments. The White-tailed Eagle was adopted into this programme in 1989 as an indicator for effects from pollutants. The history behind this is that the eagle was the first species to signal the alarming state of the Baltic Sea, by showing a dramatic decrease in productivity that was linked to contaminants, especially DDE and PCB (Helander er a/ 2002). Figure 6 illustrates the changes over time in brood size (number of nestlings per successful pair). There was a significant drop in the brood size already in the first half of the 1950s, but this was not really noticed until about 10 years later. The Swedish Museum of Natural History is responsible for the monitoring of the eagle under the EMP. The White-tailed Eagle also has features that are favourable in a monitoring perspective: mates of pairs are faithful to each other and to their breeding sites; breeding territories are generally SB 27 used over many generations of eagles providing good opportunities for long trend studies; territorial adults on the coast are mainly sedentary and thus reflect the regional contaminant situation; and a large proportion of the breeders are ringed, helping the study of individual birds over time. The monitoring of reproduction focuses on 2 robust parameters: @ the annual proportion of reproducing pairs @ the distribution of brood sizes, based on >4 weeks old nestlings All dead eggs, shell fragments and shed feathers of the adult birds are collected, and nestlings are weighed, measured (wing chord, tarsus width and depth), colour ringed and sampled for blood and feathers (upper wing coverts). So far, all collected eggs have been analysed on a regular basis for organochlorine contaminants; the blood samples have been saved for special research projects (eg Olsson et al 2000, Hailer 2006). Contaminants and reproduction The most striking thing we have seen in the White-tailed Eagle was the phenomenon of ‘dry’ eggs. In the 1960s — 1980s, most dead eggs that were retrieved from nests on the Baltic coast were strongly desiccated, whereas this was very rarely the case in Lapland. Figure 7 illustrates the relationship between desiccation index (Di) and DDE in eagle eggs from the Swedish Baltic coast. Di is calculated as weight divided by volume of the individual egg; a fresh egg would have a Di of near 1.0. There is a slow natural weight loss of eggs from evaporation through the shell (Rahn et al 1979). The weight of most of the eagle eggs collected in our studies was determined around 100 days after they were laid. Eggs with normally functioning shells would then have expected Di-values exceeding 0.7. In Figure 7, a dramatic effect on shell functionality is indicated as the DDE-levels exceed 170 ug/g and at concentrations exceeding 470 there was not a single ‘normal ‘egg. Scottish Birds (2007) Desiccation index 25 50 100 2505001000 3000 log(DDE) Figure 7. Desiccation index (Di) in relation to DDE (g/g, lipid weight) in 126 eagle eggs from the Swedish Baltic coast 1969-1997. From Helander et al (2002). Remaining effects from contamination When testing productivity of the Baltic population versus concentrations of DDE, PCB and shell parameters, the strongest correlation was with Di. There was no significant correlation between productivity and shell thickness or index. All 3 shell parameters were most strongly correlated with DDE. Some females were studied over many years after the bans of DDE and PCB were enforced in the 1970s. Although residue concentrations decreased strongly in their eggs during the 1980s, their ability to produce functioning shells did not improve and produc- tivity remained at zero. This indicates remaining effects from the previously much higher contam- ination of those old females. The presence of eggs from such females in statistical analyses of relationships between productivity versus contaminants and eggshell parameters would confuse the results and was probably contributing to the observed stronger correlation to produc- tivity for Di than for DDE, the suggested cause for the desiccation syndrome. The White-tailed Eagle in Sweden - past, present and future B))| mean productivity T T scot T T oO O—-@ 25 50 100 250 500 1000 DDE Figure 8. Mean 5-year productivity of 82 individual females in relation to DDE concen- trations (g/g lipid weight) in their eggs, grouped into 14 concentration intervals. From Helander et al (2002). Effect levels Our studies have shown strong relationships between productivity (number of young produced per pair and year) and DDE and PCB. Figure 8 shows a suggested dose response relationship for DDE, indicating normal produc- tivity (>1.0) at concentrations up to about 100 u¢/g in the egg lipids, a reduction to half at 200 ug/g, and no reproduction at >900 ug/g (for conversion to a wet weight basis, multiply by 0.05). Based on individual data, a lowest observable effect level (LOEL) of about 120 is indicated for DDE. There is a strong correlation between concentrations of PCB and DDE in the eggs. But DDE has decreased faster than PCB since the 1970s, so the possibility to separate between effects from these major contaminants has improved with time. In the late 1980s and 90s, PCB but not DDE concentrations were significantly higher in eggs with dead embryo as compared to infertile eggs, and a LOEL of 500 g/g for embryo mortality was implied for PCB. In the 1970s, the range of PCB concen- trations in eagle eggs from the Baltic coast was 52 - << = \ Ae 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 Figure 9. Successful breeding attempts on the Swedish Baltic coast (%). Pre 1954 reference line with 95% confidence limits (shaded) from Helander (2003b) 320 | 280 240 | 200 | 1965 1970 1975 1980 1985 1990 1995 2000 2005 80 a 9 Co LH tok ko id : i ia i a 1965 1970 1975 1980 1985 1990 1995 2000 2005 Figure 10. Number of pairs (left bar) and nestlings (right bar) in Sweden. Upper graph shows populations on the Baltic coast (dark grey and black) and lakes in Southern and Central Sweden (light grey, on top of Baltic bars); lower graph shows the population in Lapland. 0.5-4 times higher and the range of DDE concentrations as much as 2-15 times higher than the estimated LOELs, respectively. Thus, effects also from PCB on eagle reproduction were for a long time largely concealed by the effects from DDE. SB 27 Reproduction —much better but not good enough! A good improvement in the breeding success of the Baltic population began in the late 1980s, and by the year 2000 the proportion of successfully breeding pairs reached the background level (Figure 9). In 1998-2005, concentrations of DDE and PCB in analysed eggs were on average 25 % below estimated LOELs, but 20 % of the analysed eggs exceeded those levels. Thus, although the breeding success has improved and is no longer significantly different from the background level, this population is still in the risk zone. The number of young per productive nest also increased from the 1980s but levelled in the mid 1990s at about 1.65, that is significantly below the reference level (Figure 6). This is mainly due to smaller broods along a particular stretch of coastline, the southern part of the Bothnian Sea (mean brood size = 1.4). DDE and PCB concen- trations in eagle eggs from that region are not higher than in eggs from regions with brood sizes around 1.8 (equalling the background level). A new study is in progress in order to investigate possible effects from other contaminants to explain the observed differences in brood sizes, with polybrominated flame retardants (PBDEs) and dioxins among the suspects. Population trends and expansion Following the improvement in productivity on the Baltic coast, this population increased from 56 to 256 pairs in 1979-2005 (Figure 10). In addition, offspring from nests on the coast began to re-colonise old inland sites from 1980. This inland population had increased to 66 pairs in 2005, and has the potential for a further increase. The mean annual rate of increase of the Baltic and inland populations combined was near 10 % between 1990 and 2005. The annual rate of increase of the population in Swedish Lapland was only about 5 % over the same period, reflecting lower productivity and probably also a lower annual survival rate there. Illegal Scottish Birds (2007) The White-tailed Eagle in Sweden - past, present and future 53 persecution is still occurring at least locally in Lapland. Along with the increase in numbers we have seen an extensive geographical expansion (Helander 2003c) and the current range is beginning to resemble the historical distribution. But the White-tailed Eagle is still absent as a breeder on the Swedish west coast, a former stronghold, and in areas in Northern Sweden where it was also known to breed. Outlook for the future With a good annual increase rate and much of the historical range already reoccupied, the outlook for the White-tailed Eagie seems good. But there are things that still need to be managed for the future. Some of the goals in a national Species Action Plan for the eagle over the next 5 years are: ™ a population of 500 pairs, and resettlements on the west coast and in the county of Jaémtland in the north. B® regional protection plans should be made for each county, as a means in land use planning. ™ a network of protected breeding areas should be established all over the species’ natural distribution range in Sweden. ™ a new routine should be implemented in forestry in order to create a succession of old trees for nesting, saved as ‘eternity trees’. This will benefit other raptors too. = better routines should be enforced in forestry regarding protection zones around nests, in order to save nests sites from cutting. @ the frequent killing of eagles by trains should be reduced by more effective removal of carcasses from the railways. ™ mortality by lead intoxication through ingestion of gunshot pellets from prey should be investigated. ™ mortality from blade strikes and collisions at wind power plants should be investigated; nothing is done so far with raptors and wind parks in Sweden. ™ risk areas for wind power plants should be identified When making a Species Action Plan, you are also asked to give a vision for the future. To me, such a vision should mean that the whole natural breeding range should be recolonised; this would imply a capacity population of about 1000 pairs in Sweden. And there should be no need for feeding programmes or the building of artificial nests to support the population. Such a vision may not be realistic, but is still what should be strived for. What will really happen in the long future we cannot really anticipate - but at least for the near future I expect to see more eagles in competitive fights over the best territories, and a further growth and expansion of this eagle population in Sweden! References Berg B 1923. De sista 6rnarna. Bonniers, Stockholm. Hailer F 2006. Conservation genetics of the white-tailed eagle. Dissertation, University of Uppsala. Acta Universitatis Upsaliensis. Uppsala. Hahr T 1868. Om de tillforlitligaste medlen att utrota wara skadligare rofdjur och roffoglar. P.B.Eklund, Stockholm. Hardey J, Crick H Q P, Wernham C V, Riley H T, Etheridge B & Thompson D B A 2006. Raptors: a field guide to survey and monitoring. The Stationary Office Ltd, Edinburgh. Helander B 2003a. “Project Sea Eagle” — the first species action plan in Sweden. SEA EAGLE 2000 (eds B Helander, M Marquiss & W Bowerman) pp 13-22. Atta45 Tryckeri AB, Stockholm. Helander B 2003b. The white-tailed sea eagle in Sweden — reproduction, numbers and trends SEA EAGLE 2000 (eds B Helander, M Marquiss & W Bowerman) pp 57-66. Atta45 Tryckeri AB, Stockholm. Helander B 2003c. The international colour- ringing programme — adult survival, homing and the expansion of the white-tailed sea eagle in Sweden. SEA EAGLE 2000 (eds B Helander, 54 M Marquiss & W Bowerman), pp 145-154. Atta45 Tryckeri AB, Stockholm Helander B, Olsson A, Bignert A, Asplund L & Litzén K 2002. The role of DDE, PCB, coplanar PCB and eggshell parameters for reproduction in the white-tailed sea eagle in Sweden. Ambio 31(5):386-403. Hogdahl T 1923. Gammel-Ante. L. Hokerbergs Forlag, Stockholm. Newton A 1864. Ootheca Wolleyana. Part I. Accipitres. John van Voorst, London. Olsson A, Ceder K, Bergman A & Helander B 2000. Nestling blood of the white-tailed sea eagle (Haliaeetus albicilla) as an indicator of territorial exposure to organochlorine compounds — an evaluation. Environ. Sci. Technicol, 34:2733-2740. The Derek Ratcliffe Memorial Lectures SB 27 Rahn H, Ar A & Paganelli C V 1979. How bird eggs breathe. Sci. Amer. 240:46-55. Ratcliffe D 1967. Decrease in eggshell weight in certain birds of prey. Nature 215:208-210. von Wright W 1850. Férteckning, jemte nagra anteckningar, Ofver de i Bohuslanska skargarden observerade foglarne. Gotheb. K. Vet. O. Vitt. Samh. Handl., ny foljd 1:43-78. Bjorn Helander, Swedish Museum of Natural History Department of Contaminant Research Box 50007 SE-104 05 Stockholm, Sweden Tel.+46 (0)8 5195 4109 Fax.+46 (0)8 5195 5195 (or 5195 4256) E-mail bjorn.helander@nrm.se In memory of the late Dr Derek Ratcliffe (1929-2005) the Scottish Raptor Study Groups (SRSG) established an annual Memorial Lecture, delivered during the SRSG Conference at Battleby. The inaugural and second lectures were delivered in 2006 and 2007 respectively. Derek was one of the world’s leading raptor conservationists and made the key discovery about the link between pesticides and eggshell thinning in birds of prey. He published classic bird monographs, The Peregrine Falcon (1980, 1993) and The Raven (1997), and other books including Bird Life of Mountain and Upland (1990), Galloway and the Borders (2007), Lapland: a natural history (2005) and Lakeland: the wildlife of Cumbria (2002). He effectively led the UK statutory nature conservation movement in the 1970s and 1980s and in A Nature Conservation Review (1977) laid the foundations for the wildlife legislation and protected areas we have in Britain today. The SOC, in partnership with the SRSGs and Scottish Natural Heritage, is delighted to honour Derek’s work by publishing summaries of the memorial lectures. PK Stirling-Aird and Des Thompson Scottish Birds (2007) Trends in breeding duck populations at Loch Leven, Perth & Kinross DS) Trends in breeding duck populations at Loch Leven, Perth & Kinross ALAN LAUDER Results of annual breeding duck surveys from 1985 to 2004 are presented. Comparisons with earlier studies show that some significant population changes have occurred. Notable increases occurred with Gadwall, Eurasian Teal and Common Pochard while decreases have occurred with Common Shelduck, Eurasian Wigeon, Northern Shoveler and Tufted Duck. Possible reasons for these changes are discussed including the role of submerged macrophytes, water quality and increases in predator populations. Introduction Loch Leven lies to the east of Kinross in Perth & Kinross in east central Scotland (Figure |). At 1597ha it is the largest lowland freshwater loch in Scotland. It is shallow and eutrophic lying amidst rich agricultural land in a basin bounded by the uplands of the Lomond Hills, Benarty Hill and, further to the west, the Ochil Hills. Loch Leven is afforded legal protection under a number of designations mainly on account of its important waterbird populations notably wintering Pink-footed Geese Anser brachyrhynchus and a rich assemblage of breeding, passage and wintering waterfowl. It is a Site of Special Scientific Interest (SSSI), Ramsar site and Special Protection Area (SPA) and has been managed by Scottish Natural Heritage (SNH) as a National Nature Reserve since 1964. Loch Leven holds the largest concentration of breeding ducks on a freshwater site within Britain. St Serf’s Island holds the vast majority of the nesting ducks. The population size of all the breeding duck species were first formally estimated by researchers from 1966 to 1971 (Newton & Campbell, 1975). Regular monitoring, largely by counting territorial pairs in spring, has been carried out since then. Wright (1986) gave an update at that time of the status of the population, relying on spring counts of pairs and males for most species. Wright assessed the population to be at least as large as it was in Newton and Campbell’s time, numbering around 1000 pairs of 6 main species, mostly Tufted Duck Aythya fuligula and Mallard Anas platyrhynchos. Since 1986, a number of potentially significant developments have occurred. These include improvements in water quality, conservation management of loch shore habitats and changes in the size and distribution of breeding gull populations (Laridae). | Kinross | L iG = 3 Saree eo N | y 1 RAS 500 metres RSPB Vane Fa R. Leven Figure 1. Location map of Loch Leven. ° Scart ls. . Sx Castle Is. ree © Reed Bower 56 Methods This paper compares data generated by surveys from the periods 1966 to 1971 and 1980 to 1984 to data collected between 1985 and 2004 and examines population trends since 1980. Methods of surveys in 1966-71 and from 1980-1984 are given in Newton & Campbell (1975) and Wright (1986) respectively. Method A: Counting pairs The method of counting pairs and males used by Wright (1986) has been followed annually since 1981 with only slight amendments in coverage to encompass newly created wetland areas, primarily at Vane Farm. In brief the method requires 2 observers in a slow moving boat and in calm conditions, recording all pairs, males and groups of ducks encountered around the shoreline of the loch and islands, plotting these on a large scale map. In addition lagoons and pools away from the main loch shore are counted by viewing from vantage points. This method broadly follows that recommended in Gilbert er al (1998) though it was adapted for the very large size and open nature of the loch by relying on good viewing conditions and utilising boat transport. Difficulties arise when attempting to count Tufted Duck and Mallard using this method, due to the large number of birds present, the nature of large groups of Tufted, widespread use of sites away from the loch by Mallard and the occurrence of, sometimes large influxes of presumed migrant Tufted Ducks well into May. This method was therefore not used to count these 2 common species. Method B: Counting nests Full sweep searches of all suitable nesting habitat on the main nesting island were carried out in 1981 (Wright, 1981). This was repeated using similar methods in 1993, 1994, 1998 SB 27 (except for Tufted Duck) and 2003. This required a team of between 3 and 8 people walking in a line intensively searching all potentially suitable duck nesting habitat on St. Serf’s Island, identifying and recording all duck nests and their contents including old or predated nests. The total population was then determined by multiplying the number of nests found by a conversion factor. The conversion factor was derived from weekly cumulative nest totals observed by Newton & Campbell (1975) and is illustrated below: Predicted total population=n{ 100/p} n is the total number of nests detected on the sweep search, and p is the mean percentage of the total nests detected by the equivalent week in the period 1966-70. Thus, the figure derived was the predicted total number of nests for a species in any year. The survey visits were timed for Mallard (third week in April) and for Tufted Duck (second week in June) to coincide with the peak numbers initiated nests (Mallard) and the peak incubation period (Tufted Duck). Method C: Sample nest counts In 1998, 1999, 2002 and 2004 randomised sample quadrat surveys were carried out to estimate the size of the Tufted Duck population. This required searching a number of 25mx25m randomly selected quadrats within the known duck nesting area. The number of quadrats searched was equivalent to approximately 10% of the total colony area. The total population was then estimated for that week and the conversion factor applied as in method B. In addition, for scarce breeding species such as Northern Pintail Anas acuta, Garganey Anas querquedula, Goosander Mergus merganser and Ruddy Duck Oxyura jamaicensis, casual records gleaned from bird reports and unpublished sources have been used to document the size of the population. Scottish Birds (2007) Statistical analysis Population trends over time were analysed using the S-Plus 2000 statistical package (Mathsoft Inc, Seattle, WA). As pairs counts and nest surveys result in observed count data and thus the data appeared to follow a Poisson distribution, log- linear regression models were used to determine the significance of trends in the population. Results Common Shelduck Tadorna tadorna Wright (1986) described a marked increase from an estimated population of around 10 pairs in the period prior to 1970 up to a mean of 48 (range 39-56) in 1980 to 1984. Since then Shelduck numbers have varied considerably with both the highest and lowest pairs counts occurring in this period. Despite the 2 high counts of 56 and 62 in 1998 and 1999 respectively, the overall trend was one of significant decline (Figure 2). Eurasian Wigeon Anas penelope This was the fourth commonest breeding duck in Newton & Campbell’s time with a mean of 31 pairs (range 23-40). It had been elevated to third place with a mean of 39 pairs (range 35-46) during 1980-1984. The situation changed dramat- ically between 1985 and 1986 with a decline from 44 pairs in 1985 to just 8 pairs in 1986. Numbers only slightly recovered thereafter and in at least 2 years since 1985 no pairs were confirmed on the loch though some years were also not surveyed (Figure 2). The decline was found to be statis- tically highly significant. Gadwall Anas strepera Newton & Campbell (1975) found a mean Gadwall population of 36 pairs in 1966-1970 (range 24-47), Wright (1986) reported similar numbers with a mean population in 1980-1984 of 37 pairs (range 25-47). Between 1980 and 2000 the population has undergone a statistically significant increase (Figure 2) and _ the no. pairs > Trends in breeding duck populations at Loch Leven, Perth & Kinross D7 Figure 2. Changes in major breeding duck populations at Loch Leven 1980-2004. 60 significant decline P<0.001 1980 1985 1990 1995 2000 year Figure 2a. Common Shelduck. significant decline P<0.001 No. Pairs N a 1980 1985 1990 1995 2000 Year Figure 2b. Eurasian Wigeon. No. pairs 2 2 ¢ . significant increase ° 1980-2000 P<0.001 1980 1985 1990 1995 2000 Year Figure 2c. Gadwall. population between 1990 and 2000 has been at least 50 pairs annually. Since 2000, numbers appear to have declined to early levels with 23 to 29 pairs between 2002 and 2004. These are still within the range seen since 1980 however. 58 significant increase P<0.001 4 Obs. No. Pairs ° 0 e 6 #6 6.8 e e 1980 1985 1990 1995 2000 Year Figure 2d. Eurasian Teal. est no. pairs 200 No change P>0.05 Year Figure 2e. Mallard. Eurasian Teal Anas crecca Breeding Teal numbers at Loch Leven have historically been difficult to determine but in the prel970 period the mean population was 3 pairs (range 0-8). In 1980 to 1984 Wright (1986) again also found a mean of 3 pairs (range 0-7). Since 1985, numbers of breeding Teal increased signif- icantly (Figure 2), with a mean of 7 pairs (range 0-22). Despite this recorded population change Teal nests were never found on the main nesting area on St. Serf’s Island from 1981 onwards, despite periodic extensive searches for all ducks (pers obs and Lauder & Brooks, 1998) though nests are occasionally found on the loch shore including one to the west (W Wilson pers comm) and older records from the North east shore (GA Wright pers comm) SB 27 Mallard Full surveys carried out in 1993, 1994 and 1998 produced estimates of 389, 401 and 612 pairs respectively. A sample quadrat survey in 2004 estimated 313 pairs. The mean of these 4 years is 429 pairs. Newton & Campbell (1975) found a mean of 441 pairs (range 312 — 664) in the period 1966-1970. Wright (1986) conservatively estimated the population in 1981 to be in the range 420-450 pairs. Data from this study indicates that the population has shown no significant change since 1981 (Figure 2) and is also comparable to the mean Figure found in Newton & Campbells study. Northern Pintail Pintail was absent as a breeder during Newton & Campbell’s 1966- 1970 study and Wright’s 1980 -1984 study, despite previously being a regular breeder prior to 1952 with up to 20 pairs in the 1930’s (Allison, Newton & Campbell, 1974). Sporadic pairs have appeared since and breeding was proved in 1992 when a female with a brood of 7 young was seen on the south shore of the loch near the mouth of the Gairney Burn. Northern Shoveler Anas clypeata Newton & Campbell (1975) found a mean Shoveler population of 6 pairs during 1966-1970 (range 3-10), Wright (1986) reported larger numbers with a mean population in 1980-1984 of 17 pairs (range 9-30). The 30 pairs level was reached in 1984 and 1988 but out with those higher years the population has largely remained under 20 pairs and has shown a significant decline since 1980 (see Figure 2). In 2004 no birds were recorded on pairs counts but 11 nests were found during sweep searches, with 5 broods being seen on the loch later in the summer thus pairs count data may under represent the population in at least some years. Scottish Birds (2007) Common Pochard Aythya ferina Pochard was never a common breeding species at Loch Leven. During Newton & Campbell’s study no nests were found and Allison et al (1974) refer to breeding as never having been proven though mentions ‘a few pairs recorded as breeding from 1886’ Wright’s 1980 to 1984 surveys found no confirmed pairs and no known breeding occurred. Breeding was first confirmed in 1988 with a nest found on St Serf’s Island (G A Wright, pers comm). Since then breeding was initially sporadic but increasing numbers of pairs present gave rise to further confirmed breeding records with nests found and broods increasingly seen. The mean number of pairs recorded annually since 1985 was 5 (range 0-14). The increase recorded was found to be highly significant. Tufted Duck Tufted Ducks have probably been at least as numerous as Mallard since Newton & Campbell’s time. The 1966-1970 mean was 368 (range 250 - 485) and Wright (1986) felt they were at least as numerous during his survey in 1981, with an estimated population of 556 nests, which exceeds the upper limit of the 1966-71 range. Between 1993 and 2004, 7 surveys were undertaken and a mean estimate of 327 pairs was found (range 190 - 463). When tested, a statistically significant decline was found since 1981 all but one of the estimates remain above Newton and Campbell’s minimum estimate which may indicate little overall change since the 1960s. Other species Other duck species to have established or re established themselves as breeders since 1985 are Garganey, Common Goosander and Ruddy Duck. Garganey are recorded increasingly frequently on the floods and pools at Vane Farm RSPB Reserve and have shown signs of breeding in at least 2002 and 2003 though no broods or nests were found (Brooks et al 2004) Trends in breeding duck populations at Loch Leven, Perth & Kinross 59 ° significant decline P<0.001 Obs. No. Pairs 1980 1985 1990 1995 2000 Year Figure 2f. Northern Shoveler. significant increase ci P<0.001 * Obs. No. Pairs 0 # @ # @ @ ¢ @ «6 C) 1980 1985 1990 1995 2000 Year Figure 2g. Common Pochard. 2 significant decline P<0.001 1980 1985 1990 1995 2000 Year Figure 2h. Tufted Duck. Goosander was recorded during the breeding season at times (Allison, Newton & Campbell, 1974) but was not suspected as a breeder by Wright (1986). Since 1985 the species has increased its frequency as a breeder and since 1991 has appeared near annually with at least one brood in most years. 60 Occasional males or pairs of Greater Scaup Aythya marila and Red-Breasted Merganser Mergus serrator were seen in_ potentially suitable habitat in some springs from 199] onwards but have never been proved to breed. A non breeding flock of around 50 Common Goldeneye Bucephala clangula occur in summer but have not been shown to breed despite the presence of a range of natural nest sites and the ~ erection of many nest boxes. Ruddy Duck definitely bred at least once during the period since 1985 as a brood was seen in 1992 and pairs were regularly seen in spring and summer from 1991 onwards. They have become less common since around 1998. Discussion The pairs counts undertaken since 1980 were thought to be broadly representative of the total populations of most duck species nesting at Loch Leven, except for the commonest 2 species, Mallard and Tufted Duck. Occasional apparent anomalies occurred during surveying, for example in 2004 when no Shoveler pairs were observed during the pairs count but 11 nests were found on St. Serf’s Island during the census of Tufted Duck and subsequently 5 broods were observed. The trend is probably still representative of a real decline in the Shoveler population as shown by the statistical analysis of pairs count data as pairs counts are widely regarded to provide a repeatable and relatively accurate measure of the populations of most breeding species holding territories on water bodies (eg Koskimies & Poysa, 1985, 1987 and 1989, Koskimies & Vaisanen, 1991, Poysa, 1996 and Gardarsson, 1979). One caveat however is that the detection rates of pairs remains the same and this is largely dependent upon suitable counting conditions occurring in spring, observer coverage remaining the same and observer ability remaining consistently high. SB 27 At Loch Leven, with the existence of long term datasets using this simple pairs counting technique, it would seem sensible to continue its use as the main method for estimating population size and abundance of most duck species on Loch Leven. To enhance the method additional calibration would lend greater confidence in the technique by comparing the pairs counts against season total nest counts if these were carried out over a series of years, though this would require very significant manpower resources to achieve. Sample period nest counts of the 2 main species, Mallard and Tufted Duck, may well have still been valid in 1981 when Wright undertook the work for his study. Since then apparent changes in climate, loch water level regime, and subtle habitat or species associations may have altered the timing of breeding and hence caused a potentially significant source of error in the population estimates. This could be reduced in future by resurveying the colony using Newton & Campbell’s method and deriving new weekly nest totals for future use. Four duck species breeding at Loch Leven have shown statistically significant population declines since 1980: Common Shelduck, Eurasian Wigeon, Northern Shoveler and Tufted Duck. A further 5 species have shown increases, 3 of which were statistically significant; Gadwall, Teal and Pochard. The only species to show no significant population change was Mallard. The decline in the Eurasian Wigeon population reflects a perceived national trend where breeding numbers on lowland sites have declined (C Mitchell pers comm). This contrasts with the total national population which has remained fairly constant since the 1960s (Owen, 1993). Declines found in Shoveler and Tufted Duck, while statistically significant, fall within the range of variation in the population since Scottish Birds (2007) 1966. In addition the methodological difficulties may be a result of a change in the timing of the peak incubation period, or in the case of Shoveler, a change in the detection rate of pairs due to both changes in observer and a shift in the distribution of birds to more vegetated lagoons developed adjacent to the loch. Shoveler have also have shown a national decline since 1968- 72 (Mitchell, 1993, Boyd et al, 1994) Increases in the Gadwall population are in line with the continuing national increase as described by Fox (1988). More recent declines are likely to be temporary and may be due to local changes as for Shoveler. Duckling predation by an increasing population of Lesser Black-backed Larus fuscus and Herring Larus argentatus Gulls has been suggested as potentially affecting the duck population. Evidence to support this has been anecdotal thus far and is based on casual observations collected by a number of observers over many years rather any systematic recording. Dabbling duck species are thought to be the less susceptible to predation by gulls as they use the cover of thick emergent vegetation on the loch shore and along ditches and inflow burns (Clark et al, 1987). Shelduck and diving ducks are more susceptible to predation but recent analyses (Lauder, 2006) suggest the overall impact of gull predation is likely to be lower than perceived from casual observations. There is no evidence to suggest that large gulls predate duck nests on St Serf’s Island. Other potential aerial predators, eg Corvids, have not significantly increased their populations on the island and are unlikely to have had any negative impact, particularly given the buffering effects of more available gull nests (Liordos, 1996). There are normally no ground predators on the island and any influxes of rats or mink are quickly controlled once detected. The nest failure rate at Trends in breeding duck populations at Loch Leven, Perth & Kinross 61 least for Tufted Duck has remained fairly constant since Newton & Campbell’s study (Liordos, 1996 and Lauder, in prep). The availability of nesting habitat has not declined and may in fact have increased with recent reductions in summer grazing resulting in new areas of Deschampsia caespitosa dominated cover. The nesting association of ducks with Black- headed Gulls Larus ridibundus at Loch Leven is well known and has been described by various authors eg Allison et al (1974) and Liordos (1996). At least since 1990 the colony of Black- headed Gulls on St Serf’s Island has been dynamic in its size and location and appears to have responded to the increasing presence of large gulls. It appears that competition for nest sites has moved Black-headed Gulls to new areas. Ducks continue to nest in suitable habitat in association with both large gulls as well as Black-headed Gulls. Shelduck nests in burrows on St. Serf’s Island are relatively safe from predation. However, Western Jackdaw (Corvus monedula) which also utilises burrows for nesting on St. Serf’s Island, may have an impact on Shelduck through competition or nest predation. No nest data has been collected and thus the level of impact remains unknown. Water quality in Loch Leven has undergone changes over the past 100 years. In modern times it has been the subject of a great deal of press coverage and subsequent efforts to improve it (LLCMP, 1999). Until around 1996 the loch suffered increasing eutrophication with increases in phytoplankton, resultant reduced light penetration and reduced growth of submerged macrophytes (LLCMP, 1999). Subsequent efforts to reduce nutrient inputs have resulted in slight improvements in water quality and _ light penetration (A Kirika pers comm) with resultant benefits to aquatic plant and invertebrate 62 communities. The exact effects of changing water quality on breeding duck populations are poorly understood but broadly those species which feed on the seeds or leafy matter of aquatic plants at some time of the year eg Teal (Cramp & Simmons, 1977) and Pochard (Stewart & Lauder, 1997), and fish eaters such as Goosander, are likely to have benefited and this is reflected in increasing populations of these species. The Loch Leven population retains its status as the largest inland concentration of breeding ducks in Britain. However, management will be required in future to halt or reverse the observed recent declines in some species. Management to benefit many key species would entail maintenance of suitable nesting areas and the associated Black-headed Gull — colony, improvement in potential brood rearing areas by expanding areas of emergent vegetation and provision of further sheltered lagoon areas and continuing efforts to improve water quality to benefit invertebrate and submerged plants. Acknowledgements Data collection and analysis was funded by SNH. Thanks to Paul Brooks at SNH for much discussion and helpful comments on drafts of the text, to Jeremy Squire at SNH for collection, inputting and provision of the data used since 2000 and to Ken Shaw who gave data, comments and discussion on breeding ducks at RSPB Vane Farm. Thanks are also due to all those who have ably collected data and have written about the ducks of Loch Leven before, not least Alan Allison, Colin Campbell, Ian Newton and Gordon Wright. Willie Wilson of Kinross Estate has been helpful in discussing old records. Dr Claire McKeever provided much needed help and advice with statistical analysis. SB 27 References Allison A, Newton I & Campbell C R G 1974. Loch Leven National Nature Reserve: a study of waterfowl biology WAGBI, Chester. Brooks P M, Devine M, Lauder A W, Shaw K D & Wilson A S 2004. Bird Report 2003: Loch Leven NNR including Vane Farm RSPB Reserve RSPB/SNH, Kinross. Boyd H, Quinn J & Owen M 1994. An Overview of Wildfowl Research Surveys and Monitoring at Loch Leven NNR Wetlands Advisory Service draft report to SNH. Clark N A, Varley T A, Evans J & Wright G A 1987. Breeding success of Mallard Anas Platyrhynchos at Loch Leven National Nature Reserve Bird Study 34: 129-134. Cramp S & Simmons K E L 1977. The Handbook of the Birds of Europe, the Middle east and North Africa, The birds of the Western Palearctic (Vol 1) Oxford University Press, Oxford. Crawley M J 2002. Statistical Computing: An Introduction to Data Analysis Using S-Plus. John Wiley. Fox A D 1988. Breeding status of the Gadwall in Britain and Ireland British Birds 81:51-65 Gardarsson A 1979. Waterfowl populations of Lake Myvatn and recent changes in numbers and food habits Oikos 32: 250-270. Koskimies P & Poysa H 1985. Monitoring Waterfowl populations in Finland; method- ological remarks Lintumies 20:270-279. Koskimies P & Poysa H 1987. Vesilinnuston seuranta ja laskentamenetelmat Suomen Riista 34: 31-41. Koskimies P & Poysa H 1989. Waterfowl censusing in environmental monitoring: a comparison between point and round counts Ann. Zool. Fennici 26: 201-206. Koskimies P & Vaisanen R A eds 1991. Monitoring Bird Populations Zool. Mus., Finnish Mus. Nat. Hist., University of Helsinki, Helsinki. Scottish Birds (2007) Lauder A W 2006. The Impact of Large Gulls on the Breeding Duck Population at Loch Leven NNR, Perth & Kinross Scottish Natural Heritage Commissioned Report (unpublished report). Lauder A W & Brooks P M 1998. The 1994 and 1998 Surveys of Breeding Duck, St Serf’s Island, Loch Leven National Nature Reserve internal SNH report (unpublished report). Liordos V 1996. A study of Nesting Tufted Duck Aythya fuligula on St. Serf’s Island Loch Leven National Nature Reserve: Factors involved in determining nesting density and success with particular regard to the importance of the Black-headed Gull Larus _ ridibundus. Unpublished MSc _ thesis, University of Edinburgh. Loch Leven Catchment Management Project 1999. The Loch Leven Catchment Management Plan LLCMP Steering Group, Perth Mitchell C in Gibbons D W, Reid J B and Chapman R A eds 1993. The New Atlas of Breeding Birds in Britain and Ireland:1988- 1991 T & AD Poyser, London. Newton I & Campbell C R G 1975. Breeding ducks at Loch Leven, Kinross Wildfowl 26: 83-102. Owen M in Gibbons D W, Reid J B & Chapman, R Aeds 1993. The New Atlas of Breeding Birds in Britain and TIreland:1988-1991 T. & A.D. Poyser, London. Poysa H 1996. Population estimates and the timing of waterfowl censuses Ornis Fennica 73:60-68. S-Plus 2000. Guide to Statistics Vol 1 Data Analysis Products Division, Mathsoft, Seattle, WA, USA. Stewart B & Lauder A W 1997. Gizzard contents of Pochard, Tufted Duck and Goldeneye from Loch Leven, Kinross, in winter 1994-95 Scottish Birds 18:28-35. Wright G A 1981. Duck Breeding Survey of St. Serf’s Island, Loch Leven internal NCC report unpublished. Trends in breeding duck populations at Loch Leven, Perth & Kinross 63 Wright G A 1986. Breeding Wildfowl of Loch Leven National Nature Reserve Scottish Birds 14: 39-43. Alan Lauder, 4 Braemar Grove, Dunblane, FK15 9EF email: alan.lauder@rspb.org.uk Revised manuscript accepted March 2007 64 SB 27 Low tide counts on the Firth of Tay NORMAN ELKINS Monthly Wetland Bird Survey (WeBS) counts are made throughout coastal Britain but are chiefly of birds roosting in estuarine habitats at high tide. As the importance of feeding sites within estuaries cannot be determined from these counts, a programme of counts ensures that each major estuary is counted at low tide every few years. Substrates, and therefore available food for waterfowl, vary considerable across an estuary, so that monitoring of the location and activity of feeding birds is vital to determine the conser- vation value of its component parts. In the winter of 2006-07, The Firth of Tay, part of a Special Area of Conservation (SAC), was the subject of comprehensive counts under the WeBS Low Tide scheme. Counts for this winter period are compared to previous counts and the results confirm findings from high tide roost counts that certain species of waterfowl, notably Dunlin, have undergone considerable declines. Introduction As part of the rolling programme of Wetland Bird Survey (WeBS) Low Tide counts on Britain’s estuaries, organised by the British Trust for Ornithology (BTO) for the WeBS partnership (BTO, RSPB, WWT and JNCC), the Firth of Tay was counted in its entirety during the winter of 2006-07. Previous counts had been made during the winters of 1993-94 and 1996-97 (see Musgrove et al 2003). The aim of Low Tide counts is to determine those sections of an estuary most used by each species, rather than the total numbers. However, the counts do give an idea of the numbers involved, since they are made on specific predetermined dates or within one or 2 days of this. Methods Counts were made near or at low tide on specified dates in November and December 2006 and January and February 2007. The estuary was divided into 69 sections and counts were made from both north and south shores between the mouth of the River Earn and Buddon Ness/Tentsmuir. All shorebirds and wildfowl were counted, plus divers, grebes, Grey Heron Ardea cinerea and Great Cormorant Phalacrocorax carbo. All sections were counted, with the exceptions of Dundee harbour and Buddon Ness itself. Results Most species favoured particular feeding sites within the estuary. Great Cormorants were found mainly at Tentsmuir Point and near the rail bridge. Grey geese Anser spp. concentrated on the mudflats east of Mugdrum Island with Common Shelduck Tadorna tadorna in the Mugdrum area and at Tayport but spreading out as numbers built in late winter. Of the dabbling duck, Eurasian Wigeon Anas penelope favoured downstream shores from Broughty Ferry and Tayport, mainly on the south shore, whereas Eurasian Teal Anas crecca were found mostly in Invergowrie bay and north of Mugdrum. Mallard Anas platyrhynchos were much more widespread, but concentrated in the upper reaches near and downstream of Mugdrum, between the rail bridge and Lucky Scalp and off Monifieth. Common _ Eiders Somateria mollissima favoured the main channel Scottish Birds (2007) downstream from Lucky Scalp while Common Goldeneye Bucephala clangula were mostly found in the channel between Mugdrum and Flisk. Red-breasted Mergansers Mergus serrator were mainly off Tentsmuir Point, with some in the river between Broughty Ferry and Tayport, and a few upriver off Flisk. Not surprisingly, waders were found in good feeding areas on the mudflats and on mussel beds. Eurasian Oystercatchers Haematopus ostralegus concentrated between Tayport and Tentsmuir Point and also off Barnhill but considerable numbers also used the north shore between Invergowrie and Templehall. Ringed Plovers Charadrius hiaticula were almost entirely found at Tentsmuir Point but Grey Plovers Pluvialis squatarola were spread from there along to Tayport. The very small numbers of Northern Lapwing Vanellus vanellus were located in Invergowrie Bay with even smaller numbers of European Golden Plover Pluvialis apricaria off Powgavie. Both Red Knot Calidris canutus and Sanderling Calidris alba continued to favour the Lucky Scalp and Tentsmuir Point area. Dunlins Calidris alpina were found mainly east of Tayport. Bar-tailed Godwits Limosa lapponica fed along both shores of the outer estuary east of Broughty Ferry and Tayport with a small flock at Invergowrie. Eurasian Curlews Numenius arquata were widespread, inhabiting all shores with the largest numbers between Kingoodie and Powgavie. Common Redshank Tringa totanus were ubiquitous in small numbers, with the main flock at Invergowrie. Ruddy Turnstones Arenaria interpres were confined to the outer estuary, particularly east of Tayport. A small minority of relatively empty sections were uncounted during at least one of the winters and, for direct comparison, Table | shows the counts of the 60 sections covered in every winter. Low tide counts on the Firth of Tay 65 Discussion The Low Tide counts show a similar trend to high tide roost counts, namely a_ significant decrease in the numbers of species for which the Firth of Tay is deemed especially important (see Elkins & Lynch 1997, Elkins 2006). These include Bar-tailed Godwit and Common Redshank, but the largest numerical decrease has been for Dunlin. Sanderling was the only wader that increased in numbers. The figures for Great Cormorant and wildfowl also show decreases, especially for Tufted Duck Aythya fuligula, of which none was present in the most recent winter. The variation in both Pink-footed Geese Anser brachyrhynchus and Greylag Geese Anser anser illustrate local movements between the estuary and feeding sites on surrounding agricultural land. Likewise, the discrepancy in the Common Eider counts is due to the difficulty of counting mobile birds moving in and out of the estuary. Recent counts at high tide suggest that there are at least 10.000 Common Eider still present in the Outer Tay. Mute Swans Cygnus olor are at their fewest in midwinter and Whooper Swans Cygnus cygnus also winter in only small numbers. High tide wader roost counts also indicate that a proportion of some feeding populations roost outwith the Firth. Even when this is taken into account, there are some species that are notably fewer than in the 1990s. Wintering European Golden Plovers have diminished greatly. Low tide flocks were common in the early 1990s, but are now very unusual. Even inland flocks seem scarcer, a fact endorsed by the low numbers of the species on both the Non Estuarine Waterbird Survey (NEWS) and the Winter Plover survey, both carried out during this same winter in east Fife. There have been some marked changes in distri- bution since earlier counts. The decline in Mallard appears to have been mainly in Invergowrie Bay where considerable numbers were once found. A similar situation applies to 66 SB 27 Table 1. Peak mid winter low tide counts in the Tay Estuary. Figures represent the highest monthly count between December and February. Those in brackets are peak high tide roost counts for waders in the same months. Species Great Cormorant Mute Swan Whooper Swan Pink-footed Goose Greylag Goose Common Shelduck Eurasian Wigeon EurasianTeal Mallard Tufted Duck Common Eider Common Goldeneye Red-breasted Merganser Eurasian Oystercatcher Ringed Plover European Golden Plover Grey Plover Northern Lapwing Red Knot Sanderling Dunlin Bar-tailed Godwit Eurasian Curlew Common Redshank Ruddy Turnstone 1993/94 231 14 4 2 150 80 122 4 729 7 no count 19] JO 2264 (1905) S173) 350 (249) 215 (233) 687 (134) 84 (67) 44 (75) 3769 (4657) 1593 (1367) 430 (180) 889 (961) 49 (42) 1996/97 174 16 8 93 83 56 29 16 948 i735 12253 220 60 2239 (1880) 188 (145) 775 (0) 159 (187) 615 (/68) 305 (480) 113 (106) 5195 (1581) 2238 (2305) 735 (210) 1203 (633) 28 (46) 2006/07 118 20 21 40 1587 (982) 29 (32) 44 (0) 179 (110) 77 (93) 42 (51) 145 (84) 980 (449) 1000 (588) 496 (329) 413 (565) 17 (48) Table 2. Low tide counts of waders made in the inner Tay between the rail bridge and Port Allen for 4 winter periods. Species Eurasian Oystercatcher Ringed Plover European Golden Plover Northern Lapwing Dunlin Bar-tailed Godwit Eurasian Curlew Common Redshank 1994 1997 1263 14 2007 822 0 Scottish Birds (2007) Low tide counts on the Firth of Tay 67 Tufted Duck, Common Goldeneye, Dunlin and Common Redshank, and flocks of the latter 2 species are a shadow of their former size. The former Tayport flock of European Golden Plover has now disappeared and the large inner estuarine flocks of Northern Lapwing have also declined significantly. Both could be due to milder winters allowing them to feed inland more often. Low numbers of feeding waders on the expanse of mud in the western part of the inner estuary is thought to be due to poor feeding opportunities (Elkins & Lynch 1997). The bulk of the feeding flocks use the eastern portion and Table 2 shows counts in the inner estuary for 4 winters. Included in Table 2 is a count made in late January 1989 (Laing & Taylor 1993) and the peak January/February low tide counts in each of 1994, 1997 and 2007. The declines in European Golden Plover, Northern Lapwing, Dunlin and Common Redshank are apparent and match those found on high tide counts. Acknowledgements The logistics of organising this type of survey are considerable. I am grateful to the BTO for the opportunity to participate in the survey and for providing data for previous winters. I wish also to thank David Bell for detailed statistics for the winter of 1996-97. Dave Ferguson and Bruce Lynch kindly found extra counters for their areas. Those who took part were: George Adam, Dave Arthur, David Bell, Paul Blackburn, Dan Carmichael, Graham Craig, Cliff Davies, Dave Ferguson, Alistair Godfrey, David Hill, Garden Johnston, Alan Leitch, Niall Lobley, Bruce Lynch, Ian Montgomery, David Shepherd, Donald Stewart, Jean Stewart, Paul Taylor, Andy Wight and the author. I thank them all and it is testimony to their commitment that all counts were made within one day of the due date despite some miserable weather, with several finding few birds to console them. References Elkins N 2006. Waterfowl counts on the Tay Estuary. Scottish Birds 26:30-38 Elkins N & Lynch B M 1997. Waterfowl] counts on the Tay Estuary, 1985-1995. Scottish Birds 19: 36-54. Laing S & Taylor N W 1993. The status of autumn passage and winter wader populations on the inner Tay Estuary, 1971-89. Perthshire Society of Natural Sciences 1993: 14-29. Musgrove A J, Langston R H W, Baker H & Ward R M (eds) 2003. Estuarine Waterbirds at Low Tide: the WeBS Low Tide Counts 1992/93 to 1998/99. WSG/BTO/WWT/RSPB/JNCC. Thetford. Norman Elkins, 18 Scotstarvit View, Cupar, Fife KYI5 5DX Revised manuscript accepted March 2007 68 SHORT NOTES Gannet laying dates My note (Scottish Birds 26:50) suggested that in 2005 the Atlantic Gannets Morus bassanus on the Bass Rock laid significantly later than in preceding years. On July 15 2006 a visit to the Rock revealed that, judged by eye, the average age of visible chicks in the chapel area was around 26 days (for age criteria see Nelson J B 2002, The Atlantic Gannet second edition, Fenix Books Ltd, Great Yarmouth). None were much older than 5 weeks, which is the age at which the black tips of the growing tail feathers begin to show. An average age of 26 days suggests a mean laying date as least as late as the end of the first week in May. Any pairs which may have had eggs or tiny chicks which I did not see would make the MLD even later. A date of approximately 7 May is again significantly later than in the recent past. The chapel area is not, in my experience, the place in which the oldest Bass chicks occur, so it is possible that the MLD on other parts of the rock, especially the NW slopes, could have been slightly earlier, but this would not alter the general conclusion. It was noticeable that an unusually high proportion of nests which were being attended by one or both adults did not hold a chick. Past experience strongly suggests that they were highly unlikely to have lost their chicks, in which case the conclusion is that they had not laid. These apparently non breeding pairs taken together with the compara- tively late MLD could suggest that something is happening to the relationship between the Bass Gannets and their food supply. Unfortunately it is impossible to judge whether food is being affected by factors other than, or in addition to, the increasing population of Gannets, although I am strongly inclined to dismiss Gannets as the principal factor affecting fish numbers and, through them, gannet breeding biology. SB 27 Notwithstanding the important ‘unknowns’ in the above scenario it is worth recording both the lateness of the MLD and the associated speculation, in case really substantive changes occur in future years, in which case this record could be relevant. It would be useful to be able to compare the growth rate of Bass Gannet chicks at the present time with the rate which I recorded in the early 1960s (Nelson 1964, Ibis 106: 63-77). Whilst it would be somewhat difficult to establish current growth rates (simply because of logistics) it would be perfectly feasible to determine the fledging period, that is between hatching and leaving the nest, by using the live cameras in the Scottish Seabird Centre in North Berwick. Given an adequate sample this could demonstrate both the range and the average length of the fledging period for comparison with earlier records. The fledging period relates to the chick’s food intake and thus to the apparent food supply. This and other comparable observational work would be a valuable use of the extraordinary but greatly under used facilities at the Scottish Seabird Centre. J Bryan Nelson, Mine House, Auchencairn DG7 IRL Revised manuscript accepted August 2006 Great Spotted Woodpecker in Red Squirrel’s drey After strong westerly gales on the night of 31 December 2006 a Red Squirrel’s Sciurus vulgaris winter drey was found blown out of a Serbian Spruce Picea omorika at Cluny House Gardens, Aberfeldy on | January 2007. The drey was composed typically of layers of interwoven twigs, mainly birch and dead spruce, and an inner thick bundle of fresh lichens and mosses. Scottish Birds (2007) Short Notes 69 Buried and rolled up within this was the decomposing complete body of an adult Great Spotted Woodpecker Dendrocopus major. Feathers and fur have both been recorded as materials used by Red Squirrels to line their dreys (Holm, J. The Red Squirrel. Shire Natural History 2005) but there is no reference to a complete body being used. Wendy Mattingley, Cluny House, Aberfeldy, Perthshire, PH15 2JT, matcluny@aol.com Revised manuscript accepted January 2007 Hen Harriers’ bowing behaviour and cocks’ nests Watson (1977, The Hen Harrier Berkhamsted) could only cite a brief description by Delamain (1932, Why Birds Sing London) of bowing by a male Hen Harrier Circus cyaneus ‘repeatedly lower his empty his white breast towards his talons then standing erect again, exhibited his white breast’. Subsequently Dickson (1982, British Birds 75:329-330) recorded bowing behaviour in west Galloway in 1978 and 1979. This note gives 19 additional occurrences of this behaviour and suggests that bowing by Hen Harriers may not be uncommon especially when females are scarce in the local population. Most cocks’ nests were built in April-May. When a female was attracted to such a nest the males would bow apparently in greeting. The majority of adult males bowed from one to 30 times. Immature males - recognised by the grey patches on the greater coverts, yellow eye colour and smaller size - bowed from 2 to 20 times (Table 1). Delamain (1932) states the male ‘then standing erect again, exhibits his white breast’. The opposite occurred at breeding areas in west Galloway. The adult males showed their white rumps to the females 76 times. Immature males did so 25 times; on only 48 occasions the adult males faced the females. When a male did not attract a female, he remained in the area of the cock’s nest for several weeks either adding nest material or skydancing; on one occasion in 1988 an adult male carried out 185 successive dives. The most times that an immature male bowed was 20 in 1979 after delivering prey to an incubating female who was mated to an adult male. Females attracted to a cock’s nest usually built their own nest about 10-15m away; one female built her nest about 250m from the cock’s nest (Dickson 2006, Scottish Birds 26:49-50). A cock’s nest which was built behind a stone wall for 3 years was occupied on the fourth year by a pair of harriers; the female added nest material to the cock’s nest and laid 4 eggs; this was the only case of a female building on top of a cocks nest. Immature males sometimes built up to 3 other nests in the vicinity of the original one but none were successful in attracting females. In 1983 a female laid 3 addled eggs and in July, when the nest should have contained young, the adult male bowed 5 times to the female which was still sitting on the 3 eggs; she continued to incubate for 62+ days. In 1984 an adult bowed to a female at a cock’s nest 30 times because of intrusion by another unmated male from a neighbouring site 1.5km away. In 1988 the most nest material that a male added to a cock’s nest was 41 pieces in 35 minutes; on one occasion he lifted a large wad of material then dropped it, but this male did not breed. In 1997 a 5 year old tagged female was present in a breeding area. The next day a male flew up from a cock’s nest, tail fanned, and landed about Im from an untagged female and bowed once facing her. Fifteen minutes later the male landed beside the tagged female but it was the untagged female who bred laying 5 eggs. All known Hen Harrier breeding areas were located in west Galloway from 1968 to 1999. Although there appears to be much suitable ground throughout the district, breeding has SB 27 70 pag paoiq jou pid paoig Jou pid paoig Jou pid paoig Jou pid pag pacig Jou pid pag pag pag pag paoiq ou pid paoiq you pid paoiq you pid pooiq 10U pid pooiq ou pid paciq ou pid pooiq ou pid paoiq ou pid poig poig pooig }OU pig paoiq ou pid pag pag SSIIINS SUIPIIIG ayeul wu] Mm AN od A ft J[BUIIJ 0} BTR sUON QTBUIDJ 0} OTR SUON QUON Q]eULdJ 0} BTR JTBUIDJ 0} BTR QTBUIDJ 0} BTR d]BUIDJ 0} BTR JTBULIJ 0} BTR SUON aTBULd} OF SPRL LULU] dUON D]BULDJ 0} BTR sUON d[VULIJ 0} BTR DRUID} 0} JR J]BULDJ 0} DTP DRUID} 0} JTRY JUON a]RULd} OF SPRUE LULU] D[BUIDJ 0} OTR SUON dBW} 0} OTR JUON AINOIALYIG SUIMOYG oP oyeu WW] SPIN S[RIN S[PW SPN o[PN SPIN S[PIN S[RIN S[PIN ayeuu WW] a A fat S[PIN SPN S[PIN S[PIN =a fal S[PIN 9k S[PIN S[k oyeuu WWW] S[k S[PIN WING s}sau .syI0D — — eS eed — en oe oN ee a on a n.d a lo OME i ool Da i — —_— JUS L661 9661 v66l 8861 9861 S86l vs6l C86l C861 [86l O86l 6L6l 8L6l SL6l 6961 IVI_ 6661-696] (DMO]]VY ISA4{ Ul SVAAD SUIPIAAG JD SJSAU ,SYIOI JD ANOIADYAG BUIMO| ,SAIIAAD HY UIF *{ I|QeL Scottish Birds (2007) Short Notes Fi been confined to about 70 sq km. The majority of nests were located between the 130 and 230m contours, separated by an average distance of 5.5 km, with the exception of peak ‘vole years’ in 1978-79 and 1997-98 when the average distance was 1.3km. Of 174 nesting areas monitored in 1968-1999, 28 (16%) were cocks’ nests built by adult males, 25 (14%) and immature males 3 (2%) and bowing behaviour (19) were correlated with them. A minimum of 95 males, 20 immature males and 82 females were noted during this study suggesting that females were in the minority. Montagu’s Harriers Circus pygargus built cocks’ nests where males outnumbered females (Weis 1923, Life of the Harrier in Denmark. Copenhagen) while Simmons et al (1987, Reproductive behaviour of Circus cyaneus in North America and Europe: a comparison. Ornis Scandinavica 18:33-41) considered that ‘male built platforms’ were a possible ‘frustration’ or displacement function by males that could not attract a mate. R C Dickson, Lismore, New Luce, Newton Stewart, Dumfries & Galloway DG8 OAJ Revised manuscript accepted August 2006 Hen Harriers driving Goshawk away from a roost At around 1930 hrs on 12 September 1997, I arrived at my usual position to overlook a Hen Harrier Circus cyaneus roost in Galloway. The roost was characteristically (for the region) in an open, wet flow of Phalaris and Juncus on the edge of a pre thicket forestry plantation, which I had monitored regularly since discovering it in September 1993. It was a mild, dry evening with a light, southwest wind and visibility was good. No harriers were over the roost when I arrived, and I began to check the usual pre roosting perches on dykes and fence posts etc. From a distance of 750m and using binoculars, I discerned the pale front of what appeared to be a perched medium sized raptor showing against lower foliage of the dark conifer background. Without a telescope I could not immediately confirm the identification. About half an hour later I became aware of first one and then 2 adult male harriers approaching the roost from the northeast, about 20m and 50m above the ground respectively. However, they continued over the main settling area and began to dive at the perched bird, *yickering’ loudly. They kept this up for a minute or so, usually pulling out of each dive a few metres above the intruder. I became aware of a third adult male harrier over the roost. The perched raptor then made off low through the conifers to its rear, confirming my initial suspicions that it was a Northern Goshawk Accipiter gentilis. The harriers did not pursue the accipiter, but neither did they settle to roost. Although I did not see the Goshawk return, the 3 harriers circled “yickering’ for several minutes, then circled ever higher and eventually soared off to the northeast in the fading light. The nearest alternative roost I was aware of was some 6 km away in the same direction. I saw an adult male harrier roost at the site the following evening, 2 adult males on 17 September and one on each of 24 and 28 September respectively. Since then, the roost has been occupied by harriers on an annual basis, but I have not seen Goshawk again in the area. I have watched Hen Harriers at various roosts for over 20 years, and have often seen them dive at each other, and occasionally at a variety of other birds including Short-Eared owl Asio flammeus and Common Buzzard Buteo buteo. Calling away from the nest, however, is quite rare and [I have never heard anything approaching the squeal call described at winter roosts by Harold (1989. Squeal call of the hen harrier at winter roost. British Birds 82: 93-96) and Clarke (1990. Harriers of the British Isles. qe Shire, Princes Risborough), which is a confrontational intra specific call. The calling heard on this occasion was the characteristic ~yickering’ commonly issued by birds disturbed at or near the nest and described by Watson (1977. The hen harrier. Poyser, Berkhamsted). Successive attacks by a Goshawk on up to 4 ringtail Hen Harriers near a communal roost on a moor in Yorkshire on 17 February 1980, were reported separately by Marshall (1983. Prolonged aerial encounters between Hen Harrier and Goshawk. British Birds 76: 448-449) and Watson (1986. Prolonged aerial encounters between Hen Harriers and Goshawk. British Birds 79: 89-91). Watson considered there to be a territorial aspect to the attacks since Goshawks and Hen Harriers both hunted over the area. Several instances of late afternoon interactions between individual Goshawks and groups of up to 15 harriers have been noted in Suffolk on the Hen Harrier Winter Roost Survey. On all these occasions the harriers eventually settled in their usual roosts (Roger Clarke pers comm). Kropp and Munch (1979. Observations at the roosting sites of overwintering Hen Harriers Circus cyaneus in the Rench Plain, Central Baden. Okologie der Vogel :1:165-179) reported repeated Goshawk disturbance of a Hen Harrier roost in a winter rape Brassica rapa sativa field in southern Germany. During the filming of Skydancer, a 1993 RSPB film on Hen Harriers, photographer Mark Smith captured a remarkable sequence of disturbance by an immature Golden Eagle Aguila chrysaetos of a Highland roost of some 12 harriers, which rose en masse and chased the eagle off. Goshawk and Golden Eagle are 2 powerful avian predators, which could be potential threats to roosting birds on the ground. I am grateful to Roger Clarke for commenting on an earlier draft of this note. SB 27 Chris Rollie, RSPB Scotland, Old School, Crossmichael, Kirkcudbrightshire, DG7 3AP Revised manuscript accepted November 2006 Conflict between Long-tailed Tit and Blue Tit at nest site The area known as the Castle Grounds in Stornoway is the only significant area of deciduous and mixed woodland in the Western Isles. Some woodland birds such as Blue Tit Parus caeruleus, have become established breeders whereas others, such as the Long-tailed Tit, Aegithalos caudatus, which first bred in 1996, have held on precariously with only 2 birds known to be present at the beginning of 2000. On 2 May I located a Long-tailed Tit in the Castle Grounds carrying nesting material. I was puzzled, however when a Blue Tit chased the bird on 2 occasions that same day and on 3 May. On 5 May I saw the Long-tailed Tit carrying more nesting material, this time without harassment and over the following 2 days I was able to trace the nest site to a small Scots Pine Pinus sylvestris . The nest site was about 3 metres above the ground with the entrance formed by a horizontal split in a short stump protruding from the trunk. On the following days I observed the Long-tailed Tit collecting more nesting material and on 8 May it was seen entering the nest cavity with a small insect in its bill, suggesting that its mate was incubating eggs. Two days later on 10 May, I returned to find one of a pair of Blue Tits perched beside the nest entrance pulling out material. A Long-tailed Tit appeared and was chased by both Blue Tits. The Long-tailed Tit then chased one of the Blue Tits until both birds disappeared from my view. A few minutes later one of the Blue Tits and the Long-tailed Tit appeared about 5 metres from Scottish Birds (2007) Short Notes 73 the tree with legs entangled. The Blue Tit appeared victorious and the Long-tailed Tit flew off. However, it returned a few minutes later only to be set upon again by one of the Blue Tits. After more mid air combat both birds fell to the ground a few metres in front of me. The Long- tailed Tit was on its back on the ground protecting itself with outstretched legs from the Blue Tit. This altercation lasted about 5 seconds before both birds flew off. After about 20 minutes one of the Blue Tits entered the nest cavity and began to pull out more material from the Long-tailed Tit’s nest. Thereafter, only Blue Tits were seen at the nest site. On 19 May one of the pair entered the cavity carrying nesting material. On 29 May I observed both birds with food in their bills. On 9 June one was seen carrying a faecal sac out of the nest confirming that young were present. The young could clearly be heard on 18 June and by 21 June the Blue Tits had left presumably with their brood. The site chosen by the Long-tailed Tits for nesting was unusual for a species more likely to build a hanging, or supported nest, outside the trunk. The aggression between the Long-tailed Tit and the Blue Tit witnessed on 2, 3 and 10 May could indicate that the Blue Tit was trying to reclaim its former nest site. A pair of Blue Tits nested at this location for the next 4 years until 2004. The tree was partly destroyed in storms in early 2005, the trunk of the tree breaking at the split where the Blue Tits had nested. A single Long-tailed Tit was seen on 26 July and then in mixed tit flocks on 10 separate occasions until the end of the year. The species has only been a sporadic visitor since then and has returned to its previous status as a nonbreeder in the Western Isles. Bob Wemyss, Willowglen House, Willowglen Road, Stornoway, Isle of Lewis, HSI 2EP Revised manuscript accepted February 2007 Interactions between birds and a day flying bat Whilst visiting the Dun Troddan broch in upper Glen Beag, near Glenelg (57 11° 40” N, 05 35’ 08” W), on 2 September 2006, we watched several Barn Swallows (Hirundo_ rustica) feeding on insects in the clearing above the broch between 2 large trees. A Spotted Flycatcher (Muscicapa striata) was also seen feeding, using both the broch and one of the trees as perches. After a few minutes of observation, during a lull in bird activity at approximately 1235hrs, a bat, presumed to be a Pipistrelle (Pipistrellus pipistrellus) on the basis of flight pattern and size, appeared above the broch apparently also feeding. For several minutes the bat continued flying unhindered in the area above the broch, until several Barn Swallows returned to the area. These flew very close to the bat on at least 5 occasions, causing it to take evasive action by rapidly altering its flight path and height. In addition, the Spotted Flycatcher was also seen to fly out towards the bat from its perch on at least 3 occasions, again causing the bat to take evasive action. We are unaware of previous records of interactions between day flying bats and insectivorous birds, although attempted predation of bats by raptors has been recorded (eg Shimmings 1985 British Birds 78:109). Competition, of an unspecified nature, between insectivorous birds and bats has been suggested as a reason for bats predominantly flying nocturnally, despite lower abundance of insects at night (eg Speakman et al 2000 Oikos 88:75- 86). The observations provided here suggest that interference competition between bats and insectivorous birds of disparate taxonomic groups, mediated through direct antagonistic interactions, may be a significant component of such competition. 74 SB 27 AR Lyndon, 4 Marchbank Way, Balerno, Midlothian, EH14 7EX M Jackson, 25 Jordan Lane, Edinburgh, EH10 4QZ Revised manuscript accepted October 2006 Survey of roof nesting gulls in Caithness The number of Herring Gulls Larus argentatus breeding in Caithness has been in decline for at least the past 3 decades. Operation Seafarer (1969- 70) found a total of 22,483 apparently occupied nests (aon) and this had fallen to 10,033 aon by the time of the Seabird Colony Register Survey in 1985-88 and had further decreased to 3,743 aon by 1998-2002 when the Seabird 2000 Survey was carried out (Mitchell ef al, 2004, Seabird Populations of Britain and Ireland, T & A D Poyser, London). Our own observations show that almost all colonies in the county have reduced numbers and a few have disappeared altogether. The only exceptions appear to be the county’s 2 principal roof nesting colonies, one on_ the buildings of the Dounreay nuclear establishment and the other in Wick. These appear to have been established in the first half of the 1980s but were not fully surveyed until 1994 when there was a total of 131 aon roof nesting birds, nearly all Herring Gulls but including a few Lesser Black- backed Gulls L fuscus, and Common Gulls L canus (Table 1). Only a partial survey of these birds appears to have been undertaken as part of Seabird 2000 and, as it was our impression that these colonies were thriving, we undertook a full survey during the 2006 breeding season. The results of the survey are summarised in Table 1. In total we located 210 + 10 aon Herring Gulls in 4 colonies. The survey in Wick was carried out mainly on 27 May 2006, with a number of further checks over the following 4 weeks. Nests were mostly straightforward to locate from street level or from vantage points around the town, but we suspect that a few were overlooked. The total includes one nest on the roof of the Sheriff Court which was only visible from one particular window at the top of the building. Overall 60 nests (52%) were located on chimney stacks, 22 (19%) on flat roofs, 20 (17%) on sloping roofs, 9 (8%) in the V between 2 sloping roofs, 4 (3%) on the tops of walls of ruined buildings and one (1%) on a ledge. The main concentration of nests was around the harbour and the principal commercial district of the town in Lower Pulteneytown (45 aon) with smaller numbers in the town centre (17 aon) and the residential areas of Hillhead (21 aon), Upper Pulteneytown (31 aon) and Broadhaven (2 aon). Compared with the position in 1994, this represents a spread northward from Lower Pulteneytown into the town centre and Hillhead and south into Upper Pulteneytown. A figure for the number of birds breeding at Dounreay was estimated from information supplied by Johnson Controls, the contractors responsible for the maintenance of the buildings there. They undertook a campaign to control the number of gulls breeding there during 2006, and pricked a total of 238 eggs. At least 3 nests containing a total of 5 chicks were missed, so a minimum of 243 eggs were laid. Unfortunately no record was kept of the clutch sizes but all nests contained either 2 or 3 eggs, and we estimate that 243 eggs is equivalent to about 80- 100 aon (i e assuming a mean clutch size of 2.5 - 3.0). At the end of the breeding season some 70-80 nests were identified, a figure which is not inconsistent with the figure estimated from the egg count, allowing for those destroyed or lost due either to the weather or the birds themselves. No doubt a few nests were overlooked and taking this, and these 2 estimates into account, we think the total number of roof nesting birds here was probably 90 + 10 aon. Scottish Birds (2007) Short Notes Vd Table 1. Roof nesting Herring Gulls in Caithness, 1994-2006. Colony Map ref Dounreay NC9866/67, NC9967 Thurso ND1168 Castletown ND1967 Wick Airport ND3653 Wick ND3650/51, ND3751 Total a Wick, Caithness Bird Report, 56-58. Not counted, presumed nil. In addition 1 aon Common Gull. Not counted; status unknown. p- 0g nO (Gs eS Number of breeding pairs (aon) 19942 2000° 2006 58° ca.504 90 + 10 0 e 3 0 e l 7 g 0 66 h 116 131 : P0410 Data from H Clark and E W E Maughan, 1994, Survey of roof nesting gulls at Dounreay and Data from Mitchell et al, 2004, Seabird Populations of Britain and Ireland. One Herring Gull paired with a Lesser Black-backed Gull; in addition 12 aon Common Gull. Also 86 aon Common Gulls (13 aon on roofs, 73 aon on ground nearby). Not counted, but definitely some birds breeding (Clark & Sellers, unpublished data). Table 2. Productivity of Herring Gulls nesting in Caithness in 2006. Colony Map ref Habitat Productivity chicks per pr nests Wick ND3650/51, roofs Up) a1 ND3751 Castle of Old Wick ND3648 cliff ledges 2.00 27 Badbea ND0920 boulder beach 1.83 89 In addition to these 2 main colonies we found a minimum of 3 aon in Thurso, one on a chimney stack, a second actually in a chimney pot, and a third on a flat roof. All were in the town centre. There are few really suitable vantage points here and it is possible that others were missed. A number of loafing birds were seen, but we were unable to establish unequivocally that these were birds associated with active nests. Finally there was one aon in Castletown, located on a chimney stack. We checked all other coastal villages in Caithness including the Vulcan Naval Reactor Test Establishment and Forss Business Park but found no evidence for any breeding gulls, nor were there any in Halkirk. The 1994 survey found birds nesting on the tops of the Second World War bomb magazines at the north side of Wick Airport and about | km north of the northern edge of Wick town, but we found none here in 2006. 76 At the end of June 2006 we undertook a further survey of the Wick birds to provide an estimate of chick productivity, with the results shown in Table 2. For comparison we include results from a coastal colony at the Castle of Old Wick, a kilometre or so south east of the centre of Wick, and another from a large but declining coastal colony at Badbea, in the south of Caithness. Chick productivity at Dounreay and Wick was higher than at Badbea, for which the produc- tivity was typical of other coastal colonies in the south of Caithness (R M Sellers, unpublished data). The colony at the Castle of Old Wick was intermediate between these 2 groups, perhaps because the birds there were able to take advantage to some extent of the scavenging opportunities in Wick. We have no direct evidence of this at the Castle of Old Wick colony, but at another coastal colony at Papigoe, a kilometre or so east north east of Wick, regurgitates obtained during ringing operations contained amongst other things the remains of cooked human foodstuffs. This survey shows that the numbers of roof nesting gulls in Caithness has increased by about 60% in the 12 years since the earlier survey, the colonies at Dounreay and Wick being the only ones of any size in Caithness to have increased during this period, so far as we are aware. By contrast the small roof nesting colony at Wick Airport has disappeared. Birds definitely bred here in 1995 (ca 10 pr Herring Gull, 2 pr Great Black-backed Gulls and | pr Common Gull), but breeding appears to have ceased a year or 2 later. Despite its proximity to the runways, which at their closest are only about 250 m away, we understand from the airport authorities that no direct attempts have ever been made to discourage nesting here. However, around this time gulls had taken to roosting on the runways at night and as a preventative measure gas guns were used to scare the birds away. This appears to have been successful and, though the evidence SB 27 is circumstantial, we suspect that the increased levels of disturbance may also have resulted in the breeding colony being deserted. We are indebted to Mike Rennie, Sandy Moran and the staff of Johnson Controls for making available the information on gulls at Dounreay, and UKAEA for their permission to publish this data. Our thanks also to Matt Parsons and JNCC for providing information from Seabird 2000. Hugh Clark, 3 Lindsay Place, Wick, Caithness, KW1 4PF Eric W E Maughan, Burnside, Reay, Caithness, KW14 7RG. Robin M Sellers, Crag House, Ellerslie Park, Gosforth, Cumbria CA20 IBL. (e-mail:sellers@craghouse7 freeserve.co.uk) Revised manuscript accepted October 2006 Late breeding season in Northern Gannets at Hermaness, Shetland in 2006 Nelson (Scottish Birds 2006 26:50. Mean laying dates in the Northern Gannet) noted that the Gannets Morus bassanus on the Bass Rock were significantly later breeding in the 2005 season and stated that this had not previously been observed before at that colony. On 26 June 2006, together with other members of the Shetland Ringing Group, I visited the Gannet colony at Hermaness, Unst intending to ring a sample of chicks. Unfortunately, few chicks were old enough to ring and many nests had newly hatched chicks or well incubated eggs, indicating that most females had laid about 3 weeks later than usual. Scottish Birds (2007) The colony has been visited on similar dates or slightly earlier in many previous years since 1979, (but not 2005) and such a late season has not been recorded previously, the chicks normally being ready to ring in large numbers by the third week of June. Nelson suggested that the late laying *...could potentially be linked to the relative paucity of sand eels Ammodytes spp within the foraging range of Bass Gannets. These fish are an important part of the Gannet’s diet prior to egg laying’. The effects of the lack of availability of sand eels are well documented in other seabirds but there is little information on the food of Shetland Gannets during the prebreeding season, although there is information on diet collected during ringing visits. It is worth noting that in 1981 sand eels made up 90% of the diet of Gannets at Hermaness but this had fallen to 15% by 1986, with Herring Clupea harengus and Mackerel Scromber scrombus being the most important species. By 1997, sand eels were again available and made up 50% of the diet (The Birds of Shetland. 2004. Helm. London). In 2003, 2004 and 2006 sand eels were not recorded in the diet at all, with the main species being Mackerel and Herring with small numbers of whitefish Gadus spp. There is no evidence that breeding success has been affected by this change in diet away from sand eels. Given that the laying dates for this species have been regarded as very consistent from year to year, it would be valuable to have information on the phenology of the breeding season for Northern Gannets in other colonies and in other years. J David Okill, Heilinabretta, Trondra, Shetland, ZE1 0XL Revised manuscript accepted October 2006 Short Notes 77 Eurasian Sparrowhawk attacking Red Squirrels It is well known that Eurasian Sparrowhawks Accipiter nisus will attack birds at garden feeders (eg Wilson and Weir 1989, Scottish Birds 15:126-130). Newton (1986, The Sparrowhawk, Calton) could only document a few Red Squirrels Sciurus vulgaris as prey of Sparrowhawks in Europe, mainly in forest plantations, while no Sparrowhawks have been reported attacking Red Squirrels at garden feeders. From November 2005 to May 2006 up to 3 Red Squirrels fed at nut feeders and fat balls on Rowans Sorbus aucuparia and Hawthorns Crataegus mongyna in a garden in Galloway. Eight direct attacks were made on squirrels by a male Sparrowhawk attempting to make a kill with its feet and legs extended but none were successful; to evade capture the squirrels quickly climbed a tree and lay flat, hugging a branch. During the same months, 30 attacks on birds by a male Sparrowhawk were recorded of which 6 (20%) were successful: 2 Chaffinches Fringilla coelebs, 2 European Goldfinches Carduelis carduelis, one European Greenfinch C chloris and one Eurasian Siskin C spinus. R C Dickson, Lismore, New Luce, Newton Stewart, Dumfries & Galloway DG8 OAJ Revised manuscript received August 2006 78 SB 27 Dr Pamela Mary Collett 1925 — 2006 The SOC has lost a strong supporter in Pam who died on 13 March 2006. Less than 3 months earlier she had celebrated her 80th birthday with friends and had been ‘on good form’, reminiscing over old times and birdwatching trips both at home and abroad. After studying medicine in London, Pam McMorran worked at a paediatric hospital in Norwich where, through friendship with a staff nurse from the north, she was introduced to Caithness. She moved to Thurso in the late 1950s and was a highly respected GP for over 30 years, continuing as police doctor for several years after retiring from general practice in 1990. Apart from birds, Pam had a sound knowledge of wild flowers and her later holidays had been in search of these; she claimed that her eyesight was not as good as it used to be, and plants didn’t fly away. She also loved music, had learned some Gaelic, the better to communicate with some of her patients, and was active in country dancing. She loved dogs and apart from her last year or so she had always shared her life with a dog. After retirement, and right up until her death, she helped at Riding for the Disabled. She also had a fine library of bird books which she bequeathed to the SOC. Pam must have joined the SOC soon after arriving in Thurso, perhaps about the time of the birth of Scottish Birds; certainly the first time she appeared in print was with a short note on a Marsh Harrier in Caithness dated 5 March 1961, (Scottish Birds 1: 426-427). In that, she mentioned her familiarity with the species in Norfolk so her passion for birds must have come early in her life. In 1970, when I met her, she was in the driving seat of birdwatching in Caithness, organising the cover for the first breeding atlas as BTO local representative. She was also the local representative of the RSPB and the Local Recorder from 1970-1983. During these years she also produced the Caithness Bird Report and many of us remember her cajoling to submit our records at the end of each year, threatening to omit them if we missed her deadline. Of course she never did. All this she somehow fitted around her busy professional commitments, working nights and weekends as well as days. Pam had a love of islands and early on she joined groups surveying the birds on Stroma, the Pentland Skerries and Sule Skerry. She was possibly the first woman ever to land on Sule Stack; the Duchess of Bedford in 1914 had not been able to land (Scottish Birds 1: 266). She visited St Kilda and knew Fair Isle well from Scottish Birds (2007) regular visits to friends on the island. After marriage to Tony Collett in 1971 together they reached Spitzbergen, Mongolia and Peru as well as the Middle East and North Africa. Neither did they neglect islands nearer home: the writer’s tent still has a hint of eau de Fulmar after they borrowed it for a few days on North Rona. Although she continued to travel after Tony died in 1990, a spark had gone. She did, though, realise an ambition of which she was proud, to visit all the island groups on the mid Atlantic Ridge. Pam loved Caithness, the wild coast, and especially the vast peatlands, and I remember a trip with her when we surprised a Wood Sandpiper from a nest at our feet. We looked at each other, she grabbed me, and we danced around like children, having finally proved breeding in the county, before withdrawing to allow the bird to return. For Pam, the bird’s welfare was always paramount. Birds need more champions like Pam and her many friends in the SOC, and we in Caithness, miss her. My thanks to David Glass, Peter James and Sinclair Manson for memories of Pam, and to Peter James for the photograph of Pam en route to the Pentland Skerries in 1965. Stan Laybourne i) CORRECTION Crossbill numbers in old pinewoods on upper Deeside and Speyside In our paper in Scottish Birds 26, 2006, pp 51- 55, it was invalid to use proportionate change on scores. This does not affect conclusions, but we make corrections below. Page 51, second paragraph, line 10: ‘This also held with change, ranking the change in score from year | to year 2, eg +2 for 3 rising to 5, or = 3.tor 5 falling 10.2 (n= 17,7r,=—0.25). Page 51, right column, second paragraph, line 3: ‘likewise for change’, and line 4: ‘r, =—0.01’. Table 2. Title and Correlations 2 and 4, alter ‘proportionate change’ to ‘change’. In Correlation 2, change 0.09 to 0.07. In Correlation 4, change —0.15 to —0.06. Proportionate change is valid for use on the index in Correlation 6. A Watson & M Marquiss 50 Advice to contributors Authors should bear in mind that only a small proportion of the Scottish Birds readership are scientists and should aim to present their material concisely, interestingly and clearly. Unfamiliar technical terms and symbols should be avoided wherever possible and, if deemed essential, should be explained. Supporting statistics should be kept to a minimum. All papers and short notes are accepted on the understanding that they have not been offered for publication elsewhere and that they will be subject to editing. Papers will be acknowledged on receipt and are normally reviewed by at least 2 members of the editorial panel and, in most cases, also by an independent referee. They will normally be published in order of acceptance of fully revised manuscripts. The editor will be happy to advise authors on the preparation of papers. Reference should be made to the most recent issues Of Scottish Birds for guidance on style of presentation, use of capitals, form of references, etc. Papers should be typed on one side of the paper only, double spaced and with wide margins and of good quality; 2 copies are required and the author should also retain one. We are also happy to accept papers on disk or by email at: mail@the-soc.org.uk, stating the type of word processing package used. If at all possible please use Microsoft Word . Contact the Admin Officer on 01875 871330 for further information. SB 27 Headings should not be underlined, nor typed entirely in capitals. Scientific names in italics should normally follow the first text reference to each species unless all can be incorporated into a table. Names of birds should follow the official Scottish List (Scottish Birds 2001 Vol 22:33-49). Only single quotation marks should be used throughout. Numbers should be written as numerals except for one and the start of sentences. Avoid hyphens except where essential eg in bird names. Dates should be written: ...on 5 August 1991]...but not ...on the Sth... (if the name of the month does not follow). Please do not use headers, footers and page numbers. Please note that papers shorter than c700 words will normally be treated as short notes, where all references should be incorporated into the text, and not listed at the end, as in full papers. Tables, maps and diagrams should be designed to fit either a single column or the full page width. Tables should be self explanatory and headings should be kept as simple as possible, with footnotes used to provide extra details where necessary. Each table, graph or map should be on a separate sheet, and if on disc each table, graph, map etc should be on a separate document. Please do not insert tables, graphs and maps in the same document as the text. Maps and diagrams should be either good quality computer print out and in black and white (please do not use greyscale shading) or drawn in black ink , but suitable for reduction from their original size. Contact the Admin Officer on 01875 871330 for further details of how best to lay out tables, graphs, maps etc. The Scottish Ornithologists’ Club was formed in 1936 to encourage all aspects of ornithology in Scotland. It has local branches which meet in Aberdeen, Ayr, the Borders, Dumfries, Dundee, Edinburgh, Glasgow, Inverness, New Galloway, Orkney, St Andrews, Stirling, Stranraer and Thurso, each with its own programme of field meetings and winter lectures. The Waterston Library at the Club’s headquarters at The Scottish Birdwatching Resource Centre, Waterston House, Aberlady, East Lothian EH32 OPY is the most comprehensive ornithological library in Scotland and is available for reference 7 days a week. A selection of Scottish local bird reports is held at headquarters and may be purchased by mail order. Check out our website for more information about the SOC and other bird related organisations: www.the-soc.org.uk An annual issue of Scottish Birds is published each year, in June. Scottish Bird News is published quarterly in March, June, September and December and the Scottish Bird Report is annual. The Scottish Ornithologists’ Club publishes the annual Scottish Raptor Monitoring Report on behalf of the Scottish Raptor Monitoring Group with grant aid from Scottish Natural Heritage. It is sent to all members. Bird names used in publications follow those used in the Scottish List (Scottish Birds Vol 22(1): 33-49). Copies of these reports may be purchased by non members on application to the SOC. SOC annual membership subscription rates Adult £ 25.00 Family (2 adults and any juniors living at the same address) £ 35.00 Junior (aged 17 or under) £ 10.00 Student (aged 18 to 24 and in full-time education) £ 10.00 Concession (Pensioner) £ 16.00 Concession (Unwaged) £ 16.00 Joint Concession £ 21.00 Life £500.00 Life Family £700.00 Overseas supplement £ 10.00 All subscriptions may be paid by Direct Debit and Gift Aided. Subscriptions paid by Direct Debit greatly assist the Club. Please ask for an application form by telephoning 01875 871330 or visiting the club’s website at www.the-soc.org.uk SMITHSONIAN TT Scottish Birds _ Iii Volume 27 June 2007 3 9088 Contents Main Papers Amendments to the Scottish List: species and subspecies - Ronald W Forrester 1 The seasonal and altitudinal distribution of birds in a Scottish Borders valley - DG Pyatt 15 Autumn skua migration at Hound Point, Lothian - Christopher J McInerny & Mervyn Griffin 24 Numbers, distribution and habitats of Black Grouse leks in Lochaber - Alison MacLennan & Ron Summers 32 An unsuccessful attempt to establish Northern Goshawks in Argyll - Steve J Petty 40 The White-tailed Eagle in Sweden — past, present and future - Bjorn Helander 46 Trends in breeding duck populations at Loch Leven, Perth & Kinross - Alan Lauder 55 Low tide counts on the Firth of Tay - Norman Elkins 64 Short Notes Gannet laying dates - J Bryan Nelson 68 Great Spotted Woodpecker in Red Squirrel’s drey - Wendy Mattingley 68 Hen Harriers’ bowing behaviour and cocks’ nests - R C Dickson 69 Hen Harriers driving Goshawk away from a roost - Chris Rollie 7 Conflict between Long-tailed Tit and Blue Tit at nest site - Bob Wemyss 12 Interactions between birds and a day flying bat - AR Lyndon & M Jackson Ne Survey of roof nesting gulls in Caithness - Hugh Clark, Eric W E Maughan & Robin M Sellers 74 Late breeding season in Northern Gannets at Hermaness, Shetland in 2006 - J David Okill 76 Eurasian Sparrowhawk attacking Red Squirrels - R C Dickson T. Dr Pamela Mary McMorran 1925 — 2006 - Stan Laybourne 78 Crossbill numbers in old pinewoods on upper Deeside and Speyside - A Watson & M Marquiss 19 Advice to contributors 80 Front Cover Gadwall by Edmund Fellowes Published by the Scottish Ornithologists’ Club, The Scottish Birdwatching Resource Centre, Waterston House, Aberlady, East Lothian EH32 0PY. © 2007 [_ AO Se Ce) dS ‘o) Scottish © Birds “ THE JOURNAL OF THE SOC Vol 28 June 2008 Rookeries in Caithness in 2007 Behaviour and diet of non breeding Snowy Owls on St Kilda Post breeding movements of Sandwich Terns in the Firth of Forth Wintering wader surveys on the Isle of Tiree, Argyll a ' A a Golden Eagle chick in nest © Clive R McKay Scottish Birds (2008) Scottish Birds — The Journal of the SOC Editor: Dr Stan da Prato. Assisted by: Dr I Bainbridge, Professor D Jenkins, Dr M Marquiss, Dr J B Nelson and R Swann. Business Editor: Wendy Hicks, Officer Manager, SOC, Scottish Birdwatchers’ Centre, Waterston House, Aberlady, East Lothian EH32 OPY. Scottish Birds, the official journal of the SOC, publishes original material relating to ornithology in Scotland. Papers and notes should be sent to The Editor, Scottish Birds, SOC, Scottish Birdwatchers’ Centre, Waterston House, Aberlady, East Lothian EH32 OPY. Scottish Birds is published each year in June. Scottish Birds is issued free to members of the SOC, who also receive the quarterly newsletter Scottish Bird News, the annual Scottish Bird Report and the annual Scottish Raptor Monitoring Scheme report. Published by: The Scottish Ornithologists’ Club, Scottish Birdwatchers’ Centre, Waterston House, Aberlady, East Lothian EH32 OPY. Scottish Charity Reg. No SC 009859 Design by: Pica Design, 51 Charlton Crescent, Aboyne, Aberdeenshire AB34 5GN. Printed by: Meigle Colour Printers Ltd, Block 1, Tweedbank Industrial Estate, Galashiels TD1 3RS 2 SB 28 Rookeries in Caithness in 2007 ROBIN M SELLERS, HUGH CLARK & STAN LAYBOURNE A full survey of Caithness rookeries was undertaken during the 2007 breeding season. This found a total of 34 rookeries varying in size between one and 791 nests, the majority built in stands of Sycamore, and containing a total of 6556 nests. This represents an increase of 7% over the previous survey of the county’s Rooks in 1975. There was, however, much individual variation between rookeries in how numbers had changed between the 2 surveys. One large 1975 rookery was completely deserted and 3 others showed declines of over 50%. Human persecution was certainly responsible in one instance and disturbance of some kind was implicated in the others. Introduction Rookeries are either absent or occur at much lower densities in the Highlands than elsewhere in Scotland, except on the coastal hinterland along the north western side of the Moray Firth between Inverness and Wick. Between the latter place and Thurso densities are more typical of those in the Lowlands, and Caithness represents the most northerly part of Britain with good numbers and high densities of breeding Rooks Corvus frugilegus (Gibbons et al 1993, Thom 1986). The last complete survey of Caithness rookeries was carried out in 1975 as part of the national census undertaken in that year (Castle 1977, Sage & Vernon 1978), since when at least one large rookery is known to have disappeared, whilst Manson (2002) mentions that there is anecdotal evidence from local landowners that numbers generally have decreased in recent decades. Given that there are reported to have been recent declines in Sutherland (Bremner & Macdonald 1996) and East Ross (McGhie 2000), we undertook a complete survey of Caithness rookeries during the 2007 breeding season. Methods Locating rookeries is fairly straightforward but to ensure as complete coverage as possible we checked all stands of mature broad leaved trees in Caithness including those planted around farm buildings, some 69 sites in total. We placed particular emphasis on all sites used in the 1975 survey, and a number of others identified in preliminary surveys carried out in the breeding seasons of 2004, 2005 and 2006. We also checked some, but by no means all, of the conifer plantations that have sprung up in the county in recent decades, although there is no history of Rooks using such sites in Caithness. Given that we were unable to find any nests in those plantations that we did check, we felt justified in not investigating them all. Isolated nests or small groups of nests can be difficult to find, and, with the exception of the isolated nest at Shebster (see below), all single nests that we checked during the survey were those of either Carrion Crows C corone or Hooded Crows C cornix or their hybrids. It 1s possible that one or 2 others were overlooked, but we think it very unlikely that this had a significant effect on the final population estimate. With 2 minor exceptions, the number of nests in each rookery was determined in the period 19-22 April 2007, representing a reasonable compromise between counting early enough to avoid problems of visibility after the leaf buds have fully opened, and late enough to include 3 Rookeries in Caithness in 2007 Scottish Birds (2008) Rook with nestlings © Bobby Smith the last nests to be constructed. The 2 exceptions were the small rookeries at Shebster and Gerston which were counted in the last week of April and the first week of May respectively. For comparison the 1975 survey was carried out mainly in the third week of April, with a couple of counts made in the second and fourth weeks of April, and a few in May. So far as practical, counts of nests in broad leaved trees were made from below, whilst those in conifers were made from a more oblique angle or, where practical, looking horizontally into the tops of the trees. We have assumed that all nests counted were actively occupied and report the counts as ‘nests’. Whenever possible we _ sought information from local residents about the recent history of their rookeries, particularly with respect to disturbance and/or persecution as well as nuisance levels and other imputs. For the purposes of analysis we have defined a rookery as any group of Rook nests separated by more than 500 m from the next nearest group. This differs from the definition used in the 1975 national survey where separation by 100 m was the criterion employed, but reflects the results of more recent research on Rook ecology (eg Griffin Table 1. Rookeries in Caithness in 1975 and 2007 SB 28 & Thomas 2000, Mason & Macdonald 2004). We have, however, been careful to identify those groups of nests separated by 100 m to permit comparison with the results of the 1975 survey. A detailed inventory of the material collected during the survey, together with a list of other sites checked and a summary of the 1975 results is given with the Supplementary Material (see Sellers et al 2007), copies of which have been lodged with the SOC’s Waterston Library in Aberlady and the Carnegie Library in Wick. Results The 2007 survey of breeding Rooks in Caithness found a total of 6556 nests in 34 rookeries; other details are summarised in Table 1. Sage & Vernon (1978) considered that their totals for the 1975 National Rookery Survey were probably an underestimate of the true population, possibly by as much as 10%. Although we were able to make virtually all our counts at about the optimum date, there are several other sources of error, not least the difficulty of distinguishing individual nests in large, dense colonies, and we suspect that our count has an uncertainty only a little better than that of the 1975 survey. Rookery Map ref Number of nests a 1975 2007 Barrock House ND2862/2863 7167 =~) 20% a Barrogill Mains ND2973 22» (2) 166 —(Q) Bower ND2363 ) 28 (1) Braal Castle ND1360/1460 520 -iZ) AST QZ) Buckies ND1163 31- (1) 161 (1) Calder Mains NDO959 198 (1) 184 (1) Castletown ND1967/1968/2067/2068 756 (6) 7A _ VO) Clayock ND1759 0) i fell) Durran Mains ND1962 0 19 CR) cio -aei(cls) Loch Scarmclate ND1859/1860 628 (4) 0 c Lybster ND2436/2536 es aan a) 2NBt (2) Lynegar ND2256/2257 150. @) Sil Ely) Mains of Watten ND2556 5o7? CE) 251. (1) Oldhall House ND2056 304 =s(1) oo) ACL) Olrig ND1866/1867 601 (3) 165- -@) d Reaster House ND2564 0 30.—Ss (1) Roadside ND1560 0 id > (hb) Shebster NDO163 0) ae) Skinnet ND1261 0 29 (1) Stirkoke Wood ND3251 0 259% =(ele) Swiney ND2335 0 29. (1) Thurdistoft ND2067 Ae nk) 64 (1) Thurso ND1167/1168 540 (4) 480 (4) Upper Dunn ND1956 oat oly) 0 Watten ND2454 52: = (lh) 194 (1) Wester Watten ND2355 95 (1) 162 (1) Westfield ND0664 492 (3) 190 (1) Wick ND3650/365 1 220. (7) 292 2.8) e Total nests 6124 6556 Total rookeries 24 (48) 34 (52) (a) The figures in brackets show the equivalent number of rookeries assuming that a rookery is defined as a group of nests separated by at least 100 m from the next nearest group; (1) implies that the rookery is the same whichever definition is used. (b) Rookery deserted part way through 2007 breeding season, probably due to disturbance resulting from construction of a footpath immediately below the rookery. (c) In 1975 one rookery (100 m separation definition) straddling the boundary between ND1859 and ND1860 was counted as 2 rookeries; we have preferred to consider it as a single rookery (100 m separation definition). (d) There was a shift of several hundred metres in the location of the Olrig rookery between the two surveys. (e) 2-3 nests in Thurso Road, Wick were knocked down 3-4 weeks before the survey; we assume that these birds moved to breed elsewhere in Wick and therefore they have not been included in the total for Wick. Rookeries (500 m definition) varied in size between one nest and 791 nests (range 2-767 nests in 1975), with the 3 largest holding 2015 nests (31% of all nests) and with 86% of nests occurring in the top 50% of rookeries. The largest individual group of nests (that is separated by at least 100 m from the next nearest group) in 2007 was the Forss House rookery with 743 nests, whilst the largest in 1975 was the section of the Thurso rookery around Ormlie Lodge and Miller Academy with 471 nests. There was a substantial decrease in the mean size of rookeries (SOO m definition) between the 2 surveys from 255 nests in 1975 to 193 nests in 2007; the mean size of the rookeries on the 100 m definition, however, remained almost the same at 128 nests in 1975 and 126 nests in 2007. Rookeries were mainly distributed in a broad band across northern Caithness, the majority of ] - 50 nests @ 51-200nests | | 200+ nests Figure 1. Distribution of Rookeries in Caithness in 2007 (Extent of arable and stock rearing land based on Omand 1972; + shows site of deserted Loch Scarmclate rookery) SB 28 them being found between Thurso and Wick (Figure 1). There was a clear association with the principal arable and stock rearing parts of the area; none was associated with hill pasture, moorland or the flow country of south and-west Caithness. The mean nearest neighbour distance between rookeries was 2.3 km (range 0.7-6.1 km). The coastal strip south west of Wick held just 3 relatively isolated rookeries, all near Lybster, some 15 km from the next nearest in Caithness, that at Hempriggs House, and 27 km from the next nearest to the south west, one of c 50 nests at Helmsdale in East Sutherland. On the north coast of Caithness east of Dunnet Head there was just a single rookery, that at Barrogill Mains, occupying the only stand of broad leaved trees in north Caithness between Dunnet Head and Duncansby Head. Similarly the lack of rookeries along the coastal strip in the north east of the county between Duncansby Head and Wick reflects the absence of any broad leaved trees here. It is also noteworthy that the few rookeries in the north west of Caithness were somewhat more widely spaced than those in either the central or eastern part of the county. The majority of nests were in Sycamores, one of the commonest and most widespread broad leaved tree species in Caithness. At least 8 other tree species were used for nesting as indicated in Table 2. Of these spruce, Wych Elm and Beech were the most important, accounting for about 11% of nests. The majority of rookeries were built in trees planted around farmsteads or the larger type of country house: there were 26 such rookeries, and they contained 64% of all nests. A second group of rookeries containing 28% of nests was found in small stands of trees in the county’s main population centres: Castletown, Halkirk, Thurso, Watten, Wick and one part of the Lybster rookery. The only ones in woods away from human habitation were those at Stirkoke Wood, Knockdee and Olrig, which together accounted for 8% of nests. No rookeries Scottish Birds (2008) Rookeries in Caithness in 2007 7 Table 2. Tree species used by Rooks nesting in Caithness in 2007 Species Sycamore Acer pseudoplatanus Spruce Picea spp a Wych Elm Ulmus glabra Beech Fagus sylvatica Ash Fraxinus excelsior Horse Chestnut Aesculus hippocastanum Pine Pinus spp b Silver Birch Betula pendula Total Total nests % 5676 86.5 295 4.5 221 3.4 203 oul 84 3 51 0.8 pa | 0.3 5 0.1 6556 (a) Includes both Norway Spruce P abies and Sitka Spruce P sitchensis. (b) Includes some Scots Pines P sylvestris were located on high ground and the majority were at altitudes of less than 50 m asl. The highest was the isolated nest at Shebster at 80 m asl; the next highest were the neighbouring rookeries at Lybster and Swiney, both 75 m asl. Discussion Population trends The figure of 6556 nests for the size of the breeding population of Rooks in Caithness found in this survey is some 7% higher than that found in 1975, and suggests that fears concerning population declines in the area are unfounded. Indeed, the population appears to have increased gradually throughout the past half century, for a survey in 1945-46 found a total of 5371 nests (Sage & Vernon 1978) and another in 1971 some 5700 nests (D Stark quoted in Manson 2002). Overall these figures represent an average growth rate of 0.4% per annum. No complete census of Caithness Rooks was undertaken in the 19th century, but from the partial information summarised by Harvie- Brown & Buckley (1887) it appears that there were fewer rookeries then and that individual rookeries were smaller. The largest they were aware of was one at Westfield with 250 pairs, whilst the majority of those for which they quote a size held between 20 and 200 pairs. Limited though the evidence is, it appears that the Caithness Rook population may have been on the increase for as much as a century and a half. Rookery distribution In both the 1975 and 2007 surveys rookeries were found only in the main arable and stock- rearing part of the county which runs roughly east to west across the northern half of the county and down the county’s east coast (Fig 1). Even within this zone, however, they were absent from peripheral areas, for instance around Reay in north west Caithness or the coastal strip in the south east of the county, from areas where broad leaved trees were absent for instance between Wick and Duncansby Head and between Hempriggs and Lybster or in short supply such as the north coast between Dunnet Head and Duncansby Head. Perhaps the most striking feature of rookery distribution in Caithness, however, is their close association with country houses or farmsteads. Partly, of course, this simply reflects the availability of places for nesting, but many appear to be of considerable age (well over a century), and it may be, as Cocker (2007) has described, that the pattern was established in Victorian times when Co SB 28 Table 3. Changes in the size of Caithness rookeries 1975-2007 Change in rookery Number of rookeries size 1975-2007 Not in 1-50 nests 51-200 nests 200+ nests Total existence in 1975 in 1975 in 1975 in 1975 deserted - 0 ] 2 >20% decrease - ] 0) 4 5) <20% change - 0 Z 3 5 >20% increase - 12 0 0 12 established 12 - - [2 it was fashionable for the landed gentry to plant trees close to their residences specifically to attract Rooks. Changes at individual rookeries 1975-2007 The small overall change in the population between the 1975 and 2007 surveys belies much larger changes at individual rookeries, as summarised in Table 3. Between the 2 surveys 2 were lost including the very large one at Loch Scarmclate and 5, including 4 large ones showed reductions of well over 20% (>50 % in the case of 3 of the 4 large rookeries). On the other hand 12 of the rookeries found in 2007 had been established since the previous survey, and, though many were fairly small, they included the substantial one at Stirkoke Wood (259 nests in 2007). There had, moreover, been increases of over 20% at another 12 rookeries. In general we have little direct evidence on the causes of the declines; however, the large reduction in numbers at Barrock House has undoubtedly been due to human intervention for control measures have been undertaken here in recent years, in response to the noise and mess that the birds have created, and we suspect that this is true of several of the other rookeries that have decreased markedly in size, especially those such as Westfield, Olrig and Lybster which are close to human habitation. The Westfield rookery appears to have been deserted for a while in the late 1980s for no breeding Rooks were found in NDOS, the 10 km square in which Westfield is located, at the time of the 1988-91 Breeding Atlas Survey (cf Gibbons er al 1993). However there was a rookery here in 1991 according to the Caithness Bird Report for that year; perhaps the year in which it was re established. The growth of the Forss House rookery (first recorded in 1975 when there were 3 nests - see Table 1) may be related to the disruption at Westfield. By contrast disturbance at the Braal Castle rookery, which was subject to much shooting in the 1970s and 1980s (S Laybourne, pers obs), did not result in any significant long term population decline. Other, more recent cases of disturbance have been recorded from Lybster, where a small, slightly separate part of the rookery in the centre of the village was eliminated after felling of the trees there between the 2006 and 2007 breeding seasons, and from Halkirk where part of the rookery was similarly treated in the same winter period. These birds may well be those which formed the small rookery at Gerston, which is only about 700 m away. Furthermore that part of the Thurso rookery in Sir John’s Square was eliminated following pollarding of the trees there in c 2005. Considerable disturbance to those parts of the Thurso rookery around Ormlie Lodge and Miller Academy took place in the late Scottish Birds (2008) Rookeries in Caithness in 2007 ) 1980s when the trees here were heavily pruned and some cut down. We are also aware that shortly after our counts at the Roadside rookery in 2007 several trees containing Rook nests were cut down. Something similar happened in Thurso Road, Wick, when 2-3 nests were knocked down some 3-4 weeks before we carried out our survey there. Whether human disturbance was responsible for the desertion of the Loch Scarmclate rookery, which was located in woodland some distance from the nearest houses or farm buildings, is less clear. No shooting has taken place here during the past 2 decades or so according to the local landowner. It is noteworthy, however, that several new, comparatively small rookeries have been established within a few kilometres of the site, including those at Clayock, Knockdee, Durran Mains and possibly at Roadside too, the sort of pattern to be expected if disturbance was responsible for the desertion of the Loch Scarmclate rookery. Indeed careful examination of the data in Table | reveals that within c 10 km of 4 of the 5 sites at which there have been large declines in numbers, the Rook population has shown net increases of at least the same magnitude. The exception concerns the relatively isolated group of 3 rookeries near Lybster. If there has been some shift of birds from these to other rookeries then it was presumably to the next nearest group, those around Wick, a minimum of 15 km away. Consistent with this several rookeries in the latter area have shown substantial increases in the recent past. We conclude that disturbance if not outright persecution continues to be a feature of the lives of Caithness Rooks. Persecution by man is by no means a new phenomenon in Caithness. Harvie-Brown & Buckley (1887) noted that gamekeepers... in Caithness, almost more than any other county in Scotland, wage war upon egg-destroying vermin. They go on to note that the rookery at Braal Castle was persecuted in the nineteenth century to the extent that the birds abandoned the site and bred in the heather a short distance away. The surprise perhaps is that no less than 11 of the 15 rookeries mentioned by Harvie- Brown & Buckley are still in existence and 2 others are within a kilometre or so of existing rookeries and were probably precursors of them; it is probably no coincidence that both the original sites were at large country houses. We cannot be certain that all 11 of these rookeries have been continuously occupied ever since, or, like that at Braal Castle, have been vacated for a time following persecution and subsequently recolonised. Persecution remains a threat to the Rooks of Caithness, but at its current levels it appears not to be having any significant impact on breeding numbers. Within the constraints imposed by the availability of suitable breeding and feeding places, it appears, however, to be an important factor in defining rookery size and distribution. Acknowledgements We are indebted to the many landowners who gave us permission to counts Rooks on their properties, to Eric Maughan for drawing our attention to the single Rook nest at Shebster and Donald Omand for pointing out to us the small rookery at Gerston. References Bremner D M and Macdonald M 1996. Decline of Sutherland rookeries. Scottish Birds 18:248. Castle M E 1977. Rookeries in Scotland 1975. Scottish Birds 9:327-334. Cocker M 2007. Crow Country. Jonathan Cape, London. Chapter 13. Gibbons D W, Reid J B and Chapman R A 1993. The New Atlas of Breeding Birds in Britain and Ireland: 1988—1991.T & AD Poyser, London. Griffin L R and Thomas C J 2000. The spatial distribution and rook size of (Corvus 10 frugilegus) breeding colonies is affected by both the distribution of foraging habitat and by intercolony competition. Proc Roy Soc Lond B 267:1463-1467, Harvie-Brown J A and Buckley T E 1887. A Vertebrate Fauna of Sutherland, Caithness and West Cromarty. David Douglas, Edinburgh. Omand D (ed) 1972. The Caithness Book. Highland Printers, Inverness, p 178. Manson S A M 2002. A History of Caithness Birds 1769 to 2001. Mason,C F and Macdonald S M_ 2004. Distribution of foraging rooks, Corvus frugilegus, and rookeries in a landscape in eastern England dominated by winter cereals. Folia Zool, 53:179-188. McGhie H 2000. East Ross Rookeries in 1998-99. Scottish Birds 21:36—-42. Sage B L and Vernon J D R 1978. The 1975 national survey of Rookeries. Bird Study 25:64-86. Sellers R M, Clark H and Laybourne S 2007. Rookeries in Caithness in 2007 - Supplementary Material. Thom V M 1986. Birds in Scotland. T & A D Poyser, Calton. SB 28 Robin M Sellers, Crag House, Ellerslie Park, Gosforth, Cumbria CA20 1BL (e-mail: sellers@craghouse7 freeserve.co.uk) Hugh Clark, 3 Lindsay Place, Wick, Caithness KW1I 4PF Stan Laybourne, Old Schoolhouse, Harpsdale, Caithness KW12 6UN Revised ms accepted December 2007 Scottish Birds (2008) Behaviour and diet of non breeding Snowy Owls on St Kilda I] Behaviour and diet of non breeding Snowy Owls on St Kilda W TS MILES & S MONEY We report numbers of individual Snowy Owls, their behaviour, and diet on Hirta, St Kilda, between late May and early August 2007. Five different individuals were identified and movement of different individuals to and from Hirta was regular. Compared with previous records, sightings in 2007 were very frequent and the number of different birds recorded was high. Favoured roost sites were perches sheltered by natural or ancient artificial stone structures, with good vantage. Territorial, courtship or nesting behaviour was not observed on any occasion. Prey species recorded in the diet were few. Most commonly found were remains of the endemic subspecies of St Kilda Field Mouse and adult Atlantic Puffins, including a 27 year old ringed bird. A Great Skua chick was the only other prey species found in pellets. Introduction Snowy Owls Bubo scandiacus are scarce vagrants to Scotland (Forrester et al 2007). Records are almost annual and are most frequent from the Northern and Western Isles (Scottish Bird Reports 1970-2001, Thom 1986). Individuals have arrived in all months of the year, but there is a clear peak in April and May (Forrester et al 2007). Long staying Snowy Owls are not infrequent, but the only records of nesting in the UK are a pair which bred annually on Fetlar, Shetland, between 1967 and 1975 (Tulloch 1968, Sharrock 1976, Pennington et al 2004). Diet of these birds was studied by pellet analysis and consisted primarily of Rabbits Oryctolagus. cuniculus and wader chicks (Robinson & Becker 1986). Worldwide, Snowy Owls prey mostly on small mammals, although feeding on birds is not at all uncommon (eg del Hoyo et al 1999, Hakala et al 2006). There have been very few studies of the diet of non breeding Snowy Owls in Scotland. Systematic collection of pellets is difficult in this situation, as migrant owls may not stay for long, can range over very large areas, and numbers of individuals present are not always easy to assess (Scottish Bird Reports 1970-2001). Ageing and sexing single owls in the field is not always straightforward, except in the case of adult males (Forrester et al 2007). On St Kilda, 12 Snowy Owls have been recorded in 9 of the 45 years from 1962 to 2006 (Harris & Murray 1978, Murray 2002, Murray pers comm 2007). In 2007, there were exceptional numbers of Snowy Owls on St Kilda, involving several different individuals. This study was carried out on Hirta, St Kilda, between late May and early August 2007, and aimed to make an accurate record of the number of individuals present, to observe their behaviour and use of habitat, and to record their diet. Methods The study was conducted on Hirta, the largest island in the St Kilda archipelago (57°49°N, 08°35’W), an area of 628.5h with elevation to 426m. Habitat is primarily vegetated sea cliffs and maritime heath and grassland, dotted by ruins of many hundreds of cleits - stone shelters historically used by St Kildans for drying and storing seabirds. Data were collected between 20 May and 6 August 2007. The number of individual Snowy Owls on Hirta was assessed by direct observations, detailed field notes of plumage, and digital photographs of all birds 12 encountered, also used to help judge birds’ sex and age. Particular attention was paid to the exact positions and extent of black spots and barring in the plumage. Positions of roosting birds were noted daily, as an indication of where best to search for indigestible prey remains (regurgitated as pellets) and of habitat use by roosting Snowy Owls. Observations of other Snowy Owl behaviour, such as interactions with each other and with other species, were made incidentally and recorded by detailed field notes and, where possible, digital photography. SB 28 Diet was assessed from pellets, collected from the areas on Hirta where owls were seen to roost. Roosts were systematically checked for pellets every 6-8 days, even in periods when no owls were known to be present on Hirta. Pellets ranged in condition from warm, wet and slimy (very fresh) to dry, bleached and cracked (at least a few days old). Distinction of Snowy Owl pellets from those of Great Skua Stercorarius skua and Great Black-backed Gull Larus marinus was very easy, based on pellet size, Snowy Owl pellets were more than twice the size of pellets dropped by skuas and gulls, cylindrical compared to the oval shape of those of skuas and gulls and the texture of feather and Table 1. Summary of different individual Snowy Owls seen on Hirta, St Kilda, between 24 May and 6 August 2007 Bird Duration seen on Hirta / Sex & age Summary of plumage features used to distinguish individual | 24 May—31 May (M_ Ist year) i) 4 June—19 June F adult 3 4 June—29 June M_ adult ~ 8 July—1 August M_ sub adult 5 10 July—5 August M_ sub adult Black spotting on forehead and crown. Nape, neck & neck sides clean white. Large white bib. Dense, narrow black barring on under parts. Heavy black barring across scapulars, mantle, back and wing feathers. Black spotting on forehead, crown, nape and neck sides. Black barring on under parts. Small white bib. Very large black sub terminal crescents on tips of almost all feathers across mantle, back, scapulars and wings. All plumage clean white except for occasional very small black spots on outer primaries and very occasional black spots on scapulars. Extremely faint and sparse narrow brown barring on lower belly. All plumage clean white except for very infrequent small black spots and bars on scapulars, mantle, back and wing feathers. Heavy black barring on tertials. Faint brown spotting on forehead. Crown, nape, neck & neck sides clean white. Very large white bib. Dense, very faint, narrow brown barring on under parts. Black barring across scapulars, mantle, back and wing feathers. Scottish Birds (2008) fur remains tended to be more finely ground and compacted in Snowy Owl pellets. Skulls were absent from many pellets, and so regurgitated remains were mostly identified from a combination of tarso-metatarsi, pelvises, jaws, vertebrae, claws, feathers, fur and skin remains. Age classes of bird prey were determined, where possible, by comparison of the size, shape and skin colour of relatively undigested and complex remains, such as complete leg and foot arrangements from differently aged Atlantic Puffins Fratercula arctica. Presence in a pellet of one or more identifiable remains of an individual animal was considered representative of one occurrence as prey, identical remains of 2 individuals of the same species representative of 2 occurrences, etc, even if other major skeletal elements were missing. The proportion of total prey, expressed as percentage mass of all individuals recorded from pellets for each prey species, was calculated using mean adult and unfledged juvenile weights published by Boyd (1956) and Cramp ef al (1985). Results By comparison of field observations, notes and photographs, a total of 5 different Snowy Owls were identified on Hirta between 24 May and 5 August 2007 (Table 1). The study lasted 77 days and owls were seen on 63 days. Two birds were present together from 4 to 19 June (an adult male and adult female) and 10 July to | August (2 sub adult males), but on other dates only single birds were seen. Display or nesting behaviours (eg Murie 1929, Sutton & Parmelee 1956, Watson 1956, Tulloch 1968) were not observed on any occasion. Birds’ age and sex were judged by reference to information on plumage characters from previous studies and photographic identifi- cation resources (eg Josephson 1980, Cramp et al 1985, British Birds Interactive 2007). In most cases we were confident in determining males from females. Age was more difficult to assess from plumage. Two birds did not show adult Behaviour and diet of non breeding Snowy Owls on St Kilda 13 Figure 1. Adult male Snowy Owl (Individual 3, Table 1) at a typical roost site on Hirta sheltered by boulders and high on the ridge of a hill. Feet and talons were occasionally seen used in defence against Great Skuas plumages but younger plumages were not easily assignable to an exact age class (eg first year, second year, etc), so were categorised ‘sub adult’. Age and sex is parenthesised for Individual 1 (Table 1) because it showed plumage characters almost entirely typical of a first year male yet not absolutely distinct from characters shown by some females. Snowy Owls were most frequently seen roosting on, or within, stone structures: either natural crevices among crags and boulders (eg Figure 1) or perched within the ruins of cleits. Only occasionally were Snowy Owls seen roosting away from these very sheltered habitats, when perched out on relatively open grassland in shallow dips or hollows in the ground. Roost sites were relatively few, some were heavily used, and they were localised, mostly away from 14 the coast (Figure 2). By far the majority were high up on the slopes of hills, in good vantage positions. Only once was a bird seen roosting at the base of a hillside, within one of the cleits in Village Bay. On the occasions when 2 Snowy Owls were seen on Hirta simultaneously, roost sites were sometimes relatively close together, down to a minimum estimate of 10 metres. No aggressive or territorial behaviour was observed between individuals and they seemed highly tolerant of each other. Interactions between Snowy Owls and other bird species were seen infrequently and rarely involved owls that were roosting. Snowy Owls were occasionally seen in flight during daylight and were then often mobbed by Great Skuas, Ravens Corvus corax and Hooded Crows Corvus cornix. Great Skuas far outnumbered corvids on Hirta, and Snowy Owls were mobbed relatively heavily by skuas, sometimes involving over 20 individuals (never more than 10 Crows or 2 Ravens) chasing any Figure 2. Distribution of Snowy Owl roost sites observed on the islands of Hirta and Dun, St Kilda, between 24 May and 6 August 2007. Size of circles indicates the number of times roost sites were seen in use SB 28 one owl. Surprisingly, actual contact between birds during mobbing was very rare. Mobbing of a roosting Snowy Owl was seen on only one occasion and involved a Great Skua pair mobbing a male owl (Individual 4) perched in a relatively exposed position in open grassland, presumably within the skuas’ nesting territory. A total of 24 pellets were found and all contained remains of at least one identifiable prey species (Table 2). Remains from more than one prey species were found in 4 pellets, 3 containing a mixture of adult Atlantic Puffin and St Kilda Field Mouse Apodemus sylvaticus hirtensis remains and one a mixture of adult Atlantic Puffin and juvenile Great Skua remains. Prey species found most frequently in pellets were St Kilda Field Mouse (32 individuals from 14 pellets) and adult Atlantic Puffin (12 individuals from 12 pellets). Least frequent were remains from one juvenile Atlantic Puffin and from one juvenile Great Skua. Although mice predom- inated in the diet in terms of numbers, the much larger size of puffins means that the Snowy Owls obtained a far greater proportion of prey, in terms of total mass of individuals consumed, from puffins than from mice (Table 2). Discussion The number of individual Snowy Owls recorded during the study was surprisingly high, relative to previous records on Hirta since 1962. These, and the duration of sightings on Hirta, were indicative of a relatively high turn over and movement of Snowy Owls to and from St Kilda between late May and early August in 2007. Outside of the period of this study, there were records of Snowy Owls on St Kilda in 2007 in April, early May, late August and September. However, those records were not detailed or systematic in recording the identification of individuals, duration of stay, behaviour, or diet, so are extremely difficult to relate to this study. They do, however, Scottish Birds (2008) Behaviour and diet of non breeding Snowy Owls on St Kilda He) Table 2. Summary of diet of non breeding Snowy Owls on Hirta, St Kilda, from 24 May-6 August 2007 Prey species Number of pellets (n=24) in which prey species occurred St Kilda Field Mouse 14 Apodemus sylvaticus hirtensis Atlantic Puffin (adult) 12 Fratercula arctica Atlantic Puffin (Guvenile) l Fratercula arctica Great Skua (juvenile) | Stercorarius skua emphasise the exceptionally frequent occurrence and movement of Snowy Owls on St Kilda in 2007. Identification of individual owls was only possible in this study from daily observations, detailed field notes, and digital photographs of all birds encountered. Without these, it 1s possible that numbers may have occasionally been underestimated in the past. When identifying individuals, careful consid- eration was given to effects of plumage bleaching, wear and moult, especially as the study progressed into July and August, when these processes have greatest effect (Josephson 1980, Cramp et al 1985). Even so, plumage details of Snowy Owls encountered in this study, particularly the exact position, shape and size of dark spots and bars, appeared highly specific to individuals. This supports observations of individual variation from other studies and Scottish records of Snowy Owls (Tulloch 1968, Josephson 1980, Forrester et al 2007). It is possible that comparison of detailed photographs of Snowy Owls’ plumage could be used with caution to identify individuals and their movements within the UK. This would be particularly useful in areas with relatively Number Proportion of total prey of (% no (% mass individuals individuals) individuals) 32 69.6 16.8 12 26 63.5 | 22 36 l pp) 16.1 frequent records of Snowy Owls, for example to identify inter island movements and numbers of Snowy Owls within the Western Isles. Despite identification of individual Snowy Owls in this study from plumage, we still found it difficult to age and, to a lesser extent, sex birds on plumage criteria. Sexual dimorphism was especially difficult to judge in the field. However, size differences between the sexes and plumage differences between age classes are not always discrete (Earhart & Johnson 1970, Josephson 1980). Roost sites sheltered by natural and artificial stone structures were apparently favoured above roost sites on more open ground. This may have been due to more sheltered sites affording protection from very bad weather conditions which are frequent on St Kilda and from skuas and corvids which mob the owls. Large areas of the open grassland and maritime heath on Hirta are occupied by breeding Great Skuas (>180 pairs) and owls may have been deterred from using these areas by the highly aggressive behaviour of skuas defending their nesting territories. Snowy Owls seen being mobbed in 16 flight and, on one occasion, on the ground by skuas, did not seem very reactive to the treatment, and mostly avoided dive bombing from skuas simply by briefly ducking down out of the way. However, no owl was ever seen in a position very close to skua eggs or chicks. Defensive behaviour by adult skuas is usually most intense when the clutch and chicks are under greatest threat (Furness 1987). Perhaps owls are more reactive to this degree of mobbing intensity, and are then deterred effectively. Only very rarely was actual contact observed between a Snowy Owl and skuas mobbing it, and only when owls were in flight. On these occasions the owls’ reaction was spectacular. After the moment of contact, and typically when the next mobbing dive was made, the owl would flip over, momentarily fly upside down and bare or swipe its talons up at the attacker. This usually caused mobbing to cease immediately. The only other major response seen to be made by Snowy Owls in response to mobbing behaviour was loud wing clapping in flight. This, however, was apparently very rare and only seen on 3 occasions. Owls were occasionally flushed accidentally by humans from _ particularly secluded roost positions. In this case, they usually moved to a new roost position close by, did not fly far and never to a different island in the archipelago. The owls were silent in flight and never heard making any vocalisations. Roosting Snowy Owls were generally very visible, despite their sheltered locations, but it is conceivable that owls occasionally roosted undetected on Hirta and therefore the duration that individuals were present may be underes- timated. The likelihood of such inaccuracy is probably low, however, given that after the first day that any individual owl was not seen on Hirta, it was never sighted again during the study period. Snowy Owls were never seen on Soay, Dun or Boreray, but observer coverage was incidental and limited by access restrictions and suitable vantage points from Hirta. SB 28 Other than sheep and humans, the St Kilda Field Mouse is the only terrestrial mammal present on Hirta, and it was not surprising that the species formed a high proportion (69.6%) of the total individuals found in Snowy Owl pellets. The mouse is an endemic subspecies to the archipelago but its population size is not well known. Snowy Owls have been shown to favour mammalian prey if available (eg Murie 1929, Gross 1944, Robinson & Becker 1986) so the fact that adult puffins formed the majority of prey by mass may indicate that density of mice on Hirta was inadequate to allow owls to feed entirely on a mammalian diet. It should be noted, however, that the sample size of pellets was small although representative of several owls. The small number of pellets found at roost sites also suggests that prey remains may have been regurgitated away from these areas, possibly on the hunting grounds where they were likely to he undiscovered. Frequency of body parts of prey found in pellets was variable between prey species. Skeletal remains of mice were representative of the entire body (including skulls and jaws), while skeletal remains of puffins were much less representative, typically comprising complete leg and foot arrangements, ribs, spine and occasional other body parts, but never remains of the head. This supports other studies that suggested prey handling by Snowy Owls differs according to prey species (eg Wiggins 1953, Williams & Frank 1979). In this case, mice were apparently swallowed whole but puffins were swallowed in pieces less than, or equal to, body size minus the head. One of the greatest surprises of the study was a metal BTO ring found on a puffin tarsus in a Snowy Owl pellet. This puffin had been ringed as a newly fledged juvenile on Hirta in 1980. Surprisingly, Snowy Owl pellets did not contain remains from any of the other 17 species of seabird that breed on Hirta, apart from a single Great Skua chick. Prey selection by Snowy Owls of burrow nesting and nocturnal seabirds, such as alcids and storm-petrels, has been recorded in Scottish Birds (2008) Behaviour and diet of non breeding Snowy Owls on St Kilda 1, North America (Williams & Frank 1979). Puffin colonies on St Kilda are in close proximity to very large colonies of Leach’s Petrels Oceanodroma leucorhoa, European Storm-petrels Hydrobates pelagicus and Manx Shearwaters Puffinus puffinus, so perhaps remains from other species would have been found in Snowy Owl pellets had the sample size been larger in this study. Foraging behaviour of Snowy Owls was never directly observed. Owls were seen roosting at all times of the day and it seems likely that prey was caught mostly at night. Predation of puffins, however, may have occurred more towards dusk and dawn, when puffins are more active at their colonies than they are during the night, when most are underground or out at sea (Harris 1984). We found no evidence of Snowy Owls attempting to catch nocturnal seabirds, such as storm petrels and shearwaters. Predation of nocturnal seabirds by Great Skuas is generally a very unusual occurrence, but is relatively common on St Kilda where skuas feed extensively on Leach’s Petrels (Votier et al 2006). The owl pellet containing skua remains was found before most juvenile skuas on Hirta had fledged. Finding remains from a juvenile Great Skua in a Snowy Owl pellet was therefore surprising, as when young skuas are under threat from predators the parents are generally adept at defence. It begs the question of whether this juvenile skua was eaten because it had been left undefended at night by parents away hunting storm petrels. Species of prey found in Snowy Owl pellets in this study are only seasonally abundant on Hirta, as in winter the majority of seabirds are absent and mouse numbers are much reduced (Quine 2000, Mitchell et al 2004). Previous records of Snowy Owls on St Kilda include one from 14-28 November 1962 (Harris & Murray 1978, Murray 2002), which presumably would have had to survive mainly on mice, as few seabirds would be present on the archipelago in November. From pellet analysis in future, it would be interesting to know what exactly is eaten by Snowy Owls present on the islands in winter. Acknowledgements WTSM was funded by NERC and NTS through a CASE studentship to Glasgow University. SM was employed by NTS in 2007 as the St Kilda Seabird and Marine Ranger. Thanks to Martin Culshaw, Johanne Ferguson, Christina McAvoy and Roddy McCleod for access to their digital photographs and references, as well as to Terry Fountain - for many images, including Figure 1. Thanks to Stuart Murray for his helpful suggestions and St Kilda Snowy Owl records from 2000 to 2006. Helpful comments on the draft were given by Bob Furness, Richard Luxmoore and Roger Riddington. References Boyd J M 1956. The St Kilda Field Mouse (Apodemus sylvaticus hirtensis), population in the village area, Hirta, May 1955. Oikos 7: 110-116. Cramp S, Brooks D J, Dunn E, Gillmor R, Hollom P A D, Hudson R, Nicholson E M, Ogilvie M A, Olney P J S, Roselaar C S, Simmons K E L, Voous K H, Wallace D I M, Wattel J, Wilson M G 1985. The Birds of the Western Palaearctic. Vol. IV. Oxford University Press, Oxford. del Hoyo J, Elliot A, Sargatal J eds. 1999. Handbook of the Birds of the World. Vol.5. Barn-owls to Hummingbirds. Lynx Edicions, Barcelona. Dennis R H, Hogg R H, Brown A, Forrester R W, Bates D J, Wood A D, Murray R eds. 1970-2001. Scottish Bird Report. Issues from 1970 to 2001. The Scottish Ornithologists’ Club, Aberlady. Earhart C M & Johnson N K_ 1970. Size dimorphism and food habits of North American owls. Condor 72: 251-264. Forrester R W, Andrews I J, McInerny C J, Murray R D, McGowan R Y, Zonfrillo B, Betts M W, Jardine D C, Grundy D S (eds) 2007. The Birds of Scotland. The Scottish Ornithologists’ Club, Aberlady. 18 Furness R W 1987. The Skuas. T & AD Poyser, Calton. Gross A O 1944. Food of the Snowy Owl. Auk 61:1-18. Hakala A, Huhtala K, Kaikusalo A, Pulliainen E, Sulkava S 2006. Diet of Finnish Snowy Owls Nyctea scandiaca. Ornis Fennica 83:59-65. Harris M P 1984. The Puffin. T & AD Poyser, Calton. Harris M P & Murray S 1978. Birds of St Kilda. Institute of Terrestrial Ecology, Cambridge. Josephson B 1980. Aging and Sexing Snowy Owls. Journal of Field Ornithology 51: 150-159. Mitchell PI, Newton S F, Ratcliffe N, Dunn T E 2004. Seabird Populations of Britain and Ireland. A & C Black Publishers, London. Murie O J 1929. Nesting of the Snowy Owl. Condor 31: 3-12. Murray S 2002. Birds of St Kilda. The Scottish Ornithologists’ Club, Musselburgh. Pennington M, Osborn K, Harvey P, Riddington R, Okill D, Ellis D, Heubeck M 2004. The Birds of Shetland. A & C Black Publishers, London. Quine D 2000. St Kilda. Colin Photography Ltd, Grantown-on-Spey. Riddington R (current ed.) 2007. British Birds Interactive. CD Archive 1907 to 2007: Snowy Owl photographic references. Birdguides, www.birdguides.com Robinson M & Becker C D 1986. Snowy Owls on Fetlar. British Birds 79: 228-242. Sharrock J T R 1976. The Atlas of Breeding Birds in Britain and Ireland. British Trust for Ornithology, Thetford. Sutton G M & Parmelee D F 1956. Breeding of the Snowy Owl in Southeastern Baffin Island. Condor 58: 273-282. Thom V M 1986. Birds in Scotland T & A D Poyser, Calton. Tulloch R J 1968. Snowy Owls breeding in Shetland in 1967. British Birds 61: 119-132. Baxter SB 28 Votier S C, Crane J E, Bearhop S, de Leén A, McSorley C A, Minguez E, Mitchell I P, Parsons M, Phillips R A, Furness R W 2006. Nocturnal foraging by great skuas Stercorarius skua: implications for conservation of storm- petrel populations. Journal of Ornithology 147: 405-413. Watson A 1957. The behaviour, breeding, and food-ecology of the Snowy Owl Nyctea scandiaca. Ibis 99: 419-462. Wiggins I L 1953. Foraging activities of the Snowy Owl (Nyctea scandiacus) during a period of low Lemming population. Auk 70: 366-367. Williams P L & Frank L G 1979. Diet of the Snowy Owl in the absence of small mammals. Condor 81: 213-21. Will T S Miles, Institute of Biomedical and Life Sciences, Graham Kerr Building, University of Glasgow G12 8QQ. willtsmiles@hotmail.com Sarah Money, Raintree House, Church Lane, Drayton St Leonard, Wallingford, Oxfordshire OX10 7AU. Revised ms accepted February 2008 Scottish Birds (2008) Responses of owls and Kestrels to induced spatio temporal variation 19 Responses of owls and Eurasian Kestrels to natural and human induced spatio temporal variation ERKKI KORPIMAKI*, HARRI HAKKARAINEN, Introduction Most owl species and also many raptors including Eurasian Kestrels Falco tinnunculus (hereafter kestrels), buzzards and harriers feed mainly on voles of the genera Microtus and Clethrionomys in North Europe (Marti et al 1993, Korpimaki and Marti 1995, Valkama er al 2005). Three to 5 year population cycles of voles are characteristic in Northern Europe, where these cycles cover thousands of km? in pristine arctic and boreal ecosystems, and in man made forest plantations and agricultural areas. These northern multi annual population cycles of voles are different from those in temperate areas of Europe for the following reasons. First, the amplitude of the cycles is much higher in the north (50—500-fold) than in the temperate (10- fold) areas of Europe. Second, the spatial synchrony extends over 80—600 km in the north, whereas it has been documented to extend over only 8—20 km in the south. Third, there are steep summer declines of voles in the north but not in the south. Fourth, in the low phase of the cycle, vole densities are an order of magnitude smaller in the north (<1 vole per ha) than in the south (25-50 per ha) (Korpimaki et al 2004, 2005b). Fifth, population densities of herbivorous voles and mice, and even of insectivorous shrews, fluctuate in close synchrony in North Europe (Korpimaki et al 2005a). In addition, population densities of voles often fluctuate in synchrony with small game animals, such as forest grouse and hares, in particular in northern parts of Finland and Sweden (Lindén 1988). A 3 year population cycle of voles has been prevalent in our study area, the Kauhava region in TONI LAAKSONEN & VILLE VASKO western Finland, during the last 30 years (Korpimaki et al 2005a). In the low phase of the cycle, vole densities are very low during the breeding season of owls and kestrels but start to slowly recover in late summer. In the increase phase of the cycle, vole densities are intermediate during the egg laying period of owls and kestrels but fast increases occur in the course of the summer. In the decline phase of the cycle, vole densities are still intermediate in early spring but decline to low numbers at the end of the nestling and fledging periods of owls and kestrels (Fig | in Korpimaki and Hakkarainen 1991). Multi annual, high amplitude population cycles of voles in Northern Europe are mainly driven by delayed density dependent mortality, caused by mammalian and avian predators and by a shortage of winter food of voles (Korpimaki and Norrdahl 1998, Korpimaki et al 2002, 2005a, Huitu ef al 2003). These high amplitude cyclic fluctuations also induce highly varying food situations with predictable fat and lean periods for birds of prey subsisting mainly on voles. To cope with this natural temporal variation in main_ food abundance, birds of prey should be able to take full benefit from fat periods by producing large clutches and raising large broods with high quality offspring, whereas they should be able to survive over lean periods by shifting to alternative prey and/or by moving over long distances to find a breeding site with plenty of voles. Predators may respond to fluctuations in prey abundance either numerically or functionally (Solomon 1949). The former response is due to changes in natality, mortality, immigration and emigration, and is largely determined by the 20 mobility (natal and breeding dispersal), survival, reproductive rate and generation time of the predator. The availability of the main and alternative prey, the ability to shift to alternative prey, inter specific competition for food, and intra guild predation may determine the functional response of the predator (Andersson and Erlinge 1977, Korpimaki and Norrdahl 1989b, 1991). We have been mainly studying responses of Tengmalm’s Owls Aegolius funereus, Short- eared Owls Asio flammeus, Long-eared Owls A otus and kestrels to these highly fluctuating densities of their main prey (voles). In addition, we have also done research on numerical and functional responses of larger birds of prey, 20 18 Vv Microtus voles ° Bank voles | Le ®~ No ofowlnests 14 ? 12 10 oOo NS SF OD “wel? aio ec ae : SB 28 including the Ural Owl Strix uralensis, the Eagle Owl Bubo bubo, the Common Buzzard Buteo buteo and the Northern Goshawk Accipiter gentilis, that have a more diverse diet. The main focus of the studies on these large birds of prey has been the interactions that they have with their main and alternative prey species, such as small mammals and small game animals (forest grouse and hares) (Korpimaki et al 1990, Korpimaki and Norrdahl 1997, Reif et al 2001, 2004, Tornberg et al 2005, 2006, Valkama et al 2005). An additional focus has been intra guild predation and competition among different sized avian predators, including their competitive and predatory interactions with mammalian predators (Korpimaki 1987a, Korpimaki and Norrdah! 1989a, Hakkarainen and Korpimaki e / 30 r 25 20 Oo z 152 1?) pi ” 10 5 0 1974 1977 1980 1983 1986 1989 1992 1995 1998 2001 2 Year Figure 1. Number of Tengmalm’s Owl nests per 100 nest boxes and density indices of Microtus voles (the Field Vole M agrestis and the Sibling Vole M rossiaemeridionalis) and Bank Voles during 1973-2006. Data from Korpimdki (1987b), Laaksonen et al (2002) and E Korpimaki (unpublished) Scottish Birds (2008) 1996, Suhonen et al 2007). Here our main aim is to review our research on numerical responses of Tengmalm’s Owls and kestrels, for which we have long term data on variation in breeding densities, reproductive success, dispersal and survival. In addition, we will review factors affecting lifetime reproductive success (LRS) of Tengmalm’s Owls and _ interactions’ of Tengmalm’s Owls with their larger (Ural and Eagle owls, goshawks) and smaller (the Pygmy Owl Glaucidium passerinum) allospecifics. Main results Breeding densities Owls and kestrels mainly responded numerically to highly fluctuating food conditions. Breeding 12 o Responses of owls and Kestrels to induced spatio temporal variation —t— No of kestrel nests | ~-Y— Microtus voles Zi percentage of Tengmalm’s Owls in 500 nest boxes varied from 1% to 33% during 1973—2006 (Figure 1). Breeding density of kestrels varied from 0.9 to 11.7 nests/10 km? during 1977-2006 (Figure 2), that of Short-eared Owls from 0 to 11.5 nests/10 km? during 1977-2006, and that of Long-eared Owls from 0 to 4.0 nests/10 km? during 1977-1988 (Table 2 in Korpimaki and Norrdahl 1991, Table 5 in Korpimaki 1992a, and unpublished). Breeding densities of these 3 owl species and kestrels were closely positively correlated with the density indices of voles in the prevailing spring (Korpimaki 1994). The number of non breeding Tengmalm’s, Short-eared and Long-eared Owls and kestrels was very low (Table 2 in Korpim&ki and Norrdahl 1991). 60 S]SOU |a11Sey 10 ] 7 T 1977 1980 1983 1986 1989 1992 1995 1998 2001 T T 2004 Year Figure 2. Number of Kestrel nests found in the Alajoki region (47 km’), Kauhava and Lapua, western Finland and density indices of Microtus voles (the Field Vole M agrestis and the Sibling Vole M rossiaemeridionalis) and Bank Voles during 1977-2006. Data from Korpimaki and Norrdahl (1991) and E Korpimaki (unpublished) Pad Plated Sharceared Owl © Mikio Hanninen The age structure of breeding population also differed substantially among the vole cycle phases in Tengmalms’s Owls and kestrels: hardly any one year old Tengmalm’s Owls were breeding in the low phase of the vole cycle, whereas they were the majority of breeders in decrease years, after a year of successful reproduction in the increase phase (Laaksonen et al 2002). The percentage of one year old breeders was higher also in the kestrel in decrease years of the vole cycle (mean 17.8% of males and 28.2% of females) as compared to low and increase phases (6.0% and 5.7% of males, and 21.6% and 17.6% of females, respectively, Laaksonen et al 2004). Reproductive success, dispersal and survival Yearly mean clutch size of Tengmalm’s Owls varied from 3.5 to 6.5 eggs (Korpimiéki and Hakkarainen 199] and unpubl) and that of kestrels from 4.3 to 60 eggs during 1977-2006 (Korpimaki and Wiehn 1998 and unpublished). Yearly mean clutch sizes of owls and Kestrels were closely correlated with vole density indices in current spring (Korpimaki and Hakkarainen 1991, Korpimaki and Wiehn 1998). Hatching spans of Tengmalm’s Owl broods averaged 6-7 days and increased with clutch size (Valkama et al 2002). The most common hatching span of kestrel broods was 2—3 days (Wiebe et al 1998). When SB 28 controlling for clutch size, the degree of hatching asynchrony of owl and Kestrel broods varied in the course of the vole cycles: chicks of the broods hatched more synchronously in low vole years than in increase and decrease vole years (Wiebe et al 1998, Valkama et al 2002). Breeding dispersal of female Tengmalm’s Owls from the breeding site of the previous year was more extensive in the decrease than in the increase and low phases of the vole cycle, but this was not found for owl males (Korpimaki 1993), which mostly occupied the same territories after their first breeding attempts (Korpimaki 1988b). Analyses of long term dispersal and survival data from kestrels showed largely similar results: more females returned to breed close to (<5 km) the previous year breeding site in the increase than in the other phases of the vole cycle, but no cycle phase related differences were found in males (Korpimaki et al 2006, Vasko 2007). Annual adult survival of male Tengmalms’s Owls varied from approximately 25% to approximately 75% and was closely positively related to vole density indices in winter (Fig | in Hakkarainen et al 2002). Juvenile survival of Tengmalm’s Owls was apparently higher in the increase phase than in the other phases of the vole cycle, because the proportion of fledglings that in subsequent years recruited to the breeding population was twice as high for young fledged in the increase as in the other phases of the vole cycle (Korpimaki and Lagerstr6m 1988). Similar results were also obtained for recruitment of Kestrel fledglings: approx. 19% of clutches laid in the increase phase of the vole cycle produced recruits to the breeding population in subsequent years, whereas the corresponding percentage was only 8% in decrease phase and 3% in the low phase of vole cycle. Male Tengmalm’s Owls that initiated their breeding lifespan in the increase phase of the vole Scottish Birds (2008) Responses of owls and Kestrels to induced spatio temporal variation 23 cycle had higher lifetime reproductive success (LRS) than those initiating their career in the decline phase (Korpimaki 1992b). LRS of male owls was reduced in territories with a higher proportion of farmland (Laaksonen ef al 2004), mainly because their fledgling production in any one breeding attempt was reduced in territories with high proportion of farmland in years when vole populations were declining (Hakkarainen ef al 2003). LRS of male owls increased with the proportion of old growth forest in the territory, which appeared to be due to higher numbers of breeding attempts in these territories (Laaksonen et al 2004). Survival of male owls increased with the cover of old forest, although the extent of old forest within owl territories was relatively small (mean approximately 12%, range 2—37%) in our study area. This association however varied among years and appeared especially in increasing vole years (Hakkarainen et al 2008). Higher survival in old forests is likely to be due to better protection against larger birds of prey (eg Ural Owls and Goshawks), and/or to better availability of alternative prey (eg Bank Voles Clethrionomys glareolus, shrews, Willow Tits Parus montanus and Crested Tits P cristatus, etc), particularly in winter. Intra guild predation and inter specific competition Eagle, Ural and Tengmalm’s Owls coexist in North Europe and mainly feed on small rodents. Eagle and Ural Owls can even kill smaller birds of prey but cannot enter the small entrance hole nest boxes of Tengmalm’s Owls. This raises the question, whether inter specific competition due to larger owl species reduced breeding density and reproductive success of Tengmalm’s Owls. Breeding Tengmalm’s Owls occupied control nest boxes in areas of no permanent Eagle and Ural Ow/l territories, and nest boxes within Eagle Owl territories more than nest boxes within Ural Owl territories. Most breeding attempts of Tengmalm’s Owl near Ural Owls already failed during the ee G Plate 2. Tengmali’s Owl © Erkki Korpimdki r * ss Plate 3. Male Kestrel © Erkki Korpimdki 24 courtship period. The observational data also revealed that the breeding frequency of Tengmalm’s Owl nest boxes was reduced nearby (<2 km) Ural Owl nest sites. The mean start of egg laying was delayed by 11 days near Ural Owl nests, where male Tengmalm’s Owls were also younger and mated more often with short winged females than further away from Ural Owl nests. Our results suggest that inexperienced male Tengmalm’s Owls are forced to establish their territories in the vicinity of Ural Owls where they also attain subdominant females. The areas near Ural Owl territories appeared to be suboptimal habitats for Tengmalm’s Owls. Therefore, Ural Owls reduce the breeding population size of Tengmalm’s Owls by reducing the suitable habitats (Hakkarainen and Korpimaki 1996). Tengmalm’s and Pygmy Owls are the only common birds of prey which prey on small mammals and passerine birds in our study region in winter and they have broadly overlapping diets (eg Kellomaki 1977, Korpimaki 1988a, Suhonen 1993). Pygmy Owls store small mammals and birds in holes and nest boxes during late autumn and winter (Schénn 1980, Solheim 1984). To test whether competitive and predatory interactions among smaller owls also limit food store size in late autumn and winter, we erected Pygmy Owl nest boxes (45-mm entrance diameter) for food hoarding close and far (>2 km) from Tengmalm’s Owl nest boxes (>80 mm entrance diameter). We found food stores of Pygmy Owls in similar frequency both near and far from Tengmalm’s Owl nest boxes (41 vs 42% of plots), but in near plots the number and biomass of prey stored by Pygmy Owls were significantly lower (Suhonen ef al 2007). These results suggest that there is competition for food between these 2 small owl species and/or that food storing behaviour of Pygmy Owls is interfered by larger Tengmalm’s Owls that can also sometimes even kill Pygmy Owls. These 2 studies are the first experimental SB 28 demonstrations that competitive and predatory interactions induced by larger owls may decrease the fitness of smaller birds of prey. In- particular, competitive and _ predatory interactions by Ural Owls have harmful effects on Tengmalm’s Owls and thus decrease the habitat quality of smaller Tengmalm’s Owls. Tengmalm’s Owls in turn probably decrease the habitat quality of the smaller Pygmy Owls. Conclusions and future prospects Temporal variation in vole abundance was the main determinant of breeding density and success, quality of offspring, survival of adult males, breeding dispersal distances, recruitment rate of offspring and LRS of Tengmalm’s Owls. In addition, temporal variation in vole abundance also largely determined the breeding density and success, quality of offspring, breeding dispersal distances and recruitment rate of offspring of kestrels, but long term data from adult survival and LRS of kestrels still wait for further analyses. Therefore, these 2 bird of prey populations offer a prime example of how predators can adapt to largely fluctuating food conditions that vary in a predictable manner in Northern Europe. Human induced spatial changes in_ the environment (ie clear cutting of old growth coniferous forests by modern forestry) may decrease food availability for Tengmalm’s Owls and alter inter specific interactions among birds of prey because less refuge sites remain available. In particular, intra and inter specific competition for food may strengthen and intra guild predation among birds of prey may become more common due to the changes in the landscape, which further reduces the habitat quality and population size of smaller forest dwelling owls. Since reduction in the area of old coniferous forests decreased the survival and LRS of Tengmalm’s Owls, we predict long term declines of Tengmalm’s Owl populations in Scottish Birds (2008) Responses of owls and Kestrels to induced spatio temporal variation pA) northern European boreal forests. This was also found in nation wide monitoring study of birds of prey in Finland (Honkala and Saurola 2006). On the contrary, large scale clear cutting of North European boreal forests increases the grassy habitat for voles (Hakkarainen ef al 1996), which could benefit kestrels that hunt mainly in open country. A long term increase in population size of kestrels found in nation wide monitoring study of birds of prey in Finland (Honkala and Saurola 2006) may thus be partly explained by changes in habitat structure. These results show that recent human induced large scale habitat manipulations can substantially alter the breeding population sizes and _ inter- specific interactions among birds of prey, and thus have profound effects on the composition of assemblages of birds of prey. Acknowledgements We would like to thank the PhD students and postdoctoral researchers Petteri IImonen, Vesa Koivunen, Kai Norrdahl, Pasi Tolonen, Jari Valkama, Michael Griesser, Juan A Fargallo, Karen L Wiebe, Jiirgen Wiehn and Robert L Thomson, as well as field technicians and voluntary assistants Mikko Hast, Ossi Hemmink1, Mikko Hanninen, Timo Hyrsky, late Sakari Ikola, Jorma Nurmi, Reijo Passinen, Ville Poyri, Erkki Rautiainen, Jussi Ryssy and Rauno Varjonen for great help in the collection of long term data. The studies were financially supported by the Academy of Finland, the Emil Aaltonen Foundation, the Finnish Cultural Foundation and the Jenny and Antti Wihuri Foundation. References Andersson M & Erlinge S 1977. Influence of predation on rodent populations. Oikos 29591-5997. Hakkarainen H & Korpimaki E 1996. Competitive and predatory interactions among raptors: an observational and experimental study. Ecology 77:1134—-1142. Hakkarainen H, Koivunen V, Korpimaki E & Kurki S 1996. Clear-cut areas and breeding success of Tengmalm’s owls Aegolius funereus. Wildlife Biology 3:253-258. Hakkarainen H, Korpimaki E, Koivunen V & Ydenberg R 2002. Survival of male Tengmalm’s owls under temporally varying food conditions. Oecologia 131:83-88. Hakkarainen H, Mykra S, Kurki S, Korpimaki E, Nikula A & Koivunen V 2003. Habitat composition as a determinant of reproductive success of Tengmalm’s owls under fluctuating food conditions. Oikos 100:162-171. Hakkarainen H, Korpimaki E, Laaksonen T, Nikula A & Suorsa P 2008. Survival of male Tengmalm’s owls increases with cover of old forest in their territory. Oecologia (in press). Honkala J & Saurola P 2006. Breeding and population trends of common raptors and owls in Finland in 2005. Yearbook of Linnut magazine 2005:9-22. Huitu O, Koivula M, Korpimaki E, Klemola T & Norrdahl K 2003. Winter food supply limits growth of northern vole populations in the absence of predation. Ecology 84:2108—2118. Kellomaki E 1977. Food of the Pygmy Owl Glaucidium passerinum in the breeding season. Ornis Fennica 54:1-29. Korpimaki E 1987a. Dietary shifts, niche relationships and reproductive output of coexisting Kestrels and Long-eared Owls. Oecologia (Berl) 74:277-285. Korpimiéki E 1987b. Selection for nest-hole shift and tactics of breeding dispersal in Tengmalm’s owl Aegolius funereus. Journal of Animal Ecology 56:185—196. Korpiméki E 1988a. Diet of breeding Tengmalm’s Owls Aegolius funereus: long-term changes and year-to-year variation under cyclic food conditions. Ornis Fennica 65:21—30. Korpimaki E 1988b. Effects of territory quality on occupancy, breeding performance and breeding dispersal in Tengmalm’s owl. Journal of Animal Ecology 57:97-108. 26 Korpimaki E 1992a. Diet composition, prey choice and breeding success of Long-eared Owls: effects of multiannual fluctuations in food abundance. Canadian Journal of Zoology 70:2373-2381. Korpimaki E 1992b. Fluctuating food abundance determines the lifetime reproductive success of male Tengmalm’s owls. Journal of Animal Ecology 61:103-111. Korpimaki E 1993. Does nest-hole quality, poor breeding success or food depletion drive the breeding dispersal of Tengmalm’s owls. Journal of Animal Ecology 62:606-613. Korpimaki E 1994. Rapid or delayed tracking of multi-annual vole cycles by avian predators? Journal of Animal Ecology 63:619-628. Korpimaéki E & Hakkarainen H_ 1991. Fluctuating food supply affects the clutch size of Tengmalm’s Owl independent of laying date. Oecologia 85:543-552. Korpimadki E & Lagerstro6m M 1988. Survival and natal dispersal of fledglings of Tengmalm’s owl in relation to fluctuating food conditions and hatching date. Journal of Animal Ecology 57:433-441. Korpimaki E & Marti CD 1995. Geographical trends in the trophic characteristics of mammal- and bird-eating raptors in Europe and North America. Auk 112:1004—1023. Korpimaki E & Norrdahl K 1989a. Avian predation on mustelids in Europe 1: occurrence and effects on body size variation and life traits. Oikos 55:205-215. Korpimaki E & Norrdahl K 1989b. Predation of Tengmalm’s owls: numerical responses, functional responses and dampening impact on population fluctuations of voles. Oikos 54:154-164. Korpimaki E & Norrdahl K 1991. Numerical and functional responses of Kestrels, Short- eared Owls, and Long-eared Owls to vole densities. Ecology 72:814—826. Korpimaéki E & Norrdahl K 1997. Can the Alternative Prey Hypothesis explain the SB 28 occurrence of short-term population cycles of small game in Finland? Suomen Riista 43:72-84. Korpimaki E & Norrdahl K 1998. Experimental reduction of predators reverses the crash phase of small-rodent cycles. Ecology 76:2448-2455. Korpimaki E & Wiehn J 1998. Clutch size of kestrels: seasonal decline and experimental evidence for food limitation under fluctuating food conditions. Oikos 83:259-272. Korpimaki E, Huhtala K & Sulkava S 1990. Does the year-to-year variation in the diet of eagle and Ural owls support the alternative prey hypothesis? Oikos 58:47-54. Korpimaki E, Norrdahl K, Klemola T, Pettersen T & Stenseth NC 2002. Dynamic effects of predators on cyclic voles: field experimen- tation and model extrapolation. Proceedings of the Royal Society of London B 269:991-997. Korpimaki E, Brown PR, Jacob J & Pech RP 2004. The puzzles of population cycles and outbreaks of small mammals _ solved? BioScience 54:1071—1079. Korpimaki E, Norrdahl K, Huitu O & Klemola T 2005a. Predator-induced synchrony in population oscillations of co-existing small mammal species. Proceedings of the Royal Society of London B 272:193-202. Korpimaki E, Oksanen L, Oksanen T, Klemola T, Norrdahl K & Banks PB 2005b. Vole cycles and predation in temperate and boreal zones of Europe. Journal of Animal Ecology 74:1150-1159. Korpimaki E, Thomson RL, Vasko V & Laaksonen T 2006. Breeding dispersal of Eurasian Kestrels in a temporally varying environment. Journal of Ornithology 147:45-46. Laaksonen T, Korpimaéki E & Hakkarainen H 2002. Interactive effects of parental age and environmental variation on the breeding performance of Tengmalm’s owls. Journal of Animal Ecology 71:23-31. Laaksonen T, Hakkarainen H & Korpimaki E Scottish Birds (2008) Responses of owls and Kestrels to induced spatio temporal variation 2], 2004. Lifetime reproduction of a forest- dwelling owl increases with age and area of forests. Proceedings of the Royal Society of London B 271:S461-S464. Laaksonen T, Lyytinen S & Korpimaki E 2004. Sex-specific recruitment and brood sex ratios of Eurasian kestrels in a seasonally and annually fluctuating northern environment. Evolutionary Ecology 18:215—230. Lindén H 1988. Latitudinal gradients in predator-prey interactions, cyclicity and synchronism in voles and small game populations in Finland. Oikos 52:341-349. Marti CD, Korpimaéki E & Jaksic FM 1993. Trophic structure of raptor communities: a three-continent comparison and synthesis. Current Ornithology 10:47-137. Reif V, Tornberg R, Jungell S & Korpimaki E 2001. Diet variation of common buzzards in Finland supports the alternative prey hypothesis. Ecography 24:267-274. Reif V, Jungell S, Korpimaki E, Tornberg R & Mykra S 2004. Numerical response of common buzzards and predation rate of main and alternative prey under fluctuating food conditions. Annales Zoologici Fennici 41:599-607. Schénn S 1980. Der Sperlingkauz. Die Neue Brehm-Biicherei, Wittenberg-Lutherstadt. Solheim R 1984. Caching behaviour, prey choice and surplus killing by Pygmy Owls Glaucidium passerinum during winter, a functional response of a generalist predator. Annales Zoologici Fennici 21:301—308. Solomon ME 1949. The natural control of animal population. Journal of Animal Ecology 18:1-35. Suhonen J 1993. Predation risk influences the use of foraging sites by tits. Ecology 74:1197-1203. Suhonen J, Halonen M, Mappes T & Korpimaki E 2007. Interspecific competition limits larders of pygmy owls. Journal of Avian Biology 38:630-634. Tornberg R, Korpimaki E, Jungell S & Reif V 2005 Delayed numerical response of goshawks to population fluctuations of forest grouse. Oikos 111:408-415. Tornberg R, Korpimaéki E & Byholm P 2006. Ecology of the Northern Goshawk in Fennoscandia. Studies in Avian Biology 31:141-157. Valkama J, Korpima&ki E, Holm A & Hakkarainen H 2002. Hatching asynchrony and brood reduction in Tengmalm’s owl Aegolius funereus: the role of temporal and spatial variation in food abundance. Oecologia 133:334—-341. Valkama J, Korpimaki E, Arroyo B, Beja P, Bretagnolle V, Bro E, Kenward R, Manosa S, Redpath SM, Thirgood S & Vinuela J 2005. Birds of prey as limiting factors of gamebird populations in Europe: a review. Biological Reviews 80:171-—203. Vasko V 2007. Breeding dispersal of Eurasian kestrels in a temporally and_ spatially fluctuating environment. M*Sci _ thesis, Department of Biology, University of Turku, Finland, 62 pp. Wiebe KL, Korpimaéki E & Wiehn J 1998. Hatching asynchrony in Eurasian kestrels in relation to the abundance and predictability of cyclic prey. Journal of Animal Ecology 67:908-917. Erkki Korpimaki*, Harri Hakkarainen, Toni Laaksonen & Ville Vasko, Section of Ecology, Department of Biology, University of Turku, FIN-20014 Turku, Finland *Corresponding author, e-mail: ekorpi@utu.fi, web site: http://users.utu fi/ekorpi/index.htm Revised ms accepted February 2008 28 SB 28 Post breeding movements of Sandwich Terns in the Firth of Forth | JOHAN BOS Many Sandwich Terns spend the post breeding period in the Firth of Forth. By reading the inscriptions of rings with the help of a telescope, information about the origin and movements of 45 birds was obtained in the period 2000-2004. Most of these birds were ringed at the Farne Islands or further north in Scotland, from Orkney to Fife. The remaining birds were ringed in Ireland, the Netherlands, Belgium and Denmark. Introduction The Sandwich Tern Sterna sandvicensis is a regular breeding bird along the coasts of Scotland. It is, of course, a migratory bird, and its wintering areas are generally found along the coasts of western Africa (Wernham ef al 2002). The breeding period covers usually the months from May till July. It is well known that the terns, before they migrate south to their wintering quarters, spend time in the northern zones, presumably taking advantage of food supplies. This period, referred to as the post breeding period, may span_ several months, with the last birds leaving in October. The Firth of Forth is one of these areas visited by many Sandwich Terns in the post breeding period, in numbers exceeding thousands at peak times (Thomson eft al 2003). Large groups of terns can gather at communal roosts, but often small flocks of fledglings and adults can be seen fishing at sea. Family ties are strong and are maintained along the way to the wintering quarters, and it is very common to see adults feeding their young in this period. Presumably the Forth is particularly attractive to Sandwich Terns because of the combination of food supply and good roost sites, such as islands and lagoons. Sandwich Terns do not breed regularly anymore in the Forth, and numbers have fluctuated: 100-500 nests/pairs in 1988-94 (Murray et al 1998), O-5 in 1995-98 (Murray et al 1998, Kelly 1999), 122 in 1999 (Kelly 2000) and ca 300 pairs in 2000 (Thomson 2003). When I noticed that Sandwich Terns in the Forth used regular roosts during incoming tides, and that many of the terns were ringed, I saw an excellent opportunity to discover more about their origin and movements. Were these really local birds, on their way to the African coast, or were they coming from further afield? How long did they stay? Did they always use the same roost? By reading the inscription of the rings with the help of a telescope, one could find out such information in a relatively straightforward way. Methods I carried out observations over 5 seasons, from 2000 to 2004, usually in the period July to November, along the south coast of the Forth from Cramond to Musselburgh. Observations were made during incoming tides, usually one or 2 hours before the tide reached its highest point. The locations I regularly checked were: Cramond, at the the sewage outfall area, Musselburgh at the mouth of the River Esk and the large ash lagoon, and Joppa Rocks, where most of the data for this study was gathered. Scottish Birds (2008) During incoming tides, the Joppa Rocks were regularly used as a temporary roost by Sandwich Terns, often accompanied by waders. The brick wall that separates the sea from the main land made an excellent hide to get a good view of the birds without disturbing them. The roosting birds stayed here only a few hours, just until the rocks disappear under the water, and then moved to other roosts such as the Musselburgh ash lagoon. The rocks are close enough to read the inscriptions on metal rings with a telescope; I used a Swarovski AT80 HD with 20-60 zoom. The inscriptions were immediately jotted down in a notebook At home all the inscriptions were entered in a database with information about the age of the bird (juvenile or adult), the position of the ring (left/right, upside down or right side up, presence of colour rings). Rings whose the inscriptions could be read in full were reported to the BTO who forwarded ringing details. The total number of observation days was 70. These were distributed over the 5 year period as follows: 2000; 17, 2001: 17, 2002: 9, 2003: 9, and 2004: 18. Most of the visits took place in September (31), then October (18) and August (14), followed by July (4) and November (3). Results Origin I obtained resighting data from 45 different Sandwich Terns with metal rings (Table 1). From 32 of these, the inscriptions could be completely read and for these birds the ringing information was retrieved via the various national ringing centres. From 13 birds the inscription of the rings could only be partially read. The majority of the rings, 39 of 45, were BTO rings used by ringers in Britain or Ireland. In addition, there were 3 birds with rings from Belgium, 2 with rings from the Netherlands and one from Denmark. Post breeding movements of Sandwich Terns in the Firth of Forth a? Table 1. Completely and partially read rings on Sandwich Terns Origin Complete Partial Total Britain/Ireland 30 9 39 Belgium pi .) Netherlands 0 2 2 Denmark l 0 ] The 30 terns with the BTO rings were ringed in England (23), Scotland (6), and Ireland (1). The English ones were mostly ringed in Northumberland, about 100 km away at the Farne Islands (11) and Coquet Island, Amble (10), and in Cleveland (2). The Scottish terns were ringed in Orkney (2), Grampian (2), Highland (1), and Fife (1). The Irish bird was ringed at Lady’s Island Lake in Wexford. Figure 1 shows a map with the ringing locations of all the Sandwich Terns resighted in Lothian. There was a total of 51 resightings, because 6 birds were seen twice. Most of the resightings were in the month of September (33), followed by October (14), July (2), and August (2). The first and last sightings in a calendar year were respectively on 22 July 2001 and 27 October 2001. The number of resightings distributed over the 5 years of study were: 2000 (21); 2001 (17); 2002 (4); 2003 (7); 2004 (2). In the first 2 years of the study Sandwich Terns seemed more numerous in the Forth than in the following years. Site Fidelity Only 6 birds were seen more than once, and no birds were seen more than twice. In all of these cases the period between the 2 resightings was very short. Four times the same tern was seen a day later at the same site, one time 6 days later (2 and 8 September), and one time 10 days later (4 and 14 October). Never has the same 30 Sandwich Tern been seen in different years. This suggests that the Sandwich Terns do not stay long at one location. Distance In what follows the distance between the ringing and resighting locations is measured along a straight line which may not be the actual path the bird took. The smallest distance was 58 km, covered by a Sandwich Tern ringed at Tentsmuir Point in Fife in 1987 and resighted in Joppa in 2001. The largest distance, 854 km, was covered by a Sandwich Tern ringed as a nestling in Denmark in 1992 and seen in Joppa in 2004. As 5 0: SB 28 adult Sandwich Terns need not breed near the colony where they themselves fledged it is more meaningful to look at distance information obtained by resightings of birds ringed as nestlings in the same year. Here the 3 largest distances were 309 km (South Ronaldsay, Orkney), 459 km (Lady’s Island Lake, Ireland), and 660 km (Zeebrugge, Belgium). Age In Britain and Ireland, about 97% of ringed Sandwich Terns later recovered were ringed as nestlings (Wernham et al 2002). The data gathered from the Forth comprised 29 terns that were 5 0° ae 10° Figure 1. Ringing locations of Sandwich Terns resighted in the Forth of Firth in the period 2000-2004. Size of marks are proportional to number of birds [n=32] Scottish Birds (2008) ringed as nestlings, and their ages at resighting are shown in Table 2. The oldest, 19 years old, was ringed as a nestling at the Farne Islands on 20 July 1984, and resighted at the Joppa Rocks on 9 August 2003. This bird was about to make its journey to Africa for the twentieth time. Table 2. Age distribution of resighted Sandwich Terns in the Forth of Firth Age Qaawe- 45 > 12 14 15.916 19 Numbers#9Oe1-2 1 172 1 2 1 = =1 Colour Rings Several Sandwich Terns were seen with a combination of colour rings, usually in addition to their metal rings. None of the colour rings had readable inscriptions, and the combinations of colours used did not identify individuals. Moreover, it turned out to be difficult to discover the origin of the few colour ringed Sandwich Terns that were observed apart from colour ringed birds from Grampian and the Farne Islands. The 19 year-old Sandwich Tern ringed at the Farne Islands had 2 blue celluloid colour rings which, when I saw it nearly 2 decades later, were almost completely faded to a light grey color. I understand newer colour rings hold their colour for longer. Discussion The Firth of Forth is well known as an area which attracts large numbers of Sandwich Terns in the post breeding period. Wernham et al (2002) noted that post fledging dispersal around the coasts of Britain and Ireland and across the North Sea to the Netherlands and Denmark commences in late June, and that during July and August dispersal occurs in both directions between the Netherlands and Britain. Murray et al (1998) report that terns visit the Forth from as far as the Sands of Forvie to the north and the Farne Islands to the south. Post breeding movements of Sandwich Terns in the Firth of Forth sd The data presented in this short paper confirm this, but also show that some terns come from the Netherlands, Belgium, and Ireland. Observations of tern roosts can clearly provide much interesting information on migrating terns. Unpublished observations mainly of colour ringed Sandwich Terns from the Moray Firth to the Scottish Border showed a similar pattern of origins to this study (A Smith pers comm). Acknowledgements Thanks to Mark Grantham (BTO) for answering all my questions on ringed terns accurately and promptly. Further thanks to Raymond Duncan (Grampian), Roderick Thorne (Orkney), and Chris Redfern (Farne Islands) for information about terns that they ringed and I resighted. I also thank Alastair Smith for helpful comments that improved this article. Last but not least, 1am extremely grateful to Malvina Nissim who accompanied me on many a ring reading trip and provided invaluable feedback to previous versions of this paper. References Wernham C, Toms M, Marchant J, Clark J, Siriwardena G & Baillie S (eds) 2002. The Migration Atlas. T & AD Poyser. Kelly D J (ed) 1999. Lothian Bird Report 1998. Lothian Branch, SOC. Kelly D J (ed) 2000. Lothian Bird Report 1999. Lothian Branch, SOC. Murray R, Holling M, Dott H & Vandome P 1998. The Breeding Birds of South-East Scotland. A Tetrad Atlas 1988-1994. The Scottish Ornithologists’ Club, Edinburgh. Thomson I M (ed) 2003. Lothian Bird Report 2000. Lothian Branch, SOC. Johan Bos, Via Nosadella 34, 40123 Bologna BO Italy Revised ms accepted March 2008 32 SB 28 Wintering wader surveys on the Isle of Tiree, Argyll JOHN BOWLER, CARL MITCHELL, ROGER BROAD, ALAN LEITCH & DAVID STROUD Two surveys of the entire coastline of the Isle of Tiree including the coastal sections of the Sleibhtean agus Cladach Thiriodh (Tiree Wetlands and Coast) Special Protection Area (SPA) were conducted on 5-10 February 2006, by a team of 15 fieldworkers, using the same methodology as in previous counts. Results of the surveys are detailed here and placed in the context of the results of 5 similar midwinter coastal wader counts conducted on the island since 1985. Differences in count totals between the 2 surveys were small. Maximum whole island counts of key species in February 2006 included: 1191 Ruddy Turnstone, 648 Ringed Plover, 489 Sanderling and 368 Purple Sandpiper. Tiree still holds nationally important numbers of these 4 key wintering wader species. In contrast to counts from sites in Eastern Scotland, none of these species showed significant declines between 1985 and 2006. Tiree remains the premier wintering site for Ruddy Turnstones and Purple Sandpipers in the UK. Taking the maximum whole island count for either survey, the area within the SPA held 5721 waders in 2006 including 1006 Ruddy Turnstone, 578 Ringed Plover, 458 Sanderling and 305 Purple Sandpiper. Introduction The island of Tiree is the westernmost of the Inner Hebrides and is a flat, low lying island, roughly 20 km long and 5 km wide, which has long been known for its important bird populations. Its sensitively managed croftland is home to over 25% of the Corncrakes Crex crex in the UK, whilst its machairs and wetlands are home to nationally important breeding numbers of waders and internationally important numbers of wintering Greenland White-fronted Geese Anser albifrons and Barnacle Geese Branta (Bowler & Hunter 2007). The importance of its coastline for wintering waders is perhaps less widely known but its mix of broad sandy bays containing rocky outcrops, shelving rocky shores, numerous offshore skerries and sheltered inlets, provides a winter habitat for a wide range of wading birds and other waterfowl (see Madders & Moser 1989, Newton 1989). Five midwinter counts of the waders and waterfowl of the entire Tiree leucopsis coastline were made between 1985 and 2000. The results of these surveys were compiled in various reports to Scottish Natural Heritage and the findings were used to designate the most important sections of the coastline as part of Sleibhtean agus Cladach Thiriodh (Tiree Wetlands and Coast) SPA in 2001. In the light of recent reported declines in some key non estuarine wader species throughout the UK (see Rehfisch et al 2003), a repeat survey was organised in February 2006 involving 2 complete counts of the entire Tiree coastline. Methods For each count, the island was divided into 37 sections (see Fig 1) as in previous surveys and walked in stages by surveyors. Counts were carried out with care to reduce disturbing birds and to avoid duplication of results both within and between sections. A complete count of the island was conducted on 5—7 February and a second complete count on 8-10 February. All counts Scottish Birds (2008) ‘99° Wintering wader surveys on the Isle of Tiree, Argyll 33 25 Figure 1. Map of Tiree showing the 37 count sections were conducted within 3.5 hours of high tide and under favourable weather conditions: wind less than force 4, not during heavy or continuous rain and with good visibility throughout. Unlike mainland sites which are usually counted either side of low tide, the Tiree coast is best counted either side of high tide, as at low tide many extensive offshore skerries become exposed and offer feeding opportunities for waders which are difficult to count by a land based observer. All waders encountered feeding or roosting below high water mark were recorded for each section. Additional waders that were feeding or roosting just inland of a count section but which were Clearly associated with the shore, such as Ruddy Turnstone Arenaria interpres, Dunlin Calidris alpina and Ringed Plover Charadrius hiaticula, were also counted and coded separately as ‘inland’. Such inland counts of predominantly coastal wader species were included in the totals in order to obtain a closer estimate of the true island totals for these species. For each species the highest whole island count of the 2 surveys was taken as the most accurate count. Due to the tidal cycle, Soa (section 7), an offshore island, could only be counted on 10 February during the second survey. The number of birds encountered on that date was also included in the counts for the first survey (see Discussion). Additional species encountered on the surveys such as ducks, divers and Grey Herons Ardea cinerea were also recorded for each section. Additional counts were carried out on a selected number of sections on 11-13 February. The counts were not used in the calculation of whole island census totals, due to the possibility of movement of birds between sections, however they provided a useful check on the maximum number of birds using a particular section. 34 Results The highest whole island counts encountered on either survey are shown in Table |. A maximum total of 6,696 waders of 16 species was recorded, the second highest total count for the island following the peak count of 6,848 waders in February 1995. Numerically the most abundant species were Ruddy Turnstone and European Golden Plover Pluvialis apricaria with over 1000 birds, accounting for 17% and 15% of the total respectively. They were SB 28 followed by ~ Eurasian Oystercatcher Haematopus ostralegus, Ringed Plover, Northern Lapwing Vanellus vanellus, Eurasian Curlew Numenius arquata, Sanderling Calidris Alba and Common Redshank Tringa totanus, which accounted for 13%, 9%, 9%, 7%, 7% and 7% respectively of the total. Five species: Bar- tailed Godwit Limosa lapponica, Grey Plover Pluvialis squatarola, Eurasian Woodcock Scolopax rusticola, Red Knot Calidris canutus and Jack Snipe Lymnocryptes minimus were Table 1. Maximum whole island counts of coastal waders on Tiree, February, 2006. Species present in nationally important numbers are in bold (survey number in brackets) Max Count (survey) Eurasian Oystercatcher 889 (1) Ringed Plover 648 (2) European Golden Plover 1023 (1) Grey Plover 26 (2) Northern Lapwing 607 (1) Sanderling 489 (2) Purple Sandpiper 368 (2) Dunlin 270 (1) Bar-tailed Godwit 46 (1) Max Count (survey) Eurasian Curlew 493 (2) Common Redshank 483 (1) Ruddy Turnstone 1191 (1) Woodcock 4 (1) Snipe 173 2} Red Knot 1 (1) Jack Snipe 1 (1) Total 6718 Table 2. Differences in totals derived from 2 surveys of waders on Tiree in February, 2006. Species present in nationally important numbers are in bold. Grey Plover, Red Knot, Eurasian Woodcock, Jack Snipe and Bar-tailed Godwits have been excluded due to small counts (less than 100 birds involved). *counts affected by substantial numbers inland. **see Discussion Species Survey 1 Common Snipe * 99 Purple Sandpiper ** 263 Ringed Plover 607 Eurasian Curlew 473 Sanderling 468 Eurasian Oystercatcher 889 Common Redshank 483 Ruddy Turnstone 119] Dunlin 270 Northern Lapwing * 607 European Golden Plover * 1023 Survey 2 Difference 17S +779% 368 +40% 648 +7% 493 +4% 489 -4% 857 -4% 449 -7% 1059 -11% 22 -18% 257) -58% 225) -78% Scottish Birds (2008) present in small numbers and each accounted for less than 1% of the total wader count. Of these, Eurasian Woodcock and Jack Snipe are not especially associated with the shore and their all island wintering totals are likely to be higher. Common Greenshank Tringa nebularia was the only species not recorded in 2006, but this species is less than annual in winter on Tiree (Bowler & Hunter 2007). Numbers of Northern Lapwing, European Golden Plover and Common Snipe Gallinago gallinago have varied greatly over the years on the coastal counts. These 3 species have a predominantly inland feeding distribution on the island and the proportion recorded at the coast is likely to vary considerably between counts. The count results from the 2 surveys carried out in 2006 were generally very similar (Table 2), partic- ularly when the 3 predominantly inland wader species (European Golden Plover, Northern Wintering wader surveys on the Isle of Tiree, Argyll 35 Lapwing and Common Snipe) are excluded for analysis. The only coastal species to show a major difference between counts was Purple Sandpiper Calidris maritima (an increase of some 40% between the 2 survey periods - see Discussion). The counts made in 2006 are shown in relation to previous whole island surveys in Table 3. Key Species Accounts Ringed Plover Table 3 shows the whole island counts since 1985. The highest count was recorded in 1985 (987) but there has been no significant trend since the mid 1980s (rs=-0.314, ns). Up to 2004/05, Britain had experienced 16 years of decline in this species (Banks et al 2006), with a slight upturn in the national index in 2005/06 belying the underlying downward trend (Musgrove et al 2007). The 2006 count of 648 birds was below the threshold for a site holding internationally important numbers (730 birds), Table 3. Summary of whole island counts of coastal waders on Tiree, 1985 to 2006 1985 1986 Eurasian Oystercatcher 369 345 Ringed plover 987 555 European Golden Plover 59 9 Grey Plover 8 6 Northern Lapwing 69] 176 Sanderling 396 305 Purple Sandpiper 119 189 Dunlin 340 295 Bar-tailed Godwit 60 37 Eurasian Curlew 496 347 Common Redshank 503 pe) Ruddy Turnstone 996 861 Eurasian Woodcock ] 4 Common Snipe 73 33 Common Greenshank 0 0 Red Knot 0 0 Jack Snipe 0 0 Total 5068 3453 1995 1998 2000 2006 Average 807 714 443 889 595 958 534 561 648 707 9] 783 201 1023 361 0 12 45 26 16 16 449 1055 607 499 964 371 589 489 519 377 262 263 368 263 853 295 609 270 443 158 D2 69 46 65 780 901 845 493 643 611 415 415 483 453 1096 905 858 1191 985 0 | 0 4 l 137 183 319 75 153 0) l 0 0 0 0 0 l I 0 0 0 0 | 0 6848 5858 6273 6696 5704 36 but Tiree still holds nationally important numbers of this species. In the most recently published WeBS counts (winter 2005/06, Musgrove et al 2007), Tiree was the third most important wintering site in the UK for this species. Ringed Plovers were encountered on 59% of the 37 sections, showing a strong affinity for sandy bays (Fig 1 and Table 4). The average flock size on survey 2 was 34 birds. The highest count on a single section was 172 birds at Gott Bay (section 8b) on 8 February (survey 1). Sanderling The highest count was recorded in 1995 (964) but there has been no significant trend since the mid-1980s (rs=0.37, ns). Nationally, numbers have remained relatively stable since the mid 1970s (Musgrove et al 2007). The 2006 count of 489 birds was below the threshold for a site holding internationally important numbers (1200 birds), but Tiree still holds nationally important numbers of Sanderling. In the most recently published WeBS counts (winter 2005/06, Musgrove et al 2007), Tiree was the eighth most important wintering site in the UK for this species. Sanderling were encountered on 38% of the 37 sections, showing a strong affinity for sandy bays (Fig 1 and Table 4). The average flock size on survey 2 was 41 birds. The highest count on a single section was 106 birds at Traigh Sorobaidh (section 14) on 13 February (extra counts). Purple Sandpiper The highest count was recorded in 1995 (377), a marked increase compared to the mid 1980s. However, over the longer term, there has been no significant trend since the mid 1980s (rs=0.65, ns). Nationally, numbers have signif- icantly declined since the early 1980s (Musgrove et al 2007). The 2006 count of 368 birds was below the threshold for a site holding SB 28 internationally important numbers (750 birds), but Tiree still holds nationally important numbers of Purple Sandpipers. In the most recently published WeBS counts (winter 2006/07, Musgrove et al 2007), Tiree was the most important wintering site in the UK for this species. Purple Sandpipers were encountered on 84% of the 37 sections, showing a strong affinity for rocky shores (Fig | and Table 4). The average flock size on survey 2 was 16 birds. The highest count on a single section was 91 birds at Rubha Chraiginis (section 11) on 8 February (survey 2). Ruddy Turnstone The highest count was recorded in 2006 (1191 birds) but there has been no significant trend since the mid 1980s (rs=0.14, ns). Nationally, numbers declined significantly after the mid 1980s (Banks et al 2006) but appear to have stabilised since 2004/05 (Musgrove et al 2007). The 2006 count of 1191 birds was above the threshold for a site holding internationally important numbers (1000 birds in 2004/05) but below the revised threshold of 1500 birds set in 2007. In the most recently published WeBS counts (winter 2005/06, Musgrove et al 2007), Tiree was the most important wintering site in the UK for this species. Ruddy Turnstones were encountered on 92% of the 37 sections during either survey, and were found in most coastal habitats (Fig 1 and Table 4). The average flock size on survey 1 was 37 birds. The highest count on a single section was 317 birds at Clachan Mor (section 22) on 5 February (survey 1). Individual section maximum counts Using count data obtained from the 2 whole island surveys, plus the additional counts from 8 of the sections, Table 4 shows the maximum count in each section for 4 species of waders. Scottish Birds (2008) Wintering wader surveys on the Isle of Tiree, Argyll 37 Table 4. Maximum counts for 4 species of wader from individual sections. For section locations see Figure 1. Highest counts are in bold. * indicates the count was made during the additional count period (11-13 February). Sections in the Sleibhtean agus Cladach Thiriodh (Tiree Wetlands and Coast) SPA are indicated Section Ringed Plover Sanderling Purple Sandpiper Ruddy Turnstone 1 SPA * 62 * 95 * 67 2a SPA 3 33 54 2b SPA 3 SPA 4 30 4 6 37 5 SPA a0 40 6 10 6a SPA 5 83 6b SPA 6 7 SPA 25 44 8a SPA 58 93 4 26 8b SPA 172 79 13 54 9 5 2 a7 AT 10 SPA TT 30 4 21 11 SPA 91 81 12 SPA 9 95 4 13 SPA 20 28 69 14 SPA 94 * 106 «7 24 15 *4 * 32 16 SPA 96 3 89 17 SPA (part) 43 23 30 96 18 15 67 19 20 5 3 21 SPA 2 15 29 22 SPA 6 38 317 23 SPA 2 35 24 SPA 74 67 23 70 25 SPA 3 2 D 31 26 SPA 18 5 135 27 SPA 50 * Al 12 * 59 28 6 29 3 30 i 21 31 46 32 33 13 31 15 17 34 SPA Iz 30 7 SB 28 Table 5. Maximum counts of other species encountered during the wader surveys Species Count Common Shelduck 138 Mallard Anas platyrhynchos 269 Northern Pintail Anas acuta Z Northern Shoveler Anas clypeata 28 Eurasian Wigeon 483 Eurasian Teal 21 Common Eider 372 Long-tailed Duck 17 Red-breasted Merganser a2 Waterfowl All waterfowl encountered either on the shore or offshore (within approx 400m) were recorded during the surveys. The maximum counts derived from either survey are shown in Table 5. Apart from the species obviously associated with inshore coastal waters in winter (eg Great Northern Diver European Shag Phalacrocorax aristotelis, Common Eider Somateria mollissima, Long- tailed Duck Clangula hyemalis and Black Guillemot Cepphus grylle) most species counted are also encountered inland on Tiree. Therefore, the counts of the remaining species are additional to inland counts. However, a number of these counts are significant in an island context. The Common Shelduck Tadorna tadorna count (138 birds) was the highest ever winter count recorded on Tiree, that of Grey Heron (30) was the second highest ever recorded, whilst those of Eurasian Wigeon Anas penelope (483) and Eurasian Teal Anas crecca (521) were the highest ever recorded coastal counts for these species. Gavia immer, Discussion Counts of the 4 most important coastal wader species showed good correlation between the 2 surveys, with counts of 3 of the species being within 11%. The exception was Purple Sandpiper, with the second survey finding 40% Species Count Great Northern Diver 19 Great Cormorant 66 European Shag 128 Black Guillemot 3 Grey Heron 30 more birds than the first, a difference of 105 birds. Comparisons of field notes after the counts showed that 9 birds were seen at section 11 (Rubha Chraiginis) on the first survey and 91 birds on the second, and it is thought that these latter birds were missed on the first survey by feeding out of sight on a tidal islet. Correcting for this undercount would have brought the 2 counts to within 7% of each other (345 compared with 368, a difference of 23 birds). Unfortunately, due to the tidal cycle, section 7, the Isle of Soa, could only be visited during the second survey. Rather than omit counts from the second survey, the counts encountered during that visit were also included in the totals for the first survey. In comparison to other sections, the number of birds involved was very small. For the 4 most important species, the counts included: 44 Ruddy Turnstone (4% of the overall total), and 25 Purple Sandpipers (7% of the overall total). No Sanderling or Ringed Plover were seen on Soa. Notably fewer Sanderling and Dunlin were encountered in 2006 compared to the mid 1990s, although there was no significant long term trend for the numbers of Sanderling. Numbers of Dunlin on the survey were low in comparison to counts obtained of single flocks of the species earlier in the winter. For example, 260 Dunlin Scottish Birds (2008) Wintering wader surveys on the Isle of Tiree, Argyll 39 were counted at Gott Bay (section 8b) alone on 21 January 2006 and there were 198 at Sorobaidh Bay (section 11) on 13 December 2005, and it seems likely that some birds may have been feeding at inland sites, and therefore missed, during the survey. An alternative suggestion is that some waders that normally winter on Tiree could have shifted to feed on the adjacent island of Coll. In order to check this, Simon Wellock and Val Conway were asked to undertake an assessment of the beaches at the west end of the neighbouring island of Coll concurrently with the second round of survey work on Tiree. Only small numbers of shorebirds were located on the Coll beaches, including 47 Ruddy Turnstone, 105 Ringed Plover, 92 Sanderling and 7 Purple Sandpiper. These are typical figures for wintering shorebirds on Coll and suggest that there had been no major influx of birds from Tiree during the survey period. Additional counts on a selected number of sections on 11-13 February did not record notably larger counts of waders than on the 2 whole island surveys. This also suggests that no large flocks of Sanderling or Dunlin were missed during the first 2 surveys. Tiree still holds nationally important numbers of 4 key wintering wader species. In a national context, treating Tiree as a single site, the island is the most important site in the UK for wintering Purple Sandpiper and Ruddy Turnstone, the third most important wintering site in the UK for Ringed Plover (after the Thames estuary and Hamford Water) and the eighth most important site for wintering Sanderling. While there was variation between years, none of the 4 key species showed a significant trend in numbers between 1985 and 2006, which contrasts with the results from other Scottish sites. On the Moray Firth for example, numbers of Purple Sandpiper and Ruddy Turnstone decreased significantly between 1988 and 2003 (Kalejta-Summers 2006), whilst on the Tay estuary, Dunlin have undergone a consid- erable decline since 1997 (Elkins 2006, 2007). The decline in numbers of Purple Sandpipers and Ruddy Turnstones on the Moray Firth was thought to be related to poor breeding success resulting in low recruitment in the late 1980s and early 1990s (Summers et al 2005), and 1s part of a UK decline in these species, recorded at 16% for Ruddy Turnstone and 21% for Purple Sandpiper between 1984/85 and 1997/98 on a survey of 38% of the UK’s non estuarine coastline (Rehfisch et al 2003). Reported declines in numbers of Ruddy Turnstone and Purple Sandpiper in East Lothian between the 1970s and 1990s were even higher at 73% for Ruddy Turnstone and 88% for Purple Sandpiper (Dott 1997) indicating that these declines may be more severe in southeast Scotland. The fact that numbers of both species have not declined significantly over the same period on Tiree, suggests that either recruitment has been higher for west coast wintering populations of these species, or that some redistribution has occurred on the wintering grounds. Numbers of Ringed Plover showed no significant trend over the period 1985—2006, although counts in 1985 and 1995 were markedly higher than in subsequent years. These relatively stable numbers contrast with the national trend, which indicates a decline of more than 30% between 1988/89 and 2005/06 (Musgrove et al 2007) and a decrease of 15% on non estuarine coasts between 1984/85 and 1997/98 (Rehfisch et al 2003). Numbers of Sanderling similarly showed no significant trend over the period 1985-2006, which reflects their national index, which has remained relatively stable since the mid 1970s (Musgrove et al 2007). The coastal counts of European Golden Plover, Northern Lapwing and Common Snipe underes- timate the numbers of these species wintering 40) on the island, as they predominantly feed on inland grasslands. All island counts in recent winters have revealed that Tiree annually holds nationally important numbers of European Golden Plover (3000—4500 birds in mid winters 2004-2007) and regionally important numbers of Northern Lapwing (2500-4500 birds in mid winters 2004-2007, Bowler & Hunter 2007). Numbers of wintering Common Snipe on the island are poorly known but as Tiree’s snipe shooting is still renowned as amongst the best in Europe, these are likely to be high and of national importance. Of the other species, Ruddy Turnstone, Eurasian Oystercatcher, Ringed Plover, Eurasian Curlew, Common Redshank and Dunlin also feed on short turf inland pastures on Tiree to a varying extent during the winter and coastal counts in a given winter may miss some birds. However, these species are predominantly coastal in distribution and the counts should record the majority of birds present. Some species can be present in even higher numbers on the island during spring and autumn migration periods. Recent high counts include an estimated 2000-3000 Ringed Plovers present on 16 May 2005, some 1500-2000 Sanderling in mid-August 1995 and a minimum of 2010 Sanderling on 17 May 2006 including a flock of 1320 birds at Gott Bay alone (Bowler & Hunter 2007). Similarly, numbers of passage European Golden Plovers have peaked in recent years at around 6,500 birds on the island during spring passage in March/April and on autumn passage in September/October (Bowler & Hunter 2007). Acknowledgements We are greatly indebted to all the coordinators and fieldworkers of the earlier surveys not named elsewhere, including Gwen Evans, Greg Corbett and Andy Hachenburg, and to RSPB Scotland and Scottish Natural Heritage for supporting the work financially. Special thanks SB 28 go to Ross Lilley (SNH) for arranging funding for the survey work in 2006. Many thanks to the additional fieldworkers who helped with the 2006 surveys: Sue Agnew, Clifton Bain, Jan Baylis, Raymond Duncan, Ian Francis, Mick Green, Janet Hunter, Lyndsey Kinnes, Alan Lauder, Ross Lilley and Fiona Newcombe. Simon Wellock and Val Conway undertook counts on Coll. References Banks A, Collier M, Austin G, Hearn R & Musgrove A J 2006. Waterbirds in the UK 2004/05, The Wetland Bird Survey. BTO/WWT/RSPB/JNCC. Thetford. Bowler J & Hunter J 2007. The Birds of Tiree and Coll. Paircwood Publishing. Dott H E M 1997. Declines in Ruddy Turnstones and Purple Sandpipers wintering in south-east Scotland. Scottish Birds 19: 101-104. Elkins N 2006. Waterfowl counts on the Tay Estuary. Scottish Birds 26:30-38. Elkins N 2007. Low tide counts on the Firth of Tay. Scottish Birds 27:64—67. Kalejta-Summers B 2006. Changes in numbers and distribution of waders in the Moray Firth, 1988-2003. Scottish Birds 26:1-16. Madders M & Moser M 1989. Coastal waders in winter on Tiree and Coll. Pp 67—73 in Stroud D A (Ed) The Birds of Coll and Tiree: Status, habitats and conservation. Nature Conservancy Council/Scottish Ornithologists Club. Edinburgh. Musgrove A J, Collier M P, Banks A N, Calbrade N A, Hearn R & Austin G E 2007. Waterbirds in the UK 2005/06, The Wetland Bird Survey. BTO/WWT/RSPB/JNCC. Thetford. Newton S F 1989. Wintering wildfowl of Coll and Tiree. Pp 99-114 in Stroud D A (Ed) The Birds of Coll and Tiree: Status, habitats and conservation. Nature Conservancy Council/Scottish Ornithologists Club. Edinburgh. Rehfisch M M, Holloway S J & Austin, G E 2003. Population estimates of waders on the non-estuarine coast of the UK and the Isle of Scottish Birds (2008) Wintering wader surveys on the Isle of Tiree, Argyll Man during the winter 1997-98. Bird Study 50: 22-32. Summers RW, Butterfield D P, Swann RL & Insley H 2005. The decline of the Purple Sandpiper Calidris maritima in the Moray Firth, Scotland - an effect of recruitment. Wader Study Group Bulletin 106: 34-38. John Bowler, RSPB, Pairc na Coille, Balephuil, Isle of Tiree, Argyll PA77 6UE. e-mail john.bowler@rspb.org.uk Carl Mitchell, 1, Station Cottages, Kingussie, PH21 [EW e-mail carl.mitchell@wwt.org.uk Roger Broad, RSPB, 6 Birch Road, Killearn, Glasgow G63 9SQ. e-mail roger.broad@btinternet.com Alan Leitch, RSPB, The Willows, Old Finstown Road, Finstown, Orkney KW17 2E]. e-mail alan.leitch@rspb.org.uk David Stroud, JNCC, Monkstone House, City Road, Peterborough PE] LJY. e-mail david.stroud@jncc.gov.uk Received ms accepted April 2008 4] 42 SB 25 SHORT NOTES Changes in the breeding status of Little Ringed and Ringed Plovers in North Fife The first recorded breeding of Little Ringed Plovers Charadrius dubius in Fife was documented in 1989 (Scottish Birds 16: 42-43). The species was present at another site in 1990, but did not breed. For reference, I will call all sites by letters, A, B and so on. No subsequent nest was found until 1997, when a pair bred at another site C. First discovered on 25 May, 2 newly hatched chicks were ringed on 14 June, and one chick seen to fly on 7 July. The other chick was predated by a _ Eurasian Sparrowhawk Accipiter nisus. The female of this pair associated with a second male at a nearby site D. They formed a territory but no nest was found. On 13 July all 3 adults and the fledged chick were in a group at site C. The male of the successful pair and his chick were last seen on 20 July by which date the other 2 adults had departed. The nest of 1997 was in a marginal grass field, located within a tiny area of gravel, probably caused by a lorry cleaning its container. A muddy pool was situated some 200 metres away to which the chicks were led after hatching and around which they fed. In 2003 a single male came to another site E with a pair present by 2 June. The first clutch was found on 13 June, and eventually 3 pairs bred around a large lagoon. Nine chicks fledged from 12 eggs laid, despite the attentions of a juvenile Peregrine Falcon Falco peregrinus, seen to hunt adults on at least 2 occasions. The last records were on 24 August when 2 juveniles were still present. Eleven chicks were ringed under licence, but odd unringed juveniles passed through the site in late summer suggesting breeding elsewhere. In 2004 site E saw its first Little Ringed Plover arrive on 11 April, mating was seen on 3 May and the first clutch found on 30 May. Four pairs were proved to breed, 10 chicks were ringed under licence, of which 6 flew. One pair probably had early clutches destroyed in vehicle movements and the presumed replacement clutch of 2 was incubated until 21 July, then deserted. The last record of a definite breeder was 27 July when a ringed juvenile present. In 2005 at site E, the first adult was seen on 26 March, with 5 by 13 April (but only one female), 7 by 21 April and the first clutch by 10 May. Ten adults were present by 26 May, including the incubating bird. Four chicks were ringed here on 31 May still in the nest scrape and almost immediately predated. Carrion Crows Corvus corone constantly patrolled the nest site and adult Little Ringed Plovers alarm called when they appeared. A second nest was found on 8 June and a third by 10 June. By 22 June the second nest was predated and within a few days the third was also predated. Control measures of perhaps 2 pairs of crows will be required if plovers are to breed successfully at this site. No chicks fledged but a few unringed juveniles were recorded here in late summer. In 2006 at site E the first adult was seen on 7 April with 4 by 29 April. Only one pair stayed and a suspected early clutch very probably taken by crows. A replacement clutch was found on 1 July, but the eggs were gone by 9 July and no chicks seen. Both adults departed by 13 July. Also in 2006 at site F, a new site of dry gravel deposits where people walk dogs, there were 2 adults and one chick about a week old which I ringed under licence on 26 June. On 22 July the single chick could just fly. This was a considerably longer fledging period than chicks which had access to mud. On 25 July one adult and the chick were still present. The adult chased off an unringed Scottish Birds (2008) juvenile which flew in from the north and landed. On 27 July the local chick still flew poorly and was not seen to fly well until 30 July when the adult was still present. Both birds were gone by 6 August. Throughout the period of increased Little Ringed Plover breeding records the Ringed Plover Charadrius hiaticula has declined. In the Howe of Fife gravel pits it has ceased to breed regularly, from some 6—10 pairs in 1990s (Jay Ringing Group Report 1994-95 pp 13-16). Some of this may be attributed to a reduction in suitable breeding habitat, planning approval to extract minerals requiring the operators to return such workings to farmland when operations are exhausted. This has a negative effect, not only on breeding waders, but on other wildlife, which finds the disturbed, unsprayed land ideal for colonisation. Some sites on which the Little Ringed Plovers have bred would appear to be suitable for the larger species, but although occasional pairs have formed territories, no recent nests have been found. Without having conclusive proof, I suggest that amelioration of the climate is causing this decline, whilst furthering the spread of the Little Ringed Plover. The Ringed Plover continues to attempt to breed in coastal Fife, but with limited success due to human disturbance. I have spent many evenings and weekends attempting to protect breeding sites of Little Ringed Plovers and can only apologise to the few people I have asked to leave without being able to give explanations. My thanks to the landowners and their employees who have aided this research, often by creating, or not destroying, suitable habitat for these waders to breed on. I always tell them that I am the cheapest night watchman they will ever employ. DW Oliver, Dunearn, The Feus, Freuchie, Fife. KYI5 7HR Revised ms accepted April 2007 Short Notes 43 Intraspecific interaction by Hen Harriers over a cock’s nest Intraspecific interactions between harriers Circus sp are greatest early in the nesting season and decrease as the season progresses (Palmer 1988. Handbook of North American Birds, vol 4, London). In the spring of 2007 I watched an extension of this behaviour when 2 adult male Hen Harriers Circus cyaneus interacted over a cock’s nest. In April I watched a male Hen Harrier build a cock’s nest in a traditional nesting area, site 7, in Galloway. On 2 May this male flew from site 7 directly west for 2 km towards a soaring male above another traditional nesting area, site 1. Both males circled each other, closely, with the site 7 male overhead until both went out of sight in a heat haze. Thirty minutes later the site 7 male returned and landed on his cock's nest. On 15 May at 1048 hours the site 7 male flew up from the nest heading east. At 1126 hours the male from site 1 flew low directly across the moor, landed at the site 7 cock's nest and remained there. Ten minutes later the site 7 male returned and, about 30m from his nest, skydanced low down 8 times. He then swooped on his nest whereupon the site | bird flew up. The males circled each other closely, with the site 7 male overhead. Both males then soared until the site 7 male stooped on the other male a km from the cock's nest; they then drifted over a hill and out of sight. The site 7 male returned at 1158 hours and landed on a fence post in his nesting area. Neither male bred or even attracted any females. Conflicts between male Hen Harriers whose nests are close together is well known (Balfour 1962, Bird Notes 30: 145—152) but conflicts between 2 males disputing an unattended cock’s nest is unusual. RC Dickson, Lismore, New Luce, Newton Stewart, Dumfries & Galloway DGS OAJ Revised ms accepted August 2007 44 Breeding success of Golden Eagles on the Isle of Rum 1995-2007 The breeding success and productivity of Golden Eagles Aquila chrysaetos on the Isle of Rum has been monitored annually since 1995. Over this period there have been 3 occupied home ranges. The results of the monitoring are presented in Table 1. These results show a_ considerable improvement in productivity of Golden Eagles on Rum since the early years of the island becoming a nature reserve in 1957. Corkhill (1980, Scottish Birds 11:33-43) calculated a figure of 0.29 young per occupied territory per year for 4 occupied home ranges in the years 1957-76. The figure presented in Table 1 of 0.69 young fledged per occupied territory is higher than that found for the whole of Britain (0.36) in the national survey of 2003 (Eaton et al 2007, Bird Study 54:212—220) and is indeed better than the figure for the area with the highest productivity in that survey of 0.68 young fledged per pair for the Eastern Highlands. In 2003 all 3 pairs of Golden Eagles on Rum fledged a single chick each. The reason for improved productivity of Golden Eagles on the Isle of Rum remains unclear. The low productivity from 1957-76 SB 28 was thought to be due to pollutant burdens carried by the eagles because of their tendency to feed on seabird prey (Corkhill 1980, Furness et al 1989 in Raptors in the Modern World ed Meyburg and Chancellor. WWGBP: Berlin, London and Paris: 495-500). Unfortunately detailed analysis of prey remains and pollutant burdens for the Rum Golden Eagles has not been carried out in the years 1995-2007 . The improved productivity could be due to a change in dietary habits or a reduction in the pollutant burdens of the eagles’ seabird prey. It would be interesting to carry out further research to investigate the causes behind the improved productivity of the Rum Golden Eagles. Sean Morris, 2 Kinloch Cottages, Isle of Rum, PH43 4RR Revised ms accepted October 2007 Table 1. Breeding success of Golden Eagles in 3 occupied territories on Rum 1995-2007 No of years No of years No of territory breeding years occupied attempt failed recorded Territory 1 13 8 3 Territory 2 13 13 3 Ternitory 3. 13 10 3 Total 39 St 9 Total Young per Young per young breeding occupied reared attempt territory per year 5 0.62 0.38 14 1.08 1.08 8 0.8 0.62 agi 0.87 0.69 Scottish Birds (2008) Survey of roof nesting gulls in East Sutherland A survey of roof nesting gulls was carried out in Caithness in 2006 and found a total of 210 + 20 apparently occupied nests (aon), all Herring Gulls Larus argentatus (Clark, Maughan and Sellers 2007, Scottish Birds 27:74—76). The birds were distributed in 4 colonies, 2 small and apparently newly formed ones, a third established at least 15 years ago, and a fourth, the largest of them all, of similar age but which had been overlooked during the national seabird survey (Seabird 2000) in 1998-2002 (Mitchell et al 2004, Seabird Populations of Britain and Ireland, Poyser, London). During the course of the Caithness survey it became apparent that there were more colonies of roof nesting gulls in East Sutherland than had been found during Seabird 2000 (a single, mixed colony at Ardgay; Mitchell et al 2004, M Parsons, pers comm). Accordingly I undertook a survey of roof nesting gulls in East Sutherland during the 2007 breeding season. Short Notes 45 As with the 2006 Caithness survey, roof nesting gulls in East Sutherland were surveyed in 2 phases, an initial one during the last week of May to locate colonies and make counts of sitting birds when nests are generally at their most conspicuous, and a second survey about 4 weeks later to identify any additional nests based on the presence of chicks, and determine productivity. Chicks frequently move a short distance from their nests and in so doing reveal the presence of nests which may not have been visible during the earlier survey. The survey found a total of 88 aon of roof nesting gulls (71 aon Herring Gulls, 12 aon Common Gulls L canus, and 5 aon Great Black- backed Gulls L marinus) in 7 colonies: further details in Table 1. Common Gulls were found at a total of 5 sites, but at none of these did the total number of roof nesting birds exceed 4 aon; 3, those at Ardgay, Bonar Bridge and Lairg, were exclusively Common Gulls, whilst the other 2, at Embo and Dornoch, were within roof nesting Herring Gull colonies. In addition the colonies at Ardgay and Dornoch had some Table 1. Roof nesting gulls in East Sutherland in 2007 Colony Map ref Number of breeding pairs (aon) Common Gull Herring Gull Great B-b Gull Brora a NC9003/9004 0 24 b 2 Golspie NC8200, NH8299/8399 0 18 3 Embo NH8192 l 3 0 Dornoch NH7989/8089 4c 26 0 Ardgay NH5990/6090 4d 0 0 Bonar Bridge NH6091 | 0 0 Lairg NC5706 Z 0 0 Total 12 71 5 (a) In 2006 there were a minimum of 17 aon Herring Gulls and | aon Great Black-backed Gulls in this colony (based on single survey carried out at the end of June). (b) In addition 3 aon nesting on the ground. (c) In addition at least 3 aon nesting on the ground. (d) In addition at least 8 aon nesting on ground in railway yard. 46 ground nesting birds associated with them, and this may have been the case at Bonar Bridge and Lairg, both places where Common Gulls breed on the ground nearby. Four roof nesting Herring Gull colonies were found, one small one at Embo, and 3 larger ones at Brora, Golspie and Dornoch. The colonies at the first 2 of these three larger colonies had a few Great Black- backed Gulls associated with them. The nest sites used by the birds were quite varied. Of the 12 roof nesting Common Gulls, 5 had made their nests on chimney stacks, 4 on flat roofs and 3 in chimney pots. The Herring Gulls had built their nests as follows: 28 (39%) on chimney stacks, 22 (31%) on flat or gently sloping roofs, 8 (11%) on steeply sloping roofs usually behind standpipes, chimney stacks or other roof structures, 5 (7%) in the ‘V’ between two sloping roofs, 4 (6%) on flat coping stones and 4 (6%) in chimney pots. The Great Black-backed Gull nests had been built on flat or gently sloping roofs (4 nests) and on a steeply sloping roof (one nest). Nests in the 3 larger colonies were quite widely distributed, typically around a core area with a higher density of nests. In Golspie this was SB 28 centred on the High School, the whole colony covering an area of roughly 14 ha. Quite a number of houses to the south of the High School had netting over their chimneys or were fitted with spikes to deter gulls from nesting, usually a sign that nesting has been taking place for some time, and local residents confirmed that birds have bred on the roofs here since the early to mid 1990s. The position in Brora was similar. Gulls formerly bred on the roof of the old Hunter’s Mill on the north west side of the town, and colonised the remainder of the town when the mill was demolished around 2000. There remains a concentration of breeding birds in the vicinity of the mill site, including several pairs with nests on the ground despite efforts in recent years to discourage this by pricking the eggs. This colony currently extends over an area of 59 ha. In Dornoch the main concentration of nests was in the town centre, but the colony extends throughout much of the town covering about 39 ha. Only a few buildings showed any signs of measures to discourage nesting. Nevertheless local residents say that gulls have nested on the roofs here since at least the early 1990s. The small colony in Embo, by contrast, was only established about 5 years ago. Table 2. Productivity of Herring Gulls nesting in East Sutherland and Caithness in 2007 Colony Map ref Colonies in East Sutherland Brora a NC9003/9004 Golspie NC8200, NH8299/8399 Dornoch NH7989/8089 Colonies in Caithness Wick ND3650/51, ND3751 Castle of Old Wick ND3648 Badbea ND0920 (a) The equivalent figure for 2006 was 2.15 chicks per successful pair (n = 13). Habitat Productivity chicks per pr No. nests roofs 1.67 15 roofs 222 9 roofs 2.00 s) roofs 1.94 19 cliff ledges 1.56 18 boulder beach 1.06 17 Scottish Birds (2008) Productivity estimates for the 3 main Herring Gull colonies are shown in Table 2, together with some data from roof nesting and coastal colonies in Caithness collected in 2007 for comparison. As in the earlier study, the roof nesting colonies were’ generally more productive than those on the coast. Seabird 2000 reported only one roof nesting colony of gulls in East Sutherland, that at Ardgay and good numbers of Herring Gulls (33 aon) and Common Gulls (13 aon) as well as single nests of Lesser Great Blacked Gull L. fuscus and Great Black-backed Gull. There had evidently been a dramatic change in the birds’ fortunes by the time of the present study when just 4 aon Common Gulls were found. Of the 4 Herring Gull colonies found in 2007 at least 3 (Brora, Golspie and Dornoch) appear to have been formed in the 1990s, and were evidently overlooked at the time of the Seabird 2000 survey; the fourth, that at Embo has been established since Seabird 2000 was carried out. Ackowledgements My thanks to the many people in East Sutherland who shared information with me about their birds, allowed me onto their properties to count birds and in one instance even lent me a ladder so that I could examine a nest on a roof. I am indebted to Simon Foster, Kenny Graham, Hugh Insley, Al McNee, Matt Parsons, David Paterson and Bob Swann for answering my queries about roof nesting gulls in East Sutherland, and to JNCC for allowing access to the Seabird 2000 results. Robin M Sellers, Crag House, Ellerslie House, Gosforth, Cumbria CA20 IBL (e-mail: sellers@craghouse7 freeserve.co.uk) Revised ms accepted November 2007 Short Notes 47 Breeding dates and success of Bass Rock gannets in 2007 There is a suggestion (Nelson J B 2006; 2007 Scottish Birds 26:50; 27:68) that Northern Gannets, Morus bassanus, on the Bass Rock are nowadays laying later than in former years and that, apparently, a higher proportion of established pairs are failing to breed even though they hold on to their sites. In 2007 JH used the pictures transmitted by cameras on the Bass Rock to the Scottish Seabird Centre to obtain information on laying dates, breeding success and the proportion of adult non breeders. Six sample plots, each containing between 17 and 32 apparently occupied sites (AOS), in total 135, were chosen within the well established breeding ranks. They were all reasonably visible on the screens in the centre. Each plot was sketched to show prominent features as land marks and the sketches were used during each observation period. Within each plot JH tried to record the dates on which eggs were laid and on which the chicks fledged; 32 sets of observations were made between late April and mid October mostly late in the afternoon when there were fewer visitors at the Centre as it was impracticable to bar visitors from using the cameras. Each observation period took 1-2 hours during which a variable number of the plots were assessed. The dates on which eggs were laid and chicks hatched were estimated by working back from the first occasion on which a chick could be aged, using criteria in Nelson (1964 Ibis 106: 63-77 and 1978 The Gannet, T&A D Poyser. Berkhamsted). The results are given separately for each plot since to amalgamate would obscure variation and hinder plot specific comment. 48 SB 28 Table 1. Results from sample monitoring plots Date of | Dateof Number Number Date last Fledging Productivity firstegg lastegg knownto fledged fledged success have hatched Plot 1 20 AOS) c25 Apr cl3May 9 9 cl Oct 100% 45% Plot 2 I7AOS = c25Apr_ cl6May 8 8 cl Oct 100% 47% Plot 3 21AOS c28Apr cl6May 8 7 cl Oct 87% 33% Plot 4 22 AOS c28Apr — cl0May 9 9 el Oct 100% 41% Plot 5 . 32 AOS” c28 Apr cl9May~ 18 13 cl Oct 72% 43% Plot 6 23 AOS c28Apr = cl9May 14 10 c5 Oct 713% 43% Note a. Productivity is chicks fledged from occupied sites; not breeding success which is chicks fledged from eggs laid. Note b. 2 adults died in May and no breeding occurred on these sites. Note c. Excluding a late replacement egg laid c8 July. Thus from 135 AOS a minimum 66 chicks hatched (48%) out of which 56 fledged (81%); overall productivity (fledged from occupied sites) was 41%. The figures from the individual plots show that: @ Only 48% of AOS hatched a chick. On the reasonable assumption that egg loss in 2007 was about normal for gannets (ie very low) it seems that a high proportion of site owning pairs did not attempt to breed. @ Of those that did lay, a normally high proportion succeeded in rearing a chick to fledging. Plots 5 & 6 were exceptional in being near the base of the east cliffs and somewhat vulnerable to heavy seas. @ As in 2005 and 2006, the mean laying date in 2007 was significantly later than in the period 1960-90. These observations appear to support the suggestions (Nelson Joc cit) that Bass gannets are breeding later in the year and that a higher proportion of site owning and established pairs are not breeding than used to be the case. This is a critically important observation since it raises the possibility that a higher proportion of adults than formerly are below the threshold of fitness for breeding. It would be highly desirable to obtain information on this via an index of fitness such as body weight and fat content. Scottish Birds (2008) On only one occasion was a chick seen to be left unattended, which indicates that protection of site/chick is a high priority even when it may be that the availability of food within normal foraging distance is sub optimal. If there had been a great deal of human intrusion into the colony with consequent egg loss then the figure for productivity and the assumption about withholding breeding would be meaningless, but there is no reason to assume such intrusion. J F Hunt, 20 York Road, North Berwick, East Lothian EH39 4LX J Bryan Nelson, Mine House, Auchencairn DG7 IRL Revised ms accepted January 2008 The first Lesser Grey Shrike in Scotland: clarification and correction One important aspect of The Birds of Scotland (Forrester R W, Andrews, I J, McInerny C J, Murray R D, McGowan R Y, Zonfrillo B, Betts M W, Jardine D C & Grundy DS (eds) 2007, The Scottish Ornithologists’ Club, Aberlady) was the inclusion of full collection details of the first few occurrences of all rare and other noteworthy species. For ‘firsts’, in particular, it is important that the documented record is accurate and here we correct an error concerning the first Lesser Grey Shrike Lanius minor. The Birds of Scotland gave as the first occurrence an adult male collected by Jerome Wilson on 25 May 1928 on Fair Isle; this specimen is held at the National Museums Scotland (NMS) (NMS.Z 1928.58). The account also dismissed another Fair Isle record from Short Notes 49 1913, citing apparent confusion with Woodchat Shrike Lanius senator, and quoting The Birds of Shetland (Pennington M, Osborn K, Harvey P, Riddington R, Okill D, Ellis P & Heubeck M, 2004, Christopher Helm) as the source. During the compilation of The Birds of Shetland it was not always possible to refer to original sources. The main sources of historical data for Fair Isle were William Eagle Clarke’s Studies in Bird Migration (1912), which pre dates this record, Peter Davis’ list which appeared in Ken Williamson’s Fair Isle and its Birds (1965) and Nick Dymond’s The Birds of Fair Isle (1991). The latter 2 sources only summarised records for Lesser Grey Shrike (and, indeed, Woodchat Shrike) and these totals did not tally with the number of records traced. Davis listed 12 records of Lesser Grey Shrike, whereas 13 were located. Jerome Wilson’s short note on the 1928 event was accompanied by an editorial statement noting it as the second Scottish occurrence with the first also at Fair Isle on 3 June 1913 (Wilson J, 1928 Scottish Naturalist 1928: 127). Later, Baxter and Rintoul also noted the latter as first record, with a Woodchat on 4 June (Baxter E V & Rintoul, L J, 1953, The Birds of Scotland, Oliver & Boyd). During the editing of The Birds of Shetland, however, no further details of the 1913 record could be traced and it was wrongly assumed that the close juxtaposition of occurrences of both Lesser Grey and Woodchat Shrike on Fair Isle on 3 and 4 June 1913 had led to confusion. In fact the separate occurrences of these birds were explicitly recorded in Clarke’s paper on migrants at Fair Isle in 1913, which had been overlooked in the compilation of The Birds of Shetland (Clarke W E, 1914, On some migratory birds observed at Fair Isle during the spring and autumn of 1913. Scottish Naturalist 1914: 53-55). An adult female Lesser Grey was collected on 3 June and crucially noted as the 30 first for Scotland, with an adult female Woodchat collected on the following day. Both these specimens entered the collections of the NMS and were mounted for exhibition. Remarkably, both came to light in the British Bird Hall in January 2008 when specimen lists were being prepared in advance of closure of public galleries as part of redevelopment of NMS. Due to the vagueness of labelling, the historical significance of the mounts had been hitherto missed. This provenance is further confirmed by other literature sources in NMS. Therefore the correct attribution of the first Lesser Grey Shrike in Scotland 1s: Adult female, collected by Jerome Wilson on 3 June 1913 at Fair Isle. Specimen at NMS (NMS.Z 1913.145.5). Robert Y McGowan, Department of Natural Sciences, National Museums Scotland, Chambers Street, Edinburgh, EH1 1JF Mike Pennington, 9 Daisy Park, Baltasound, Unst, Shetland, ZE2 9EA Christopher J MclInerny, 10 Athole Gardens, Glasgow, G12 9AZ Revised ms accepted April 2008 SB 28 Merlin impaling prey on barbed wire fence On 19 June 2007 I watched a male Merlin Falco columbarius hunting over moorland in Galloway. He stooped on a Meadow Pipit Anthus pratensis and caught it. The Merlin flew down to a fence post and managed to impale the Meadow Pipit on to the spikes of the barbed wire fence similar to a Great Grey Shrike Lanius excubitor wedging prey. The Merlin then partly plucked the Meadow Pipit for 5 minutes. He flew up with the prey and soared for over 2 km before diving down toward his breeding site. Hunting perches used in the summer by Merlins include fence posts in Galloway (Dickson 2002, British Birds 95: 141-142) but for a Merlin to impale prey on a barbed wire fence 1s unusual behaviour. R C Dickson, Lismore, New Luce, Newton Stewart, Dumfries & Galloway DG8 OAJ Revised ms accepted March 2008 Scottish Birds (2008) Robert Grier Caldow 1922-2006 Bob Caldow was born in Troon on 29 September 1922, and educated at Marr College in that town. He subsequently won bursaries to Glasgow University, his studies being interrupted by war service, which he undertook at the Nobel Division of ICI between 1944 and 1947. Following the end of hostilities Bob completed his chemistry degree, and then worked in industry until deciding to pursue a career in teaching. His first placing was at Allan Glen’s School in Glasgow between 1964 and 1972; from there he progressed to the post of Assistant Head Teacher at Paisley Grammar, remaining there until retirement in 1986. Obituaries Si Bob was a talented and popular teacher, and also passed on his passion for birds to many of his students through the school bird club which he organised. Ill always remember them attempting to squeeze as many bodies as possible into a makeshift hide on the saltmarsh at Erskine. A sense of fun was an important ingredient which Bob added to these outings. He was elected Glasgow Branch chairman, then a Council member of SOC in 1966. Bob brought warmth to these posts which endeared him to everyone. As youngsters, George White, myself and the rest of the Paisley Moss watchers hung on his every word; this was our Mr Caldow that we turned to for advice when anything unusual turned up on our local patch. He was never too busy to spend time on the phone, or to meet us down at ‘the moss’ to confirm our latest sighting. Bob was a regular at the SOC conference, an event to which he eagerly looked forward, renewing friendships and exchanging the past year’s experiences over a pint. Bob will be fondly remembered. He was an unassuming man; not many are aware that Bob Caldow was the principal advocate who convinced George Waterston to establish the RSPB Reserve at Lochwinnoch. For much of the 1960s the Barr Loch was Bob’s local patch and he dreamed of restoring the ancient drainage system to recreate the glory days when the loch was a huge marsh, with breeding Black-necked Grebes amongst the vast colony of Black-headed Gulls. It will be a lasting tribute to Bob if the RSPB finally achieves this goal. Bob had a= strong and unashamedly sentimental sense of history, and an open admiration for the pioneers of Scottish ornithology; he considered himself privileged to count a number of them amongst his friends. In particular he had tremendous respect for the late Professor Maury a2 Meiklejohn and one of Bob’s lasting achievements was to edit a Clyde SOC tribute to that great elder statesman of ornithology. Published in 1994, it was composed of cuttings collected by Bob from many years of Saturday Glasgow Heralds in which the professor wrote an evocative and often humorous. diary of his ornithological excursions, over the initials MFMM. Bob Caldow’s health had been in decline for a number of years and he became seriously ill early in 2006. He died on 26 August 2006. He is survived by his beloved wife Sheena, and sons Martin and Richard. Tain Gibson Russell Young 1936-2007 Russell Johnston Young, who was Club Secretary from Oct 2001 until August 2005, died on 19 August 2007 aged 71 after admission to hospital. Russell’s time as Secretary covered a substantial part of the period of move from Regent Terrace to Waterston House. The resulting protracted and onerous secretarial burden was met with Russell’s characteristic diligence which ensured that the burden was carried and discharged effectively. In earlier life, Russell had studied for and made a career change from the textile industry of his native Borders - he came from Greenlaw - to primary school teaching. He became headmaster of Strathdevon Primary School, Dollar, a position he held until early retirement for health reasons in 1989. His new career however had realized his innate vocation of teaching. That was especially so of natural history for which he often exceeded the allotted time in his SB 28 curriculum engendering great response from his pupils. He was always pleased beyond measure when those pupils told him of following up by observation or practice some subject of his natural history teaching. He was rewarded in that way many times. For Russell, however, teaching was not limited to classroom and pupils. Anywhere, anytime he readily imparted knowledge of the wild or of other subjects. His responses were easily kindled, presented with a smile and humour, and were immediately engaging. He was welcomed and listened to in any company and came into his own on club branch outings; what he said was remembered. His interests went beyond ornithology and for him were gateways to knowledge. He was a dedicated angler, with abundant craft to counter the guile of the trout, not least in the flies that he made by his own hand. He had a knowledge of fish biology and ecology derived especially from his local River Devon and neighbouring waters. He had been a water bailiff on that river. Botany was in his repertoire. So was gardening, which he practised extensively and successfully. He was a member of his church choir, the Kirkcaldy Gilbert and Sullivan Society and the East Fife Male Voice Choir and a painter of rural subjects. Fundamentally, Russell was a man who cared. That was manifest in his attention to detail , in the way he shared his interests, in his unfailing consideration to all in his circle, and in his unstinting helpfulness to the needs of acquain- tances less fortunate than himself. He had a long involvement with the SOC attending the Stirling (now Central) branch where he inevitably became chairman. He was the driving force behind the branch’s weekend survey of the birds of the River Devon in 1977, Scottish Birds (2008) Obituaries 53 a project which drew club members from outside areas, on which a report was produced, and which was repeated in subsequent years. In Russell’s passing has fallen a bright leaf from the canopy of Scotland’s ruralists. That fall saddens all who knew him but they remember a friend of strong and kind character and retain the inspiration of his cheering presence and helpful word. Sandy Mitchell Jeff Watson 24 December 1952-19 September 2007 The world’s acknowledged expert on Golden Eagles, Dr Jeff Watson, died of cancer at his home in the Black Isle last September. Jeff dedicated a third of his all too short life to studying Golden Eagles, and in 1997 published the definitive book on these magnificent birds, The Golden Eagle. Just days before he died, he wrote the foreword to the second edition, which will be published next year. Raised in Galloway, Jeff’s interest in birds was inspired by his father, bird artist and ornithologist Donald Watson. Jeff attended the local primary school at Dalry, Edinburgh Academy, and then Aberdeen University where he to took an honours degree in zoology, graduating in 1974. Moving to the Seychelles, in the Indian Ocean, Jeff spent 4 years researching the endemic Seychelles Kestrel, for which he was awarded a doctorate by Aberdeen University in 1977. This led to work for the WWE on the conservation of endangered, endemic landbirds. Jeff’s work on these islands continued through field visits and active collaboration with conservation groups. On his return to Britain, Jeff worked for a short time for the Scottish Wildlife Trust. In 1981 he joined the Nature Conservancy Council researching Golden Eagles, in particular the impacts of land uses such as forestry, farming and red deer management on the birds. This developed into a pioneering study of eagles in 9 ecological regions, which set the highest of standards for understanding the effects of land uses on raptors. Following in the footsteps of the great eagle fieldworkers such as H B Macpherson, Seton Gordon and Adam Watson, Jeff had the essential skills of tenacity, robustness, patience, determination and an ability to cope with weeks of solitude. For years his research base was a camper van, and, with his wife Vanessa, he walked hundreds of miles each year in difficult terrain. Jeff became the world’s foremost authority on the Golden Eagle. He had personal knowledge of most of their nesting territories in Scotland. He also visited other parts of the bird’s world range to advise on _ conservation and management issues. His monograph on _ the 54 SB 28 Golden Eagle, published by Poyser/Academic Press, was praised throughout the world. Beautifully written, it was translated into a Japanese edition last year. The book and Jeff’s more recent research have led to many research papers on the conservation and population ecology of Golden Eagles, including DNA fingerprinting of much of the breeding population, and a conservation framework for the species which has attracted international interest. Until days before his death, Jeff corresponded with workers in Kazakhstan, the Republic of Ireland where the reintroduction programme has benefited from the donation of chicks from Highland eyries, Alaska, Idaho and the Appalachian Mountains of North America. During the last 10 years, Jeff held down the challenging role of Director of Operations (North) in Scottish Natural Heritage (SNH). He had lead responsibility for the designation, conservation and management of Scotland’s internationally important Special Protection Areas and Special Areas of Conservation. He led the programme of work which overhauled Scotland’s series of National Nature Reserve, and led an ambitious programme of monitoring habitats, species and landforms across Scotland. The recent batches of Scotland’s SPAs classified for raptors are a fine legacy, which involved SNH working closely with land managers, raptor fieldworkers and the RSPB. Jeff also led the SNH team which worked closely with the Scottish Executive in support of the passage of the Nature Conservation (Scotland) Act 2004, giving rise to wildlife conservation legislation which has set new standards for the rest of Europe. He represented the UK as a Council Member of Eurosite, and was a Trustee of the National Biodiversity Network. An outstanding photographer, Jeff produced evocative landscape pictures which have been compared favourably with the hill and raptor paintings by his father. Despite all of these accomplishments, Jeff remained modest, indeed humble, and was genuinely embarrassed if praise came his way. His dealings with the politics of conservation were, at times, troubling, and he often spoke of the agonies of compromising in the face of development pressures. However, many battles were won, often due to Jeff's calm mastery of the facts and cogent arguments. His final battle was with cancer, which he lost after a courageous and determined fight, during which he showed supreme dignity. Last year, Jeff was awarded the prestigious Conservation Medal by the RSPB, and the Scottish Raptor Study Groups have established an award in the name of Jeff and his father. Jeff is survived by his wife Vanessa, son Ronan and sisters Pamela, Kate and Louise, who cherish special family days watching Golden Eagles amongst the rugged mountains of the Highlands. Des Thompson & John Lister-Kaye Scottish Birds (2008) Advice to contributors Authors should bear in mind that only a small proportion of the Scottish Birds readership are scientists and should aim to present their material concisely, interestingly and clearly. Unfamiliar technical terms and symbols should be avoided wherever possible and, if deemed essential, should be explained. Supporting statistics should be kept to a minimum. All papers and short notes are accepted on the understanding that they have not been offered for publication elsewhere and that they will be subject to editing. Papers will be acknowledged on receipt and are normally reviewed by at least 2 members of the editorial panel and, in most cases, also by an independent referee. They will normally be published in order of acceptance of fully revised manuscripts. The editor will be happy to advise authors on the preparation of papers. Reference should be made to the most recent issues of Scottish Birds for guidance on style of presentation, use of capitals, form of references, etc. Papers should be typed on one side of the paper only, double spaced and with wide margins and of good quality; 2 copies are required and the author should also retain one. We are also happy to accept papers on disk or by email at: mail@the-soc.org.uk, stating the type of word processing package used. If at all possible please use Microsoft Word . Contact the Admin Officer on 01875 871330 for further information. ee) Headings should not be underlined, nor typed entirely in capitals. Scientific names in italics should normally follow the first text reference to each species unless all can be incorporated into a table. Names of birds should follow the official Scottish List (Scottish Birds 2001 Vol 22:33-49). Only single quotation marks should be used throughout. Numbers should be written as numerals except for one and the start of sentences. Avoid hyphens except where essential eg in bird names. Dates should be written: ...on 5 August 1991...but not ...on the Sth... (if the name of the month does not follow). Please do not use headers, footers and page numbers. Please note that papers shorter than c700 words will normally be treated as short notes, where all references should be incorporated into the text, and not listed at the end, as in full papers. Tables, maps and diagrams should be designed to fit either a single column or the full page width. Tables should be self explanatory and headings should be kept as simple as possible, with footnotes used to provide extra details where necessary. Each table, graph or map should be on a separate sheet, and if on disc each table, graph, map etc should be on a separate document. Please do not insert tables, graphs and maps in the same document as the text. Maps and diagrams should be either good quality computer print out and in black and white (please do not use greyscale shading) or drawn in black ink , but suitable for reduction from their original size. Contact the Admin Officer on 01875 871330 for further details of how best to lay out tables, graphs, maps etc. 56 The Scottish Ornithologists’ Club was formed in 1936 to encourage all aspects of ornithology in Scotland. It has local branches which meet in Aberdeen, Ayr, the Borders, Dumfries, Dundee, Edinburgh, Glasgow, Inverness, New Galloway, Orkney, St Andrews, Stirling, Stranraer and Thurso, each with its own programme of field meetings and winter lectures. The George Waterston Library at the Club’s headquarters at the Scottish Birdwatchers’ Centre, Waterston House, Aberlady, East Lothian EH32 OPY is the most comprehensive ornithological library in Scotland and is available for reference 7 days a week. A selection of Scottish local bird reports is held at headquarters and may be purchased by mail order. Check out our website for more information about the SOC and other bird related organisations: www.the-soc.org.uk An issue of Scottish Birds is published each year, in June. Scottish Bird News is published quarterly in March, June, September and December and the Scottish Bird Report is annual. The Scottish Ornithologists’ Club publishes the annual Scottish Raptor Monitoring Scheme report on behalf of the Scottish Raptor Monitoring Group with grant aid from Scottish Natural Heritage. It is sent to all members. Bird names used in publications follow those used in the Scottish List (Scottish Birds Vol 22(1): 33-49). Copies of these reports may be purchased by non members on application to the SOC. SOC annual membership subscription rates Adult £25.00 Family (2 adults and any juniors living at the same address) £ 35:00 Junior (aged 17 or under) £ 10.00 Student (aged 18 to 24 and in full-time education) £ 10.00 Concession (Pensioner) £ 16.00 Concession (Unwaged) £ 16.00 Joint Concession £ 21.00 Life £500.00 Life Family £700.00 Overseas supplement £ 10.00 All subscriptions may be paid by Direct Debit and Gift Aided. Subscriptions paid by Direct Debit greatly assist the Club. Please ask for an application form by telephoning 01875 871330 or visiting the club’s website at www.the-soc.org.uk SB 28 et rgyll © Edmund F lowes Scottish Birds Volume 28 June 2008 Contents Main Papers Rookeries in Caithness in 2007 - R M Sellers, H Clark & S Laybourne 2 Behaviour and diet of non breeding Snowy Owls on St Kilda- WTS Miles & S Money 11 Responses of owls and Eurasian Kestrels to natural and human induced spatio temporal variation - E Korpimdki, H Hakkarainen, T Laaksonen & V Vasko 19 Post breeding movements of Sandwich Terns in the Firth of Forth - J Bos 28 Wintering wader surveys on the Isle of Tiree, Argyll - J Bowler, C Mitchell, R Broad, A Leitch & D Stroud 3 Short Notes Changes in the breeding status of Little Ringed and Ringed Plovers in North Fife - D W Oliver 42 Intraspecific interaction by Hen Harriers over a cock’s nest - R C Dickson 43 Breeding success of Golden Eagles on the Isle of Rum 1995-2007 - S Morris 44 Survey of roof nesting gulls in East Sutherland - R M Sellers 45 Breeding dates and success of Bass Rock gannets in 2007 - J F Hunt & J B Nelson 47 The first Lesser Grey Shrike in Scotland: clarification and correction - R Y McGowan, M Pennington & C J McInerny 49 Merlin impaling prey on barbed wire fence - R C Dickson 50 Obinaares NSTITUTION LIBRARIES Js MM | | | Rohe Grier Callow LI Russell Young 52 Jeff Watson D2 Advice to contributors 55 Front Cover Snowy Owl by Paul Hackett Published by the Scottish Ornithologists’ Club, Scottish Birdwatchers’ Centre, Waterston House, Aberlady, East Lothian EH32 OPY. © 2008 Scottish Charity Reg. No SC 009859 2000 and 2001 Scottish Birds - the Journal of the SOC Supplement to Volume 24 (2003) Scottish Birds -The Journal of the SOC p ¢ Editor: Dr S da Prato “‘ Assisted by: Dr I Bainbridge, Professor D Jenkins, Dr M Marquiss, Dr J B Nelson, and R Swann Business Editor: Caroline Scott, Admin Officer, SOC, Harbour Point, SOC Newhailes Road, Musselburgh EH21 6SJ Scottish Birds, the official journal of the SOC, publishes original material relating to ornithology in Scotland. Papers and notes should be sent to The Editor, Scottish Birds, SOC, Harbour Point, Newhailes Road, Musselburgh EH21 6SJ Scottish Birds is published each year in June. Scottish Birds is issued free to members of the SOC, who also receive the quarterly newsletter Scottish Bird News, the annual Scottish Bird Report and the an- nual Raptor Round Up. Published by: The Scottish Ornithologists’ Club, Harbour Point, Newhailes Road, Musselburgh EH21 6SJ Printed by: Meigle Colour Printers Ltd, Block 11, Tweedbank Industrial Estate, Galashiels TD1 3RS 2000 & 2001 Raptor Round Up Contents Indroduction European Honey Buzzard Red Kite White-tailed Eagle Eurasian Marsh Harrier Hen Harrier Northern Goshawk Eurasian Sparrowhawk Common Buzzard Golden Eagle Osprey Common Kestrel Merlin Eurasian Hobby Peregrine Falcon Barn Owl Tawny Owl Short-eared Owl Pernis aviporus Milvus milvus Haliaeetus albicilla Circus aeruginosus Circus cyaneus Accipter gentilis Accipter nisus Buteo buteo Aquila chrysaetos Pandion haliaetus Falco tinnunculus Falco columbarius Falco subbuteo Falco peregrinus Tyto alba Strix aluco Asio flammeus 27 Af} I) 2000 & 2001 Raptor Round Up Introduction This report covers 2000 and 2001. Another round up for 2002 will be produced in this format. It is expected that compilation of future reports will then be the responsibility of the recently established Scottish Raptor Monitoring Group. This group’s aim is to promote effective monitoring of all species of birds of prey, of owls and Ravens in Scotland. It intends to consolidate good data standards and to make better use of the information for the understanding and conservation of these birds. The group consists of an equal partnership between Scottish Natural Heritage, Scottish Raptor Study Groups, Royal Society for the Protection of Birds Scotland, British Trust for Ornithology Scotland, Joint Nature Conservation Committee, Rare Birds Breeding Panel and Scottish Ornithologists’ Club. The group has appointed a Raptor Monitoring Officer who will collect and collate data from all sources on our birds of prey, owls and Ravens. This data will be published in an annual report from 2003, giving a full analysis of the population monitoring of these birds. Persecution remains a serious threat to birds of prey in Scotland. At least 93 Red Kites are known to have been killed by feeding on poisoned bait in the 10 years from 1989 to 1998. These losses continue. In 2000, 4 Kites were found poisoned with carbofuran, while another 5 were discovered in 2001. Other kinds of persecution are less easy to substantiate. For example, it has been suggested that the numbers of Hen Harriers and Goshawks breeding in Scotland are severely restricted by the illegal human destruction of nests and birds; and Peregrines and Golden Eagles are absent from many formerly occupied home ranges in areas managed for driven grouse shooting. Persecution is one of a number of factors thought to govern the distribution and success of raptor populations in Scotland. The Scottish Raptor Monitoring Group aims to identify these. Monitoring in 2001 was badly affected by access restrictions caused by foot and mouth precautions. Dumfries and Galloway, South Strathclyde and Lothian and Borders groups could not visit many home ranges and survey work in these areas was limited or absent. Jon Hardey This report was produced by Jon Hardey for the Scottish Ornithologists’ Club & SNH. The data are from the annual reports of the Scottish Raptor Study Groups, the Red Kite and Osprey newsletters. The report has been supported by grant aid from Scottish Natural Heritage. Illustrations are by Andrew Stevenson and Lorna Hardey. The photograph of a Goshawk on the front cover is by Laurie Campbell (courtesy of SNH), from the forthcoming book ‘Birds of Prey in a Changing Environment’. Please note that where ‘-’ appears in a table, no data is available. Tables only show data collected by the Scottish Raptor Study Groups. Other data may be available elsewhere. 2000 & 2001 Raptor Round Up 3 European Honey-buzzard Pernis apivorus Two pairs were monitored in northern Scotland in 2001. Both reared 2 young. Small satellite transmitters were fitted to both chicks from one nest. These chicks were followed on their migration. The larger chick moved through Kintyre and western England to south-east Spain and into Morocco before the signal failed. The smaller chick migrated into Ireland before moving out into the Atlantic. The last signal was from 100 km SSW of Cape Clear in Eire. Fuller details can be obtained at www.roydennis.org.uk Honey Buzzards were reported from central Scotland in both years. No nests were located. 4 2000 & 2001 Raptor Round Up Red Kite Milvus milvus Table 1. Breeding success of Red Kites in Scotland, 2000-2001 Region Year Pairs Pairs Pairs Pairs Chicks fledged located laying hatched successful eggs Highland 2000 a] 33 EB 9 76 2001 40 34 31 31 77 Tayside 2001 ] 1 - 0 0 Central 2000 8 7 - 5 10 2001 13 1] - a 17-18 TOTAL 2000 45 40 - 36 86 2001 54 46 - 38 94-95 In 2001, a minimum of 5 kites were known to have been killed by the poison carbofuran. This agricultural insecticide is widely used in illegal poisoned baits. Illegal poisoning is the greatest cause of mortality in the Scottish Red Kite population (Etheridge 2001). Between 1989 and 1998, analysis has revealed that 37.5% (93 birds) of kites that were released or bred in north Scotland, were killed by feeding on poisoned baits. This level of persecution has slowed the expansion of the Scottish Red Kite population. In 2000, 3 Red Kite chicks in north Scotland and a breeding female from central Scotland died from secondary poisoning by rodenticides. These deaths occurred after the kites had scavenged dead rats that had been legally controlled. Following publicity on the effects on kites and other birds by this secondary poisoning and the publication of recommendations on the safe use of rodenticides, there were no recorded deaths of either adult or young kites in 2001. This is very encouraging. Farmers and others involved in the control of rats and mice are to be congratulated for adopting safer methods for the use of these anticoagulants. The re-introduction of Red Kites continued in central Scotland in 2000 with 20 young being imported from Germany and successfully released. This was the final year for the project in this area. In 2001, 13 kites that had been illegally removed from the wild in Germany, were made available for the Scottish reintroduction project. Six of these were released in central Scotland with the other 7 going to the new release site in Dumfries and Galloway. Twelve chicks from nests in the Highlands and 14 from nests in the Chilterns were also released at this site in early August. These young were all removed under licence from nests that contained 3 to 4 young. Reference Etheridge B 2001. Red Kite, North Scotland. Newsletter 2001. Royal Society for the Protection of Birds, Inverness. 2000 & 2001 Raptor Round Up b) White-tailed Eagle Haliaeetus albicilla Table 2. Breeding success of White-tailed Eagles in Scotland, 2000-2001 Year Pairs located Pairsknownto Pairs known to Pairs known Minimum lay eggs hatch eggs to fledge number of young young fledged 2000 21 19 11-13 8 |p 2001 24 li) 10 1 1] 2000 was the second most successful year since reintroduction began. Of the 21 pairs located, only 8 went on to fledge a total of 12 young. One nest was robbed by humans but all the other failures are believed to be natural. Twenty four home ranges were occupied in 2001. Ten were successful, rearing 11 young. There is concern that the number of chicks raised annually has not increased as more pairs begin to breed. Eurasian Marsh Harrier Circus aeruginosus Marsh Harriers bred successfully in Orkney at 2 sites for the first time in 2001. The 2 males were bigamous. The 4 females all reared young. A single pair in the North-east bred in both years, rearing young in 2000 and failing in 2001. In Tayside, 5 nests were located in 2000 with 12 young fledged. However there were only 3 in 2001 with a single nest fledging 3 young. 6 2000 & 2001 Raptor Round Up Table 3. Breeding success of Marsh Harriers in Scotland, 2000-2001 Region Year Nests Nests with Number of fledged young _ fledged young Orkney 2001 4 4 9 North-east 2000 l l 4 2001 l 0 0 Tayside 2000 5 4 12 tr S =) —_— ww — Oso —————— SS yy i ae XA) Se 4 vf fe / Wee abe “es * % LEO, Greens oc; Hen Harrier Circus cyaneus Orkney In 2000, 45 females and 20 males were located. Twenty-three of the females bred, rearing 28 young from 18 successful nests. Breeding success was better as chicks were fed at some of the nests as part of a research programme and good weather in the early part of the season. There was considerable chick loss (approximately 34%) between the ringing stage and fledging. In 2001, more breeding attempts were recorded but productivity was lower than in 2000. Highland Harriers have declined in the Highlands since the survey in 1988 and 1989 from 85 pairs to only an estimated 30 pairs in the 1998 survey. The nests surveyed in 2000 were very successful with 17 of the 19 nests monitored being successful. Two young out of a brood of 4 were shot by a gamekeeper. 2000 & 2001 Raptor Round Up The birds monitored in 2001 reflected the previous year’s success with 14 nests fledging 43 young. The 4 nests which failed were all due to natural causes with 3 broods being predated. It is believed that the absence of Hen Harriers from large tracts of suitable breeding habitat is due to persecution. North-east Hen Harriers remain at low density in north-east Scotland. Only 13 and 14 breeding attempts were made in 2000 and 2001 respectively. It is estimated that the region should hold between 60 and 80 breeding females. Breeding success improved in 2000 with 11 nests out of 33 fledging young. This success was maintained in 2001 with 38 young fledging from 11 successful nests. Most of the successful nests were on estates actively involved in a protection scheme organised by Scottish Natural Heritage, Grampian Police and the raptor study group. Table 4. Breeding success of Hen Harriers in Scotland, 2000-2001 Region Orkney Highland and west Moray North-east Tayside Uists Argyll South-west Scotland Lothian and Borders TOTAL Year 2000 2001 2000 2001 2000 2001 2000 2001 2000 2001 2000 2001 2000 2001 2000 2001 2000 2001 Females with nests Nests fledging young 18 15 Li 14 11 11 114 Minimum number of young known to fledge 28 31 47 43 & 2000 & 2001 Raptor Round Up Tayside The number of Hen Harriers in Tayside is gradually declining with large tracts of apparently suitable habitat devoid of harriers. Coverage in 2000 was poor. Less than half the monitored nests fledged young. Breeding success was low on grouse moors. In 2001, coverage was again poor. A large number of pairs breeding on grouse moors again failed. Central Three pairs of breeding harriers were reported from 3 separate areas in 2000 with one pair probably being successful. In 2001, 2 pairs fledged young in the east (3 young) and west (2 young) of the group’s area. A separate study area in west Dunbartonshire was occupied by 2 pairs of harrier in 2000 and by 3 pairs in 2001. These birds fledged 2 young from 1 nest in 2000. The results are not recorded for 2001. Argyll In 2000, breeding success was better on Islay and Cowal than in other areas. Three nests in Kintyre/ Knapdale all failed. The Rhinns of Islay was surveyed in 2001 for Scottish Natural Heritage. Twenty-two home ranges were identified with breeding confirmed at 18 with 50 chicks estimated to have fledged. Harriers were recorded at another 19 home ranges with 14 successful pairs fledging 42 young. Polygamy was suspected at 6 ‘sites’. Each female breeding attempt has been counted as a separate home range. Uists In 2000, Hen Harriers were suspected to be breeding on 14 home ranges of which 9 were monitored, fledging 26 young. Successful breeding was recorded at 14 home ranges in 2001, fledging at least 27 young. Breeding was suspected at another 11 locations. South-west Scotland Human persecution was strongly suspected to be the cause of failure at 15 nests in 2000. In south Strathclyde, 71% of the nests failed despite excellent weather and a high vole population. In 2001, many home ranges could not be visited as access was restricted by foot and mouth precautions. In 2000 & 2001 Raptor Round Up 9 south Strathclyde, 13 occupied home ranges were found. Nests were located on 7 of these with 3 successful nests fledging 11 young. One failure was caused by predation by a fox but human interference was strongly suspected at the other 3 nests. Lothian and Borders In 2000 only 5 pairs were reported compared to 10 in 1999. No data from 2001 owing to Foot and Mouth restrictions. Northern Goshawk Accipter gentilis Table 5. Breeding success of Goshawks in Scotland, 2000-2001 Region Year Home ranges Occupied home _Pairs fledging Minimum checked ranges found young number of young fledged Highland 2000 8 4 3 7 2001 8 5 3 6 North-east 2001 28 17 15 4() Tayside 2000 2 2 2 4 2001 5 5 4 10 Central 2000 l l 0 0 Dumfries and 2000 16 10 8 iif Galloway Lothian and 2000 56 4] 29 82 Borders 2001 46 36 26 73 TOTAL 2000 83 58 42 110 2001 87 63 48 129 10 2000 & 2001 Raptor Round Up This species is under-recorded in most areas. Breeding success is restricted by persecution in all areas. This is preventing expansion of Goshawks into new home ranges. In Highland, only 3 pairs fledged young in either year from 8 home ranges checked. Occupancy was low (61%) in the North- east but the breeding success of the pairs located was good (2.35 young fledged/ occupied home range). In Tayside, the small number of home ranges checked were successful. In central Scotland, a nest that later failed was found in 2000. These birds were not present in 2001. In Dumfries and Galloway in 2000, occupancy was low (62%) with the birds being moderately successful (1.7 young fledged/ occupied home range). No home ranges were visited in 2001 as access was restricted because of the foot and mouth outbreak. In Lothian and Borders there were several instances of recorded nest persecution in both years. Despite Foot and Mouth, the majority of sites were still monitored in 2001 owing to the good nature of Forest Enterprise and other landowners and the care taken by raptor workers entering and leaving forest areas. Eurasian Sparrowhawk Accipter nisus Orkney Four sites were checked in 2000. None fledged young. In 2001, Sparrowhawks were more successful with 4 sites fledging a minimum of 6 young. Birds failed at another 3 sites. Highland Sparrowhawks in the Highlands had a lower productivity (3.13 young fledged/occupied home range) in 2001 than in 2000 (5.00 young fledged/occupied home range). The sample size was small. Table 6. Sparrowhawk breeding data, Highland, 2000-2001 Year Home ranges Home ranges _Pairsknownto __ Pairs known to Young checked occupied lay eggs fledge young fledged 2000 6 6 6 6 30 2001 8 8 | 7 25 Lothian and Borders In 2001, Sparrowhawks in Edinburgh were surveyed. Fieldworkers were prevented from visiting other raptors as access was restricted due to foot and mouth. 2000 & 2001 Raptor Round Up I] Table 7. Mean number of young produced from all clutches of different sizes including those that produced no young (from Edinburgh 2001 G. D. Smith) Initial clutch size 1 2 3 4 5 6 Number 0 0 4 18 d 2 Mean brood size at hatching 0 0 ps 261 4.40 5 Mean brood size at fledging 0 0 175 2.10 3220 4.50 This survey by G. D. Smith shows a drop in productivity (2.08 young fledged/occupied home range) from M. McGrady’s work from 1986 to 1989 (2.50 young fledged/occupied home range). The biggest losses were of young between hatching and fledging with 2 nests robbed of 4 young each and another 4 young found dead at the nest site with the other 2 missing and presumed eaten. Common Buzzard Buteo buteo Orkney Only 1 nest was located in 2000. It had a clutch of 3 eggs. It is not known if this nest was success- ful. A second pair was seen but no nest was found. In 2001, 3 home ranges were occupied. Only one was successful. The number of fledged young is not known. At the second home range, a nest was refurbished but not used, while at the third, no nest was found. Highland In 2000, 60 pairs were monitored in East Ross. An unusually large number did not lay (10), or failed just after laying. This was assumed to be linked to prolonged rain in late April. Another 13 pairs failed. Productivity was low (1.3 young fledged/occupied home range). In the other main study area in north and west Sutherland, productivity was higher at 1.68 young fledged/occupied home range from the 25 ranges monitored. Long dry spells of sunny weather may have assisted this improved output. In Skye and Lochaber (n=17), productivity was even higher with 1.76 young fledging/ occupied home range. The greatest output was in the Black Isle, mid Ross and Badenoch and Strathspey, where 2.05 young fledged/occupied home range (n=22). In 2001, 51 pairs in East Ross had a higher productivity (1.35 young fledged/occupied home range) than 2000. The north and west 12 2000 & 2001 Raptor Round Up Sutherland population (n=15) was again successful with a productivity of 1.87 young fledged/ occupied home range. Productivity fell in Skye and Lochaber (n=18) to only 1.28 young fledged/ occupied home range. The Badenoch Buzzards were also less productive with 1.85 young fledged/ occupied home range. Overall, Buzzards were slightly less successful in 2001 (1.50 young fledged/ occupied home range) than in 2000 (1.55 young fledged/occupied home range. Central A post fledging survey of Common Buzzards in 2000 around Doune, revealed 11 successful nests rearing 22 young. In 2001, 12 occupied home ranges were found in the same area. The outcome of two of these is not known but the other 10 all reared young. Productivity for the Doune population in 2000 was 2.0 young fledged/occupied home range and in 2001 was 2.7 young fledged/occupied home range. Table 8. Breeding success of Common Buzzards in Scotland, 2000-2001 Region Year Home ranges Pairs laying Pairs Minimum occupied eggs fledging number of young young fledged Orkney 2000 | I - : 2001 3 1 1 I+ Highland 2000 130 119 96 202 2001 102 97 82 153 Central 2000 11 11 11 22 2001 12 : 10 pH Argyll 2000 82 - 66 128 2001 47 42 39 72 Uists 2000 19 - - 4] 2001 2A - - 38 Lothian and Borders 2000 43 - 30 67 2001 Zz ps 1 1 2000 & 2001 Raptor Round Up 13 Argyll In 2000, Common Buzzards breeding in Cowal produced 1.9 young fledged/occupied home range (n=31) with 2 pairs failing on eggs and | with young. On Colonsay, 10 pairs out of 17 apparently failed or did not breed. The remaining 7 pairs bred rearing 9 young with a productivity of 0.52 young fledged /occupied home range. Buzzards on Bute were more successful with 2.21 young fledged/ occupied home range (n=14). Data from SW Mull records that 20 successful pairs reared 30 young. In 2001, 18 occupied home ranges were monitored in the same area. All fledged young (1.61 young fledged/occupied home range, n=18). Fifty home ranges were checked on Colonsay. Twenty were occupied with 7 pairs presumed to have fledged 13 young (0.65 young fledge/occupied home range). Only 5 sites were monitored in Cowal (1.4 young fledged/occupied home range) but Buzzards were known to have occupied another 21 home ranges, most of which were believed to be successful. On Bute, 8 home ranges all fledged young with 2.38 young fledged/occupied home range. Lothian and Borders In the Borders in 2000, Buzzards were reported from 44 new tetrads. This brings the population closer to the estimated maximum of 1000 pairs. In a study area around Fountainhall, 26 home ranges were identified giving a density of 0.43 pairs/km?. Nineteen nesting attempts were recorded with 10 successful pairs rearing 18 young (0.95 young fledged/occupied home range). In Mid and West Lothian, 24 home ranges were monitored. Four failed and the remaining 20 reared 49 young (2.04 young fledged/occupied home range). In 2001, access was restricted by foot and mouth precautions. In the Fountainhall study area, 29 home ranges were identified (0.48 pairs/km’). Only two nests were visited, one reared a single chick and the other failed. Golden Eagle Aquila chrysaetos Highland 2000 was a successful year for Highland Golden Eagles. The poorest breeding success was in Lochaber (0.30 young fledged/pair) and in the east of the region (0.48 young fledged/pair) where it was linked to poor spring weather, land use changes and persecution. Birds on Skye (0.55 young fledged/pair) were more successful than in neighbouring Lochaber although this was below the annual mean for the island. The eagles in West Inverness bred well (0.64 young fledged/pair) but productiv- ity was exceptional in NW Sutherland and Wester Ross (0.92 young fledged/pair). It is possible that these birds benefited from good spring and summer weather. In 2001, overall breeding success (0.58 young fledged/pair) was above the annual mean of 0.50 young fledged/pair. The birds in West Inverness (0.75 young fledged/pair), in NW Sutherland and West Ross (0.81 young fledged/pair) and in the eastern part of the region (0.71 young fledged/pair) had a successful year. Lochaber eagles continue to have poor success (0.32 young fledged/pair) with the birds on neighbouring Skye being less productive than usual (0.46 young fledged/pair). As part of the Golden Eagle re-introduction programme in Ireland, six chicks were removed, under licence, for translocation. 14 2000 & 2001 Raptor Round Up North-east Breeding success was low in 2000, with only 5 pairs from 19 occupied home ranges fledging 6 young (0.32 young fledged/pair). Five pairs were known to fail with eggs. The other pairs may not have laid. This may be linked to poor weather in the spring. In 2001, 4 home ranges were occupied by single birds. Three of these home ranges are on grouse moors. Six pairs are known to have failed with eggs. Five pairs fledged 7 young (0.40 young fledged/pair). Tayside The eagles were less successful than usual west of the A9 in 2000 with only 2 sites fledging 3 young (0.33 young fledged/pair). Coverage east of the A9 was poor with data only collected from 4 successful home ranges. In 2001, 11 home ranges were checked with 6 pairs rearing 8 young (0.80 young fledged/ pair). West of the A9, 8 home ranges were occupied and 3 pairs fledged 4 young (0.50 young fledged/pair). There is concern about the number of home ranges not producing young. Central In 2000, only 4 pairs fledged young in the 10 occupied home ranges (0.4 young fledged/pair). Two home ranges were occupied by single birds in 2001, with the other 7 which were checked occupied by pairs. Four pairs successfully reared 5 young (0.71 young fledged/pair). Disturbance by hang gliders may be a problem at one nest site. Argyll 2000 was a more productive year (0.37 young fledged/pair) than 1999 (0.30 young fledged/pair) but the number of occupied home ranges fledging young (32.2%) was still well below the mean for the last 5 years (38.8%). The 2001 season was slightly better with 0.39 young fledging/pair and 33.3% of occupied home ranges fledging young. The majority of nests were on crags but 2 pairs bred in trees, a Scots pine and an ash, while one pair bred in a nest on the ground at the top of coastal slope. 2000 & 2001 Raptor Round Up 15 Table 9. Breeding success of Golden Eagles in Scotland, 2000-2001 Region Year Home ranges Home ranges Pairs Minimum checked occupied fledging number of young young fledged Highland’ 2000 95 95 44 52 2001 96 96 42 55 North-east 2000 23 19 3) 6 2001 22 18 5 7 Tayside 2000 16 13 6 9 2001 22 18 9 12 Central! 2000 10 10 4 4 2001 9 9 4 5 Argyll 2000 62 59 19 22 2001 54 54 18 21 Uists” 2000 22 Pi. 9 11 2001 23 23 (5) (7) Dumfries and Galloway 2000 2 2 1 1 Lothian and Borders 2000 2 2 1 2 2001 2 Z 2 g) TOTAL 2000 210 201 80 96 (excluding Uists) 2001 205 196 80 102 1. The number of home ranges checked includes only those which were occupied in Highland and central Scotland. 2. Data from the Uists were not included in the totals as not all occupied home ranges were checked for fledged young. Uists In 2000, 22 occupied home ranges were identified in the Uists and Barra. Successful breeding was known to occur at 9 sites which reared 11 chicks. This was the most successful year since monitor- ing began in 1988. Golden Eagles were present on 23 home ranges in 2001 with breeding attempted on at least 13. Two sites were disturbed by an egg collector. One of these still managed to fledge a chick. Five home ranges are known to have fledged 7 young. 16 2000 & 2001 Raptor Round Up Dumfries and Galloway Two pairs attempted to breed in 2000. Both laid 2 eggs. One pair hatched 2 chicks, one of which died. The other fledged. The eggs of the second pair failed to hatch. Lothian and Borders As usual, in 2000 only one pair was successful, although two young fledged. The success of the second pair in 2001 was attributed to reduced disturbance owing to Foot and Mouth restrictions. Osprey Pandion haliaetus Table 10. Breeding success of ospreys in Scotland, 2000-2001 (from Dennis 2000, 2001) The areas in the table are those used by Dennis (2001). Region Year _—‘ Pairslocated _— Pairs laying Pairs Minimum eggs fledging number of young young fledged North 2000 70 61 47 101 (Highland) 2001 76 68 51 106 Aberdeenshire 2000 13 10 8 14 2001 13 i 9 20 Perthshire/Angus 2000 50 36 - 52 2001 45 37 29 60 South and West 2000 14 14 12 28 2001 19 18 15 33 Lothian and Borders 2000 1 1 l 3 2001 2 1 1 l TOTAL 2000 148 122 - 198 2001 155 136 105 220 North (Highland) Productivity was higher in 2000 (1.44 young fledged/pair) than in 1999 (1.41 young fledged/pair). One pair lost its eggs, either to a pine marten or to humans. Another young pair lost their eggs in high winds. Six pairs failed when small young died. These were all in Moray and it may have been caused by a lack of flounders in Findhorn Bay. They all migrated successfully. Six more pairs were 2000 & 2001 Raptor Round Up 17 located in 2001 than in 2000, with 5 more pairs laying eggs. At least 3 probably 4, nests were predated by pine martens. Three nests were robbed by egg collectors. Seven pairs failed to hatch their eggs. One of these was probably due to interference by a close nesting pair of Common Buzzards. A nest in Sutherland fledged 4 young, the first record of this brood size for the Highlands. Productivity was similar to 2000 with 1.43 young fledging/pair. In both years, twelve young were collected under licence and released at Rutland Water. Aberdeenshire (North-east) In 2000, productivity was low with 1.08 young fledging/pair. It is suggested that the failure of 2 pairs was caused by high winds. 2001 was similar with 3 pairs failing after strong winds. Productiv- ity was higher at 1.54 young fledged/pair. Perthshire and Angus (Tayside) In 2000, the increase in the number of pairs continued with 5 more pairs located giving a total of 50 pairs. A pair built a nest on a building for the first time. Three nests were robbed of eggs. Productiv- ity was 1.04 young fledged/pair. The number of pairs located fell by 5 pairs in 2001. Productivity was higher at 1.33 young fledged/pair. An adult Osprey was illegally trapped and had to be put down. West and South (Central and Argyll) The number of pairs located increased by 2 in 2000. These birds were very successful and produced a mean of 2.00 young fledged/pair. One home range was unoccupied. A chick ringed in Argyll in 1998 was recovered in Gambia in November. The ring was found inside a crocodile. In 2001, productivity was lower at 1.74 young fledged/pair but these Ospreys were the most productive in Scotland for the second year. Five new pairs were located. References Dennis R 2000. Ospreys 2000. Highland Foundation for Wildlife, Nethybridge. Dennis R 2001. Ospreys 2001. Highland Foundation for Wildlife, Nethybridge. 18 2000 & 2001 Raptor Round Up Common Kestrel Falco tinnunculus Ayrshire In 2000, the first hen laid on 17 March. This was the earliest record for 29 seasons. Other birds didn’t lay until much later with another hen laying the first egg of a clutch of 6 on the 31 March. The majority of clutches were started in April with two late clutches in the second week of May. It was a productive season with high brood survival which was expected as field voles were at a peak in their cycle. Fieldwork in 2001 was curtailed by access restrictions related to the outbreak of foot and mouth. Eggs were laid later than 2000 with the first egg being found at a lowland home range on 15 April. Occupation of home ranges and clutch size were slightly down on 2000 but vole numbers were still high and the Kestrels had a good season. Table 11. Breeding success of Kestrels in the Ayrshire Study Area, 1999-2001 Year 2001 2000 1999 Number of home ranges checked 24 4] 40 % occupation 70 76 52 Number of clutches known 15 22 10 Average clutch size 5.00 5.70 5.00 % of eggs which hatched 76 85 90 Number of results known 7 27 17 Average young per breeding pair 3.80 4.00 2.76 Average young per successful pair 4.60 5.65 4.30 Number of breeding attempts failed 3 4 6 % failed 23 15 35 % brood survival 98 93 91 Orkney In 2000, five nests were reported. These raised 14 young from 4 nest sites. One ground nest was predated. Six nests were recorded in 2001, with another 6 sites being occupied. Three of the nests failed with 11 young being raised from the other 3. It is believed that 2 of the other occupied sites may have fledged young. 2000 & 2001 Raptor Round Up 19 Highland The downward trend in the number of recorded nests continued in 2000. Mean clutch size (4.4 eggs, n=11) was the lowest recorded. In 2001, 11 home ranges were again recorded. Clutch size (4.9 eggs, n=11) was higher than the previous year. Productivity was also higher with 4.2 young fledging/ occupied home range in 2001 compared to 3.5 young fledged/occupied home range in 2000. More work is required on Highland Kestrels to ascertain if the population has declined in recent years or if the lower number of records reflects less interest in the species. Table 12. Breeding success of Kestrels in Highland, 2000-2001 Year Home ranges occupied Pairs fledging young Fledged young 2000 11 10 38 2001 11 11 46 Uists Eighteen pairs were recorded in both 2000 and 2001. There is no data on the breeding success. Argyll In 2000, only 6 out of 20 known home ranges on the Ross of Mull were occupied. A pair bred on Bute and fledged 3 young. Sixteen home ranges were occupied in south-west Mull in 2001. Six of these were successful but the number of young reared is not known. Data from 11 home ranges recorded that 23 young fledged. Table 13. Breeding success of Kestrels in Argyll, 2000-2001 Year Home ranges Pairs fledging young Minimum number of occupied young known to fledge 2001 11 10 23 20 2000 & 2001 Raptor Round Up Merlin Falco columbarius Table 14. Breeding success of Merlins in Scotland, 2000-2001 Region Year Home ranges Home ranges Pairs fledging Minimum checked occupied young number of young fledged Shetland 2000 - 10 8 vA| 2001 - 8 6 16 Orkney 2000 - 15 4-8 13-24 2001 - 14 10 34 Highland 2000 76 43 30 109 2001 61 35 21 73 North-east 2000 86 37 27 102 2001 70 39 35 104 Tayside 2000 86 38 25° 75 2001 56 33 15 43 Argyll 2000 7 3 3 11 Uists 2000 - 9 9 31 2001 - 9 9 21 Dumfries and 2000 3 9 5 10 Galloway Lothian and 2000 54 26 16 a7 Borders 2001 25 ao 11 30 TOTAL 2000 - 187 127-131 435-446 2001 - 160 107 321 ' the number of young fledging from 4 home ranges is not known 2000 & 2001 Raptor Round Up 21 Shetland Fewer occupied home ranges were located and productivity was lower in 2001 (2.0 young fledged/ occupied home range) than in 2000 (2.7 young fledged/occupied home range). Orkney 2000 was the poorest breeding season for some years. Coverage was incomplete but only 15 sites showed strong signs of occupation with only 8 nests being found. Four of these nests could not be visited at fledging so the number of young known to have been reared on the islands is between 13 to 24. Only 14 occupied home ranges were located in 2001 as access was restricted early in the season by precautions against foot and mouth. Young hatched at all of the 11 nests located but at one site a whole brood was predated, while at a second, part of the brood was lost — probably by feral cats. The 10 successful nests are believed to have fledged 34 young. Productivity was 2.4 young fledged/ occupied home range. Highland In 2000, 43 occupied home ranges were located, of which 3 were occupied by single pairs. Home range occupation improved in west Moray rising from 36% in 1999 to 66%. It fell in Inverness-shire from 50% to 38%. Excessive burning of heather is believed to be a factor as this removed nest sites. The mean clutch size of 4.3 eggs (n=29) was the highest recorded for the whole area. The mean of 3.6 young fledged/successful pair was above the long-term average. In 2001, site occupancy fell in Moray to 53% and in Sutherland, from 52% in 2000 to 48%. Inverness-shire was not covered as the fieldworker left the area. The mean clutch size was 4.1 eggs (n=17) with 3.4 young fledged/ successful pair. Productivity was lower in 2001 (2.09 young fledged/occupied home range) than in 2000 (2.53 young fledged/occupied home range). North-east Despite cool wet spring weather, Merlins bred well in 2000. Occupancy was low on the lower moors on Deeside and Donside. Productivity was 2.76 young fledged/occupied home range. In 2001, breeding success was high. Weather in the early part of the season was warm and dry although it is probable that brood reductions later on were linked to wetter weather in June affecting more exposed sites. Productivity was lower (2.67 young fledged/occupied home range) than in 2000. Tayside Merlins had an extremely good season in Angus in 2000 with 44 young fledging from 18 occupied home ranges. Perthshire was less successful. Only 31 young fledged from 24 occupied home ranges. There is evidence to suggest that there is a continued decline in breeding pairs in the west of the area. In 2001, the Angus Merlins were less successful with only 15 young known to have fledged from 14 occupied home ranges. The reasons for this reduced success are unclear. Perthshire Merlins had improved breeding success with 28 young fledging from 21 occupied home ranges. The productivity for the whole area was lower in 2001 (1.30 young fledged/occupied home range) than in 2000 (2.21 young fledged/occupied home range) which reflects the poor season in Angus. 22 2000 & 2001 Raptor Round Up Argyll Very few sites were monitored in 2000. Occupancy was low. Productivity was high with all three occupied home ranges fledging young. Coverage in 2001 was lower than in 2000 with only 5 home ranges being checked. Three were occupied but they were not followed up and it is not known if young fledged. Uists Merlins bred successfully at 9 home ranges with birds being seen at another 8 where the outcome was not known. The 9 pairs were successful fledging 21 young. In 2001, Merlins again bred successfully on 9 home ranges. They were recorded from another 6 home ranges. Dumfries and Galloway Merlins could not be monitored in 2001 as the outbreak of foot and mouth prevented access. In 2000, 9 pairs fledged a minimum of 10 young. Lothian and Borders Occupancy in 2000 continued to be low in the Lammermuirs (13 out of 28 home ranges occupied) and Moorfoots (5 out of 12 home ranges with signs of occupation). Coverage was restricted in the Moorfoots as access remains a problem. In the Pentlands, 5 home ranges out of 7 monitored were occupied. Four nests were located in trees. This is a reflection of the poor heather and the availabil- ity of shelter belts. Nests in the other areas are normally on the ground. Productivity was highest in the Pentland tree nests with 14 young fledging from 5 occupied home ranges. Over the whole area, productivity was 2.19 young fledged/occupied home range. Data from 2001 was collected from a variety of sources as access was restricted by foot and mouth. The recorded productivity was 1.36 fledged young/occupied home range. This figure must be treated with caution as coverage is incomplete and only partial data were obtained. Eurasian Hobby Falco subbuteo In 2001 a pair of hobbies successfully bred in northern Scotland and reared 4 young. Birds were reported from another 2 possible breeding localities in the same year. 2000 & 2001 Raptor Round Up VES) Peregrine Falcon Falco peregrinus Table 15. Breeding success of Peregrine Falcons in Scotland, 2000-2001 Region Year Home Home ranges Pairs known to Minimum ranges occupied by a fledge young number of young checked pair or single fledged bird Highland 2000 37 26 12 25 2001 45 26 18 40 North-east 2000 110 61 23 46 2001 101 59 34 73 Tayside 2000 102 76 a1 82 2001 102 71 50 99 Central 2000 33 24 18 38 2001 20 16 16 34 Argyll 2000 17 16 13 24 2001 12 12 8 14 South Strathclyde 2000 34 21 9 23 Dumfries and 2000 109 86 48 103 Galloway Lothian and Borders 2000 64 51 23 54 2001 15 9 6 17 TOTAL 2000 506 361 197 395 2001 295 193 129 277 Shetland Peregrines are now extinct in the Shetlands. No breeding Peregrines were found in either year. The last known pair bred in 1998 and fledged a single chick. Orkney Coverage was incomplete in both years. In 2000, four nests were located. These reared 5 chicks. Peregrines may have attempted to breed at another 9 to 10 sites. Seven sites were known to have been occupied in 2001 with 4 pairs fledging 11 young. 24 2000 & 2001 Raptor Round Up Highland The high number of unoccupied sites is of concern. It is believed that a decline in available prey may be responsible for the loss of some pairs. Productivity was low in 2000 with many pairs failing to rear young (54%). Brood size of successful pairs was high (2.08 young/successful pair) compared to 1998 and 1999 (1.26 young/successful pair). Brood size for successful pairs in 2001 (2.2 young/successful pair) was slightly higher than in 2000 with fewer pairs failing to fledge young (31%). North-east Occupation of home ranges on grouse moors continues to decline with 46% of the home ranges that were checked occupied in 2000 and only 33% in 2001. It is believed that this decline is linked to illegal persecution. Productivity was also low with 0.50 young/occupied home range in 2000 and 1.06 young/occupied home range in 2001. This compares unfavourably with other inland home ranges associated with other land uses, which reared 1.57 and 1.40 young/occupied home range in 2000 and 2001 respectively. Occupancy was also higher with 56% in 2000 and 76% in 2001 of home ranges occupied. The lower occupancy in 2000 was linked to atrocious weather early in the season. Coastal Peregrines bred poorly in 2000 with only 3 young fledging from 15 occupied nest sites. It is believed that the majority of nests failed during a gale in late April. In 2001, these Peregrines reared a record number of young with over 21 young flying from 18 occupied nest sites. Tayside Productivity was low in 2000 across the whole region (1.08 young/occupied home range) compared to 2001 (1.39 young/occupied home range). In north-west Perthshire, a huge area now only supports a few breeding pairs. Persecution is not suspected but a lack of prey may be the problem. The number of coastal pairs increased. 2000 & 2001 Raptor Round Up de) Central In central Scotland, productivity was higher in 2001 (2.13 young/occupied home range) than in 2000 (1.52 young/occupied home range). The settled weather from early spring through to late June may be the reason for this improvement. Argyll The majority of the Argyll data came from one area with only 3 nests in 2000 and 4 in 2001 being in other areas. Productivity was lower in 2001 (1.17 young/occupied home range) than in 2000 (1.50 young/occupied home range). Uists Peregrines were believed to be present and breeding at 10 home ranges in 2000. Birds were located at 12 home ranges with breeding proved at 8 in 2001. Lothian and Borders 2000 was a poor season for Peregrines in south-east Scotland with only 45% of occupied home ranges fledging young (1.06 young/occupied home range). This may be due to poor weather. Failure due to persecution was suspected in several areas. Coverage was poor in 2001 as foot and mouth restrictions prevented access. Only 15 sites were checked with 9 occupied. South Strathclyde and Dumfries and Galloway Persecution is still a problem on grouse moors and in Ayrshire close to pigeon lofts. Productivity in 2000 was similar to 1999 (1.22 young/occupied home range in 2000, 1.19 young/occupied home range in 1999). Occupancy was higher in 2000 (80%) than in 1999 (69%). No sites were checked in 2001 as access was restricted by the outbreak of foot and mouth. 26 Barn Owl Highland 2000 & 2001 Raptor Round Up Tyto alba In Highland, records were received from 5 nest sites in 2000. All were occupied with 4 successful pairs fledging 9 young. Twelve nests were reported in 2001. Two nests failed. The remainder fledged 23 young. Two other known nest sites were not used. The other records came from Argyll where 54 nests were found in 2000 and 32 in 2001 (Table 16). Table 16. Breeding success of Barn Owls in Argyll, 2000-2001 Area Cowal Mull Kintrye/Knapdale Other Year 2000 2001 2000 2001 2000 2001 2000 2001 Nests with eggs 11 Nests fledged young 27 19 Minimum number of young fledged 36 11 24 i 2000 & 2001 Raptor Round Up 7, Tawny Owl Strix aluco Highland Records were received for 43 nests in 2000 with 36 pairs successfully rearing 67 young (1.6 young fledged/occupied home range). Most of these records came from a study area in East Ross (n=36). Twenty nine pairs were successful, fledging 53 young (1.5 young fledged/occupied home range). In 2001, only 35 nests were recorded, This is partly due to lower occupancy rates and partly due to foot and mouth restrictions. Twenty-six of these nests were successful fledging 54 young (1.5 young fledged/occupied home range). In the East Ross study area, only 20 owls were known to lay with 12 successful nests fledging 30 young. This gave the same productivity as 2000. In the Inner Moray Firth, 11 nests were monitored with 10 successful nests rearing 18 young (1.6 young fledged/ occupied home range). Argyll Only one record was received in 2000. A pair fledged a single chick on Bute. In 2001, for the whole of Argyll, 21 pairs were known to fledge 29 young. Short-eared Owl Asio flammeus The only records received were from Argyll. In 2000, a pair bred at RAF Machrinanish. Elsewhere in Kintyre, 2 to 3 pairs were recorded as breeding successfully in Glen Barr. On the Ross of Mull, there were 6 pairs and in Cowal, 7 to 8 pairs were located with at least 2 successful. In 2001, pairs were reported from Bute (1 pair fledged at least 1 chick), Islay (1 pair fledged 4 young) and Cowal where 5 pairs were located. Four of these were successful rearing over 9 young. - SE ) i PT Gs oN ae ae ASA 4 — ~~ ay : Z pa — Mh june a in i o hy he Fe a re o : ee oe A of a “« \ \& iy. " < te, na be “7 . oe yh, M4 in, y ‘ \ P co setts ~ Ney DAY .”. “¢' ‘\ Ae PIR 28 2000 & 2001 Raptor Round Up ————————— The Scottish Ornithologists’ Club was formed in 1936 to encourage all aspects of ornithology in Scotland. It has local branches which meet in Aberdeen, Ayr, the Borders, Dumfries, Dundee, Edinburgh, Glasgow, Inverness, New Galloway, Orkney, St Andrews, Stirling, Stranraer and Thurso, each with its own programme of field meetings and winter lectures. The Waterston Library at the Club’s headquarters at Harbour Point, Newhailes Road, Musselburgh, EH21 6SJ is the most comprehensive ornithological library in Scotland and is available for reference during office hours (Monday to Friday 0930hrs - 1630hrs - but please phone beforehand). A selection of Scottish local bird reports is held at headquarters and may be purchased by mail order. Check out our website for more information about the SOC and other bird related organisations: www.the-soc.org.uk An annual issue of Scottish Birds is published each year, in June. Scottish Bird News is published quarterly in March, June, September and December and the Scottish Bird Report is annual. The Scottish Ornithologists’ Club publishes the annual Raptor Round Up on behalf of the Scottish Raptor Study Groups with grant aid from Scottish Natural Heritage. It is sent to all members. Bird names used in publications follow those used in the Scottish List (Scottish Birds Vol 22(1): 33-49). Copies of these reports may be purchased by non members on application to the SOC. SOC annual membership subscription rates Adult £25.00 Family (2 adults and any children under 18) £35.00 Junior (under 18, or student under 25) £13.00 Pensioner/unwaged £16.00 Pensioner Family (2 adults living at one address) £21.00 Life £500.00 Life Family £700.00 All subscriptions may be paid by Direct Debit and Gift Aided. Subscriptions paid by Direct Debit greatly assist the Club. Please ask for an application form by telephoning the Admin Officer at the above address or visiting the clubs website at www.the-soc.org.uk Produced by the SOC on behalf of the Scottish Raptor Study Groups with financial support from Scottish Natural Heritage ISSN 0036 9144 SCOTTISH NATURAL HERITAGE © 2003 IM 3 9088 01458 5574 SMITHSONIAN INSTITUTION LIBRARIES eet Lady Et Fae tgburrite CoN : ‘ aes = > 7 " be ey ; . 7 a aeey Fatt fe hats CBS c i de a : ‘ < © i s “ 7 % (eee : > 3 : oe err ey fg var APRNA ARETE ale be nea eed hen pee?