sain Gee Lo ween mute ee ere aR Sib LEV EL Ae ASA, Prateek WANA Maa Ys a8, sittary ab eNee nim gh ne se NS LVEN NS atvelee Seat eh coe EE LMM DSN AWE, Bes asP ah dee knee Ee ra eel ae the dh £ oui Neva h a ee beens ba aeaa ad Wy ee am, ae ry VSP ARAD r ge ones ; ENON ALN TA SIL aaritd soe ik hp Borge rrarh Py i rac aetna bate halt 3 HUE SIN ANS yr eS PUNE NOS TN ay eke Perey de " Neale {e5 4 i Sahib ipa far ap 8 en br Henig sd eal edith rerken aioe (bain ? UE bde any Vets Les Apart ps para a ig ee a i ce ee fy, ABTA Cad MAS ae Ps BA ANAM UE SS ANSE Ob og HIN PLAIN ke ed h ‘ i Pe Wem N GH aN Ned hh yg) WF WWE Pd own te!) oft ts La Won optoam Dany Snag aoe cP ds ae mctin regen ‘4 SM sles cage ee ie gays Feb tN crete yy Min Sia BAPitaat Billie Tor xen p phe ere la dette 2 ati aii SValhin 2, ier a eile x fh Oe is eee ater: 2 Oy WARES SEN, datas Urenrnrren SRN oh re La Dede atta Pas y LEME I Oh ao ge DEVS eee Se Reta eae ato Nalin Aube ly Yet Sythe fee a desee ‘ sents AS (DS Ado IVY es yas Maa awe ee Stary y Mle a egre oeg or ee iytaat ak Fadia) oy gn ake oP aren BENrHt ‘ He : ae Ants Weary bas : fe be AS AO ALE aw ath rey ere red UE SE veme Me gt ye : pee eraser PAGES, Son lh ahie ated sak Pile ia HCO RIC Wiis IRTP LEC ET teed we ae ere ner NEE ELEM ATE OEM A tity os Lay PND eel OG d t : meg A eb a ei al DA Puree Ge teeta ON pnt ery wate ys 3 SOME ee Cer he ene tty ed rx Peer eat a! MeO mi pelevad nts 1 PS : we K 4 Wad IM Ue Dg sr a PORE a ESA TAS The SINE a IAMS LEN TS, Neg Sess beaibic reisiaby / DOS TN IN TRAY AS OLIVE MED EN ENE aN Le Aye AGEN TR eats Pe EAN EN OREN, PAN AL ETE he SHES EAN INDY ANE Pe ys a tae ¥ heathy Vue ® ak Mag, 04 ee ta UAT eWeek edt ary tg, PY LVR a Seite nN IS Ete Nel SW SILS Vg PI Ay ha! SIENA Se AN MENA dpe BS Rts Og tH LENA DTN Ge SESS bend AL, Ah eC Te Phen. Moey an! La 10M ot fultt IR hel D ca el a! ee he ne atte Fitna PY EBL oral Pay “ SW Mb 5 ee pat) eee te ha ad ee eee Wa S Whe eta eth I goa, R Fe a NUMA ohh gy ly wth ak be oy Lig nas Hiset aA Sates : 3 ‘ J . i 4 at Me Bi Wa Hi m8 - , a 7 bh b BEETS ANS Ms eS SEs a) Berl Mali * o Peep th ETON eM eat Nagy 1 ; “ toe Ef tdS ones 4 ‘ WEN Gale tow EL f ie Kees OOM AEE IN VE Fal ed Ie og mE A Pe : OWE AE Re es Pitot 6) ah SN EOIE Sete ag fy Ad Sa al ipa a c ate i s Mae egal bal ING a LUEd one oe, i as Eh ery E Pav tether E 2 Edie SU arate ade. Sa 3 DIN these Dua “Y y Dan 1h Vleet fea ad Mme! wiht navies AV : ; : LoiMN ake gag Nea ref d-* tees : Y Det eeee ‘ . b Aaur Wee wpe Se ere iz es ahs ser atababere HURL m ate satan ge Sone: , ; Phe pay I ‘i K Pe é cade dA % 4 AMARA EVERY X it ee ee ns , I 2, ATA ne R > ANGST eT ale Hp ae Om ge k he HLA Panay E sridhseh oder ‘ Sn pair ila i\ ’ ua a uy pL aS eee : BieL a tHSONy es \ JUL 24 $997 ~~ caste ; C/BRARIES 1s ra L ee 7B : < Scottish Birds The Journal of the Scottish Ornithologists’ Club Editor: Dr. S. da Prato Assisted by: Professor D. Jenkins, Dr J.B. Nelson and Dr. |. Bainbridge Business Editor: The Secretary, S.O.C., 21 Regent Terrace, Edinburgh EH7 5BT (tel. 0131-556 6042). Scottish Birds, the official journal of the Scottish Ornithologists’ Club, publishes original material relating to ornithology in Scotland. Papers and notes should be sent to The Editor, Scottish Birds, 21 Regent Terrace, Edinburgh EH7 5BT. Two issues of Scottish Birds are published each year, in June and in December. Scottish Birds is issued free to members of the Scottish Ornithologists’ Club, who also receive the quarterly newsletter Scottish Bird News and the annual Scottish Bird Report. These are available to non- members at a Subscription rate (1992) of £30. 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 21 Regent Terrace, Edinburgh EH7 5BT is the most comprehensive ornithological library in Scotland, and is available for reference during office hours (Monday to Friday, 9.00 - 17.00 hrs). (But please phone beforehand). A comprehensive stock of Scottish local bird reports is held at headquarters and may be purchased by mail order. SOC annual membership subscription rates ACUI occ cassegaocnchesacseanedbenssanctionescoasnssnaedieieincss e¥aausie va uecdei edach saat eee eeee eee CeCe Renee 15100) Family (husband & wife; & childrem under 18) .........5..0.2)...cc.cc000 se eeee ee eee £22.00 Junior (under 18, or student under 25) ..........ccc.ccc.scheeescncgasaseaneee eae eee eee eee £6.00 PONSIONEH/UNWAQE ........cccsccccssneascacsoncssesscaeseseecseideescnacentisesscs Gs ded eee eee eee eee £8.50 PONSIONCK FAMILY .o.c....20..cccckessccoesseonsesdeneeesosuesssiecnesshduausendeas sess sete eee eee eee EEE ERE EEEEEEE £12.00 LAPS vssinn Souhuneu shoes -cocsaceneuwaeeneucdune se dacuyauenoucqecelesee ns succes te Rete et £300.00 Life FRAIMNILY: ousisaiasesshacasccscaveatdesesdelanasceates taugasalesnieasgeonedlansteas Sekai tele EE EE eae eee eeeE Eee £450.00 All subscriptions may be paid by Banker’s Order and Covenanted. Subscriptions paid this way will greatly assist the Club. Published by the Scottish Ornithologists’ Club, 21 Regent Terrace, Edinburgh EH7 SBT Printed by Douglas Gauld & Co Limited, 22 Tannadice Street, Dundee DD3 7QH Scottish Birds (1997) 19: 1-9 Distribution and abundance of Twites wintering in Caithness H CLARK & R M SELLERS The Twite is one of the commonest small seed eating passerines in Caithness in the non breeding season. Five censuses carried out in the winter months between 1992 and 1994 found between 2,136 and 6,882 birds, typically in flocks of <300 birds, occasionally up to about 1,300 birds. Their wintering range extends over some 615 km? of agricultural land mainly between Thurso and Wick. The principal habitats used were weedy Turnip fields, in which Charlock Sinapis arvensis was the main foodplant, uncut Rape, especially in 1992 when poor weather prevented the Rape crop being harvested, and Rape stubbles. The results emphasise the importance of Caithness as a wintering area for this British Red Data List species. The conservation aspects of the results are discussed. Introduction The Twite Carduelis flavirostris is one of only 2 passerines for which the British breeding population is of international importance (Batten etal 1990). Within Britain the species breeds in 2 main areas, the Peak District and southern Pennines in England and in northern and western Scotland from the Mull of Kintyre to the Western Isles and through the Highlands to the Northern Isles (Orford 1973, Sharrock 1976). Small numbers also breed in several other parts of Scotland and in Ireland and a few pairs nestinWales. Elsewhere in Europe the species’ breeding range is restricted to Norway and the Kola peninsula in northern Russia. The Peak District population winters mainly onthe east coast of England from the Humber to Kent and in the Low Countries (Davies 1987, 1988). In Scotland, Twites are virtually absent from high ground in the winter months and appear to be restricted primarily to coastal areas at this time of year (Lack 1986). Despite the obvious conservation interest in the species, many aspects of the Twite’s biology remain unknown. We report here on a study of Twites wintering in Caithness, one of the more important wintering areas in Scotland. Materials and methods Information on the distribution, flock sizes and habitats used by Twites in Caithness in the winter months was obtained principally through 5 surveys of the area carried out on 14-15 November 1992 (a small area in the extreme south of Caithness was not covered until 16 November), 28-29 December 1992, 4-5 December 1993, 3-4 December 1994 and 29-30 December 1994, together with other observations we have made over the past 15 years and records published in the Caithness Bird Report 1983-94 inclusive. Each of the 5 surveys was carried out over 2 consecutive days, and involved checking fields and other land visible from all public roads in Caithness. Typically one day was devoted to north east Caithness (roughly north of the A882 Thurso- Wick road), and the other to the south west of the area. A number of checks made mainly along the boundary between these 2 areas 2 H Clark & R Sellers showed no evidence of a substantial redistribution of birds between one day and the next and, in the short term atleast, Twites appear to return daily to favoured feeding sites. We believe, therefore, that any double counting of birds will have been minimal. Most fieldwork was done by car, but, as necessary, Turnip fields, stubbles and weedy areas were checked by foot. Twites are restless birds and, whilst not easy to locate when feeding, soon take flight revealing their presence. Twites appear to be less active in the afternoon and we found that flocks began to break up in mid afternoon as the birds went to roost. Fieldwork was, therefore, restricted to the period 0900-1400 hr GMT. It became evident during the first survey that birds were restricted almost exclusively to fields of Turnips Brassica napa and Oil Seed Rape Bnapus, so, forthe subsequent surveys, we attempted to survey all such fields in Caithness . Most were also covered in the first survey, but a few were undoubtedly missed; the count for the first survey will, therefore, be an underestimate. Information on the distribution of Turnip and Rape fields SB 19(1) was kindly supplied by Mr P Miller, who, inthe course of his work with the Scottish Office Agriculture and Fisheries Department, visits farms throughout Caithness. The surveys should, in principle, have given a direct estimate of the total number of Twites in Caithness, but no doubt an unknown, but small, percentage will have been missed. Weather for the surveys was generally good. For all Twite flocks located the number of birds present was estimated and the habitat they used recorded. Twites sometimes occur in mixed flocks with other finches, buntings, etc, and, in these cases, we estimated the number of Twites from the total flock size and the proportion of Twites present. For ringing purposes birds were caught in single shelf nets set up in Turnip or Rape fields or in 4 shelf nets set by hedges bordering these fields. Birds were attracted to the general trapping area by playing a tape recording of Twite song. The majority of the birds ringed were caught at Bruan, Killimster and Lynegar. TABLE 1 Flock sizes of Twites wintering in Caithness 1992-94. Mean Survey date flock size 1-30 14/15 Nov 92 125 10 28/29 Dec 92 344 3 4/5 Dec 93 82 12 3/4 Dec 94 101 14 29/30 Dec 94 Sa 5 All combined LZ 44 (39%) Number of flocks of size shown 31-100 101-300 301-1000 >1000 4 3 2 0 4, 7 4 2 8 6 0 0 9 5 2 0 5 8 1 0 30 29 9 2 1997 Results Formation and size of flocks Twites occur in Caithness throughout the year. In the breeding season they are well dispersed throughout the area and are rarely to be seen other than in small groups which we take to be family parties. Flocks begin to form as early as July, but it is not until late August that the majority of birds are to be found in flocks. Flocks start to break up in February and few have been recorded in Caithness after March. Flock sizes vary considerably but, in midwinter, typically number no more than about 300 birds. The distribution of flock sizes recorded in our 5 surveys is Summarised in Table 1; mean flock sizes varied from 82 to 344 birds with about 40% of flocks containing <30 birds, about 65% <100 birds and 90% <300 birds. Averaged over our 5 surveys, about 5% of the population were in flocks of <30 birds, 12% <100, 53% <300 and the remaining 47% in flocks of >300. The largest flocks recorded in Caithness were 1,280 birds at Alterwall and 1,100 birds at Northfield, both in December 1992 (as part of our surveys), 1,000 birds at Gills in August 1987 (R M Sellers, unpublished observations) and 1,000 at Loch Watten in November 1990 (Caithness Bird Report 1990). The majority of flocks we recorded appeared to consist solely of Twites; the remainder were mixed flocks with Greenfinches C chloris, Linnets C cannabina, Chaffinches Fringilla coelebs, and, less often, Redpoll C flammea, Brambling F montifringilla, House Sparrows Passer domesticus, Reed Buntings Emberiza schoeniclus and Yellowhammers E citrinella. Twites were generally the single most numerous species in these mixed flocks and were one of the commonest small seed eating passerines in Caithness in the winter months. Twites wintering in Caithness 3 Abundance, distribution and habitats The results of the 5 main surveys are summarised in Table 2. Between 19 and 30 flocks were found per survey, with the total number of birds recorded varying between 2,136 and 6,882 birds. There was a marked increase in the number of birds between the first and second 1992 surveys, which were carried out 6 weeks apart, though the number of flocks increased by only one. We suspect that this increase in the number of birds was a consequence of an influx of birds. By contrast, there was only a small decrease in the number of birds between the 1994 surveys, though the number of flocks decreased by about a third. TABLE 2 Numbers of Twites wintering in Caithness 1992-94. Survey date Flocks Birds 14/15 Nov 1992 19 2368 28/29 Dec 1992 20 6882 4/5 Dec 1993 26 2136 3/4 Dec 1994 30 3036 29/30 Dec 1994 19 2873 The winter range of Twites in Caithness is shown in Fig 1, which is based on all observations from our surveys, together with records from the Caithness Bird Report for October-March inclusive. Sightings are confined mostly to land below 200 metres above sea level and occur almost exclusively on the arable and stock rearing land between Reay, Thurso and Wick and Dunbeath. The flow country of south and west Caithness is difficult to survey, but, on the basis of fieldwork along the Causey Mire (the A895 between Georgemas and Latheron), in the vicinity of Loch More, near Broubster, between Camster 4 H Clark & R Sellers and Badlipster, near Houstry, and between Dunbeath and Braemore, we are satisfied that the absence of records from this area represents a genuine absence of birds. On the basis that the winter range coincides with that of agricultural land, as shownin Fig 1, we estimate that the area usually occupied by Twites in Caithness in the winter months is approximately 615 km2. Only 6 sites (32%) occupied during the 14-15 November 1992 survey still held birds 6 weeks later; for the 3- 4 December 1994 survey, 14 sites (487%) still held birds by the time of the next survey 4 weeks later. The majority of the sites vacated held flocks of <30 birds and/or, in the case of the 1994 surveys, were sites where stubbles SB 19(1) had been ploughed between the 2 surveys. The habitats from which Twites were recorded in Our Surveys are summarised in Table 3. The most important by far were weedy Turnip fields (Charlock Sinapis arvensis being the most common weed), Rape stubbles and, in 1992 especially, uncut Rape as wet weather delayed harvesting of the Caithness Rape crop in 1992 and many fields were simply left uncut, or were cut after most seed pods had spilt their contents on the ground. Small numbers of birds were recorded from a number of other habitats, including the sand dune system at Dunnet Links (dominated by Marram Grass Ammophila arenaria), weedy Figure 1 Winter range of Twites in Caithness. Shaded area shows agricultural land (after Omand 1972). Largest flock size for each site shown; all records refer to period October-March inclusive; includes data from Caithness Bird Report and special surveys. Dunnet ~ John O’Groats £ Flock size 2714-30 © 31-100 @ 101-300 @® 301-1000 @ >1000 1997 ground (fields, roadside verges etc) and Barley stubbles. Flock sizes were appreciably larger on Rape and Rape stubbles than on Turnips andthe other habitats (Table 3). Our fieldwork has shown that Turnip fields, uncut Rape and Marram Grass are also used by some birds for roosting overnight. We have not systematically collected information on foodstuffs but, during the course of the study, recorded Twites taking seeds of the following plants: Charlock (husked seeds of this plant were visible in the crops of many of the birds we caught), Thistle Carduus spp, Dock Rumex spp, and Rape. There is also a record from the late summer of 1992 of Twites feeding on the putting greens of Reay Golf Course where they were taking the seeds of Annual Meadow Grass Poa annua (J Gunn, pers comm). Ringing recoveries Despite having ringed 608 Twites in Caithness, about 37% with colour rings, only 4have been retrapped, recovered or seen again away from where ringed. These involved an adult female ringed at Killimster, Caithness in December and found dead near Poolewe, Wester Ross (165 km WSW) the following May, a first year male ringed at Lynegar, Caithness in November and recovered on Lewis in the Western Isles (180 km W) in July 3.5 years later, a first winter female ringed at Lynegar, Caithness in January and found dead the following April at Dunbeath, Caithness (29 km SSW) and a first year male ringed at Northfield, Caithness in December and found tangled in sheep’s wool at Hempriggs, Caithness (2 km SSE) in June 2.5 years later. The third and fourth of these seem to have been birds which bred and wintered in Caithness, whereas the first and second were ones which bred in north west Scotland and wintered in Caithness. Twites wintering in Caithness 5 Discussion Population size The Twite is one of the most abundant small seed eating passerines which winter in Caithness with a population varying between about 2,000 to 3,000 birds in typical years and up to 7,000 birds in exceptional years in the period 1992-94 according to our surveys. The Winter Atlas (Lack 1986) gives an estimated British and Irish midwinter population of 100,000-150,000 birds (based on a breeding population of 20,000-40,000 pairs and “a net input of 50,000-100,000 birds by December’). Birds wintering in Caithness thus represent roughly 2% in typical years, and up to 5% in exceptional years, of the British wintering population, emphasising the importance of the area for Twites. The first, third, fourth and fifth of our surveys found roughly the same number of birds (2,000-3,000) and provide a baseline against which future changes can be compared. We found a substantial increase in the number of birds present between the first and second 1992 surveys when food was unusually abundant. We suspect that this was the result of an influx of birds from outside Caithness, or passage birds staying to winter. Flock of Twites in Caithness typically hold up to 300 birds but, on occasion, numbered over 1,000 birds. These large flocks appear to be amongstthe biggest ever recorded in Scotland (Thom 1986), but flocks up to 2,500 strong have been seen on the Wash (Davies 1987, 1988) andup to 3,000 on the Continent (Cramp & Perrins 1994). Habitats In Caithness, Twites are confined almost SB 19(1) 6 H Clark & R Sellers OSI 6y SP pL Gce 28 @ZIS 4OO}} UBS (%t) OSt (%t) 8bL (%1) Lz (%22) 9€8E (%Er) LSrZ (%1€) €6ES [B}0] 0 (%1'0) € (%€) OOF (%81) O€S 0 (%8Z) Ovez 6 99d 0€/6Z 0 (%2) GZ (%L) €€& (%SS) ~ve89l (%Z) OlZ (%ve) vEeOL 76 98d +/E (%Z) OSL (%E) OZ (%S) 801 (%2S) 6ZIL 0 (%22) 6G¢ €6 98d G/Pp 0 0 0 0 (%88) 2909 (%Zlt) 028 26 98 62/82 0 0 0 (%61) Ebr (%0S) S8LL (%1€) OZ 26 AON SI/rh ‘SPIg JO ON (%t) - (6) *c (%p) (%61) 22 (%81) 12 (%~S) 29 soyep |IV 0 (%S) I (%G) 1 (%91) € 0) (%rl) vi 76 99d 0€/62 0 (%€) | (%Z) 2? (%L1) (%Z) ¢ (%29) O02 v6 99 P/E (%b) - (%p) IL (%8) 2 (%SE) 6 0 (%0S) EL €6 98q S/P 0 0 0 ) (%GZ) SI (%Gz) ¢ 26 99 62/82 0) 0 0 (%92) G (%12) ¥ (%€S) OL 26 AON SL/L ‘SHOO|J JO ON giqqnis punos6 ajqqnis edey Aayeg Wee Apsa\\ edey ynoun sdiuin | syeyiqeH ayep Aavins 'y6-Z661 Ssauyyed ul Bulsajuim sayimy Aq pasn sjeyqeH € AIGVL roy exclusively to farmland in the winter months, the important habitats being weedy Turnip fields, Rape stubbles and, mainly in 1992, fields of uncut Rape. Rape is a new crop in Caithness. The birds were exploiting spring sown Rape, which, in 1991, amounted to only 5 ha rising to 491 ha in 1992, the increase apparently being due to a change in agricultural subsidies. Over the same period, winter sown acreage decreased from 234 ha in 1990/91 to 35 ha in 1991/92. We have no figures for seasons before 1991 but understand that the areas sown with Rape were extremely small. As noted above, much of the 1992 spring sown crop was not cut, or was cut late, because of wet weather at harvest time; the abundance of Rape seed in the winter of 1992/93 in Caithness was, therefore, wholly novel, but our observations serve to highlight the ability of Twites to exploit new food sources when they become available. Spring sown Rape appears to be establishing itself as a crop in Caithness with 274 hain 1993 and 488 ha in 1994, and, even if the crop is successfully harvested, Rape stubbles have the potential to supply a significant part of the species’ winter food requirements. Weedy Turnip fields, however, appear to be the traditional wintering habitat and our preliminary observations suggest that they are also important in the Dornoch Firth and Beauly Firth areas. In addition, Twites are recorded as using Turnip and stubble fields in north east Scotland (Buckland, Bell & Picozzi 1990), weedy Turnip fields in Morayshire (Cook 1992) and, on Islay, as using stubbles as well as taking small seeds from farmland weeds and maritime flowers (Elliot 1989). Origins Although there are only 4 ringing recoveries available, they suggest that the Caithness wintering population is drawn from both the Twites wintering in Caithness 7 local breeding population and that of north west Scotland. There is a passage of Twites on Fairlslein both autumn and spring (Dymond 1991) as wellas anumber of ringing recoveries showing movements between Shetland and Orkney. We consider it likely that some birds from the Northern Isles also winter in Caithness. There appears to be a small passage across the northern North Sea. Small numbers of Twites are seen each year on North Sea oil rigs (North Sea Bird Club Reports) and it is tempting to see this as evidence that Norwegian birds move into Scotland, though they could be Scottish birds moving to the Continent. This is supported by the recovery of a bird ringed on Fair Isle in July 1953 onaship off Heligoland the following October. By contrast, an extensive ringing programme in Norway in the 1960s failed to generate asingle recovery anywhere in Britain, though there were over 50 along the North Sea coast from Denmark to France (Bernbhoft-Osa 1965). On the evidence currently available, we suggest that the Caithness population is drawn almost entirely from the northern Scottish breeding population. Conservation The Red Data Book for birds in Britain (Batten et al 1990) stresses the importance of saltmarsh as wintering habitat for the Twite and thatamongthe main threats to the species’ survival are loss of habitat and a reduction in the food supply at wintering sites. These comments appear to be based on the English population for which the main foodplants in winter are Marsh Samphire Salicornia spp vand Sea Aster Aster tripolium . The former does not occur in Caithness, nor in several of the other important wintering areas in Scotland, and the latter is relatively scarce in Caithness and is restricted to coastal cliffs (Perring & Walters 1990, Bullard et a/ 1977) 8 H Clark & R Sellers which Twites do not appear to use in the non breeding season. Our results suggest that the weeds of Turnip fields, especially Charlock, are important in Caithness and that this may well be the case in several other parts of Scotland. The acreage of Turnips, and, by implication, that of the associated weed flora, has decreased by a factor of about 8 over the past half century (Table 4), and this trend gives some cause for concern as to future winter food supplies in the area. A similar trendis apparentin Sutherland (Omand 1991). Our results show that Twites can adapt to other foodstuffs if circumstances permit. The abundance of Rape seed in Caithness in 1992 was probably unique, though the continued availability of Rape stubbles may be important. Itis clear that the abundance of Turnip fields and Rape stubbles in Caithness needs to be monitored carefully and that further studies are required elsewhere in the main Scottish wintering areas to define the habitats used and the extent to which these are under threat. Scottish Twites have been described as largely sedentary (eg Batten et a/ 1990, Jardine & Reid 1993), and, whilst this is partly true, TABLE 4 Turnip acreages in Caithness over the past half century .4 Date Area of Turnips (ha) 1937 3662 1961 2743 1971 1121 1983 707 1991 4396 a Data from Scottish Office Agriculture & Fisheries Department b After 1983 Turnip fields < 2 ha did not have to be reported; this figure thus slightly underestimates the true figure SB 19(1) evidence is mounting that some do move and that the pattern of these movements may be quite complex. The breeding and winter distributions show a substantial redistribution of birds between the breeding and non breeding seasons (Jardine & Reid 1993, and Lack 1988), and the results of this study are evidence of movements across the north of Scotland between the west coast and Caithness. The conservation of these birds thus depends on protecting both the breeding areas in north west Scotland and wintering grounds in Caithness. Acknowledgements Many people have assisted with the collection of the information reported here and we owe a tremendous debt to them all. In particular, we would like to express our thanks to the farmers who permitted us to count or catch Twites on their property, past and present members of the Thurso Branch of the SOC for counting Twites over many years, to Peter Miller for providing information on the location of Turnip and Rape fields in Caithness and for help with the surveys, to M Drummond of the Scottish Office Agriculture & Fisheries Department, Thurso for data on the acreages of Turnips and Rape, to Neil Darroch and Julian Smith for help with the fieldwork, to Jimmy Gunn for advice about plants in Caithness and to Baz Hughes for comments on an earlier draft of this paper. We also wish to acknowledge our thanks to the SOC for the award of an endowment grant in support of this work, the British Trust for Ornithology for access to their ringing recoveries, and the British Museum (Natural History) and the Zoological Museum, Oslo for permitting access to their collections of skins. 1997 Twites wintering in Caithness 9 References BattenLA, Bibby CJ, ClementP, Elliot G D & Porter R F 1990. Red Data Birds in Britain. Poyser, London. Bernhoft-Osa A 1965. Om Birgiriksens Carduelis flavirostris trekk. Stavanger Museums Arbok, 35, 109-118. Buckland ST, Bell MV & PicozziN 1990. The Birds of north-east Scotland. North-east Scotland Bird Club, Aberdeen, p 418. Bullard E R, Butler J K, Gunn J M & Hewison V 1977. The Wild Flowers of Caithness. Caithness Field Club. Cook M 1992. The Birds of Moray and Nairn. The Mercat Press, Edinburgh. Cramp S & Perrins C M (eds) 1994. The Birds of the Western Palearctic, Vol. 8. Oxford University Press, Oxford. Davies M 1987. Twite and other wintering passerines on the Wash saltmarshes. The Wash and its Environment, Nature Conservancy Council, Peterborough, pp 123-132. Davies M 1988. The importance of Britain’s Twites. RSPB Conservation Review. 2:91-94. Dymond J N 1991. The Birds of Fair Isle. Elliott RE 1989. Birds of Islay. Christopher Helm, London, p 187. Jardine D C & Reid J B 1993. Twite in The New Atlas of Breeding Birds in Britain and Ireland: 1988- 1991 (ed D W Gibbons, J B Reid & RA Chapman), T & AD Poyser, London, p 418. Lack P 1986. The Atlas of Wintering Birds in Britain and Ireland. Poyser, Calton. Omand D (ed) 1972. The Caithness Book. Highland Printers, Inverness. Omand D (ed) 1991. The Sutherland Book. Northern Times, Goispie, p 250. Orford N 1973, Breeding distribution of the Twite in Central Britain. Bird Study 20: 51-62, 121-126. Perring F H & Walters SM 1990. Atlas of the British Flora. 3rd Ed Botanical Society of the British Isles. Sharrock J TR 1976. The Atlas of Breeding Birds in Britain and Ireland. British Trust for Ornithology/ lrish Wildbird Conservancy. Thom V M 1986. Birds in Scotland. T& AD Poyser, Calton. p331. H Clark, 3 Lindsay Place, Wick, Caithness KW1 4PF R M Sellars, Rose Cottage, Ragnall Lane, Walkley Wood, Nailsworth, Gloucestershire GL6 ORU Revised manuscript accepted May 1996 Wintering Twite LEE If Mike Ashley 10 Scottish Birds 19: 10-27 SB 19(1) The status of the Gannet in Scotland in 1994-95 S MURRAY & S WANLESS A census of all the Gannet colonies in the east Atlantic was carried out in 1994-95. This paper summarises the results from all the Scottish gannetries and compares them with the last major census in 1984-85. Scotland remains the stronghold of the Gannet with 12 colonies and a total population of 167,407 apparently occupied sites representing 61.1% of the east Atlantic population. Numbers were divided very unevenly between the colonies with St Kilda, the Bass Rock and Ailsa Craig together holding 76% of the Scottish population and 47% of the east Atlantic population. Since 1985 Gannets have bred at least once on Rockall (1992) and colonised Scotland’s first mainland gannetry at Troup Head (1988). The colony on the Shiant Isles was short lived. The Scottish population increased at an average rate of 2.4% pa between 1984-5 and 1994-95. Sule Stack was the only colony which did not increase over the period. Rates of increase at the other colonies varied considerably but, in general, were highest at recently founded colonies. Introduction At the time of the last comprehensive census of Gannets Morus bassanusin 1984-85, there were 12 gannetries in Scotland. The Scottish population was estimated at 132,100 occupied sites which represented almost 60% of the total for the east Atlantic and about 50% of the world population (Murray & Wanless 1986, Wanless 1987). Numbers at most colonies were increasing and the Scottish population was estimated to have shown an average rate of increase of 2.0% per annum (pa) between 1969 and 1985. We coordinated another complete survey of east Atlantic gannetries in 1994-95 and this paper details the counts from all the Scottish colonies and compares them with the results of the 1984- 85 census. Methods The locations of the colonies are shown in Figure 1. We carried out aerial surveys of St Kilda, Sule Stack, Sula Sgeir, the Flannan Isles and the Bass Rock in 1994, and Ailsa Craig and the Scar Rocks (also referred to as the Scare Rocks) in 1995. The remaining colonies were counted by other observers who are acknowledged below. The 1994-95 ‘survey was carried out using similar methods to those adopted in 1984-85 with colonies counted either from field counts made from the land or sea, or from photographs (mainly transparencies rather than prints) taken from the land, sea or air. For a few colonies such as Bass Rock, Hermaness and St Kilda different sections of the colony were counted using different methods. Details of the various techniques 1997 The Status of the Gannet in Scotland in 1994-95 17 Figure 1 The distribution of gannetries in Scotland 1994-95. The colonies are identified by the following numbers: Scar Rocks (1); Ailsa Craig (2); St Kilda (3); Rockall (4, not known if occupied in 1994/95); Shiant Isles (5 abandoned); Flannan Isles (6); Sule Sgeir (7); Sule Stack (8); Foula (9); Hermaness (10); Noss (11); Fair Isle (12); Troup Head (13); and the Bass Rock (14). used to count Scottish gannetries, and the problems associated with these methods are given in Murray & Wanless (1986). 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). Sites which were clearly occupied by nonbreeders were excluded whenever possible. In most field counts the unit was the apparently occupied nest (AON, one or 2 birds at a site with some nest material present). Sites with a chick but no obvious nest were includedin this category. For both count units we, or other observers, judged that it was always possible to distinguish directly whether a pair or a single bird was present and hence no correction for this effect was necessary for totals of either AOS or AON (cf Nelson 1978). The only exception to this was J B Nelson’s count of the Bass Rock for which full details are given in the section dealing with this colony. We emphasise that 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 also makes it impossible to 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, acombination of totals of apparently occupied sites at the majority of colonies anda 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. 12 S Murray & S Wanless Nelson (1978) recommended that counts be made in June or July. These guidelines were followed atall colonies, except St Kilda, where the count was carried out in mid May to take advantage of a spell of fine, settled weather, and Ailsa Craig and Scar Rocks where the aerial survey was delayed until early August so as to ensure that there was no disturbance of Guillemots Uria aalge. Where possible, counts for a colony are presented in terms of its constituent sections to facilitate future comparisons of population changes. Average rates of change for each colony were calculated using the equation: Po/Py =(1+ rt where ris the rate of population change, P4 is the nest or site count in 1984 or 1985, Po is the nest or site count in 1994 or 1995 and tis the number of years between the 2 counts. Counts made up to 1985 are summarised in Fisher & Vevers (1943, 1944) Nelson (1978) and Murray & Wanless (1986). Only subsequent counts are referenced in this paper. Figure 2 Counting divisions used to census the Scar Rocks in 1995. Counts of the number of AOS in each section are also shown. Summit rim North Cliff West Stack (111) SB 19(1) Results There were 12 active gannetries in Scotland in 1994-95 (Fig 1). Since the 1984-85 survey, the first colony on mainland Scotland was established at Troup Head (Matthews & North 1989) and the first recorded breeding on Rockall occurred in 1992 (Belaoussoff 1993). The lone nest site on the Shiant Isles was abandoned around 1987. The 1994-95 counts and current population trends are summarised below for each of the Scottish gannetries. Scar (Scare) Rocks An aerial survey was carried out on 5 August 1995. The colony was divided into 4 sections (Fig 2) and the 2 independent counts gave an average total of 1952 AOS with observer variation being 7.4% of the mean. The quality of the photographs of the north cliff and the plateau was excellent but only moderate for the west stack and slope. There was a substantial increase in numbers between 1984 and 1995 (Table1). The change from a land based field count of AON in 1984 to an aerial survey of AOS in 1995 will have exaggerated the rate of increase particularly since some areas of the colony are difficult to see from the land (P Collin pers comm). For these reasons, the calculated increase of 154% over 11 years, an annual rate of 8.8% pa, must be regarded as an overestimate. (26) 1997 The Status of the Gannet in Scotland in 1994-95 13 Table 1 Counts of the gannetry on the Scar Rocks 1984-1995. Year Field counts Aerial survey Source (nests) (occupied sites) 1984 770 Muray & Wanless (1986) 1987 830 Dickson (1992) 1989 700 Nelson in Dickson (1992) 1994 1200+ P N Collin pers comm 1995 1952 This survey Table 2 Counts of the number of _ Ailsa Craig The count was made from aerial photographs taken on 5 August 1995. Weather conditions were ideal and the standard of the resulting slides was excellent. For the count the colony ’ was divided into 19 sections (Fig 3) based on | 23 sections originally used by Gibson (1951) and Wanless (1979). A few sections had to be amalgamated, either because increases in adjacent sections made it difficult to distinguish the original boundaries, or because of differences in viewing angle between the previous sea based censuses andthe 1995 aerial survey. Two independent counts were made of the colony, the average of which gave a total of 32,455 AOS (Table 2) and observer variation was 5.0% of the mean. This represented an increase of 42% (3.6% _ pa) over the 1985 count of 22,811 AON. However, although the colony has undoubtedly shown a substantial increase over the period, differences in methodology - and count units used in the 2 surveys are likely to have exaggerated the magnitude of | the change. St Kilda Full details of the methodology and results for the St Kilda survey are given in Murray & _ Wanless (1996) but, in brief, an aerial occupied Gannet sites on Ailsa Craig on 5 August 1995. Sections are shown in Fig 3 and are based on those used by Gibson (1951) and Wanless (1979). Count Section (AOS) 1 946 2 187 3 441 4,5 1877 6 2097 vh 954 8 4300 9,10,11 11959 12 140 13,14 1994 15 993 16 1270 17 21 18 908 19 3410 20 531 21 250 22 0 23 ZY: Total 32455 14 S Murray & S Wanless SB 19(1) Figure 3 Counting divisions used to count aerial photographs of Ailsa Craig in 1995. photographic survey of the gannetry was carried out on 15 May 1994. Counts of Stac Lee and Stac an Armin were made entirely from aerial photographs; one section on Boreray was counted from a photograph taken from the sea and the remaining sections were counted from aerial photographs. Photographic coverage of the stacs was complete but a few sites on Boreray (c 1.4% of the total) were thought to have been missed and the count was adjusted accordingly. The overall total for St Kilda was 60,428 AOS with 14,660 AOS (24%) of the total on Stac Lee, 12,950 AOS (21%) on Stac an Armin and 32,818 AOS (54%) on Boreray. These figures represented overall increases since the last survey in 1985 of 20.7%, 8.4%, 9.3% and 33.0% respectively, equivalent to average annual rates of 2.1% pa, 0.9% pa, 1.0% paand 3.2% pa. While some, but notall, of the increase on Boreray was thought to be due to differences in count methodology, the increases on the stacs, and some parts of Boreray, were undoubtedly real and we concluded that the St Kilda gannetry had probably shown a sustained increase of c 0.9% pa over the last 35 years. Rockall No observations were made in 1994-95 buta single nest containing an egg was found on 19 June 1992 (Belaoussoff 1993). Although Gannets had previously been seen on the rock (review in Bourne 1993) this was the first recorded breeding. Shiant Isles No nests were found during a visit to the islands on 3-4 July 1995, buta single bird was seen ashore close to the Garbh Eilean site (J Love pers comm). During the 1985 survey, a single bird was recorded on a nest on Eilean Mhuire and there was another well built nest on Garbh Eilean. Only the Garbh Eilean site was occupied in 1986 and no Gannets were recorded ashore in May 1992 (S Murray pers 1997 obs). It therefore appears that, after a period of occupancy lasting from 1979 to 1986, during which there was no definite breeding record, Gannets have, for the time being at least, given up their attempt to colonise the Shiant Isles. Flannan Isles The count was made from aerial photographs taken on 15 July 1994. The colony consists of 6 separate subgroups on the island of Roareim and its islets (Fig 4). The quality of the slides was high and the average of 2 independent counts gave a total of 1,438 AOS with an observer variation of 4.7% of the mean. No birds were recorded ashore on Eilean a Gobha during the 1994 survey, but Gannets were present, but not apparently nesting, on the island’s east cliffs during visits in 1988 and 1992 (Table 3). The Status of the Gannet in Scotland in 1994-95 15 Previous counts of the colony have all been made from field counts supplemented by land or sea based photographs (Table 3). In 1992, landing was limited to the edge of Sgeir an Eoin resulting ina probable underestimate of this sub colony and the main colony. Although it is clear that the gannetry on the Flannan Isles has increased greatly between 1985 and 1994, the change in count methodology will undoubtedly have resulted in the overall increase and average rate of change, 545% and 23.0% pa respectively, being overestimated. Sula Sgeir The count was made from aerial photographs taken on 15 July 1994. Full details of the survey are given in Murray & Wanless (1994). The colony was divided into 8 sections and the standard of the count was considered to be high. The average of 2 independent counts was 10,440 AOS (Table 4) with an observer variation of 0.2% of the mean. Count Figure 4 Locations of sub colonies of Gannets on the Flannan Isles. All the subcolonies except the one on Eilean a Gobha were occupied in the 1994 survey. Counts of the number of AOS in each section are also shown. BRONA CLEIT ——— East (7) South (114) Main colony (960) Arch Stack (108) > sGEIR AN EOIN (248) EILEAN A GOBHA 500 m ———EE 16_S Murray & S Wanless SB 19(1) Table 3 Counts of the gannetry on the Flannan Isles 1985-94. Locations of the various sections are shown in Fig 4. 16 June 12 June 24 May 15 July 1985! 19882 19928 19944 (nests) (occupied (nests) (occupied sites) sites) a) Roareim East 2 c6 ? South 23 113 Main colony 223 384 539+ 960 Arch stack 23 107 Sgeir an Eoin 26 84+ 248 West stack 4 4 8 b) Eilean a Gobha East c50° 0556 0 Total 223 416 73447 1438 Notes 1 Counted from photographs taken from the land supplemented by land based field counts 2 Counted from photographs taken from the land supplemented by land based field counts (M Tasker & D Rothe, pers comm) Land based field counts (S Murray, unpublished) Aerial photographs Count unit birds Count unit occupied sites Includes 55 occupied sites N ®D OQ A © methodology for the 1985 and 1994 surveys Sula Sgeir is the only Scottish colony where was Identical. The number of AOS increased young Gannets, known as gugas, are still by 14% over the period at an annual rate of | taken for food. The current licence, which is 1.5% pa. There was no evidence of any issued by the Scottish Office Agriculture, expansion in area of the gannetry and the — Environment and Fisheries Department, is increase appearedtohaveoccurredmainlyin for 2,000 young per year but the number of sections 3 and 4 (Table 4). gugas taken is not closely monitored, neither 1997 is there any check on the number of chicks lost during the hunt. Counts of corpses in photographs of gugas prior to off loading from Sula Sgeir (Beatty 1992) suggest that, at leastinsome years, the number killed may be greater than the licence. Sule Stack The count was made from aerial photographs taken on 15 July 1994. Complete coverage of the rock was achieved and the standard of photographs was excellent. The colony was divided into 8 sections (Fig 5) and 2 independent counts gave an average total of 4,888 AOS with an observer variation of 2.8% of the mean. The North Rock remains uncolonised, while the South Rock had only 73 occupied sites but large numbers of non breeders. Although count methodology in 1985 and 1994 was directly comparable, the 1985 photographs were of only moderate standard. The possibility exists that the 17% The Status of the Gannet in Scotland in 1994-95 17 decrease between 1985-94 was atleast partly due to the 1985 total being too high (Table 5), but Sule Stack appears to be the only Scottish gannetry not to have increased in size over the last decade. Foula A iotal of 600 AOS was counted from the land and sea by R W Furness on 2 and 16 July 1994. The colony has increased by 186% over the last 10 years at an average rate of 11.1% pa. Additional counts during the period were 210 AOS/140 AON (1984), 124 AON, 151 AON, 158 AON, 220 AON and 280 AON 1987-1991 respectively (data from Seabird Colony Register). Hermaness The numbers of AON were counted from the land by H Towll (SNH) between 12 and 18 July 1994. Hidden areas were photographed Table 4 Counts of the Sula Sgeir gannetry in 1985 and 1994. Count sections are shown in Fig 3 of Murray & Wanless (1986). Occupied sites 1985 1 858 2 584 3 2394 4a , 4b 2031) 5 136 6 1532 if 1541 Helipad 68 Total 9143 1994 % Change 879 2.4 384 -34.2 2954 23.3 2038 sy 1154 57.2) 0 -100.0 1353 -11.7 1680 9.0 0 -100.0 10440 14.2 * What was Section 4 in 1985 became 4a and 4b in 1995 18 S Murray & S Wanless from the sea on 22 August and AOSs counted off the slides. Following Murray (1992), the colony was divided into 18 sections and the figures given in Table 6 are an average of 3 replicate counts by the same observer. A total of 10,640 AON was counted from the land and 1,353 AOS were counted from the photographs (Table 6) which gives an overall total of 11,993 AON/AOS. The 1984 total was originally given as 8,063 AON but this figure was subsequently revised to 8,506 AON (Murray 1992). Thus, over the last 10 years, the colony has increased by 25%, an annual average rate of 3.5% pa. During the period 1969-84, the Gannet population on Hermaness showed marked fluctuations. As the colony is notoriously difficult to count, it was unclear whether these variations reflected real changes in numbers or were due to counting error. In contrast there has been no evidence of fluctuations over the past 10 years with 9,904 AON recorded in 1986 (S. Wanless & M. Heubeck unpublished) and 10,057 AON in 1991 (Murray 1992). Count methodology was standardised in 1991 (Murray 1992) which suggests that the earlier variability was due mainly to differences in the way that counts were made. SB 19(1) Figure 5 Counting divisions used to census the gannetry on Sule Stack in 1994. Counts of the number of AOS in each section are also shown. N 100 m Table 5 Comparison of the number of occupied Gannet sites in sections of the Sule Stack gannetry in 1985 and 1994. Section 15 July 15 July % 1985 1994 change North 0 0 0 North west and north east 2240 1406 -37 East, south east, centre and top 3590 3409 -5 South 50 73 46 Total 5880 4888 -17 1997 The Status of the Gannet in Scotland in 1994-95 19 Table 6 Counts of the Hermaness gannetry in 1994. Data supplied by SNH. The sections are shown in Murray (1992). Section Land count Sea count (nests) (occupied sites) Rumblings East 736 - Vesta Skerry East 1381 - Vesta Skerry West - 623 Humla Stac North 251 - Humla Stac West 251 - Humla Stac Southwest 0 - Humla Houl North 133 - Humla Houl South 435 - Burra Stack East 335 . Burra Stack West - 238 Clingra Stack 207 - Neap North Face - 492 Neap-Soorie 3096 - Soorie-Saito 1799 - Saito 1509 - Soorie Geo North Stack 19 - Soorie Geo South Stack 35 - Neapna Stack 453 - Total 10640 1353 Overall total 11993 Noss AON (1991) and 6,856 AON (1992) although In 1994, complete land counts were made on 18-20 June and 17 July respectively and 2 replicate sea counts were carried out on 30 June by S Smith and C Barton (SNH). The methods are described in Murray & Wanless (1992) and, for both land and sea counts, the count unit was apparently occupied nests. The colony was estimated to contain 7,310 AON (Table 7). In 1984 the colony was estimated to contain 5,231 AON. Subsequent counts have been 7,218 AON (1989), 6,730 the 1989 figure is now thought to be an overestimate due to counting error (Murray & Wanless 1992). Numbers have increased steadily over the last 10 years with the colony showing an increase of 40% at an average annual rate of 3.4% pa. Fair Isle The 1994 counts of 825 nests was made on 13 June by G Thompson, Fair Isle Bird Observatory Trust. Most of the colony was 20 S Murray & S Wanless SB 19(1) Table 7 Counts of the Noss gannetry in 1994. Values for land counts are the averages of 2 surveys carried out 18-20 June and 17 July respectively; values for sea counts are averages of 2 counts carried out on 30 June. Data supplied by SNH. Section Land count (nests) Cradleholm 0 Holmoless 91 Holmoless to Geordies Hole 1569 Geordies Hole 49 Rumblewick 223 Rumblewick Face 162 Cuddacks Geo - Noup South 2659 Noup East : Noup North 817 The Rump 560 Rump North 297 Geo Heogatoug - Total counted from vantage points on the land but part of the Inner Stack of Skroo was counted directly from the sea (Table 8). The number of nests rose from 138 in 1985 to 975 in 1995, an overall increase of 600% and an average annual rate of 21.6%. Nest counts between 1986 and 1993 were 258, 304,488, 676, 643, 687, 781 and 764 respectively (Riddiford 1993, Seabird Colony Register). Troup Head The first definite breeding record for Troup Head was in 1988, although the colony could have been established in 1987 (Matthews & North 1989). Full details of changes in numbers, breeding success and breeding chronology for the colony for the period 1988 Sea count Total (nests) (nests) 1 1 17 108 - 1569 - 49 - 223 - 162 - 0 - 2659 517 517 93 910 213 773 42 339 - 0 7310 Table 8 Counts of the Fair Isle gannetry in 1994. Dats supplied by Fair Isle Bird Observatory Trust. Section Count (nests) Outer Stack 219 Inner Stack 66 Yellow Head 34 Dronger OF North Felsigeo 285 Toor ‘o’ da Ward Hill 82 Matchi Stack 30 Kame ‘o’ Guidicum 32 Total 825 1997 to 1995 have been summarised by Wanless et al (1996) but, in brief, the number of nests increased by 105% from 5 in 1988 to 530 in 1995, aspectacular rate ofincrease averaging 64% pa. Bass Rock Two counts were made in 1994. One was by JB Nelson on 13 July using the same methods as in 1984, ie mainly from photographs taken from the sea and land with areas which were inadequately covered estimated by eye, using either a comparable area of counted birds or relying upon previous knowledge of the area. Counts were made of the number of birds and these were converted to occupied sites assuming a single bird to pair ratio of 10:1. It was considered possible that the corrected total of 27,293 AOS underestimated the actual The Status of the Gannet in Scotland in 1994-95 21 total by c 10% and, thus, that the total was around 30,000 AOS. The other survey was made by ourselves from aerial photographs taken on 11 July (Murray & Wanless 1995). The colony was divided into 10 sections (Fig 6). The standard of the slides was so high that, for the majority of the colony, it was possible to separate sites with and without nest material. We, therefore, produced totals in terms of both AOS and AON (Table 9). The average of the 2 counts of AOS indicated a total of 39,751 AOS andan observer variation of 2.1% of the mean, substantially higher than Nelson’s total. One reason for this difference is likely to be that we included more non breeding and loafing birds. Typically, non breeders leave a colony during an aerial survey but, during the 1994 census of the Bass Rock, it was clear that, particularly along the landward edge of the colony Figure 6 Counting divisions used to census the gannetry on the Bass Rock in 1994. SB 19(1) ssajueM S Aq apew sjunog 2 22 S Murray & S Wanless Aeuinw § Aq apew sjunop | SaJON v8 68S1L2 LGZ6E 66S0r y'88 OO68E L6EVE JE}OL 8ce 69€ 8BZS1l EZSl v'l6 €8Sl LZvvl OL 9E1 LZES LG9CL GOLEL 206 Z61LcL 89011 6 VE LOL? O€9E B8rSE L'L6 oLLE cB8EE 8 88 G8lLv G88Z cL6Z L°98 6622 BLZ9 iE: 9€ O9SI 6LLe 6ccc vy 62 6002 v6Sl 9 9Ov- 6061 920 cSOl L°L9 OOO LZ9 G Lo GOS ELD LE9 v' 16 88S BES 74 OOl 9EVE 2989 L9cL 8°88 cLv9 6vZS € cOl OcE Gv9 LE9 8°v8 oG9 €SS ces eck Lech Lele G8Sc 1°26 888¢ ZL9¢@ L v66} ul S861 v66 1 eBueyo % (Says paidna90) (Says paidnoo0) (sais peidnos0) (Says peidnoso) (sjsau) ebeiany abeiony 2junod SISON % jjunoD pUNOD =: UOI}DS ‘9 614 ul UMOYS ase SUOIZaS JUNOD “G86L aUuNL LL pue | pue ‘P66 Ajnr LL uo yO0Y sseg ay] UO Sajis paldndz0 pue sjsau Jo siaquinu Jo sJuNOD 6 a|\qGeL 1997 (sections 7-9), very few birds were disturbed and itwas difficult to separate such individuals from true site holders. However, a second reason for the discrepancy could be that the aerial survey provided better coverage of the colony, thus reducing the chances that sections were missed. The nest count indicated a total of 34,397 AON (Table 9). Comparing this figure with the appropriate count of AOS suggested that 88.4% of AOS were in fact AON. This value should be regarded as amaximum because, while there were few problems distinguishing nests in cliff face sections (sections 2-6), itwas harder to identify nests in some of the cliff top areas (sections 1, 7-10). In 1984, Nelson (in Murray & Wanless 1986) estimated that there were 18,162 AOS on the Bass Rock. Comparing this figure with his 1994 total indicates an increase of 50%, an annual rate of 4.2% pa. Bearing in mind that the 1994 total may have been an underestimate (see above) these changes are likely to be minimum values. Comparison of our 1985 aerial survey in which 21,589 AOS were counted, with the 1994 total, indicates an even more spectacular increase of 84%, an average annual rate of 7.0% pa. Inspection of changes in each of the 10 sections indicated that numbers increased in 9 of them, the only exception being section 5 (Fig 6, Table 9) which is an area of steep, clean rock with comparatively few sites. The most obvious and spectacular expansion of the colony occurred on the northwest slope (section 9) but marked increases in colony extent were also apparent in sections 1, 3, 7 and 10. Comparison of the 1994 nest count with the 1985 site countindicated an increase of at least 59% at an average annual rate of 5.3% pa. The Status of the Gannet in Scotland in 1994-95 23 Total numbers in Scotland The counts documented above were combined to provide an estimate of the total Scottish population in 1994-95. Where there were totals for both years such as on Fair Isle and at Troup Head we took the later count. For the Bass Rock, for which there was considerable variation between the 2 independent counts (see above), we used the nest count from the aerial survey as this was considered to have the lowest error associated with it. The overall total for colonies counted using AOS was 113,177, the total for those counted using AON was 54,230 (Table 10). Combining these figures and, for convenience, expressing the total as the number of AOS, gave a grand total of 167,407 AOS. Numbers were divided very unequally between the 12 colonies with St Kilda, the Bass Rock and Ailsa Craig together holding 76% of the total. In 1984/85 the Scottish population was estimated at 132,100 AOS (Murray & Wanless 1986). Comparing the 1994/95 total with this value indicates that numbers increased by 27% at an average rate of 2.4% pa. Discussion Compared with many other seabird species, Gannets are relatively straightforward to count, being large and conspicuous, and nesting in well defined but not excessively dense groups. Nevertheless, it is inevitable that some of the observed changes in numbers will be due to counting or sampling errors rather than actual changes in abundance and variations in count methods, dates or times between the totals being compared can all contribute to counting errors (Nelson 1978). A further source of bias stems from the tendency of individual observers to count consistently high or low (Harris & Lloyd 1977). 24 S Murray & S Wanless SB 19(1) Table 10 Summary of counts of Scottish gannetries in 1994-95 and changes since the 1984-85 survey. See text for details. The possible colony on Rockall was not checked in either year. Count Colony AON AOS Scar Rocks 1952 Ailsa Craig 32456 St Kilda 60428 Shiant Isles Flannan Isles 1438 Sula Sgeir 10440 Sule Stack 4888 Foula 600 Hermaness 10640 1353 Noss 7310 Fair Isle 975 Troup Head 530 Bass Rock 34397 Total 53852 113555 Grand total 167407 This effect was highlighted in our replicate counts where SW consistently obtained higher totals than SM, even though the counts were made under identical conditions. The mean (+SD) observer variation was 3.3+2.3% (n= 7 comparisons) However, while these potential errors should be borne in mind, we consider that in many cases the error of any colony counts is likely to be below 5% and, in most cases, below 10%. With an inter census interval of approximately 10 years, small, but consistent, annual changes in numbers of 2- 3%, (the rate of increase typical of the Gannet (Nelson 1978)), result in overall numerical % change % Scottish overall pa total 1.2 154 8.8 19.4 42 3.6 36.1 21 2.1 0 abandoned 0.9 545 23.1 6.2 14 1.5 2.9 -17 -2.1 0.4 186 11.1 7.2 25 3.5 4.4 40 3.4 0.6 600 21.6 0.3 colonised 20.5 59 5.3 27 2.4 changes of the order of 22-34% which is well within the accuracy of counts at individual colonies. The results of the 1994/95 east Atlantic Gannet survey show that the sustained increases in both abundance and range which have been recorded this century have continued over the last 10 years (Gurney 1913, Fisher & Vevers 1943, 1944, Cramp et al 1974, Nelson 1978, Wanless 1987). Numbers increased by 23% from 223,400 AOS in 1984/85 to 273,803 AOS in 1994/95, an average annual rate of 2.0% pa (Wanless 1997 1987, this study). The Gannet’s range has extended south with the bizarre record of a pair building a nest on a yacht at Port Frioul in the Mediterranean in 1993 (Fernandez & Bayle 1994) and east into Russia with the establishment of a colony on Kharlov, one of the islands in the Seven Islands Reserve off the Kola Peninsula (Y V Krasnov pers comm). However, Scotland remains the species’ stronghold in the east Atlantic, both in terms of the number of active colonies (35% of the total) and population size (61%) (Table 11). Except for Sule Stack and the nest sites on the Shiant Isles which have been abandoned, counts of every Scottish gannetry were higher in 1994/95 than in 1984/85. There was no evidence that the reduction in Sandeel The Status of the Gannet in Scotland in 1994-95 25 Ammodytes marinus stocks around Shetland during the 1980s (Kunzlik 1989) had any impact on the gannetries there, even though this fish is sometimes an important prey item for Gannets during the breeding season (Martin 1989). There was, however, considerable variation between colonies in average rates of change with values being significantly higher at small colonies compared with large ones (Fig 7). Nelson (1978) calculated that the intrinsic rate of increase for the Gannet on the Bass Rock in the 1960s was about 3% pa. Assuming that current vital rates for the species are similar, this suggests that there is net immigration to colonies containing fewer than about 1,000 AOS. Table 11 Summary of results from the 1994-95 east Atlantic Gannet survey. Country Number of % total colonies Scotland 12 S513 England 1 2.9 Wales 1 2.9 Channel Islands 2 5.9 lreland 5 14.7 France 1 2.9 Norway 5 14.7 Faroes 1 2.9 Iceland 54 14.7 Russia 1 2.9 Total 34 Total % total AOS/AON' 167407 61.1 1631 0.6 26277? 9.6 5478 2.0 29948? 10.9 11628 4.2 3654 1.3 2340 0.9 25437 9.3 3 0.00 273803 Notes: ‘Data for Norway and Iceland are from Barrett & Folkestad (1996) and Gardarsson (1995) respectively. Counts for other areas were provided by RSPB, M G Hill, O Merne, P GH Evans, F Siorat, B Olsen, Y V Krasnov and ourselves Total considered a minimum due to imcomplete photo coverage No count available for Clare Island. Current status is unknown, but 7 to 9 individuals recorded in 1996( O Merne pers comm) 4 Westmanns considered to be one colony 26 S Murray & S Wanless SB 19(1) Figure 7 Relationship between the average annual rate of change in numbers between 1984-85 and 1994-95 and the size of the gannetry in 1984-85. The negative trend indicates that the rates of increase declines with increasing colony size (r= 0.88, n= 10, p<0.0071). =~ fo) 2 ite) @ o 34° =~ ~ "Seige ls 2 @/ GTS 1 ‘ey. Ve) oe << : Sak s Te @ o >> = e@ o~. e@ o ~S on! on ¢ ~s 6 5} e J ro) a 0 A t ; T T T 1 Cc 5 6 8 9 10 11 Colony size 1984/85 (log) At the beginning of the twentieth century, when Gurney (1913) made his first pioneering census, there were only 5 gannetries in Scotiand: Ailsa Craig, St Kilda, Sula Sgeir, Sule Stack and the Bass Rock. The Scottish population was thought to be about 41,300 AOS. As we approach the end of the century, numbers have increased fourfold and there are 12 active colonies. All the evidence suggests that conditions are still extremely favourable for Gannets. However, the species is potentially of conservation concern because of its highly localised breeding distribution with 76% of the Scottish population and almost 50% of the east Atlantic population concentrated in just 3 colonies: St Kilda, the Bass Rock and Ailsa Craig. Acknowledgements This survey was a partnership project and contributions towards the costs were received from the Scottish Ornithologists’ Club, Scottish Natural Heritage, the Joint Nature Conservation Committee, the British Trust for Ornithology, the Seabird Group and the Royal Society for the Protection of Birds. We hope that all the people who provided counts or information have been acknowledged at the appropriate place in the text. We thank Paul Walsh and Kate Thompson for counts from the JNCC Colony Site Register. Mike Harris, Bryan Nelson, Kate Thompson, Mark Tasker, Greg Mudge, Emma Brindley, Paul Harvey, Bob Furness, Rob Barrett, Oscar Merne and Roger Riddington criticised relevant sections of the manuscript. To all those who participated in this survey we offer our sincere thanks. In particular we are grateful for the enthusiastic cooperation and flying skills of Alan Mossman and Rod Brown of Air Alba. Thanks are also due to Stan Kerr, Digger Jackson, Mary Harman, Jim Vaughan, Peter MacGregor, Paul Collin and Bernie Zonfrillo. tare References Barrett R & Folkestad A O 1996. The status of the North Atlantic Gannet Morus bassanus after 50 years in Norway. Seabird 18:30-37. Beatty J 1992. Sula: the seabird-hunters of Lewis. Michael Joseph, London. Belaoussoff S 1993. Northern Gannet and Common Guillemot nesting on Rockall. British Birds 86:16. Bourne W RP 1993. Birds breeding on Rockall. British Birds 86:16-17. Cramp S, Bourne W RP & Saunders S 1974. The Seabirds of Britain and Ireland. Collins, London. Dickson RC 1992. The birds in Wigtownshire. G.C. Book Publishers Ltd, Wigtown. Fernandez O & Bayle P 1994. Tentative insolite de nidification du fou de bassan Sula bassana a Port- Frioul (Marseille, Bouches-du-Rhéne). Alauda 62:140-143. Fisher J & Vevers HG 1943-44. The breeding distribution, history and population of the North Atlantic Gannet Sula bassana. Journal of Animal Ecology 12:173-213; 13:49-62. Gardarsson A 1995. Fjdldi sulk vid Islandu 1989- 1994. Natturufraeoingurinn 64:203-208. Gibson J A_ 1951. The breeding distribution population and history of the birds of Ailsa Craig Scottish Naturalist 63: 73-100,159-177. Gurney JH 1913. The Gannet. A Bird with a History. Witherby, London. Harris MP & Lloyd C S 1977. Variations in counts of seabirds from photographs. British Birds 70: 200-205. Kunzlik P A 1989. Small fish around Shetland. /n Seabirds and Sandeels: Proceedings of a seminar held in Lerwick, Shetland, 15-16 October 1988, (Ed) Heubeck M pp.38-49. Shetland Bird Club, Lerwick. Martin A R 1989. The diet of Atlantic Puffin Fratercula arctica and Northern Gannet Sula bassana chicks at a Shetland colony during a The Status of the Gannet in Scotland in 1994-95 27 period of changing prey availability. Bird Study 36:1 170-180. Matthews J & North S 1989. Gannets breeding on mainland Scotland. Scottish Birds 15:132-133. Murray S 1992. Acount of the Hermaness gannetry in 1991. Joint Nature Conservation Committee Report No.49. Murray S & Wanless S 1986. The status of the Gannet in Scotland 1984-85. Scottish Birds 14: 74- 85. Murray S & Wanless S 1992. A count of the Noss gannetry in 1992 and analysis of Gannet monitoring plots on Noss NNR 1975-91. Joint Nature Conservation Committee Report No.50. Murray S & Wanless S 1994. Census of the Sula Sgeir gannetry in July 1994. Unpublished report to SNH. Murray S & Wanless S 1995. Census of the Bass Rock gannetry in July 1994. Unpublished report to SNH southeast region Scotland. Murray S & Wanless S 1996. A census of the St Kilda gannetry in May 1994. Scottish Birds 18:152- 158. Nelson JB 1978. The Gannet. T & A D Poyser, Berkhamsted. Riddiford N 1993. Recent changes in Fair Isle seabird populations. Seabird 15:60-67. Wanless S 1979. Aspects of population dynamics and breeding ecology in the Gannet (Sula bassana (L.)) of Ailsa Craig. Unpublished PhD thesis, University of Aberdeen. Wanless S 1987. A survey of the numbers and breeding distribution of the NorthAtlantic Gannet Sula bassana and an assessment of the changes which have occurred since Operation Seafarer 1969/70. Research and Survey in Nature Conservation No.4, Nature Conservancy Council, Peterborough. Wanless S, Matthews J & Bourne W RP 1996. The Troup Head Gannetry. Scottish Birds 18:214-221. S Murray, Craigie Dhu, Cardney, Dunkeld, Perthshire PH8 OEY S Wanless, Institute of Terrestrial Ecology, Hill of Brathens, Banchory, Kincardineshire AB3 4BY Revised manuscript accepted July 1996 28 Scottish Birds (1997) 18:28-35 SB 19(1) Gizzard contents of Pochard, Tufted Duck and Goldeneye from Loch Leven, Kinross, in winter 1994-95 BARRY STEWART & ALAN LAUDER We examined the gizzard contents of 3 species of diving duck which fed on the benthos of Loch Leven during winter 1994-95. Tufted Ducks and Goldeneye appeared to be mainly carnivorous and Pochard mainly herbivorous. Chironomid larvae appear to be the most important food item overall. The results are discussed in relation to previous studies. Introduction Loch Leven National Nature Reserve is internationally important for its wintering wildfowl populations (Owen et a/ 1986) and 3 diving duck species occur there in nationally important numbers - Pochard Aythya ferina, Tufted Duck A fuligula and Goldeneye Bucephala clangula (Waters & Cranswick 1993). The diet of Tufted Duck and Goldeneye is thought to be primarily Chironomid larvae and other invertebrates. Pochard are thought to be mainly herbivorous (Cramp & Simmons 1977) although some studies have shown a predominance of Chironomid larvae in their diet (Winfield & Winfield 1994). The only previous study of diving duck diet at Loch Leven suggested that both Tufted Ducks and Goldeneye fed mainly on Chironomids (Laughlin 1973). Eutrophication and associated algal blooms have become an important factor in determining aquatic plant growth in Loch Leven. The fluctuating water chemistry has also affected benthic invertebrates (eg Chironomids). Macrophyte and benthic invertebrate communities have been the subject of recent monitoring work on the loch (Murphy & Mulligan 1993, Gunn & Kirika 1994) though their importance to the diving duck populations has not been evaluated since work for the International Biological Programme (IBP) in the 1960s and 1970s. Methods Samples from birds’ proventriculus and gizzards were collected during winter 1994/ 95 from freshly shot birds at Loch Leven. After labelling, each was frozen and stored before subsequent analysis. However, many samples refer to gizzard contents only. Consequently, because softer food, such as invertebrates, is often fully digested higher in the gut, analyses of gizzard contents bias estimates towards plant matter and seeds (Swanson & Bartonek 1970). As the resulting small samples are inappropriate for statistical analysis, age and sex classes of the birds have been amalgamated. Identification and nomenclature for vascular plants follows Clapham, Tutin & Moore (1987), Chironomidae that of Bryce & Hobart (1972) and Bird (undated), Mollusca Macan & Cooper (1977) and other invertebrates Fitter & Manuel (1986). 1997 Two estimates of gizzard contents are presented: 1 The proportion of birds with >5% of each food item in the gizzard (% occurrence). 2 The aggregate percentage cry weight (Swanson et a/ 1974). This is calculated by averaging the proportion of each food item per gizzard inthe number of gizzards sampled. This gives equal weight in the analysis to each bird, and thus reduces bias, which may be introduced by individual birds feeding on large amounts of one food item or by variable amounts of total gizzard contents. Results Unidentified plant material was found in c80% of Tufted Duck and Goldeneye gizzards, and in all Pochard gizzards (Fig 1). However, Pochard gizzards contained nearly twice the amount of plant matter anda higher proportion of seed than the other 2 duck species (Fig 1, Table 1). Polygonum, Potamogetonand Carex were the dominant seeds found. Tufted Duck and Goldeneye gizzards contained mainly animal matter, mostly Chironomids and unidentified arthropods. Pochard gizzards also contained notable amounts of Chironomids, but Bryozoans were present in much higher quantities than in the other 2 species. Small amounts of leeches were present in all 3 species. Tufted Duck and Goldeneye had both fed on Trichopteraiarvae while fish were only taken by Goldeneye. Discussion Gizzard contents Gizzard analysis of birds shot apparently at random cannot be used to provide a quantitative estimate of diet, due mainly to the time lag between the intake of food and Gizzard contents of Pochard, Tufted Duck & Goldeneye L Leven, Kinross 29 subsequentanalysis of contents. This includes pre-mortem factors, such as the time delay between the cessation of feeding activity and shooting, and post-mortem factors, such as the differential digestion of soft versus harder food items. In addition, it does not necessarily follow that birds shot at a site have been feeding there. A study by Winfield & Winfield (1994) on Lough Neagh showed that only 20% of shot birds had food in the oesophagus or proventriculus, whereas 85% of birds sampled from accidental netting incidents had food there. Swanson & Bartonek (1970) showed that analysis of gizzard contents alone produced a bias toward plant material, with animals such as molluscs, insects and crustaceans almost totally digested after 20 minutes in the gizzard and with digestive juices continuing to digestfood forlong periods after death. Because gizzard contents in this study were not obtained immediately following a bird’s death, only a small amount of the gizzard contents could be identified accurately, especially in the case of Chironomids, which were mostly identified when partially digested or fragmented. Even determining family status proved difficult in many cases. Pochard Pochard feed by day and night, preferably at 1m-2.5m depth with dives lasting 13-30 secs. (Owen et a/ 1986, Willi 1970). Winfield & Winfield (1994), however, recorded birds feeding at up to 10m depth in Lough Neagh where Chironomids formed 90% of the diet. In the absence of benthic macrophytes, Pochard have also been found to feed selectively in areas rich in Chironomid larvae (Phillips 1991). These findings contradict other studies which describe the winter diet of Pochard as mainly 30 B Stewart & A Lauder SB 19 (1) Figure 1 Gizzard contents of Tufted Duck, Pochard and Goldeneye at Loch Leven, winter 1994-1995. > ag a 38 z < wi = 8 3 2 £ = @ ro) © aad @ oO f°) a ” 8 AY + (3) e $ 8 i= 2 100 uw 80 S) rT MW ow 90 ac el afl laa oO s 20 | [ a Seeds Bryozoans Unidentified Plants Leech Cocoons plants, chiefly the oospores of stoneworts, Charaand Nitella, and the seeds of submerged aquatic plants, especially Potamogeton (Owen et al 1986). Chironomids and other invertebrates are generally reported to comprise less than 15% of the diet (Olney 1968). However, the erroneous belief that Tutted Duck Po on ard Trichoptera Chironomids £ A} re Trichoptera Chironomids Unid’d Arthropods [TT Pochard are herbivorous presumably derived from earlier studies which analysed only gizzard samples. Whilst also based on gizzard analyses, this study suggests that Pochard are omnivorous at Loch Leven, taking significant proportions 1997 of both plant and animal material (mainly Chironomids and Bryozoans). A previous study at Loch Leven analysed one ‘stomach’ and also found that Chironomid larvae were taken (Laughlin 1973). We do not know whether the high levels of Bryozoans (Cristatella mucedo statoblasts) identified in Pochard during our current study were taken on purpose or were consumed inadvertently from benthic vegetation. The relatively higher proportion of plant matter in Pochard gizzards in comparison with the other 2 duck species may suggest the latter possibility. In areas where Pochard and other species feed primarily on Chironomids, they may compete with a number of fish species, such as Brown Trout Sa/mo trutta, Rainbow Trout Onchorynchus mykiss and Perch Perca fluviatilis which occur in Loch Leven and Eels Anguilla anguilla (Winfield & Winfield 1994) although the latter are thought to be scarce in the loch. Chironomids are eaten by all these species. Thorpe (1973) estimated that Brown Trout consumed no more than 6% of the Chironomid production between June and September. This level is unlikely to have increased because a recent estimate of the fish population (O’Grady et a/ 1993) indicated that the Brown Trout population has declined since Thorpe’s study and the Perch population has subsequently decreased dramatically. Annual stocking of Rainbow Trout since 1993 has done little to redress the balance in terms of fish biomass and we assume that fish predation on Chironomids, leading to competition with birds forfood, has decreased. The results of this survey indicate that, while Pochard ate fewer Chironomids and more vegetable matter than the other 2 species, Chironomids still formed a large part of their diet. Gizzard contents of Pochard, Tufted Duck & Goldeneye L Leven, Kinross 31 Tufted Duck Witherby et a/ (1940), Cramp & Simmons (1977) and Owen etal (1986) all describe this species aS omnivorous, but emphasise a preference for animal matter, especially molluscs, which usually accounts for over 80% of the diet. Between 300 and 600 pairs of Tufted Ducks are known to nest annually at Loch Leven (Lauder 1993) and Laughlin (1974) states that Chironomids formed 60% of the diet during this period. Other important items also taken then included caddis larvae and molluscs, especially Va/vata spp. Thus it appears that Tufted Ducks at Loch Leven mainly eat Chironomid larvae. When looking at the energy flow through the loch, Laughlin found that water depth and depth of food in the substrate were 2 of the major factors affecting the availability of food for Tufted Ducks and that only 20% of the total loch area was suitable for feeding birds. Laughlin (1974) also suggested that the autumnal peaks of Tufted Ducks at Loch Levren (typically c2,500 birds, but sometimes over 4,000 birds) were correlated with Chironomid production. The high numbers of Tufted Duck present in 1970, coinciding with peak biomass values for G/lyptotendipes (Maitland & Hudspith 1974), provided evidence for this, but a wide range of other factors clearly affect Chironomid and waterfowl populations. Goldeneye Goldeneye are more solitary and usually feed at 3-10m but up to 55m (Cramp & Simmons 1977, Winfield & Winfield 1994) mostly in daylight but sometimes at night (Cramp & Simmons 1977). Goldeneye are known to 32 B Stewart & A Lauder SB 19 (1) Table 1 Aggregate percentage dry weights of 9 Pochard, 11 Tufted Duck and 18 Goldeneye gizzards. Tufted Pochard Goldeneye Duck Plant material n=11 n=9 n=18 ALGAE Chlorophyta (filamentous) 0.36 0.33 BRYOPHYTA Unidentified plant material 22.45 49.11 25.61 ANGIOSPERMOPHYTA Dicotyledones Unidentified (seed) 4.55 1.67 0.33 Ranunculaceae (seed) 0.33 Leguminaceae Trifolium sp (seed) 0.06 Rosaceae Rubus fruiticosus (seed) Polygonaceae Polygonum sp (seed) 0.55 2.44 1.39 Rumex sp (seed) 0.44 0.11 Betulaceae Betula sp (seed) 0.09 Compositae Cirsium sp (seed) 0.09 0.33 Monocotyledones Potamogetonaceae Potamogeton sp (seed) 4.27 1.00 1.67 Cyperaceae Carex sp (seed) 1.00 Juncaceae Juncus sp (seed) 0.09 0.11 Gramineae (seed) 0.55 0.22 0.11 Anther unidentified 0.22 TOTAL PLANT MATERIAL 32.64 55.87 30.28 ANIMAL MATERIAL PORIFERA Spongillidae 0.22 0.56 PLATYHELMINTHES Cestoda 0.06 NEMATODA 0.06 1997 Gizzard contents of Pochard, Tufted Duck & Goldeneye L Leven, Kinross 33 ACANTHOCEPHALA 0.18 ANNELIDA : Hirudinea Erpobdella octoculata 0.91 2.33 2.44 BRYOZOA Phylactolaemata (statoblasts) Cristatella mucedo 2.82 22.00 1.89 ARTHROPODA Arachnida Hydracarina Arrhenurus caudatus & soft bodied spp 0.36 0.33 0.11 Crustacea Ostracoda 0.09 0.06 lsopoda Asellus aquaticus 0.28 Insecta Odonata Anisoptera larvae 0.06 Trichoptera Rhyacophilidae larvae 3.78 3.78 Phthiraptera 0.01 Diptera Chironomidae Tanypodinae larvae 6.89 1.11 3.22 Chironominae larvae 5.82 15.33 8.61 Chironomid mucus tubes 25.55 2.22 26.61 Unidentified larvae 0.06 Unidentified imago 0.22 Hymenoptera Evaniidae 0.09 | Braconidae 0.11 Unidentified chitinous fragments 23.73 0.56 14.17 Unidentified ova 1.00 0.11 0.11 MOLLUSCA Gastropoda Unidentified opercula 0.18 0.06 Valvata sp 0.11 0.72 Planorbis sp 0.09 Bivalvia Pisidium sp 0.09 0.06 VERTEBRATA Osteichthyes : Gasterosteiformes | Gasterosteus aculeatus 1.11 34 B Stewart & A Lauder feed actively at night at Loch Leven in illuminated water near Kinross. When feeding in flocks, birds may synchronise dives, or dive in succession with dives lasting 30-40 secs. Our results suggest that Tufted Ducks and Goldeneye at Loch Leven were mainly carnivorous but Pochard were omnivorous. Laughlin (173) obtained similar results for the former 2 species at Loch Leven, but only analysed one Pochard ‘stomach’. The range of items consumed appears to be similar for all 3 species with Chironomid larvae appearing as a major resource for each. The high percentages of plant material recorded in Tufted Ducks and Goldeneye are likely to be a result of the differential periods of digestion outlined by Swanson & Bartonek (1974) and it is possible that most of this material had been ingested incidentally or is residual. This suggestion is supported by previous analyses of diets of Tufted Duck and Goldeneye (Eriksson 1979, Laughlin 1973, Olney & Mills 1963 and Cramp & Simmons 1977), all of which reveal little evidence of plant material as a major dietary component. Chironomid productivity and diving duck numbers Laughlin (1974) suggested that winter Chironomid production was a limiting factor for the number of Tufted Duck occurring on the loch. In May 1994, Gunn & Kirika (1994) found similarly high densities of Chironomid larvae. Tanypodinae (Procladius spp) were found mainly in muddy substrates, with a mean density of 4,106 m~2, whereas Chironominae were found mainly in sandy substates at the following densities: Cladotanytarsus spp 11,086 m2 and Stictochironomus spp.2,818 m2. One genus of Chironominae, Tanytarsus spp, was also SB 19 (1) found in reasonably high densities in mud (1,219 m-2). However, the dominance of the Chironomus group in the gizzards and the frequenty of the cocoons of Erpobdella octoculata, a leech typically found on stony or sandy substrates, does suggest that most feeding occurred over sand substrates. There have been changes in the relative abundance of the constituent genera of the Chironomid community in Loch Leven since the IBP programme in the 1960/70s and producitvity has been variable in what appears to be an unstable system as a result of nutrient enrichment (Gunn & Kirika 1994). Chironomid larvae appear to be the major food resource of diving ducks at Loch Leven and the population levels of benthic feeding waterfowl may well be determined by the productivity of Chironomidae in the loch. Acknowledgments We are most grateful to Kinross Estate for allowing access to the birds for examination. We thank Paul Llewellyn for the use of his laboratory equipment and several members of WWT staff notably Baz Hughes, Carl Mitchell and Geoff Proffitt, who offered much advice and assistance. References Bird L M (undated). The identification of Chironomidae, unpublished. Bryce D & Hobart A 1972. The biology and identification of the larvae of the Chironomidae (Diptera), Entomologists Gazette, Volume 23. Clapham A R, Tutin T G & Moore D M 1987. Flora ofthe British Isles, 3rd Edition, Cambirdge University Press. Cramp S & Simmons K E L 1977. The Birds of the Western Palearctic, Volume 1, Oxford University Press. Eriksson M O G 1979. Competition between 1997 Gizzard contents of Pochard, Tufted Duck & Goldeneye L Leven, Kinross 35 freshwater fish and Goldeneyes Bucephala clangula for common prey, Oecologia (Berl.) 41:99-107. Fitter R & Manuel R 1986. Collins field guide to freshwater life, Collins. Gunn! DM & Kirika A 1994. Benthic Invertebrate Survey - Loch Leven NNR, Report to Scottish Natural Heritage, Institute of Freshwater Ecology. Lauder A W 1993. The 1993 duck breeding census of St. Serf’s Island, Loch Leven National Nature Reserve. SNH Internal Report. Laughlin K F 1973. Bioenergetics of the Tufted Duck Aythya fuliguia. PhD. Thesis, University of Stirling. Laughlin K F 1974. Bioenergetics of Tufted Duck Aythya fuligula at Loch Leven, Kinross, Proc. Roy. Soc. Edinburgh. 74:383-389. Murphy K J & Milligan A 1993. Submerged Macrophytes at Loch Leven, Kinross, Report to Scottish Natural Heritage, CREST, University of Glasgow. O’Grady M F et a/ 1993. A fish stock survey of Loch Levenand management proposals for this resourve as atrout angling fishery. Report to Scottish Natural Heritage. Central Fisheries Board, Dublin. Olney PJS & Mills DH 1963. The food and feeding habits of Goldeneye Bucephala clangula in Great Britain. Ibis 105:293-300. Olney P J S 1968. The food and feeding habits of Pochard. Biological Conservation. 1:71-76. Owen M, Atkinson-Willes G L & Salmon D G 1986. Wildfowl in Great BritainSecond Edition, Cambridge University Press. Phillips V E 1991. Pochard Aythya ferina use of Chironomid rich feeding habitat in winter Bird Study 38: 118-122. Swanson GA & Bartonek J C 1970. Bias associated with food analysis in gizzards of Blue-winged Teal, J Wildlife Management 32:739-746. Swanson G A, Krapu GL, Bartonek J C, Serie J R & Johnson DH 1974. Advantages of mathematically weighing waterfowl food habits data. J Wildlife Management 38(2):302-307. Thorpe J E 1973. Trout and Perch populations at Loch Leven, Kinross. Proc. Roy.Soc. Edinburgh (B) 74, 20:295-313. Waters R J & Cranswick P A 1993. The Wetland Bird Survey 1992-1993, Wildfowl and Wader Counts, BTO/WWT/RSPB/JNCC. Willi P 1970. Zugverhalten, Aktivitat, Nahrung und Nahrungschwerb auf dom Klingrauer Strausse haufig auftrender Anatiden, insbesondere von Krickente, Tafelente und Reiherente, Orn. Beob. 67:141-217. Winfield | J & Winfield D K 1994. Feeding ecology of the diving ducks Pochard Aythya ferina, Tufted Duck A. fuligula, Scaup A. marila and Goldeneye Bucephala clangulaoverwintering on Lough Neagh, Northern Ireland. Freshwater Biology 32:467-477. Witherby H F, Jourdain F C R, Ticehurst N F & Tucker B W 1939. The Handbook of British Birds, London. Barry Stewart, The Wildfowl & Wetlands Trust, Slimbridge, Gloucester GL2 7BT Alan Lauder, Scottish Natural Heritage, Loch Leven Laboratory, The Pier, Kinross KY13 7UF Revised manuscript accepted July 1996 36 Scottish Birds (1997) 18:36-54 SB 19 (1) Waterfowl counts on the Tay Estuary, 1985-1995 N ELKINS & B M LYNCH The Tay Estuary is an important site for both migrating and wintering waterfowl. It holds internationally important populations of wintering Eider, Redshank, and Bar-tailed Godwit. The wintering populations of Sanderling and Goldeneye achieve national importance. The estuary is a staging post for migrant waders in spring and autumn, and recent counts of migrant Sanderling, Knot, Ringed Plover and Turnstone have exceeded wintering populations, especially in spring. Mallard, Wigeon and Tufted Duck are all common wildfowl in winter, while Shelduck reach a spring peak. A substantial moulting flock of Goosander is present in late summer. This paper describes the numbers, distribution and changes in waterfowl populations on the estuary between 1985 and 1995, as shown by monthly high tide roost counts and a series of midwinter low tide counts. Introduction The Firth of Tay is the third largest tidal estuary in eastern Scotland, after the Moray Firth complex and the Firth of Forth. However, unlike the other firths, little detail has been published on its waterfowl populations. Any accounts available in local reports have concentrated on one area or another but not the whole, probably because the estuary represents the boundaries between Angus, Perthshire and Fife. With the continuous monitoring carried out over the last decade or so, there is now a clearer picture of the importance of the estuary, and this paper sets out to describe its waterfowl populations, particularly the waders. The study area The estuary has an intertidal area of 5,720 ha with a tidal reach of 53km. Its contributing rivers, the Tay and Earn, provide the greatest inflow of fresh water of any British estuary (JNCC 1994). The maximum width of the inner estuary is 5km, narrowing to 1.5km between Broughty Ferry and Tayport. The constriction is mainly due to the volcanic outliers of the Ochil Hills, which extend all the way along the southern edge of the estuary to cross into Angus at this point. The northern shore of the inner estuary is backed by a coastal plain some 8km wide with the Sidlaw Hills to the north. The tidal range at Dundee is large, being 5.2m at the highest spring tides, and 1.7m at the minimum. At low tide, the inner estuary is characterised by extensive mudflats on the northern shore, fringed by large salt marshes, with the most extensive continuous stand of Phragmites reed swamp in Britain. The inner estuarine substrates consist of coarse sediments in mid estuary and fine mud sediments along the north shore, both of 1997 which, because of their unstable nature and the large tidal range of salinities, support an impoverished variety of invertebrate fauna (JNCC 1994). However, these populations can be enormous and, from the distribution of feeding waders, we suspect that this fauna is more abundant at the eastern end. The outer estuary holds sandier intertidal flats both on the north and south shores, with considerable sublittoral populations of mussels Mytilus edulis, and several mussel beds exposed at low tide. On both shores adjacent to the estuary mouth there are extensive accreting sand dune systems. A natural division exists between the inner and outer estuaries where there is a 7km stretch of little or no intertidal flats. The adjacent shores here are mainly urban, where the rail and road bridges span the estuary. In all, the sand and mudflats cover 5,218 ha. The main high tide wader roosts are situated at Tentsmuir Point and Lucky Scalp on the south shore, and the Monifieth/Barry Buddon shore and Kingoodie/Invergowrie Bays on the north shore (Fig 1). Tentsmuir Point is a sandy promontory where roosting birds use the shoreline. Lucky Scalp is a narrow high tide islet consisting of a sand/pebble stretch covered in rough grassland and marram Ammophila arenaria, and surrounded by a pebble shore which is all but covered by the highest tides. It measures 30m in length, and between 5m and 15m in width, and lies 600m from the coastal high water mark. The main roost at Monifieth is usually on the sand spit at the mouth of the Buddon Burn, but occasionally it stretches along the sandy shore towards Buddon Ness. Due to a very busy military live firing range on the Buddon Ness promontory, eastern most roosts have, at times, proved difficult to count with only estimates obtained. There is also a satellite roost near the mouth of the Dighty Water at Waterfowl counts on the Tay Estuary, 1985-1995 37 Barnhill. The only inner estuary roosts are situated at Kingoodie and Invergowrie Bay. Here, a proportion of waders roost at Dundee airport. Subsidiary roosts can also be found on fields, including playing fields, around Tayport and Dundee, where some species feed during high tide. Methods As part of the British Trust for Ornithology’s (BTO) Wetlands Bird Survey (WeBS) (prior to autumn 1993 this was called the Birds of Estuaries Enquiry (BoEE), monthly counts of roosting waders were carried out every winter. Counts during the remainder of the year were made from 1992. During winter 1993-94, additional regular monthly counts were also made of both roosting and feeding waders at iow tide for the BTO’s Low Tide counts scheme. This study uses all counts made for WeBS between September 1985 and August 1995. In addition to waders, wildfowl were also counted ona less regular and comprehensive basis, together with some other estuarine species. High tide roost counts were made ona specified date each month between September and March inclusive, usually at spring tides near the middle of the month, although during short winter days when high tide occurred during the hours of darkness, alternative dates were chosen. The median date for all complete coordinated counts was the 14th. Onsome occasions, severe weather or lack of observer(s) disrupted the regularity and simultaneity of counts. From 1992, similar counts were also made from April to August. Monthly low tide counts were made between November 1993 and February 1994 on a chosen date. In February 1994, the date of 38 N Elkins & BM Lynch SB 19 (1) N ANGUS if Monifieth Barry Broughty soul a Links ; Dundee Ferry L-1- ~teN Invergowrie River Tay Kingoodie Airport Roa Br Lucky ag Buddon 7° tsmuirNess =. Taypon on Sse" sail \Br ue int VESoow—s’ Po, AC ee? NS ala 2 we ecm Ca 7 ee RIVER Son i zee § TAY LS ; H — Kinshaldy aa f Sole 3 St eee ‘ i Andrews Mugdrum ene FIFE j Bay Island o—.-7" : AG ie Leuchars 2 er, or “s 5 km Eder«¢ \ Newburgh SSS Say 4 e ° . Figure 1 Map of the Firth of Tay showing mean low water mark (dotted) Main roosts X Subsidiary roosts O. the low tide count fell on the same day as the high tide count, so that a unique direct comparison was available. With only 4 main high tide roosts, the logistics of providing simultaneous cover were simple. The low tide counts required more organisation; the estuary was divided into a total of 41 sections, with approximately 20 observers involved on each date. The monthly figures (Table 1) are the 10 year means of all available counts for each roost site. For comparison, casual counts between 1970 and 1985 (for which tidal state was invariably not specified) were extracted from national and local Bird Reports, while BoEE counts were also scrutinised. For the latter, counts of the whole estuary, let alone coordinated counts, were rarely achieved during this earlier period. For this reason, long term trends in population sizes were not possible to detect for many species. Boase (1970) provided useful wildfowl data for the period prior to 1970. A few observations from the 1995-96 winter counts have also been included in cases where the data show new information. The prevailing wind and weather conditions were noted on each count date. This information was supplemented by weather data from Leuchars, which was used to calculate mean values of temperature and wind speed, the atmospheric parameters assumed to be of most importance to shorebirds. These data were combined to produce a wind chill equivalent temperature for each winter. This is a partly subjective measurement calculated for suitably clothed humans! However, it is considered that the trend of these values may also be meaningful for a healthy bird insulated by plumage. 1997 Results Between winter 1985-86 and 1994-95, complete and fully coordinated roost counts were made in 48 (69%) of a possible maximum of 70 of the months between September and March inclusive. Inthe remainder, most counts were carried out within a few days of each other. The average number of waders roosting on the Tay estuary in midwinter was 10,842. This figure was derived from summing the highest monthly counts of each species between November and February. This wintering population varied from 12,789 in the winter of 1985-6 to 6,797 in 1994-95. The variation between the same months in different years was also marked. For instance, the highest count was made in September 1989, when migrants swelled the population to 13,186, but a year later only 3,960 birds figured in the September count. In winter, Dunlin comprised 30% of the mean total and Oystercatcher 20%, while Bar-tailed Godwit and Redshank accounted for 14% and 13% respectively. Fewest waders were present in May, witha mean of 619. The 2 winters of greatest wind chill were 1985-86 and 1993-94, coincident with high mean wind speeds and low mean temperatures. It is perhaps indicative of the relatively equable climate during the study period that no significant local effects on wintering populations were noted, although individual counts were affected during easterly gales accompanied by precipitation. Winters outside the study period, namely 1981-82 and 1995-96, have been severe enough at times to allow extensive ice to build up on the mudflats in the inner estuary. What effect this had on wader populations is unclear, although low wader numbers in the latter winter were second only to those of 1994-95. Waterfowl counts of the Tay Estuary, 1985-1995 39 Species accounts Oystercatcher Haematopus ostralegus Table 1 shows the rapid increase in the Tay estuary population in July, a steady over wintering population of around 1,700 with a peak in February, anda substantial exodus in March. Of the total winter roosting population, 30- 40% roost in the inner estuary between Dundee airport and Kingoodie. The remainder is found in the 3 outer estuary roosts, with apparent movement between the roosts dependent on tide conditions. The size of the roost on Lucky Scalp is highly correlated with the tidal height, with displaced birds moving to one or other of the outer estuary roosts. At high tide, small numbers sometimes feed on agricultural and sports fields surrounding Tayport, and these are included in the counts. In the low tide and high tide coordinated counts on 13 February 1994, the roosting population was 20% less than that feeding earlier in the day. This may have been due either to the birds moving out of the estuary completely, or to their movement into uncounted areas such as Buddon Ness. The largest feeding concentration was off Barnhill. Although low tide counts in the other winter months were separated from high tide counts by a week, similar reductions were recorded. Most of these decreases seemed to be confined to the outer estuary. The low tide counts in winter 1993-94 showed that the majority of birds chose to feed near the low tide mark in the outer estuary, utilising the mussel beds, and that the mean density of such birds was the highest of any of the wader species (Cranswick eft a/ 1995). Numbers using the mussel beds varied from 17% to 58% of the total population. Those birds not SB 19 (1 40 N Elkins & BM Lynch *pajunod seed Jo Jaquunu = U Ayeniqo 0} JOQquaAON ‘JUNO WNnWIxewW UBS = yeeg Ye 0} Jaquuaidas ‘Auenjse sauu! UO Hul}soo! % = % OL OL OL OL 174 v v v v Ol OL OL u QZ 0 6V Be Le O€ Sl lL 0 L cl ce vv LV Suo}SUIN | 89El vl Ol8 vecolt cSEL 6cSt 669 6ZE LS el 68Z 6¢c6 O82 vc8 yueyspesy Lvv Lv L6L 961 GSI GLE G8 80¢ 69 Go 98t OcE 6c 82 Me|IND c8r lt G LZ6 608 L66 L98 VLV €6E 6c A G6 €6€ vZOlL CEEOL MPO pojlej-ieg O€ce LZ S866 vest B8e9t IB 8cl 0c € Lob Zvyt S99 602 SILLE uljunq Gee 0 Ove 88 vol c6 €9 L 0 6 08 Let 8b OZ1 Bulepues v6E 0 Sel c6 v9 8cl LS 0 0 0 8 Elk LOL 869OZL JOU vs9 88 C9E 6SS OL2 8¢8 68S L6 L9 Lo ve 9S cOc = BLE Buimde €9¢ 0 LOL Srl ev 002 O91 vl 9 Le Z6t 691 cSt cdl JBAO|d Adi5 L8c ev Z6 Lhe 9S c6 el 0 0 0 0 OL €6 LEL JBAO0|d USP|O5 Gok 0 98 86 80L GZIL cOl GZ € £4 Le 6c v8 c8 JBAO|q pebuly EGk¢ ce EZ9OL SELL cSSlL ccvl c6cl c8cl LBP 96¢ 6LE tpl S88ZL B8SZL 1849}2918]SAO yeed % ID9G AON WO des Bbny ine une Aew idy Jew qe4 uer “666 1snbny oO] Sg6L 4aquiajdas ‘Asenjsa Ael ay] uo suojej;ndod Guljsoos sapem jo sueawl Ajy,uoyy |. B|GeL 1997 recorded on mussel beds tended also to favour the lower shore, especially in the inner estuary. As at high tide, small numbers also chose to feed in adjacent fields. As with all other waders, Oystercatchers do not feed in any great numbers on the extensive mudflats upriver from Kingoodie, where there is a scarcity of food. From mid February, small numbers can be seen upriver towards Newburgh, as they begin to move towards inland breeding sites. Between May and August, the Oystercatcher is the commonest species on the estuary, comprising up to 60% of all waders in May and June. The midsummer flocks probably consist of immature birds and non breeding adults, while post breeding birds arrive from July onwards (Cramp & Simmons 1983). In late summer, when the population is swollen by migrants, a higher proportion roosts in the inner estuary, possibly because of disturbance by holiday makers along the sandy beaches of the outer estuary (Fig 2). The peak count during the period occurred on 8 September 1991, when 3,124 birds were present. September is the month in which the species peaks over the whole of Britain (Prater 1981). Of this Tay count, only 19% roosted in the inner estuary, and 64% were present at Tentsmuir Point. Compared to winter months, this appeared to be an abnormal distribution, doubtless caused by the presence of a large number of passage migrants. Such abnormal _ distributions also occur in winter from time to | time, possible due to disturbance, especially when birds roosting on or near the airport are subject to clearance. There was no evidence of a significant change in the wintering population, although a comparison with figures for the inner estuary Waterfowl counts of the Tay Estuary, 1985-1995 41 (Laing & Taylor 1993) suggests a decrease for this area. Average maximum counts for both autumn passage and winter rose steadily here between the early 1970s and the late 1980s, but the early 1990s showed a downturn of 10-20%. This apparent decrease is supported by the fact that, elsewhere on the estuary, some previous counts far exceeded present numbers. Peak high tide counts prior to 1985 have exceeded 3,000 in all months from Augustto December, while other counts, at unspecified tidal states (but probably at high tide), have been as high as 7,500 at Tentsmuir Point on 14 October 1973. The Oystercatcher feeds readily inland, and studies have shown that snow and frost force inland feeders to unfrozen intertidal zones, while flooding due to high rainfall increases the amount of food available in inland fields. The high numbers counted in January and February 1991, and in February 1986, came inmonths of frequent frosts, while low numbers of roosting birds were counted in the exceptionally wet January of 1993. The low counts in November 1992 and 1993 followed very cold Octobers, suggesting that autumn immigrants may not have stopped but carried on southwards, although numbers recovered later in both winters. Ringing indicates that most of the wintering Oystercatchers are from the large Norwegian breeding population (Cramp & Simmons 1983, Prater 1981), with some Scottish breeders. Local August ringed birds (probably migrants) have also been recovered in late spring in the Faroes, and in December in France. One bird ringed in February 1988 in Northern Ireland was recovered on the nearby Eden estuary in December 1991. Both winters were exceptionally mild in the Tay area, indicating that the apparent change in wintering site was not as a result of severe weather. 42 N Elkins & BM Lynch 1200 800 SB 19 (1) Figure 2 Mean monthly high tide roosts counts of Oystercatchers on the Tay estuary 1985-1995 Outer estuary [_] Ringed Plover Charadrius hiaticula This species is chiefly a passage visitor to the Tay estuary (Table 1). During midwinter, a mean of 80-90 is present, with about 90% roosting in the outer estuary, mostly on the north shore. There is a sudden decrease in March. The maximum midwinter count was 197 on 14 January 1990. During the 1993-94 low tide counts, the majority of the small population fed in the outer estuary, although small numbers were recorded on the upper shore of Invergowrie Bay. As the numbers roosting invariably exceeded those feeding, it seems that some of the roosting population comes from outwith Innerestuary Mj. the estuary, probably from close by on the uncounted beaches of Buddon Ness. Indeed, marked monthly variations suggest such a movement. For example, only 7 were counted at high tide in January 1986, but the following month the roosting flock rose to 164. Both these months were characterised by high wind chill, perhaps making flocks more mobile in search of shelter. Autumn passage begins in July, reaching a peak in September. The peak high tide count occurred on 11 September 1988, when 315 were present, all on the outer estuary. The small May peak represents those recorded during the regular mid monthly counts. It does not, therefore, include late migrants which 1997 _ pass through in the last 10 days of May. The _ rapid movement of such birds means that numbers may temporarily be high and, indeed, | flock size on the outer estuary has reached 700. The mean of 8 counts between 1971 and | 1989 in the last fortnight of May was 317, with ' most flocks resting at Tentsmuir Point. A few pairs of Ringed Plovers breed among _ the dunes at the outer edge of the estuary. _ Ringed Plovers are not abundant in Britain in winter. The population consists primarily of local breeders (Prater 1981) which represent _ the most northerly wintering population of the _ species (Cramp & Simmons 1983), and which leave their wintering grounds early. Migrants come chiefly from Iceland and Greenland. _ Those passing through in May are probably _from the latter, since these are the latest to nest. Ringing returns show that migrants _passing through east Fife may take both _ western and eastern coast routes through Britain, although there is some degree of loop migration. An autumn ringed bird was _ recovered in Dumfries in May, while an autumn _ bird from Kent passed through the Tay in _ May. One September ringed bird appeared in western France exactly 3 years later. 1 | Golden Plover Pluvialis apricaria | ee | | The occurrence of Golden Plovers on the Tay | | Estuary is very variable. Flocks roost regularly ‘on the intertidal mud at low tide and can be found feeding in adjacent pastures at high _tide, although the majority leave the area altogether to feed inland. Studies of Golden | Plover wintering behaviour in Scotland (Fuller |& Lloyd 1981) have shown that about 60% /roost in coastal habitats, mostly on intertidal | [ates 70% feed on grassland but very few use | || this habitat for roosting. Waterfowl counts of the Tay Estuary, 1985-1995 43 Counts on the estuary, therefore, fluctuate markedly (Table 1), although the highest numbers are normally in November. Peak high tide numbers in this month in 1992 reached 560. No birds roost at Tentsmuir Point, the favoured high tide roosts being in fields at Tayport and in the inner estuary. At low tide, large flocks also roost both at Tayport, where up to 2,300 have been recorded on the intertidal mud in milder winters between October and February, and as far west as Newburgh. Few birds are seen during gales, when inland shelter is sought, and increases are noted when inland fields are frozen, although prolonged freezes usually result in mass emigration. However, there was no significant correlation of numbers on the Tay with weather variables. Inferences about a species with such mobile behaviour in winter are difficult to draw from estuarine counts. Few are seen in October, when the bulk of the immigrant winter population arrives, but they are present elsewhere in the area in large numbers. Itis probable thatthe wintering population consists of Scottish breeders and continental immigrants, the former departing early for their breeding grounds. Grey Plover Pluvialis squatarola Grey Plovers are both winter and passage visitors to the Tay. Table 1 shows that.a roosting population of around 170 birds is present in winter, but this is subject to considerable variation. The highest count was of 482 in February 1987, incontrastto the 20 birds present at the same time during the following winter. Another peak, of 414, occurred in November 1993. Both these peaks coincided with anticyclonic weather, when mean wind speeds were among the lowest during the study period. This may be 44 N Elkins & BMLynch significant for a species that finds it increasingly difficult to feed when winds exceed 11 m/s (Dugan eta/1981). Wind chill temperatures were not particularly low in these 2 months; thus the positive physical effects of the light winds were probably more important. The high and low tide counts in winter 1993- 94 showed that the small numbers roosting on the outer north shore equalled those feeding. However, those on the south shore fluctuated markedly, with high tide numbers exceeding those at low tide in November and February, but the reverse in December and January. This suggests that there is considerable movement between the estuary and favoured sites elsewhere in the local area. Counts at the nearby Eden estuary suggest that there is an interchange between the Tay and the Eden and, indeed, flocks have been recorded departing from Tentsmuir Point at high tide, heading south. The autumn and spring migration periods show an increase in numbers, with arrivals in August and September, and a spring peak in April. A few linger into May, while the return begins in July. All but a handful of birds roost and feed on the south shore, and the highest recorded count was 676 at Tentsmuir Point in September 1976. This site is the major high tide roost on the estuary, though casual counts reveal an occasional large migrant flock on the north shore. It has been determined that the species has increased dramatically in Britain over the past few decades (Cranswicket al 1995), and this has been mirrored by increases on the Tay. SB 19 (1) Lapwing Vanellus vanellus Like the Golden Plover, the Lapwing is a bird of inland habitats in winter, although roosting flocks can be found on exposed intertidal flats where, compared to the plovers, more feed. Thus numbers on the Tay fluctuate considerably. Peak numbers occur during the autumn migration (Table 1). The bulk of Lapwings (90%) roost on the inner estuary at both high and low tides. 2,000 in September 1987, although up to 4,000 have been recorded in the past on the extensive mudflats atlowtide. Smallnumbers The highest count was of | also roost at Newburgh at high tide, but few » utilise the outer estuary, although small numbers roost at Tayport at low tide. Most | can be found on adjacent grassland. Continental migrants augment the local populations in winter (Prater 1981)and some effects of severe weather have been noted, similar to those described for Golden Plover. Knot Calidris canutus The Knot is one of the species which has fluctuated greatly on the Tay. During the study period, numbers peaked in January and February, with 1,200 in 1987 and 770 in 1986 being the highest counts respectively (Table 1). small numbers present fed at low tide on the outerestuary. 90% ofthe wintering population | roosts at Tentsmuir Point. Autumn migration begins in late July with an_ | During | this period a few birds are found on the inner ) estuary. Asudden drop is apparent in April as migrants depart, and occasional birds occur upsurge in August and September. in summer. During the 1993-94 winter, the | |. the last 3 decades. | 1,000-1,500 were not infrequent, 1997 There is ample evidence that the number of Knot wintering on the Tay has decreased in Although in a national _ contextitnas never been an abundant species, prior to the early 1980s wintering flocks of but nowadays these numbers are unknown. Itis known that Knot suffer from a reduction in food availability during periods of low temperature (Goss-Custard et a/ 1977), but | the drop in numbers on the Tay does not _ appear to be related to the severity of the winters during the study period. In view of the _ far greater numbers wintering on the nearby Eden estuary, and also on the Montrose Basin, the Tay population can probably be counted as an offshoot of this. Tay Ringing _ Group data have shown considerable winter movement between sites on the Angus and Fife coasts. _ Our birds come from North Greenland and _ Canada, with a midwinter peak due to birds crossing the North Sea from the Waddensea after moult, returning in late winter to fatten before spring departure (Prater 1981, Tay _ Ringing Group in litt). Sanderling Calidris alba The Tay Estuary and shores of east Fife hold _ one ofthe mostimportant wintering Sanderling _ | populations in mainland Scotland. The regular | monthly counts (Table 1) reveal a wintering Sa.) + Se ee — on the Tay which reaches a peak _inDecember, with amaximum count of 750 in _ 1985. The birds are found mainly on the north | shore of the outer estuary, though greater |; numbers roost at Tentsmuir Point in late winter. The counts suggest that the wintering ' population has declined since a peak in the || mid 1980s. In the early 1970s, 1.3% of the (| | west European wintering population occurred | on the Tay (Prater 1981). Up to 1991-92, | || flocks were of national importance in most Waterfowl counts of the Tay Estuary, 1985-1995 45 winters. However, casual counts undertaken outwith the WeBS survey show that flocks of 400-600 still occur, suggesting considerable fluctuation. It may be that the uncounted shores around Buddon Ness and south to Kinshaldy provide the source of these birds, since small flocks are recorded there from time to time during winter. Certainly the species is susceptible to gales, as it feeds at the tide edge (Johnson 1985), and often moves to sandy non estuarine shores in severe weather (Moser & Summers 1987). Numbers of wintering Sanderling in the present study do show a small negative correlation with wind speed. The low tide counts in 1993-94, when the roosting population was as low as 75, showed that the birds fed in the outer estuary, where they favoured sandier substrates. Very few have been recorded from the inner estuary. As with Ringed Plover, the passage of Sanderling is not very evident from the regular monthly counts. However, substantial spring and autumn passage movements have been recorded, peaking at 450 in May, while numbers in August and September have reached 250 and 350 respectively. Parties of up to 44 have been seen in July. A colour ringing study by one of the authors (BML) between 1986 and 1990, during which 158 individuals were marked, showed that our wintering birds probably originate from the NE Greenland population. There have been 25 sightings of Tay ringed birds in May from staging sites in SW Iceland, from where the birds have departed for Greenland by 25 May (Gudmundsson 1992). The ringing study also suggests that spring migrants may stage through the Waddensee as well as Iceland, and that, in autumn, some birds route via the Baltic Sea coasts. Some colour ringed birds have returned to the Tay as early as July, with 46 N Elkins & BM Lynch the wintering birds having left by late March. Dunlin Calidris alpina This species is the most numerous wintering wader on the Tay where the peak is reach in January (Table 1). The maximum high tide roost count was 4,657 in January 1994; 70% of the population roosts in the inner estuary. In midwinter, small flocks of up to 250 have been recorded feeding on fields at high tide as far west as Newburgh, suggesting that roost counts may not encompass the whole population. Between October and February, Dunlin comprise between 35% and 50% of the total wader population on the Tay. The low tide counts in 1993-94 show that the bulk of Dunlin also use the inner estuary for feeding, mainly off Invergowrie and Kingoodie. On 13 February 1994, the number of birds counted there increased by 25% between low tide and high tide, while a large increase was also noted in the smaller flock on the south shore of the outer estuary, suggesting an influx of birds from outwith the counting area. Counts atthe Eden Estuary do indeed suggest that there is an interchange between the 2 estuaries. It is known that the density of feeding birds is related to the amount of mud substrate present, thus accounting for the high percentage feeding in the inner estuary. There was no evidence of any weather related fluctuations during the study period. Few Dunlin are recorded in summer, but the highest numbers are present during autumn migration. 5,292 was the highest count, in October 1988, but, in October 1973, a count topped 10,000 in the inner estuary. From ringing, it is evident that our birds are of the immigrant race alpina from northern Scandinavia and Russia (Tay Ringing Group in litt). These normally arrive in late October via the moulting grounds on the Waddensea SB 19 (1) (Prater 1981). The sizeable roost at Tentsmuir Pointin October and November may be related to this arrival. Some longer distance passage birds also stage through the area, as locally ringed birds in July and August have been recovered in Portugal and Senegal. As with the other common wader species on the Tay, a marked reduction in the Dunlin wintering population was recorded in 1994- 95, when the highest monthly count in winter was only 57% of the long term mean. Bar-tailed Godwit Limosa lapponica This species is one of 2 (the other being Redshank) which reaches numbers of international importance on the Tay, with maximum roost counts exceeding 1,500 in most winters. More than 1% of the west European wintering population roosts on the Tay (Prater 1981). The peak occurs in February (Table 1), with the highest count in 1992, when 2,296 birds were present. Most of the wintering and migrant populations roost on the outer estuary, divided between Monifieth and Tentsmuir Point, although the proportion at each site seems to vary according to wind strength and tidal height. Low tide counts in 1993-94 showed that the inner estuary and the shore off Broughty Ferry are both important feeding areas, although the latter held the highest density of feeding birds (Cranswick et al 1995). All feeding birds leave the inner estuary as high tide approaches, as do most of the Broughty Ferry birds, flighting to the outer estuary to roost. The Bar-tailed Godwit is one of the more mobile species on the estuary. As the tide turns, the roosting flocks are among the first to return to feeding sites, typically landing on the flats to feed before the water has receded. At Tentsmuir Point, flocks have been noted arriving from the south before high tide. 41997 lf disturbed at roost, some return south. The probability that there is some interchange between the Eden and Tay estuaries is supported by the consistent shortfall of feeding birds compared to roosting birds in the 1993- 94 winter. Bar-tailed Godwits begin to arrive in July, with a minor peak in autumn. The main arrivals on the east coast estuaries occur from November when moulted birds flight in from the Waddensee (Prater 1981). Increases are also noted in Britain when severe winters occur in that area (Cranswick eft a/ 1995). They return rapidly in March to fatten before departure for the breeding grounds. Birds ringed in November in east Fife have been recovered one month later in Portugal, indicating onward passage down the east Atlantic flyway, andin April in northern Russia, giving a clue to their origin. No longterm trends can be detected, although Boase (1970) recorded an increase after 1960 to up to 300 wintering in Invergowrie Bay. However, the winter of 1994-95 saw a marked reduction in the wintering population, with the highest monthly count being only 36% of the long term mean. Counts from the 1995-96 winter indicated a return to normal. Curlew Numenius arquata The Curlew is another species which is very catholic in its habitat requirements, being found on inland fields when unfrozen, as well as intertidal flats and rocky shores. Winter roosts on the Tay can be substantial, with up to 500-600 having been counted in the inner estuary in midwinter (Table 1). During . || thelowtide counts of 1993-94, small numbers were distributed throughout the inner and outer estuaries, but feeding predominantly at Invergowrie and Tayport. Waterfowl counts of the Tay Estuary, 1985-1995 47 The roosting population is at its maximum in late winter and early spring, building up in July with an autumn peak in September. This is similar to the pattern on other North Sea estuaries (Prater 1981), and suggests that the local population is augmented by migrants. The decrease in October is thought to be due to autumn flocks moving inland. No long term trends could be detected in this species, although there was a slow decline over the study period. Redshank Tringa totanus Like the Bar-tailed Godwit, numbers of Redshankalso reach international importance on the Tay, but only in some winters. The average midwinter roosting population is between 800 and 1,000, but occasionally exceeds 1,500. Over 70% roost in the inner estuary, with the remainder divided among the outer estuary roosts. Low tide counts in 1993-94 revealed that the vast majority feed near their roosts off Kingoodie and Invergowrie, although the highest density of feeding birds occurred near low water mark off Monifieth (Cranswick et a/ 1995). On 13 February 1994, the inner estuary flock increased by nearly 50% between low tide and high tide. Small parties also feed on adjacent pastures at high tide but, unlike Curlew, Lapwing and Golden Plover, this represents only a tiny proportion of the total wintering population. Peak numbers occur during the autumn migration, when over 1,500 are present (Table 1). The highest count was in September 1989, when 4,713 birds were counted, although casual counts have peaked at 6,900 in October, and rarely 2,500 in April. The Redshankis the most abundant wader species 48 N Elkins & BM Lynch (around 40% of the total) on the Tay during the migration months of March, April and September. Only small numbers are present from May, with arrivals from July, when British breeders move to the coasts. Icelandic birds have been caught locally in October, while birds ringed locally have been recovered in Iceland in summer, indicating the origin of most wintering birds (Tay Ringing Group in litt). A September ringed bird from the Eden estuary was also recoveredin Nigeria in January, showing that staging of Scandinavian migrants, which have the longest migration, occurs. 3000 2500 2000 1985/86 1986/87 1987/88 1988/89 1989/90 SB 19 (1) Boase (1970) recorded similar numbers on the inner estuary at migration periods, but no long term trends were detected. The marked reduction in 1994-95 in the wintering population of the more abundant species was also reflected in Redshank numbers, with the highest winter monthly count being only 40% of the long term mean (Fig 3). Turnstone Arenaria interpres Although primarily a non estuarine species, which has a substantial wintering population on the open Angus and Fife coasts, small numbers do frequent the Tay estuary. The 1994/95 1990/31 1991/92 1992/93 1993/94 Winter Figure 3 Peak winter counts of Bar-tailed Godwit and Redshank roosting on the Tay estuary, 1985-1995, Bar-tailed Godwit| | Redshank §. 1997 wintering population is small, with up to 50 birds (Table 1). The majority are found where mussel beds provide feeding, notably off Monifieth and, to a lesser extent, off Tayport. The highest count was 142 in December 1991. However, the highest mean (72) occurs in April, and, like Ringed Plover and Sanderling, there may be a considerable turnover of migrants at that time, with peaks occurring outwith the regular monthly counts. The Turnstone population wintering in the UK originates in Greenland and the Canadian Arctic, while autumn migrants include Scandinavian breeders (Prater 1981). One bird ringed in September 1985 on the Angus coast was recovered on Ellesmere Island in | the following June. Other wader species Small numbers of other species are recorded on the estuary, mainly during migration. Of these, the only regular wintering species is _ the Snipe Gallinago gallinago, which occupies salt marshes in the inner estuary, and the | smaller salt marsh by Tayport. Due to its | secretive behaviour, total numbers are unknown. Maximum numbers occur in | October, November and January, with a peak | count of 30. | Wildfowl Wildfowl figured less comprehensively in the _ counts, but sufficient data were obtained to enable an assessmentto be made of numbers | and distribution of several species. _ Mute Swan Cygnus olor | Mute Swans frequent the estuary throughout | the year and a few pairs breed. Marking with Waterfowl counts of the Tay Estuary, 1985-1995 49 unique Darvic rings has proved that the Tay Mute Swans, incommonwith others in eastern Scotland, make considerable movements up and down the east coast during the year, although many individuals are fairly sedentary (L Brown pers comm). A herd of up to 65 is present in the outer Tay between September and February, chiefly at Tayport in autumn but in the Monifieth/ Broughty Ferry area from January. A herd of up to 80 also winters upriver near Mugdrum Island or at the mouth of the River Earn, but these do not figure in the WeBS counts. Birds disperse to breeding sites in early spring, but a flock of non breeders then congregates near Broughty Ferry, to be joined by failed breeders to form a pre moult herd of 30-60 birds between April and July. In 1991, this herd apparently moved up to Mugdrum in June. Such early summer flocks are also noted on the Eden estuary and at Musselburgh on the Forth; all such flocks diminish considerably in August as birds move elsewhere to moult. From May to July in 1995, over 100 were present on the north shore of the outer Tay, mostly at Monifieth. Many of these birds may have originated from the River Earn herd, which departs at the end of April. In winter and on migration, the Mute Swan herds are occasionally joined by a few Whooper Swans Cygnus cygnus. Shelduck Tadorna tadorna This species can be seen in all months, although spring is the period during which itis mostcommon. A few pairs breed. On average of fewer than 10 birds over winter, feeding mainly in the inner estuary, with a rapid build up occurring from February. The peak is reached in April, with over 100 birds present at times between March and June, the 50 N Elkins & BM Lynch maximum being 186 in March 1991. This influx probably consists of birds en route from wintering grounds to breeding areas. In most months, the species frequents the south shore near Tayport, where small parties roost on Lucky Scalp, and the inner estuary, where the majority of the spring influx is found. A few can be located as far west as Newburgh, and, in summer, a number of broods swell the ranks. These keep to the hidden areas of the inner estuary, and may not be monitored on the WeBS counts. A rapid decrease then occurs as the birds depart for their moulting grounds. Boase (1970) recorded numbers in the 1960s of up to 300 off Kingoodie in spring, so there appears to have been a decrease on the Tay since then. Wigeon Anas penelope This is essentially a winter visitor to the estuary, arriving in September and departing in March. The peak is reached in November, with a mean of 155, but over 200 are frequently recorded between October and January. This represents a recovery in numbers since an earlier decline in mid century (Atkinson-Willes 1963). The maximum occurred in November 1994, when 262 were present. The majority feed and roost around Lucky Scalp, but small numbers also feed at low tide on the foreshore between Monifieth and Broughty Ferry. Casual counts also reveal a significant feeding flock west towards Newburgh (526 in November 1995). No Wigeon are seen between May and August. Boase (1970) recorded the species as being only ‘casual’ off Kingoodie. Teal Anas crecca The Teal is a highly mobile species in winter, dependent upon unfrozen inland waters. Nevertheless, it has shown a marked decline during the period. Prior to 1990, numbers SB 19 (1) wintering on the Tay regularly exceeded 50 birds, with over 100 recorded in January and February, while higher counts have been made in the past (Atkinson-Willes 1963). The peak count was 190 in January 1986. Since 1990, none has been seen in some months, and parties have not exceeded 45 in number. However, although at high tide they are mainly to be found around Lucky Scalp, the reedbeds of the inner estuary may also be frequented and thus they are difficult to locate. Substantial low tide feeding flocks eg 130 in November 1995, have been recorded towards Newburgh, on a par with counts given at that site by Boase (1970) in the 1960s. Any decline may be linked to the greater frequency of mild winters, when birds are able to remain inland. Mallard Anas platyrhynchos Mallard are the commonest freshwater ducks on the Tay estuary, peaking in January with a mean of 626, but with a minor peak in September, probably due to migrants. Historically, the species was very abundant (Atkinson-Willes 1963, Boase 1970), but numbers have fallen since then, with the maximum count being 1,419 in September 1986, although 1,150 were present in midwinter 1991-92 and 1992-93, and other counts have exceeded 2,000. There is a marked exodus in March and April, with an early upsurge in August. Flocks frequent Lucky Scalp at high tide, but feeding birds keep mainly to the foreshores off Invergowrie and Broughty Ferry and upriver. Tufted Duck Aythya fuligula This is aspecies whichis found almost entirely in the inner estuary, where counts have been sporadic. Nevertheless, enough data are available to show that the peak is in January, when as many as 720 have been present (1987), and the mean is over 300. Few have 1997 been recorded between March and August. This inner estuary flock was also present in the 1960s (Boase 1970) but at the same time there used also to be hundreds off Dundee and Monifieth in winter, where now there are few. Eider Somateria mollissima With peak numbers over 20,000 between autumn and spring, the Tay Eiders are of international importance. However, regular counts are not undertaken, as many feeding rafts are often out of sight beyond Tentsmuir Point, and tide dependent feeding movements present great difficulties to counters (Pounder 1971, Cranswick eta/1995). Several hundred roost on Lucky Scalp at high tide. Winter 1995-96 saw a concerted effort by local counters and staff of the Wildfowl and Wetlands Trust to determine numbers. Accurate counts are only possible under ideal conditions. This was highlighted in November 1995, when a count on the flood tide in bright sunshine and a moderate westerly wind revealed around 11,500 birds. Only 3 hours later, at high tide with the sun obscured, a freshening wind, and a redistribution of many birds into the choppier waters of mid channel, only 6,500 were located. Goldeneye Bucephala clangula This species winters in nationally important | numbers throughout the estuary, but mainly _ west of the Tay bridges. Apart from a few early individuals, birds arrive in October and November, building to a peak in January, _ although asecondary peakin March suggests || passage. The highest counts were of 344 in | January 1992, and 351 in March 1987. || Irregular counts at Newburgh have revealed flocks of up to 200, so that the monthly counts lower down the estuary monitor only part of the population. Small numbers are still present Waterfowl counts of the Tay Estuary, 1985-1995 51 in April, but the majority have left by the end of the month. Boase (1970) recorded several hundreds wintering in the early 1960s, off Invergowrie and also Dundee and Monifieth. Goosander Mergus merganser There is evidence that the large moulting flock found in late summer is a relatively recent occurrence. Previous to this study, small parties were noted both in spring and in late summer (Boase 1970), but there is also the possibility that there may have been confusion between Goosanders and Red- breasted Mergansers Mergus serrator, as all the former are redheads while moulting. The flock begins to build up in June, reaching nationally important numbers in July and August. The peak count was of 277 in August 1995. Most of the birds depart rapidly in mid September, although there were still 225 in the third week of September 1987. Occasional small parties are recorded into November, with some frequenting the upper reaches near Newburgh in winter. At high tide, most roost on Lucky Scalp, but at low tide they disperse both up and down river in smaller parties. The origin and destination of these birds is at present unknown. It is thought that the large breeding population in northern Engiand and the Borders may be the source, although a bird from the increasing Welsh population was recovered on the Eden estuary in August 1992. Other wildfowl species not monitored regularly by this study include geese, mainly Pink- footed Anser brachyrhynchus and Greylag Anser anser, which roost in large numbers (exceeding 5,000 and 1,500 respectively) from time to time on the inner estuary, and also feed in adjacent fields. Sea duck, such as Long-tailed Duck Clangula hyemalis and Red-breasted Merganser, can be found in small numbers in the outer estuary but more 52 N Elkins & BM Lynch importantly on the sea to the south. Discussion Anunderstanding of the changes in waterfowl populations on the Tay estuary cannot be achieved in isolation from other coasts and it has been noted that there is probably a large degree of movement of some species between the Tay and the Eden, and doubtless also between the Tay and Angus shores. There is also movement between the high tide roosts on the Tay. Laing & Taylor (1993) gave some trends for the inner estuary, but there is considerable evidence that for some species this cannot be meaningful in isolation from the outer estuary. Changes in species numbers have been described under the species accounts. A number of factors are probably responsible for these fluctuations: breeding success, changing winter mortality and weather probably being the most important. Local climatic fluctuations have already been described. Further afield, NW Europe, where many species stage in autumn and early winter, experienced similar temperature anomalies in most early winters to those that affect the Tay. The exceptions were in 1988- 89, when temperatures in November and December exceeded the long term mean toa much greater degree locally, and 1991-92, when the reverse was true. Overall, the wintering wader population varied between 10,000 and 13,000 over the period. However, the winter of 1994-95 saw a dramatic drop in the number of birds using the estuary, when the number fell to 6,800 (60% of the 10 year average). The 4most common wintering species were all affected, but Bar-tailed Godwit and Redshank both fell to 40% of their average (Fig 3). As the 4 species have rather different migration strategies, the reason for this SB 19 (1) decrease is unclear, although both breeding success and climatic conditions on the near continent may have played a part. Counts in the winter of 1995-96 showed that Bar-tailed Godwit numbers recovered, but populations of Dunlin and Redshank remained relatively low. Any statutory protection afforded to the estuary suchas the proposed Special Protection Area (SPA) must take into account the movements between sites within and without the estuary. At present the bulk of the estuary is covered by Sites of Special Scientific Interest (SSSIs), made up of the inner estuary (5,490 ha, including the subtidal area), Monifieth Bay (213 ha), Barry Links (1,041 ha, including the sand dune system) and most of the Tayport- Tentsmuir coast (1,044 ha). Apart of the latter is also a National Nature Reserve. Disturbance to waterfowl varies, but is generally at what seems to be an acceptable level. All roosts except Lucky Scalp are subject to disturbance by walkers, especially in fine weather. Roosting at Dundee airport can be interrupted by aircraft and bird control activities, often to the extent that the roost is completely displaced. Speedboats in the outer estuary disturb wildfowl on the water but the problem is small in winter. At low tide, there is a limited amount of bait digging, but this is thought to cause only minor damage. Any future increase, for instance large scale commercial exploitation, would create a far greater problem. Natural disturbance is confined to the highest tides, especially when onshore winds enhance tidal heights, and normal roosts are covered. Many birds then vacate the area and move to unknown destinations, although this is rare, with only 3 counts undertaken when these conditions prevailed. There was no apparent correlation between tidal height and the total roosting population. However, increased tidal heights 1997 have been shown to be detrimental elsewhere at low tide by covering feeding sites (Pienkowski & Prokosch 1982). Regular monitoring of water quality has been carried out by the Tay River Purification Board (now the Scottish Environment Protection Agency). The Tay Estuary generally has a good water quality, despite the discharge of sewerage products, mainly in the Dundee area; some untreated discharges give locally poorer quality water, but progress to improve these is in hand. Over the period of study, itis not known if changes have had any effect on the bird populations. Sewage outfalls in the Invergowrie/Kingoodie section may indeed be the reason for the large number of feeding waders along this stretch of mudflats, (Pounder 1976) and it remains to be seen whether cleaning up discharges will result in any redistribution of waterfowl. The Tay Estuary is an important link in the chain of estuaries along the east coast of Scotland. If the wildfowl, including sea duck, are taken into account, the wintering waterfowl populations of the Tay, Eden and St Andrews Bay exceeds 50,000, of which half are waders, with 3 species reaching international importance, and another 13 (7 waders and 6 ducks) being of national importance. Other waders reach national importance by virtue of their migrant populations. Acknowledgements A great debt is due to all those who put in the effort to obtain the data and who continue to count. Apart from the authors, the bulk of the counting was carried out by Dave Bell, Dave Ferguson, the late Steve Fulford, Sylvia Laing, Wendy Mattingley and Donald Stewart. We thank the following for acting as back ups and for taking part in the low tide counts: Waterfowl counts of the Tay Estuary, 1985-1995 53 GM Adam, B Anderson, D Arthur, A Barclay, | Buick, R J Burness, J Burrow, F Buxton, E D Cameron, A Davidson, R Downing, | Elkins, J Graves, K Hall, L Hatton, G Johnston, J Johnston, J Kelemen, F Littlejohn, D Malcolm, N Mann, | & J MacPherson, J McKerchar, | Morris, S Moyes, M Nicoll, W Saunders, A Scott, | Smart, A-M Smout, T C Smout, R W Summers, N TaylorandB Yule. Ourapologies go to anyone who has been inadvertently omitted. The results were enhanced by discussions with Dave Bell and Les Hatton, who also provided data from the Eden estuary. We are grateful to Allan and Lyndesay Brown, and John Kirk for Mute Swan data, the Tay Ringing Group for ringing data, the Tay River Purification Board for information on pollution, and the Meteorological Office, Leuchars, for weather data. Finally, we thank the British Trust for Ornithology for data preceding the study period, and for the opportunity to contribute to such an important survey. References Atkinson-Willes GL 1963. Wildfowl in Great Britain. HMSO London. Boase H 1970. Bird records of the Tay Area, 1961- 67. unpublished typescript. Cramp S & Simmons K E L (eds) 1983. The birds of the Western Palearctic. Vol 3. Oxford University Press. Cranswick P A, Waters R J, Evans J & Pollitt MS 1995. The Wetland Bird Survey 1993-94: Wildfowl and Wader Counts. BTO/WWT/RSPB/JNCC Slimbridge. Dugan P J, Evans P R, Goodyer L R & Davidson N C 1981. Winter fat reserves in shorebirds: disturbance of regulated levels by severe weather conditions. /bis 123: 359-363. Fuller R J & Lloyd D 1981. The distribution and habitats of wintering Golden Plovers in Britain 1977-78. Bird Study 28: 169-185. Goss-Custard JD, JenyonRA, Jones RE, Newbery PE & Williams R le B 1977. The ecology of the 54 N Elkins & BMLynch Wash Il Seasonal variation in the feeding conditions of wading birds (Charadrii). Journal of Applied Ecology 14: 701-719. Gudmundsson G A_ 1992. Flight and migration strategies of birds at polar latitudes. Ph D thesis, Dept of Ecology, Lund University, Sweden. JNCC 1994. An inventory of UK estuaries. No 86. Firth of Tay (ed A L Buck) Peterborough. Johnson C 1985. Patterns of seasonal weight variation in waders onthe Wash. Ringing & Migration 6: 19-32. Moser M E & Summers R W_ 1987. Wader populations on the non-estuarine coasts of Britain and Northern Ireland: results of the 1984-85 Winter SB 19 (1) Shorebird Count. Bird Study 34: 71-81. 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. Pienkowski M W & Prokosch P 1982. Movement patterns of waders between the coastal lands of western Europe. Seevogel 3: 123-128. Pounder B_ 1971. Wintering Eiders in the Tay estuary. Scottish Birds 6: 407-419. Pounder B 1976. Waterfowl at effluent discharges in Scottish coastal waters. Scottish Birds 9: 5-36. Prater AJ 1981. Estuary Birds of Britain and Ireland. Poyser, Calton. Norman Elkins, 18 Scotstarvit View, Cupar, Fife KY15 5DX Bruce M Lynch, 27 Luke Place, Broughty Ferry, Dundee DD5 3BN. Revised manuscript accepted August 1996 Roosting Ringed Plover David Mitchell Scottish Birds (1997) 19: 55-57 SHORT NOTES Nest site selection by Buzzards in mid Argyll Buzzard nests are sometimes sizeable and conspicuous structures. The birds are known to use a wide variety of tree and crag sites (Picozzi & Weir 1974, British Birds 67:199- 210; Fryer 1986, British Birds 79:18-28; Maguire 1979 Western Naturalist 8:3-13; Dare 1989, The Naturalist (Leeds) 114:3-31, Holdsworth 1971, British Birds 64:412-420). Ground nesting behaviour was recorded by Maguire in Kintyre, and occurs infrequently in Argyll (pers obs), but has not been recorded in any of the detailed Buzzard studies made in the south of England (Tubbs, 1974, The Buzzard David & Charles; Dare 1961, Ecological observations on a breeding population of the Common Buzzard, Buteo buteo, PhD thesis, Exeter University). The location of nests on the ground might be thought to make the nest particularly vulnerable to predators. As part of an investigation of Buzzard breeding distribution in Argyll, we considered whether there may be a shortage of suitable nest sites which led to ground nesting, and which might also limit the distribution of these birds. In this note we consider the selection of nest sites by Buzzards and their breeding success in different sites. Field work was carried out in 3 main areas of Argyll (Figure 1): north Lorn (including Glens Lonan, Feochan and Euchar to the east and south east of Oban), south Lorn (to the north of Lochgilphead) and Glen Lochy (between Tyndrum and the Lochy and Orchy rivers). Attempts were made to locate every nest within these study areas in 1989 and 1990. Each nest was usually visited a number of times each season, to obtain information on 55 clutch size, laying dates, brood size and fledging success. The laying date was usually estimated from direct observation of the date of hatching, or from measurements of the stage of growth of the chick from which its hatching date could be extrapolated. Nests were described as on crag sites where they were located on open rock faces. Bank sites were situated on vegetated ground, usually on steep banks alongside gullies, but with no exposed rock. A total of 34 nests was located in 1989 and 39 in 1990. Many occupied nest sites had those from previous years in close proximity. 68% Figure 1 Study areas. The areas for which complete coverage was obtained were: South east Mull 50 km2, North Lorn 140 km?, Glen Lochy 35 km?, South Lorn 43 km, Nest sites on the Cowal Peninsula included in the analysis with the kind permission of Steve Petty and David Anderson. Glen Lochy 56 Short Notes of nests were in trees and Table 1 shows that there was a clear preference for larger trees. Birch was undoubtedly the commonest tree species in North and South Lorn, yet was rarely used if larger trees were available. The crags used for nesting were from 5 to 50 m high, and the nesting ledge invariably had a vertical rock face immediately below the nest, although this was often only a few metes high. The nests were usually readily accessible on foot, although they were hidden by vegetation when viewed from below. There was no evidence that bank nesting was used because of a shortage of suitable trees. In all cases, bank nests were sited in woodland or wooded gullies, with apparently suitable tree sites in close proximity, often containing nests from previous years. Table 1 Numbers of nest sites occurring in trees of different species used by tree nesting Buzzards in mid Argyll. Tree species Number of nests Oak Quercus spp Spruce Picea spp 12 Birch Betula spp 4 Scots Pine Pinus sylvestris Ash Fraxinus spp Larch Larix decidua Cypress Chamaecyparis spp Douglas Fir Pseudotsuga menziesii Rowan Sorbus aucuparia Beech Fagus sylvatica =—=_ MONNOWA Table 2 shows the breeding performance of Buzzards from different nest sites. There was a significant difference in laying dates between the sites (Kruskall-Wallis one way ANOVA = 11.63, P < 0.009) and multiple comparison tests showed that clutches laid in bank nest sites were laid significantly earlier than those SB 19 (1) laid in the other nest sites. There was no significant difference in clutch size or brood size, but there was a significant difference in fledging success (K = 8.94, P < 0.03) and multiple comparison tests showed that significantly fewer young fledged from crag sites than from tree or ground sites. There was no evidence that Buzzards nesting on the ground experienced a higher level of predation than birds using tree or crag sites. Crag sites actually had a higher proportion of nest failures, although the causes could not be determined. It is reasonable to assume that nests on the ground would be exposed to a greater range of predators than those in trees or crags. There is, presumably, some advantage to ground nesting which compensates for this. Buzzard nests are conspicuous, and the first birds to lay will do so before trees are in leaf. If egg predation by corvids and other avian predators is an important cause of breeding failure, it may be that ground nests are less conspicuous at the start of the season. In addition, ground sites may be more protected from wind and rain; this would be particularly important early in the season when leafless trees provide little shelter. This may be why females which are going to breed early in the season are more likely to choose a ground nest site. By doing so they can take advantage of an early start to the breeding season without incurring the higher energy costs that are probably associated with incubation in exposed sites in unfavourable weather conditions. This may explain why ground nesting has not been reported from areas such as the south of England, where the weather is less severe early in the breeding season, and where trees are in leaf earlier in the year. We are very grateful to David Anderson, David Jardine, Mike Madders, Steve Petty, if | | } it 1997 Short Notes 57 | Table2 Breeding performance of Buzzards in mid Argyll in relation to nest site situation. | Laying date is given with reference to 1April (day one). Brood quality rank is based on | 1=brood consisting principally of nestlings of low weight for age, 2=brood consisting principally of nestlings all of expected weight for age and 3=brood consisting principally _ of nestlings of high weight for age. Nest site situation | | Breeding parameter Bank Coniferous Deciduous Crag | (n=14) trees trees (n=9) i (n=20) (n=30) _ Laying date | (Median and range) 9 (5-37) 16 (5-30) 14 (2-27) 16 (9-26) ; Clutch size | (Average and range) 2.59 (2-4) 2.01 (1-3) Boies) 2:17, (2-3) | Initial brood size | (Average and range) 2.29 (0-4) —- 1.90 (0.3) 2.19 (0-3) 2.00 (0-2) ‘| Number fledged | (Average and range) 2.13 (0-4) 1.84 (0-3) 2.11 (0-3) 1.25 (0-2) \ | Brood quality . (Mean rank and SE) 2.06 (0.25) 1.53 (0.16) 2.07 (0.14) 1.80 (0.16) | Chris Thomas and other members of the studentship to GEA with the Nature | Argyll Bird Club for their help in locatingnests © Conservancy Council (now Scottish Natural and assistance in the field, and to many Heritage) and we are extremely grateful to Dr { } landowners for granting access to their land. _ DBAThompsonandDr Colin Galbraith of the | The study was supported bya NERC CASE = JNCC for their help. Graham E Austin * and David D Houston, | | Applied Ornithology Unit, Graham Kerr Building, qt IBLS, Glasgow University, Glasgow 12 8QQ * Present address: British Trust for Ornithology, The Nunnery, Thetford, Norfolk [P24 2PU Revised manuscript accepted April 1996 58 Scotttish Birds (1997) 19: 58-63 OBITUARIES MALCOLM CASTLE 1923-1996 The death of Malcolm Castle not only means the loss of a friend, but also the end of an era for the Ayr Branch - Malcolm was the last of the founder members still coming to our meetings. Throughout this period he always maintained close links, from being the first Branch Secretary, through his chairmanship in the 1970s and, latterly, as President. His enthusiasm for birdwatching was always apparent from his questions at meetings and he was a regular on our outings, especially the longer trips. He was always keen to help with survey work. This was never clearer than in his role as the Ayrshire organiser during the 5 years of field work for the first BTO Breeding Bird Atlas. However, for many, his name will always be linked with rookeries. He first organised a survey of the county in 1966, inspired by the pioneering work of Robert Walls 10 years earlier, and then kept up the counts each decade, sometimes linking with BTO surveys, until the year of his death. Itis only a matter SB 19 (1) of months since he last sent us out with our maps and clipboards. His loyalty to Rooks combinedhis love of birds and his professional interest in agriculture, in a species that acts as an excellent indicator of the general health of the countryside. As a scientist, his many years of research at Shinfield and then at the Hannah Dairy Research Institute made him a leading expert on grassland management and silage production. This was recognised in his period as President of the British Grassland Society. | His combination of practical agricultural knowledge and concern for wildlife again came to the fore when he played a major role in the formation of the Ayrshire Farming and Wildlife Advisory Group. Away from work - and birds - Malcolm was a man of many parts. When he came to Scotland he discovered curling, an enthusiasm which remained with him for the rest of his life. After many years of breeding Jack Russell | terriers, Malcolm and Betty decided that when they moved to their retiral home at Tobergill they had room for something bigger. It was not long before their little herd of Belted Galloway cattle started to make an impact at the local shows! He also took pride in his skills as a handyman and, after visiting St | Kilda with a National Trust workparty, heliked | totell people thathehadaddedstonemasonry | to his repertoire. However, the overriding memory of Malcolm | for most of us will always be of a good friend, ; whose infectious enthusiasm for life brightened any day. Whether we remember his comments in meetings or his jokes in the minibus in Islay or Holland, we shall all miss him. Roger Hissett == | ROBERT “BOBBY” JOHN ‘TULLOCH, MBE 1929-96 '|He was a rare character, one of the kind remembered with a smile. Always known as |Bobby, he could amuse and inform without seeming to in his beautiful, soft Shetland accent. At the premiere of the RSPB film on | Shetland Land of the Simmer Dimina packed Usher Hall, Bobby walked onto the stage. There was complete silence as he looked \around the vast hall and then slowly said “I think there are more people here than in the ‘whole of Shetland’. The 2,000 people responded hugely and warmly. Born in North Aywick, his upbringing was at jone with his natural surroundings. At primary school, all the children had “pets” - gulls, \icrows or waders - all taken as chicks, carefully nurtured for weeks until they could fly. There could be problems, as when his mother found ‘young Bobby tucked up in bed with a young Raven under the covers! Bobby was “remarkably surefooted and this stemmed from when he and his school chums ran races ‘along boulder beaches, jumping from stone to stone at breakneck speed - “more interesting than a sandy beach” he told me. |One of the characteristics of Shetlanders is the love of doing something new and the (dislike of repetitive work. Bobby, a baker by ‘trade, started up atravellingshop. He bought an old van and spent a long time carefully N (painting one side with his name and pictures jof his products. However, he only did one ‘side; “too routine”, he told me, to do the other. if | as a i ————" oa oD puting the 50’s and early 60’s, all his spare me was spent studying the wildlife around aim and increasingly the RSPB used his results and expertise. In 1964, the late George Waterston made him an offer he Y ~—~< Obituaries 59 could not refuse; tobe Shetland representative forthe Society. For21 years his achievements for conservation and public relations were truly outstanding. He built up a data base of birds, developed several important nature reserves and stimulated local people to take a pride in their islands. Undoubtedly the first breeding of Snowy Owls on Fetlar (1967) was the highlight of his career. For 4 years he threw himself into organising the protection and viewing of the famous pair and he developed enormously during this time, gaining a confidence which never left him. Bobby was a remarkably talented man. He was an accomplished musician on guitar, fiddle and accordian, much in demand at any social gathering, and a self taught photographer of outstanding quality. He often processed his own film, both print and slide, and | remember being in his little dark room and seeing a lot of slides in the waste bin. “Oh, they’re not good enough” he declared. These rejects | used for many years of lecturing! Bobby’s own talks were legendary. Photographically superb, he blended these together with anecdotes and observations which were classic examples of pure entertainment and education. At SOC Conferences he was the centre of a rollicking crowd at the bar, always with laughter as Bobby was a great raconteur. On Saturday evenings the call would go out and soon he was on a chair with his squeeze-box surrounded by happy dancers. Great weekends! In 1969, Bobby started the Shetland Bird Report and 4 years later he and Dennis Coutts established the Shetland Bird Club. He was elected the Club’s President and remaind so until his death. After retiring from the RSPB, Bobby had the freedom to travel. He regularly lectured on National Trust for 60 Obituaries SB (19)1 Scotland cruises and led tours throughout the world, from the Arctic to Antarctica. Bobby enjoyed writing but his early publications were also important for bird conservation and showing his fellow Shetlanders just how vital their islands really were. Many people considered his Bobby Tulloch’s Shetland to be his finest book, giving an insight into how he viewed the natural world, illustrated with his expressive photographs. This book also reflected his deep, alround knowledge of Shetland’s natural history, particularly of Otters and seals. Several TV films would never have been possible without his intimate local knowledge. Other examples of his breadth of interest were his photographs of flowering plants and ferns, while for his own interest he built up a considerable list of fungi for the islands - all of which, he assured me, he had eaten. After an illness, Bobby died peacefully in Lerwick. His wife, Betty, had died some years previously. Everyone who had ever spoken to him and, | reckon, had even heard him speak, considered him a friend. We all miss him and we are grateful he shared his world with us. Frank Hamilton | 1997 | VERO COPNER WYNNE-EDWARDS -CBEFRSFRSE 1906-1997 1946 was a year of new beginnings. The _world was recovering from the trauma of war and hundreds of thousands of men and women | were returning home optimistically to face the future. In that dawn there was a fresh sense _ of freedom which brought with it a rediscovery of nature and a new ideology called “conservation”. Millions who had forgotten |, the sights, sounds and scents of their native | land, had a new found value of all things | natural and from their ranks came champions inthe name of natural science. One of these was Vero Wynne-Edwards, a man of great _ intellectual and physical energy, a pioneer of ‘modern marine ornithology, an explorer of / the Canadian Arctic anda fundamental thinker. _Thatyear, Wynne-Edwards was appointed to succeed Alistair Hardy to the Regius Chair of | Natural History in Aberdeen University. ‘Wynne, as he was widely and affectionately known, was 40 years of age. Like other | students of zoology of his time, at Oxford he \ did the long march through the Animal |Kingdom and graduated with First Class | Honours in 1927. Unlike most postgraduates, | who became specialists in a narrow field, he | chosea broad canvas of biological experience jand scholarship, and this determined the | wonderful personality he was to become. His | first venture was in marine research when he worked on Herring, Cod and small crustacea, » with aside interest inthe behaviour of wintering (flocks of Starlings. In 1929 he became a {lecturer at Bristol University and married y Jeannie Morris. About this time he was | studying Lapwings in a large meadow when, \| from his hide, he was horrified to see a ‘Squadron of cavalry trooping through the gate. ‘| Rushing forward he addressed the Colonel at | the head of the column. “Sir, this field is alive with nesting Lapwings and their chicks... could Obituaries 61 you please return in a month’s time?” So instantly convincing was Wynne’s plea that, to the consternation of the troopers, the Colonel turned his charger and led the squadron back through the gate. In 1930 he became an Assistant Lecturer at McGill University, Canada. He became interested in Atlantic ornithology and produced the hypothesis that seabirds were not randomly distributed across the trackless face of the ocean. His interest in alpine flora, which he had had from his childhood in his native Yorkshire Dales, blossomed in Canada with published papers on the patchy distribution of rare flora. For these works he was twice awarded the Walker Prize of the Boston Natural History Society and the Fellowship of the Royal Society of Canada. During the war, he organised a course in electronic physics for radar mechanics in the Royal Canadian Air Force and was sent to assess the fisheries of the Yukon and Mackenzie rivers. 62 Obituaries SB (19)1 After 15 years of happy and exciting life at McGill, Canada was in the blood of the Wynne- Edwards family. Their daughter Janet and son Hugh were both born in Canada and returned there to settle in later life. One of the attractions of the Aberdeen job was the fact that it provided 5 months home based skiing in most years! In 1950 Wynne returned to Canada as the zoologist on the Baird Expedition to Baffin Island. In 1946 George Waterston, Arthur Duncan and others were reviving the Scottish Ornithologists’ Club and Wynne-Edwards arrived just at the right time. He provided an instant link with centres of learning in natural science at home and abroad. He also gave professional academic backup to what was essentially an amateur movement out of which, very shortly, was to emerge the professional cadre of natural scientists who would forge the basis of nature conservation in Scotland. In the choice, training and deployment of these scientists in public life he was to play a leading part in conservation over the next 25 years. Though Wynne was a classical zoologist of the old school, devoted to scholarship, he was also a pragmatist with a rare insight into the ways in which science can be put to the service of the community. One of the many outlets of his talents and energy was the SOC. He became an Hon. Vice President and joint founder of the Aberdeen Branch of the Club, a Trustee of the Fair Isle Bird Observatory and the chief editor of the resurrected Scottish Naturalist, a connection which lasted until 1964 when he became co-editor of the new Journal of Applied Ecology. His intellectual life was greatly tempered by his desire for physical fitness, coupled with a love of the mountains and wild country. He became a life member of the Cairngorm Club and was a leading figure in the Aberdeen Branch of the Scottish Ski Club. At the age of 62 he broke the record fora traverse on foot of the 6 main summits of the Caringorms by 96 minutes. True to character, he would appear at meetings in shorts having run the distance while others travelled by car. With . brief case in rucksack, he walked across the Mounth and Lairig Ghruto meetings in Tayside and Speyside. He ran regularly on hill and shore, plungedin river and sea at all seasons, and often cycled to the mouth of the Don to have his picnic lunch while botanising and bird watching. Wynne-Edwards’ main endeavours were focussed in the building of one of the finest university departments of zoology in Europe. He took the entire establishment from its old quarters in Marischal to the new building in Old Aberdeen in 1970 and he founded the Culterty Field Station at Newburgh in 1958. The pre-eminence of Aberdeen zoology rested largely upon Wynne’s scientific reputation which was greatly enhanced by his highly controversial challenge to Darwinian orthodoxy. In his great work Animal Dispersion in Relation to Social Behaviour (1962), he proposed that animals cooperate for the benefit of the group, that they compete for territory and status rather than food, with the losers having to accept their lot. He stated that animals are not, as Darwin supposed, always striving to increase their numbers, but are instead programmed to regulate their numbers. Group selection he saw operating in the differential survival of populations. A precis of the book in the Scientific American in 1974 sold 350,000 copies. The theory was roundly rejected but Wynne was never persuaded that he was wrong. After 25 years of debate, when he was 80 years of age, he published his second work | 1997 Evolution Through Group Selection (1986). _. He took his belief in group selection to the — = ote | grave and is recognised today as one of the most important fundamental biologists of this century having “done a great service in forcing | people to think carefully about the evolution of _ social behaviour, especially altruism” (Krebs _and Davies in Behavioural Ecology: an | Evolutionary Approach, 1978, p8). _ Wynne’s contribution to science in public life was prodigious both locally and nationally. | His influence was inspirational and ranged _ from the careers of individual students and | scientists to the movements of Government _ policy, affecting the lives of millions. He was inatthe birth of nature conservation in Britain | and became a long standing member of the _ Nature Conservancy and its Scottish and | Scientific Policy Committees. __ initiated a research project on the population | ecology of Red Grouse, which led to the setting up of the research station at Hill of _ Brathens, Banchory, and this is still in | progress. |. appointment as a member of the newly created ;/| Natural Environment Research Council in | 1964. He was later to become its Chairman ,)| in 1968-71 - a very difficult time in NERC | which resulted in the split of the Nature ,_| Conservancy into the Nature Conservancy '| Council andthe Institute of Terrestrial Ecology ;)) in 1972. He was also a Vice Chairman of the In 1956 he This served as a prelude to his Red Deer Commission 1959-68, President of the Scottish Marine Biological Association 1967-73, the British Ornithologists’ Union | 1965-70 and of Section D of the British ) Association for the Advancement of Science | in 1974. He was a member of the Advisory } | | ] Obituaries 63 Committee on Fishery Research andthe Royal Commission on Environmental Pollution and a Leverhulme Fellow 1978-80. He held honorary membership of 5 biological societies. He was elected a Fellow of the Royal Society of Edinburgh in 1950, the Royal Society (London) in 1970, and was awarded 3 medals: the Neill Prize of the RSE, the Frith medal of the Zoological Society of London and the Godman-Salvin medal of the BOU. In 1973 he was appointed a CBE but many felt that he deserved a greater accolade. In the formal setting, Wynne-Edwards was the polished, dignified academic, but he had within him other dignities - the sensitive instincts, wonder and burning curiosity of the naturalist and the love and devotion of the family man with 7 grandchildren and 7 great grandchildren. His memorial takes physical form in the Department of Zoology with its Culterty Field Station in Aberdeen University. In science he will be remembered for his Theory of Group Selection, which might gain greater acceptance in future than it did in his lifetime. In education, a generation of natural scientists were fired by his example, enlightenment and romantic spirit. In human affairs, his influence will live on in that great cultural movement which seeks enduring harmony between man and nature. Information contained in this obituary has been provided by Professor David Jenkins, Dr Adam Watson and Professor Paul Racey, to whom Professor Wynne- Edwards was mentor and friend. J Morton Boyd 64 Advice to Contributors SB 19 (1) 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 will be reviewed by at least 2 members of the editorial panel and, in some cases, by anindependentreferee. They willnormally 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; 2 copies are required and the author should also retain one. We are happy to accept papers on Applemac computer discs. We cannot handle other formats because both the SOC Human induced increases of Carrion Crows and gulls on Cairngorms plateaux by Adam Watson. Scottish Birds 18(4):206 - 207 Parts of these 2 paragraphs were omitted by the printer. Since 1970, | have seen Crows on most parts of A south to Ben Macdui summit, but mainly on north parts on and near Cairn Gorm. | counted Crows and people on 30 parts. Out of 7 where | saw no Crows, 5 were among the 6 parts where | saw nobody. Parts where no Crows and people were seen lay away from the main walkers’ routes to and from Cairn Gorm, Ben Macdui and Cairn Lochan, mostly computers and those at our printers are on the Apple system. Please contact Sylvia Laing on 0131 556 6042 to discuss this. Headings should not be underlined, nor typed entirely in capitals. Scientific names in italics should follow the first text reference to each species. Names of birds should follow the official Scottish list (Scottish Birds Vol 17 : 146-159). 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........... on5 August 1991 ee but on the 5th (if the name of the month does not follow). Please note that papers shorter than c700 words will 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 shouldbe keptas simple as possible, with footnotes used to provide extra details where necessary. Each table should be on a separate sheet. Maps and diagrams should be in Indian ink and be camera ready, but drawn so as to permit reduction to half their original size. on steep rough bouldery corries. On the 30 parts for all 46 years combined, Crow and people densities were related (n-30, rs=0.69) The highest Crow density (23.2 per ha in 1971-88) was at Lochan Buidhe (next paragraph) followed by 3.4 at the March Burn nearby, Cnap Coire na Spreidhe near the Ptarmigan Restaurant (1.1), Coire Domhain (1.0, associated with garbage at snow holes), Cairn Lochan (0.7), Ben Macdui summit and west of Cairn Gorm (0.6), and lower elsewhere. Corrected copies of the complete paper can be supplied by contacting Sylvia Laing at 21 Regent Terrace, Edinburgh EH7 5BT. NEOTROPICAL BIRD CLUB Neotropical bird club launched A club has been launched to promote the study and conservation of the birds of the Neotropics (South America, Central America and the Caribbean). It is currently seeking founder members to help reach the launch budget of £2000, which is required to get the club running and to publish the two first issues of its intended journal ‘Continga’. Founder members will be asked to pay a minimum of £25, and will be formally acknowledged in the first issue of ‘Continga’. ‘Continga’ will provide a colourful and much needed forum for exchange of information on the avifauna of this extremely rich and diverse area, and will contain papers and features on the birds and their conservation as well as news of recent observations and discoveries (at present, new species are still being discovered at the rate of more than two a year). It is hoped that in due course the club will be able to provide direct funding and support for practical conservation programmes. For further details and membership forms, please contact: Rob Williams, Publicity Officer, Neotropical Bird Club, c/o The Lodge, Sandy, Bedfordshire SG19 2DL Scottish Birds Volume 19 Part 1 June 1997 Contents Main papers Distribution and abundance of Twites wintering in Caithness. H Clark & RM Sellers 1 The status of the Gannet in Scotland in 1994-95. S Murray & S Wanless 10 Gizzard contents of Pochard, Tufted Duck and Goldeneye from Loch Leven, Kinross, in winter 1994-95. Barry Stewart & Alan Lauder 28 Waterfowl counts on the Tay Estuary, 1985-1995. WN Elkins & BMLynch 36 Short notes Nest site selection by Buzzards in mid Argyll. Graham E Austin & David D Houston 55 Obituaries Malcolm Castle. Roger Hisset 58 Robert “Bobby” John Tulloch. Frank Hamilton 59 Vero Copner Wynne-Edwards. J Morton Boyd 61 Advice to contributors 64 Correction Scottish Birds 18(4) Human induced increases of Carrion Crows and gulls on Cairngorms plateaux. Adam Watson 64 Published by the Scottish Ornithologists’ Club, 21 Regent Terrace, Edinburgh EH7 5BT. © 1997 Scottish Birds The Journal of the Scottish Ornithologists’ Club Editor: Dr S da Prato Assisted by: Professor D Jenkins, Dr J B Nelson, Dr | Bainbridge, Dr M Marquiss and R Swann Business Editor: The Secretary SOC, 21 Regent Terrace Edinburgh EH7 5BT (tel 0131-556 6042, fax 0131 558 9947). Scottish Birds, the official journal of the Scottish Ornithologists’ Club, publishes original material relating to ornithology in Scotland. Papers and notes should be sent to The Editor, Scottish Birds, 21 Regent Terrace, Edinburgh EH7 5BT. Two issues of Scottish Birds are published each year, in June and in December. Scottish Birds is issued free to members of the Scottish Ornithologists’ Club, who also receive the quarterly newsletter Scottish Bird News, the annual Scottish Bird Report and the annual Raptor round up. These are available to non- members at a subscription rate (1997) of £36. 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 21 Regent Terrace, Edinburgh EH7 5BT is the most comprehensive ornithological library in Scotland, and is available for reference during office hours (Monday to Friday, 0930-1630 hrs). (But please phone beforehand). SOC annual membership subscription rates Adult £18.00 Family (2 adults and any children under 18) living at the same address £27.00 Junior (under 18, or student under 25) £7.00 Pensioner/unwaged £10.00 Pensioner Family (2 adults living at the same address) £14.50 Life £360.00 Life Family £540.00 All subscriptions may be paid by Banker's Order and Covenanted. Subscriptions paid this way will greatly | | assist the Club. Published by the Scottish Ornithologists’ Club, 21 Regent Terrace, Edinburgh EH7 5BT Printed by Douglas Gauld & Co Limited, 22 Tannadice Street, Dundee DD3 7QH —— aed - — _ Scottish Birds (1997) 19: 65-85 65 The illegal persecution of raptors in Scotland A report from the Scottish Raptor Study Groups Extended summary This document uses data from the files of the Scottish Raptor Study Groups to give examples of the scale and widespread nature of human interference with raptors. Where the data are sufficiently detailed we quantified the levels of impact by comparing breeding performance and numbers in places where raptors are apparently unmolested with places where human interference has been shown to occur. In south east Scotland human interference occurred at both lowland and upland Peregrine breeding sites, but particularly at those on or adjacent to moorland managed for grouse shooting. At the lowland sites, recorded human interference involved the robbing of about one fifth of breeding attempts with an estimated loss of 19% of the production of young. At grouse moor sites, we recorded nest robbing, nest destruction and/or the killing of adults in about half of nesting attempts reducing potential production of young by at least 33%. Taking into account the reduced occupancy of sites, the estimated loss was 52%. At other upland sites, human interference involved nest robbing in about a third of first nesting attempts resulting in an estimated loss of 3% of overall production. In the whole study area, human interference was probably responsible for the loss of about 27% of potential production of young in the years 1990-96. At Peregrine nests in north east Scotland human interference was involved in failures at 8-22% of nesting attempts each year up to 1991, and has continued since. It was frequent on 2 estates leading to the loss of about 74% of the production of young. On 2 other estates persecution was less frequently recorded and was probably responsible for a loss of 12% in production. Persecution remained unrecorded on 3 further estates. The overall loss of production of young in the area, attributable to persecution, was at least 24%. No account could be taken of losses due to lowered occupancy or reduced population expansion resulting from low production and the killing of full grown birds. In central Scotland each year a similar number of Peregrine sites were 66 Scottish Raptor Study Groups SB 19(2) checked on keepered and unkeepered ground. There was little difference in the pattern of occupancy of these sites but a large difference in their productivity, those on keepered ground producing 38% fewer young than those on unkeepered ground. This was not because keepered ground was particularly poor in food but was due to the large number of breeding attempts that failed completely at 6 sites, suggesting intensive human interference with a third of Peregrine pairs in keepered areas. Were this the case, human interference was affecting about one fifth of the Peregrine breeding population in central Scotland, reducing overall production of young by about 18% in the years from 1981 to 1996. In the Highland Council area there is circumstantial evidence of the effect of poisoning on the distribution of breeding Golden Eagles. Illegal poisoning in the northern half of Badenoch and Strathspey coincides closely with a conspicuous lack of breeding eagles in suitable habitat that has held nesting pairs in the past. A similar gap in Golden Eagle distribution occurs in east Sutherland where there have also been recent cases of poison abuse. It is likely that between 10 and 20 Golden Eagle ranges are affected. In Tayside, the breeding performance of unmolested Golden Eagles.in 15 ranges was high. Performance was lower at 3 ranges due to egg robbing, and poorest at 14 ranges where other human interference (poisoning, trapping and shooting) occurred and at 5 ranges on moorland managed for Red Grouse. On the basis of breeding performance figures for unmolested ranges we might have expected an average of 21.6 eaglets to fledge in the Tayside ranges each year, but average production was 12.3, about 43% less. In north east Scotland, 21 Golden Eagle ranges have been monitored. Fifteen are on ground mainly managed for deer and are relatively successful (0.64 young/range/annum, 1990-96). The remaining 6 are on ground managed for Red Grouse with some plantation forestry and have poor production (0.26 young/range/annum, 1990-96). Here there was intermittent occupancy by Golden Eagles in immature plumage, poisoning and nest destruction, but at 3 ranges persecution has apparently ceased with pairs surviving to breed successfully. Assuming that all ranges could produce 0.64 young/range/annum, an average of 13.4 eaglets would have fledged each year from the 21 ranges in the area, but existing production of 11.2 eaglets was 17% less. Within the existing population there are also gaps where there is suitable habitat, suggesting that at least 2 more breeding pairs could exist if the population were allowed to build up. 1997 [llegal persecution of raptors in Scotland 67 In Argyll, the occupancy of Golden Eagle ranges was high and success reasonably good but some persecution occurred. In south Argyll over a 25 year period, up to 3 of 19 ranges suffered persecution in any one year. Had such human interference not taken place we would have expected 7% more successful breeding attempts (equivalent to 19 more eaglets fledged) during this time. This is a minimal figure as only known instances of persecution were used in calculations; suspected instances were not included nor were those ranges where pairs have disappeared. On 3 moors in south west Scotland, Hen Harrier breeding success and/or numbers increased after persecution ceased. Such an increase was at odds with trends on managed grouse moors elsewhere, where harriers were formerly present. By 1996, there was decline, in places virtually to extinction, in parts of the Outer Isles, in the Ladder Hills and the Cabrach/ Fiddich/Glass area (Grampian) and in Kincardineshire. The results are discussed briefly, an appendix catalogues examples of the persecution of Golden Eagles in Tayside and another details persecution at a communal roost of Buzzards and Ravens. Introduction In the 1950s and 60s, raptors were scarce in Britain as a result of past persecution and the effect of pesticide pollution (Newton 1979). Through the 1970s and 80s raptors increased as a result of the combined effects of special protection and the alleviation of pesticide poisoning. However, despite legal protection, the killing of raptors continues, including the poisoning, shooting and trapping of full grown birds and the destruction-of nests and their contents (da Prato & Laing 1996) leading to recent population declines in some areas. Persecution occurs mainly because some people believe that raptor predation interferes with their interests. Specifically, raptors are accused of reducing the abundance of gamebirds for shooting and of killing racing pigeons and other breeds of domestic pigeon. Though less common than in the past (Morton 1995, Ratcliffe 1993) there is also still a problem with the illegal removal of eggs by egg collectors and of eggs and young to be reared for captivity. The main raptor species concerned are Hen Harrier, Goshawk, Buzzard, Golden Eagle and Peregrine, but other species such as Sparrowhawk and Raven (not a raptor) are affected, as well as some of our rarest birds of prey such as Red Kite, White-tailed Eagle, Marsh Harrier and Osprey. The Raptor Study Groups in Scotland were set up to enable liaison between fieldworkers and other interested parties, mainly to co-ordinate and standardise the monitoring of raptor populations. All the Scottish Raptor Study Groups regularly record the illegal persecution and nest robbing of birds of prey. Instances are given in every annual review, the Raptor Rounaup, formerly in Scottish Bird News and 68 Scottish Raptor Study Groups now a supplement to Scottish Birds, and the Groups are convinced that such activities are widespread and even increasing. This paper gives evidence for this and attempts to quantify the problem for some areas and species. The impact of human interference on the Goshawk population has been assessed in the past by comparing their performance in areas with proven persecution compared with areas where, inthe absence of hard evidence, it was thought that there was none (Marquiss & Newton 1982). Though still scarce in Britain, Goshawks have slowly increased and more data on the effects of persecution are now accumulating (Petty & Anderson in progress). Intensive research on Hen Harriers (RSPB and Raptor Study Groups) also compared areas with and without persecution to estimate its effect on harrier populations. The work showed that the killing of adult Hen Harriers and the destruction of their nests was widespread on moorland managed for grouse shooting. This sort of moorland is good harrier habitat so the killing there created a ‘sump’ effect, recruiting many young harriers from elsewhere and effectively limiting the whole Scottish population (Etheridge et a/ 1997). In the present paper we also compare areas or nest sites with and without persecution, reviewing recent Peregrine and Golden Eagle breeding data from specific areas where it is detailed enough to allow quantitative investigation. We provide some notes on the localised recovery of Hen Harrier numbers after cessation of persecution on 2 estates in contrast to declines elsewhere. We append documents detailing the recent history of persecution of Golden Eagles in Tayside and an instance of persecution at a communal Buzzard and Raven roost. SB 19(2) 1 Peregrines in south east Scotland, east of the A74/M74 trunk road In this area there are 67 nesting territories or ‘sites’ recorded to date (1996). We have classified these broadly into 3 categories: (i) Fifteen lowland sites surrounded by a mixture of land uses but with a high proportion of arable land (ii) Twenty six upland sites that included un- keepered heather moorland, pasture and forestry ground, and (iii) Twenty six grouse moorland sites, surrounded by or adjacent to land managed by gamekeepers for Red Grouse shooting. The number of pairs recorded on nesting territories increased over the last 7 years from 31 pairs (and one single bird) in 1990, to 44 pairs (with 3 singletons) in 1996, but peaked in 1994 at 54 pairs (with 1 singleton). As the population increased more sites became occupied and were added to the inventory of sites to be checked. A few sites remained unchecked in some years including the peak year, 1994, but this was not the reason for the recorded decline, which was largely the result of a fall in the proportion of sites occupied (Table 1.1), particularly on moorland managed for grouse shooting. The protection afforded to birds on Buccleuch Estates ground meant that some new sites became occupied there but, elsewhere on managed grouse moor, 5 sites previously holding pairs on a regular basis became deserted or occupied only sporadically by single birds. Inthe other categories of nesting territories (lowland and upland) the pattern was for the occupation of new sites and continued occupancy by breeding pairs at already established sites, only occasionally broken by gaps of one or 2 years apparently associated with a change in birds. 1997 Illegal persecution of raptors in Scotland 69 Table 1.1 The occupancy of and breeding success at 67 nesting territories of Peregrines in south east Scotland 1990-96. Year Sites Occupied Percent Known Number of Total Young Failures visited by by occupied breeding pairs young per due to pair single bypair attempts successful pair human interfernce (%) Lowland 1990 7 7 0 100 6 6 Asia (1257, 1 1991 10 7 3 70 6 4 107; tr43 1 1992 9 9 0 100 8 7 18 2.00 0 1993 11 10 0 91 8 8 155-1 .50 0 1994 15 14 1 93 12 7 22 1.57 0 1995 15 12 0 80 11 10 S2ea2 oy 1 1996 14 12 0 86 12 8 2037167 2 Total 81 71 4 88 63 50 128 1.80 5 (8%) Upland 1990 14 12 0 86 12 7, Gut o3 3 ' 4991 14 14 0 100 14 8 2A Ae 50 5 1992 14 14 0 100 12 6 13-=0:93 0 13393 17 17 0 100 17 8 19). a2 5 1994 20 20 0 100 20 15 37. leo 2 1995 17 16 0 94 15 is 37 2.30 0 1996 18 18 0 100 18 13 34 1.89 1 Total 114 111 0 97 108 70 771.59 16 (15%) Grouse moorland 1990 16 12 1 75 12 3 10 0.83 8 1991 18 16 1 89 18) 5 2) 0.56 6 # 1992 20 15 2 75 11 6 14 0.93 6 | 1993 22 16 3 73 14 4 8 0.50 6 ‘| 1994 22 20 0 91 19 9 22 1.10 5 \ 1995 21 15 1 71 14 ZA 19 127 4 1996 23 14 3 61 14 8 26 =1.86 4 Total 142 108 4 76 97 42 108 1.00 39 (40%) 70 Scottish Raptor Study Groups The breeding performance of Peregrines also varied between categories of sites, with those sites on or adjacent to grouse moorland producing far fewer young than those elsewhere, associated with the greater incidence of human interference on grouse moorland. The evidence for this interference was usually the disappearance or destruction of nest contents coupled with ground signs of a human visit. The main problem in trying to quantify this fully was that many nest failures did not have accompanying hard evidence of interference, so we could not entirely disentangle persecution from natural nest failure. We tried to estimate breeding performance in the absence of persecution, including natural failures, by comparing the SB 19(2) performance of individual sites where there was overt evidence of human interference with those where there was none (Table 1.2). There were differences in the overall breeding performance at sites in the 3 land use categories, with lowland sites producing more young than upland and grouse moor sites. There were also major differences in breeding performance associated with recorded human interference and particularly poor performance at frequently persecuted grouse moor sites. The relatively small difference in performance between upland sites with and without evidence of human interference was attributable to the low frequency of such Table 1.2 The frequency of known human interference compared with overall breeding performance at 67 Peregrine breeding sites in south east Scotland. Data are from the years 1990 to 1996 inclusive, segregated into sites in the lowlands, in upland and on grouse moor. Number % breeding Breeding of sites attempts lost performance to known human interference mean (se) young per attempt Lowland interference 20 1.66 (0.41) interference not recorded 10 0 2.51 (0.24) Upland interference 8 29 1.63 (0.19) interference not recorded 18 0 1.69 (0.28) Grouse moor interference 16 51 0.45 (0.14) interference not recorded 10 0 1.67 (0.41) A two factor analysis of variance in breeding performance showed significant statistical effects of both land use (F=5.36,p<0.01) and whether or not interference was recorded (F=7.74, p<0.01) 1997 interference, mainly by egg robbings, there and to the fact that this was often followed by a successful repeat breeding attempt. At grouse moor sites the latter was infrequent; interference was rarely followed by a repeat attempt suggesting that second attempts were also destroyed or, more likely, that the adults had been killed. The poor occupancy atthese sites and, in particular, the complete absence of adult pairs at 5 of these sites in the last 3 years, suggested that adults were being killed repeatedly or that production was so low that it started to affect recruitment in the last 3 years. Wecan estimate the minimum effect of human interference by assuming that, in the absence of such interference, those sites where it was recorded would have produced as many young as those where it was not recorded (Table 1.3). Thus we would have expected 158 young to be produced at lowland sites as opposed to the observed production of 128, 183 at upland sites as opposed to 177 and 162 at grouse moor sites as opposed to 108. Illegal persecution of raptors in Scotland 71 This estimate is minimal because it does not take into account the effect of the killing of adults or the potential effects of low production on subsequent recruitment, both of which would have depressed population increase and reduced the occupancy of known sites by breeding pairs. Bearing this in mind, amore realistic estimate of the effects of persecution on breeding production might be derived by assuming that the occupancy and proportion of pairs laying clutches at grouse moor sites would have been similar to that recorded at upland sites where killing of adults was not recorded. There, laying pairs of Peregrines were recorded at 95% of sites checked, a much higher figure than that observed at grouse moor sites (68%). If this were the case, we would estimate atleast 135 breeding attempts at grouse moor sites between 1990 and 1996, with an overall production of at least 225 young. Observed production over these 7 years was only 108 young, so the loss due to persecution was about 52% of the potential. This would mean that in the overall Peregrine population of south east Scotland the production of 413 young was only 73% of the potential (566 young) in these years. Table 1.3. Minimal estimates of the loss in breeding production of Peregrines in south east Scotland due to human interference. Number of | Young produced Potential Observed Loss nesting per attempt in production production ascribed to attempts the absence of ofyoung ofyoung human 1990-96 recorded human interference interference (% of potential) Lowland 63 2.51 158 128 30 (19%) Upland 108 1.69 183 177 6 (3%) Grouse moor 97 1.67 162 108 54 (33%) 72 Scottish Raptor Study Groups In summary, the data on breeding Peregrines in south east Scotland showed that persecution or other human interference occurred in all 3 categories of site, but particularly at sites on or adjacent to moorland managed for grouse shooting. Atthe lowland sites, recorded human interference involved the robbing of about one fifth of breeding attempts at 5 of the 15 known breeding sites, leading to an estimated loss of 19% of the production of young. At grouse moor sites, recorded human interference involved nest robbing, nest destruction and/or the killing of adult birds in about half of the nesting attempts at 16 of the 26 known breeding sites. This resulted in the loss of at least 33% of potential production of young, but this figure did not take into account the reduced occupancy of sites, particularly inthe last3 years. Assuming a breeding pair occupancy figure close to that of upland sites would increase this estimated loss to 52%. At upland sites, human interference involved nest robbing in about a third of first nesting attempts at 8 of the known 26 breeding sites, resulting in an estimated loss of 3% of potential production. In the whole study area, human interference was probably responsible for the loss of about 27% of potential production in the 7 years 1990-1996. 2 Peregrines in north east Scotland In this area, as elsewhere in Britain, breeding Peregrines have increased, numbers more than doubling from 1981 to 1991 (Hardey 1991), though less so since then. Breeding success varied between years, apparently mainly influenced by weather as much of the population breeds in mountainous areas where late snow can influence laying and rain the survival of small young. Nevertheless nest robbing and the destruction of nests and SB 19(2) adults was also a major influence. Robbing of nests (for eggs and chicks) was less frequent in the 1990s than in the early 1980s, but nest destruction and the killing of adults has increased, being recorded at between 8% and 22% of occupied sites from 1981 to 1991 (Hardey 1991) and since then. Much of this human interference took place on areas of ground managed by only a few estates. From 1992 to 1995, annual monitoring of breeding has included at least 34 nesting sites on moorland managed for Red Grouse and 32 sites in other upland areas: deer forest, hill pasture and forestry. Each year these sites were checked to see if they were occupied, whether they were successful and, if so, how many young they produced. By comparison with south east Scotland, site occupancy was low (about 70%) and overall breeding performance was poor with no more than 1.14 young per occupied upland site (Table 2.1). The breeding performance on grouse moorland averaged about a third less young produced per occupied site, compared with other upland sites, possibly the direct result of the greater levels of human interference but alternatively it could have been merely due to poor food supply there. This alternative is unlikely because those nests on grouse moorland that were successful produced no fewer young than those at other upland sites in the same year (paired comparison over the 4 years 1992- 95; t=0:57, di=3) p=0m): One of the problems with this comparison was that human interference was associated with particular estates. On some estates it was not confined to grouse moors and on others it was not practised even on grouse moor. We therefore regrouped the data into those derived from the 7 main estates (Table 2.2): 1997 Illegal persecution of raptors in Scotland 73 Table 2.1. The breeding of Peregrines at sites on moorland managed for Red Grouse shooting compared with those at other upland sites managed for deer, hill farming and forestry in north east Scotland, 1992-95. Year Sites Sites Successful Young Young per Young per checked occupied fledged successful occupied (%) (%) nest site Grouse moor 1992 34 23 (68) 2 2(9) 5 2.50 0.22 1993 39 27 (69) 13 (48) 29 2.23 1.07 1994 35 24 (69) int 429) 13 1.86 0.54 1995 34 27 (79) 9 (33) Nd, 1.89 0.63 Total 142 POA es lois tea(S1) 64 2.06 0.63 Other uplands 1992 32 22 Fa(69)hati2 (65) 25 2.08 1.14 1993 34 25 (74) 12 (48) 18 1.50 0.72 1994 33 22 (67) 11 = (50) 20 1.82 0.91 1995 35 25 (71) 11 = (44) 27 2.45 1.08 Total 134 94 (70) 46 (49) 90 1.96 0.96 Table 2.2 The breeding success of Peregrines from 1992 to 1995 on 7 different estates in north east Scotland. Sites Number Number Number Land use Recorded checked occupied successful young per human per year per year per year occupied interference (%) (%) site A 5.25 4.00 (76) 0.75 (19) 0.19 grouse frequent Be3:75 3.25 (87) 1.00 (31) 0.69 grouse none C 5.00 3.75 (75) 1.25 (33) 0.80 grouse/deer some D 10.00 6.00 (60) 1.25 (21) 0.33 grouse/forestry frequent EN750 5.50 (73) 3.50 (64) 1.23 deer none F 8.00 6.00 (75) 1.75 (29) 0.96 grouse/deer some Gi 25 5.50 (76) 3.75 (68) 1.09 deer none There was no significant relationship between the level of human interference with Peregrines on an estate and the occupancy of sites there (ANOVA,F=1.153,p=0.4) but a significant relationship with the numbers of young produced (F=7.86,p=0.04). 74 Scottish Raptor Study Groups There was no significant relationship between the occupancy of sites on an estate and the level of human interference with Peregrines on that estate. However, there were fewer successful nests and far fewer young produced per occupied site on the 2 estates where persecution was frequent (0.26 young/ occupied site) compared with those 2 where persecution was infrequent (0.88) and those 3 where it was not recorded (1.00). Human interference thus probably reduced the production of young Peregrines by 74% on 2 estates and by 12% on another 2, equivalent to a loss of 33 young over the 4 years. There were 106 young produced on these estates in this time so the overall loss attributable to persecution was about 24% of overall potential production. In summary, recorded persecution at Peregrine nests in north east Scotland involved between 8% and 22% of nesting attempts each year from 1981 to 1991, and has continued since then. It was frequent on 2 estates leading to the loss of probably 74% of the production of young. On2 other estates persecution was less frequently recorded and was probably responsible for a loss of 12% in production. Persecution remained unrecorded on 3 other estates. The overall loss of breeding production in the area attributable to persecution was at least 24%. No account could be taken of losses due to lowered occupancy or reduced population expansion resulting from low production and the killing of full grown birds. 3 Peregrines in central Scotland Peregrines have been monitored in one area in central Scotland since 1978. At that time there were only 13 pairs recorded, all in the uplands, and 6 of them on ground managed for game, mainly Red Grouse. As monitoring SB 19(2) effort increased with the progress of the study and as the population grew throughout the 1980s, new breeding sites were added until by 1996 a total of 36 different nesting territories were located, 19 of them on keepered ground. The population peaked in 1991 at 29 pairs (15 pairs and 2 singletons on keepered areas) and fell to 15 pairs and 9 singletons by 1996. Before 1981, only a few sites were checked annually but afterwards to 1996, each year there was a sample of 10 or more eyries examined. The occupancy of breeding sites (proportion of sites occupied by a pair), the average brood size of successful nests and the overall productivity (number of young produced per pair) varied from year to year so these measures of Peregrine performance were compared for keepered and unkeepered ground within years using paired ‘t’ tests (Table 3.1). Keepered ground was defined here as an area where there was at least some element of grouse as opposed to other game keepering or deer stalking. Each year a similar number of sites were checked on keepered and unkeepered ground. There was no significant difference in the pattern of occupancy of these sites but a large difference in their productivity, those on keepered ground producing 38% fewer young than sites on unkeepered ground. This was not because keepered ground was particularly poor in food because there waslittle difference in the fledged brood sizes there. Poor production of young on keepered ground was principally due to the large number of breeding attempts that failed completely. Many such failures were known to be the result of human interference, including the destruction of nests and the trapping of full grown birds, but this was not recorded at all keepered sites. Some Peregrine pairs on 1997 lllegal persecution of raptors in Scotland 75 Table 3.1. A comparison of Peregrine breeding performance between sites on keepered and unkeepered ground in central Scotland (average annual values for the 16 years 1981-96). Keepered Unkeepered Paired ‘t’ P (se) (se) (df=15) Sites checked per year 13.6 13.6 ns Occupied by pair 1OK60e(0.6) ool (0.4) ns Occupied by single bird 2.1 (0.4) 1.8 (0.3) ns % occupied 79 (4) 82 (3) ns % successful 43 (5) 63 (5) 2.96 <0.01 Mean brood size 2.09 (0.19) 2.35 (0.11) ns Young/pair 0.88 (0.11) 1.42 (0.12) 4.09 <0.001 keepered ground appeared to have been totally unmolested. Examination of the overall productivity on a site by site basis showed that productivity was much more variable on keepered ground than elsewhere. Very low production at 6 sites was sufficient to account for the overall reduced production there, suggesting intensive human interference with only a third of Peregrine pairs in keepered areas. Were this the case, human interference was affecting about one fifth of the Peregrine breeding population in central Scotland, reducing overall production by about 18% in the years 1981 to 1996. 4 Golden Eagles in the Highland Council area west of the River Spey Information on the persecution of Golden Eagles in Scotland is reviewed in Watson (1997). The evidence for poisoning in the Highland Council area is inevitably patchy andincomplete. Inrecent years, the majority of cases of illegal use of poisons in upland areas have occurred in Badenoch and Strathspey and in east Sutherland. Maps presented in Watson (1997) on pages 34 and 229 provide strong circumstantial evidence on the effect of poisoning on the distribution of breeding Golden Eagles in both these areas. A scatter of confirmed incidents of illegal poisoning in the northern half of Badenoch and Strathspey, notably in the Monadhiiath mountains, coincides closely with a conspicuous lack of breeding eagles. This is despite the fact that in terms of food and nest site availability this area is suitable for eagles and certainly has held nesting pairs in the past. Media attention has been directed at this area (Cramb 1993) when individual estates on which poisoned eagles had been found were named. A similar gap in the Golden Eagle range occurs in east Sutherland where again there have been several recent cases of poison abuse. It is likely that 10 - 20 pairs of Golden Eagles continue to be ‘displaced’ by chronic persecution, notably poison abuse, in the eastern part of the Highland Council area. 5 Golden Eagles in Tayside Tayside has good habitat for Golden Eagles and could support a large and productive population. At present, between 24 and 37 nest territories are checked annually by the 76 Scottish Raptor Study Groups Tayside Raptor Study Group but many are occupied only sporadically by mainly immature eagles. This situation has persisted for along time and characterises persistent persecution (Sandeman 1957) which still continues (Appendix |). The population west of the AQ trunk road includes 21 eagle home ranges and has been monitored since the Golden Eagle national census of 1982. Breeding performance is poor (Table 5.1) but varies much between ranges; only 5 have produced 6 or more youngin 15 years. Twelve of the ranges have been occupied consistently by adult pairs, all of them on areas managed for sheep or deer. In total, since 1982 they have produced at least 0.37 young/range/annum. This was despite one range being robbed of eggs by collectors continually, another frequently and another once. If we exclude the sites known to have been robbed, production was at least 0.42 young/range/annum. Without egg robbing we would have expected another 9 eaglets to have fledged over the 15 year period. SB 19(2) Incontrast, breeding success on the remaining 9 ranges has been very poor with only 2 young produced over the 15 year period. This was mainly because the sites were intermittently occupied and usually by birds in immature plumage. Occasionally pairs would reach maturity and build nests but within a year or so were replaced by immatures. This suggested that mortality of eagles was high. Six of these ranges were on ground managed for Red Grouse shooting. Persecution of Golden Eagles or other raptors was recorded in 6 ranges, including 2 instances of eagles having been poisoned, one of a shot bird and one instance of the removal of young from a nest. Onanother range, adead decomposed eagle was found and at another there was hearsay rumour of an eagle having been poisoned. Had Golden Eagles been allowed to mature and breed unmolested on these ranges we would have expected production to be about 0.42 young/range/annum (as above) and 57 additional eaglets might have fledged in west Tayside over this time. Table 5.1 Breeding performance at Golden Eagle ranges in 2 parts of Tayside, 1982- 96, grouping ranges according to levels of human interference. West of A9 trunk road (live prey less abundant) unmolested ranges ranges robbed of eggs ranges where actual persecution was recorded East of A9 trunk road (live prey abundant) unmolested ranges grouse moorland ranges ranges where actual persecution was recorded Number Breeding performance of ranges (young/range/annum) 9 0.42 3 O37 9 0.02 6 0.80 5 0.08 5 0.41 1997 The population east of the AQ trunk road includes 16 home ranges and has also been monitored consistently since 1982. Breeding performance in ranges where there has been no human interference is good (Table 5.1), probably influenced by a more continental climate and the high density of prey such as Mountain Hare, Red Grouse and Ptarmigan (Watson 1997). However, the overall situation is poorer because of the high incidence of persecution (Appendix 1). Over the 15 year period 1982-96, 104 eaglets are known to have fledged from the area. At 6 ranges known to be free from human interference, the average production was 0.80 young per range per annum. In contrast, 5 ranges that included much ground managed for Red Grouse shooting had poor breeding success (2 ranges reared 0.20 young per range per annum) or were occupied by a succession of immature eagles and produced no young. The remaining 5 ranges (on ground that included both deer forest and grouse moor) were almost continuously occupied by breeding pairs but there is strong circumstantial evidence that these were subjectto direct interference by gamekeepers. Their breeding production was 0.41 young per range per annum. In 1988-89 the males in one range were poisoned and shot in respective years. In another range the pair has only succeeded in rearing young in the 2 years when the eyrie was on a high cliff where deliberate interference was difficult. Their favoured eyries _ were ona lower, less weather vulnerable cliff, | but easily accessible to disturbance. The breeding attempts here failed each year when the chicks were hatching or still required brooding. Failure coincided with the head ' keeper visiting the area to control Foxes. At _ another range the eyries were in 2 glens that Illegal persecution of raptors in Scotland 77 had different keepers. The pair always failed when nesting in the eastern glen where the keeper was known for his dislike of eagles. This man’s’ beat included another eagle range with a history of persecution incidents. Considering that the productivity figures for unmolested ranges in Tayside were 0.42 young/range/year (west) and 0.80 young/ range/year (east) we might have expected an average of 21.6 eaglets to fledge in Tayside each year, but actual production averaged 12.3. Our best estimate was that human interference reduced production by 43% in these years. 6 Golden Eagles in north east Scotland In the past, the fortunes of Golden Eagles in north east Scotland have varied, this variation being associated with land use and food supply but also with the impact of persecution (Watson etal 1989). In recentyears 21 ranges have been monitored. Fifteen of these are on ground mainly managed for deer, and these are relatively successful (0.64 young/range/ annum, 1990-96). The remaining 6 are on ground managed for Red Grouse with some plantation forestry and have poor production (0.26 young/range/annum, 1990-96). These 6 ranges were for many years characterised by intermittent occupancy by eagles in immature plumage. At one range 2 poisoned eagles have been found and on one occasion when a pair survived for at least 2 years, 2 nests that had been built were destroyed. No young have been produced and in 1996 the range was occupied by 2 immature birds. At another 2 ranges there has been no record of a pair in recent years; only single immatures were seen and a dead decomposed eagle was found concealed ina 78 Scottish Raptor Study Groups peat hag. At the other 3 ranges persecution has apparently ceased with pairs surviving to breed successfully. On 2 of these ranges the keepers are well known for their positive attitude towards raptors. On the third, the birds may still be vulnerable as this range spans 3 estates. lf we assume that Golden Eagles on these 6 ranges would have been atleast as productive as others in north east Scotland (0.64 young/ range/annum) we can calculate that an average of 13.4 eaglets would have fledged each year from the 21 ranges in the area. Existing production of 11.2 eaglets is 17% less than this potential productivity. Within the existing population distribution there are also gaps where there is heather moorland suitable for Golden Eagles with potential nest sites, suggesting that at least 2 further breeding pairs might exist if the population were allowed to build up. 7 Golden Eagles in south Argyll In this area Golden Eagles have been monitored since 1964. In Argyll, the recorded range occupancy is high and success reasonably good; at least 85% of monitored ranges have been occupied in the 1990s and about 38% of pairs have produced 0.52 young/ pair/annum. However, persecution persists at least in south Argyll where monitoring has been intensive and over the last 25 years, up to 3 of the 19 ranges suffered persecution in any one year. The proportion of pairs producing young varied from year to year but overall was 0.53 in the 11 years when persecution) was recorded, about 14% less than the figure of 0.62 in the other 14 years (Man-Whitney ‘U’ test, Z=2.11, p=0.035). Had such human interference not taken place we would have expected 7% more successful SB 19(2) breeding attempts, equivalent to 19 more eaglets fledged during this 25 year period. This is a minimal figure as only known instances of persecution were used in the calculation. Other suspected instances were not included. Moreover, the calculation only includes breeding attempts, ie those where birds were known to have laid eggs. Ranges where pairs have disappeared are not included. At one range the eagles were successful in 9 years from 1970 to 1981, but, in 1983 the chick was removed and in 1984 the incubating bird was seen to be put off by a shepherd during a snowstorm. In 1986 a dead female eagle was found on the nest, andamale nearby laterthe same year. Since then there has been no nest building on this range and only occasionally have birds been seen. At another range eagles started to breed successfully but, following allegations of lamb killing, achick was found dead in the nest and, in 9 subsequent breeding attempts, the eggs were deserted, broken or removed. Ground immediately adjacent to the nest was often set on fire and one nest on bare rock was deliberately burned. Finally, after a winter when there was hearsay evidence that 2 eagles had been poisoned, the range became vacant; despite frequent sightings of birds in immature plumage a breeding pair has not become reestablished. At another range the breeding success of eagles was related to the locations where they nested. When nesting on an estate managed for sheep, breeding was successful in only the first year. The next 4 years’ attempts failed as a chick was shot on the nest and eggs were broken or removed. The pairthen moved to nesting on afforested land and ona sheep farm with a sympathetic owner and they were successful in 11 of 12 years. | Y 1997 At 2 other traditional Golden Eagle ranges, birds have failed to establish adult breeding pairs in recent years and persecution is suspected. Where this occurs it is difficult to disentangle the effects of persecution from land use change, principally afforestation, so the full impact of persecution in Argyll, and over the whole of Scotland, will remain unquantified until it stops and breeding eagles achieve their full potential distribution and breeding success. . 8 The recovery of Hen Harrier numbers following the cessation of persecution in contrast to overall | decline | We can gain some idea of the effect of the removal of raptor persecution when we | consider how rapidly Buzzards became - abundant following the replacement of poisoning by Larsen trapping (a use of legal cage traps with decoy crows) as the main method of killing crows after an anti-poisoning campaign by SOAFD and the successful prosecution of keepers. Buzzards are nowa common sight in most parts of Scotland, and may ultimately reach the sort of abundance | that is familiar to anyone who has travelled abroad in Europe. The Scottish Raptor Study Groups have mainly concentrated on the _ scarcer birds of prey and have not monitored _ Buzzards systematically over large areas. _ There are, however, good records ofincreases of the much scarcer Hen Harrier subsequent || to the cessation of persecution. _ On an estate in south west Scotland, Hen Harriers were monitored from 1984 onwards but suffered complete breeding failure. In 1988, the RSPB, following representations | from the South Strathclyde Raptor Study Illegal persecution of raptors in Scotland 79 Group, wrote to the owner regarding persecution and a meeting was held. The situation immediately improved for harriers until the estate owner died in 1992 and shooting on the estate came under new management, whereupon relations between keepers and RSPB and the Raptor Study Group significantly deteriorated. Between 1984 and 1988 there were one to 3 female harriers present each year and only 4 out of 11 nests were successful, with recorded persecution incidents including a shot female found dead on the nest and a dead female which had been poisoned with alphachlorolose. From 1989 onwards numbers improved to peak at 9 females (5 successful nests) in 1992. In 1993 no nests produced young, one did so in 1994 and none since then. Numbers of females attempting to breed on the estate initially remained the same but had fallen to only 4 by 1996 (Table 8.1, moor A). On another estate in south west Scotland, also following representation from the RSPB, persecution apparently ceased in 1990. At one moor (B) prior to 1990 there were no more than 5 females and overall breeding success was low (one successful nest from 11 attempts). After 1990 success improved immediately (4 out of 7 attempts in 1990-92) and numbers of females increased to 11 in 1994 and 14 in 1996, most of which (85%) produced young. Atanother moor (C) on this estate, the fortunes of harriers also improved after 1990. Before 1990 no more than 3 females were present in any one year and only 4 out of 7 nests produced young. After 1990, numbers increased to 10 females in 1994, 6 of which produced young. In 1994 there was some persecution on nearby ground with the destruction of several nests together with the 80_ Scottish Raptor Study Groups SB 19(2) adult females. Harrier numbers overall fell in 1995 to 5 females which bred successfully in that year, but poorly the year after (2 out of 5 nests were successful in 1996). In each of these case histories, Hen Harrier breeding success and/or numbers markedly increased after persecution ceased. Such increase was at odds with trends elsewhere on managed grouse moors where harriers were formerly present. For example, there has been a decline in part of the Outer Isles, in the Ladder Hills and the Cabrach/Fiddich/ Glass area (Grampian) andin Kincardineshire (Table 8.1). In the latter county there were at least 15 nests found in 1974 when there was relatively little persecution (Picozzi 1978) but as that study ceased and persecution increased in the surrounding area, there were only 3 nests in 1988/9 (Bibby & Etheridge 1993). A thorough search of all moorland in 1995 and 1996 failed to turn up even a single nest. Discussion and conclusions The present document lists only some of the persecution incidents from the records of members of the Scottish Raptor Study Groups. The main purpose of the paper is to relate instances from study areas where the impact of persecution can be assessed. We have shown for 3 Peregrine study areas that persecution is by no means rare. The impact of robbing by egg collectors and others appears to have waned, whereas killing of full grown birds and the destruction of nest contents by game preservers seems to have been maintained or even increased in the 1990s. It seemed that in south east Scotland about half the Peregrine breeding sites on grouse moor suffered persecution with an estimated 52% loss of productivity. In north east Scotland up to 74% of Peregrine productivity was lost on 2 estates where persecution was commonest. The equivalent figure for Table 8.1. The increase in number of Hen Harriers on 3 moors following cessation of persecution in contrast to declines elsewhere. Area Females and/or nests (years) Southwest moor A* 1-3 (1984-88) 9 (1992) 4 (1996) moor B** 5 (pre 1990) 11 (1994) 14 (1996) moor C*** 3 (pre 1990) 10 (1994) 5 (1995/96) Outer Isles 3-4 (1988/89) 0 (1995/96) Ladder Hills *** 10 (1993) 4 (1996) Cabrach/Fiddich/Glass 11 (1994) 4 (1993) 0 (1996) Kincardineshire 15+ (1974) 3 (1988/89) 0 (1995/96) t Persecution ceased in 1989, but recommenced in 1993 hy Persecution ceased in 1990 mae Most persecution ceased about 1990 1997 [llegal persecution of raptors in Scotland _81 Peregrines on keepered ground in central Scotland was 42%. In any one region onlya proportion of the Peregrine population was vulnerable so overall losses were less, though not trivial; 27% of production in the south east, 24% in the north east and 18% in central Scotland. The impact on the overall population is difficult to estimate because the killing of adult Peregrines not only reduces production but drains the pool of potential recruits, preventing population increase and perhaps, at present, causing population declines. The proportion of occupied sites has declined in at least some habitats in all 3 areas. The fortunes of the Golden Eagle population are even more affected by the loss of full grown birds because eagles take so long to mature and start breeding. In the 3 areas examined, the loss of productivity due to egg collectors was small (an estimated value of 12% in west Tayside) compared with the unquantifiable losses due to the failure of adult pairs to establish successful breeding on some ranges. This affected 9 (42%) of ranges in west Tayside, 3 (14%) of ranges in north east Scotland and 3 (16%) of ranges in south Argyll. These are minimal estimates because gaps in the existing Golden Eagle distribution, where there is suitable habitat but no record of pairs, suggest that the breeding population might be much larger were it to be allowed to build up. At present, persecution on some ranges is continuously removing the potential recruits that might otherwise settle and breed in new areas. Intensive studies of Hen Harriers (Etheridge et al 1997) has shown that this ‘sump’ effect is real, and is sufficient explanation for the speed with which numbers increased following the cessation of persecution in our examples. A further example of a potential persecution ‘sump’ for raptor populations is the killing of birds at communal roosts (Appendix 2). Anyone who travels widely abroad soon becomes accustomed to the ubiquity of large birds of prey in many countries but, until persecution ceases in Britain, we are unlikely to become accustomed to it here. Acknowledgements The information used to compile this report was collected by members of the Scottish Raptor Study Groups and is in itself a tribute to the skill, hard work and dedication of these voluntary fieldworkers. We also thank the RSPB for its cooperation in providing ancillary details of certain persecution incidents. Much of the fieldwork was done under licences to disturb Schedule 1 species of birds issued by Scottish Natural Heritage. We are grateful to the many landowners, their tenants, agents and employees for freely giving access to the ground under their jurisdiction and for information and helpful advice. Financial help was given by Scottish Natural Heritage to some Raptor Study Groups to reimburse travel expenses. Finally, we thank The Scottish Office Agriculture, Environment and Fisheries Department which encouraged us to prepare this document and made a financial contribution to its production. References Bibby C J & Etheridge B 1993. Status of the Hen Harrier Circus cyaneus in Scotland in 1988-89. Bird Study 40:1-11. Cramb A 1993. The mountains where eagles die. The Scotsman 11 March 1993. Etheridge B,Summers R W &Green R E 1997. The effects of illegal killing and destruction of nests by humans on the population dynamics of the Hen Harrier Circus cyaneus in Scotland. Journal of Applied Ecology 34:1081-1105. Hardey J 1991. Increase in the number of breeding Peregrines in North-East Scotland,1981- 1991 .North East Scotland Bird Report 1990:56-61. 82 Scottish Raptor Study Groups Marquiss M & Newton | 1982. The Goshawk in Britain. British Birds 75:243-260. Morton K (ed)1995. Raptor Roundup 1994. Scottish Bird News 38:8-12. Newton | 1979. Population Ecology of Raptors. T&AD Poyser,Berkhamsted. Picozzi N 1978. Dispersion, breeding and prey of the Hen Harrier Circus cyaneus in Glen Dye,Kincardineshire. /bis 120:498-509. da Prato S &Laing S 1996. Facts about raptors. SB 19(2) Scottish Bird News 41:6-9. Ratcliffe D 1993. The Peregrine Falcon(2nd edition). T.&ADPoyser,London Sandeman P W 1957: The breeding success of Golden Eagles in the southern Grampians. Scottish Naturalist 69.148-152. Watson A,Payne S & Rae R 1989. Golden Eagles Aquila chrysaetos:land use and food in northeast Scotland. /bis |I3| :336-348. Watson J 1997. The Golden Eagle. T&AD Poyser, London. The Scottish Raptor Study Groups c/o P Stirling-Aird, Kippenross, Dunblane, Perthshire FK15 OLQ Appendix |. Examples of persecution of Golden Eagles in Tayside There have been a number of high profile cases of Galden Eagle persecution in Tayside and adjoining areas in recent years. Bearing in mind the small chance of detecting such incidents, these examples probably represent the tip of the iceberg. Shooting 1. On 21 April 1988 an adult female Golden Eagle was found dead concealed among rocks near its nest a few miles east of Ben Lomond. An X-ray showed that it had been shot with a shotgun at very close range. The nest was later found to contain 2 cold eggs. The police interviewed the local game keeper who had been threatening to ‘get the eagle.’ 2. On7 October 1989 a group of 4 hill walkers was approaching the summit of Glas Maol, Glenshee, when they found the corpse of a recently dead Golden Eagle. The bird was retrieved by RSPB investigators and found to be an adult male which had died due to being struck by a single shotgun pellet which had lodged in its windpipe. Despite a police investigation no culprit was ever identified and no court proceedings were ever initiated. This bird had died on a National Nature Reserve. The only shotgun activity for many miles in any direction is vermin control connected to grouse shooting and grouse shooting itself. In April of the previous year the adult male in the same home range had been found poisoned on its nest (see below). In all probability the shot male was the replacement for the poisoned male. 3. On 31 May 1994 a young Golden Eagle was found dead close to the reservoir in upper Glen Lednock. This bird had also been shot. The ring it was carrying confirmed that it had fledged from an Argyllshire nest in the previous Summer, a good example of the fate awaiting birds moving into Tayside from the apparently impoverished but revealingly more productive deer forest and sheep walk territories to the west. 1997 A police investigation was again initiated but no culprit was found. 4. On30 May 1995 a RSPB contract worker carrying out an agreed visit to an estate in Glen Clova discovered the body of an eagle. On X-ray this bird was found to contain over 30 pieces of shot. A post mortem revealed that the bird had almost certainly been shot on the ground, perhaps after an initial injury. An immediate police investigation followed leading to several members of the public giving information about previously unreported persecution incidents in the area. Noone was ever charged over the shot eagle. The reaction of certain landowners was to attempt to suggest that the bird had been either atame one, or thatit had been “planted” to discredit some unnamed person. There was no evidence to back up either of these claims. Poison Poisoning still appears to be endemic in Tayside with SOAEFD, RSPB and Tayside Police aware of dozens of confirmed incidents of deliberate wildlife poisoning which have occurred in or close to Golden Eagle home ranges in the last 2 decades. The most commonly recorded victims are Buzzards, which share a liking for carrion with eagles. Golden Eagles, however, are not usually present in such accessible areas as the more ubiquitous Buzzard so the chances of finding a poisoned eagle are reduced. 1. On 1 April 1982 a contract fieldworker found one of a pair of breeding eagles poisoned by alphachloralose within a mile of its built up nest. This was at a remote location at the extreme west of Tayside at the junction | llegal persecution of raptors in Scotland 83 of 3 estates. Allestates denied any knowledge but a local shepherd/estate worker admitted to the finder of the dead bird that he used the chemical against crows. This remains the only recorded instance of persecution or poisoning anywhere near the location but it may be significant that this was the last known nesting in this home range. 2. On 14 April 1988 a Tayside Raptor Study Group member, carrying out a routine check on Golden Eagles in Caenlochan National Nature Reserve, found an adult male lying dead on its nest. After a difficult climb, he retrieved the bird which was sent for analysis. It was found to have been poisoned with alphachloralose. A field search by RSPB staff 2 days later turned up a dead hare laced with alphachloralose on a grouse shooting estate neighbouring Caenlochan. Again, all was reported to the police who interviewed the landowner and game keeper of this estate but no charges were brought. 3. In early June 1989 an adult female eagle was retrieved dead from its nest near Drumochter a few kilometres outside the Tayside area. Alsointhe nest were its 2 dead chicks.Post mortem chemical analysis confirmed that the bird had been poisoned with alphachloralose. This was reported to the police and publicised. The reaction of the landowner was to try to have the NCC withdraw the Schedule 1 disturbance licence of the RSPB contract worker who had retrieved the bird after watching the nest from a neighbouring estate by telescope. The landowner claimed that the licence was invalidated because access permission had not been granted, but how do you disturb a dead eagle and 2 dead chicks? 84 Scottish Raptor Study Groups Trapping As with poison, the use of baited traps, usually gintraps, to killeagles has along history, with many references to this practice in the literature from the early 19th century onwards. Also in common with poisoning, the practice has not died out as some may claim. 1. Following a report from a hill walker on 13 May 1986, 2 RSPB investigators located a baited gin trap ata moorland pool (a ‘drowning set’) in an eagle’s territory at Glen Tarf, near Glen Tilt. The illegal trap had caught a harmless and legally protected Black-headed Gull, no doubt unintentionally, the traditional targets of this trapping method being Fox, Wildcat and Eagle. The illegal trap was reported to the police. When approached by the RSPB, the estate’s representative said ‘it must have been an older keeper showing a young one the old methods.’ 2. On 12 May 1991 anarticlein The Observer carried a picture story on a baited gin trap found set above Glen Tilt. On being interviewed the landowner’s comment was to the effect that the incident was regrettable but understandable. The police were again informed but no charges were brought. 3. On 1 June 1991 a walker from Brighton, again in Glen Tarf, off Glen Tilt, found a live Golden Eagle dragging a gin trap behind it through the heather. Onhis return to Pitlochry he tried to report the matter but failed to leave site details or a forwarding address with the SSPCA before disappearing. Over 2 months later he wrote to RSPB HQ in England to ask if anything had been done! An immediate field investigation was carried out by Tayside Police and the RSPB but a search of Glen Tart failed to find the presumably by now long dead bird. Again the estate was investigated SB 19(2) by the police and again no charges were broughtfor lack of evidence. The estate asked for more ‘cooperation between the RSPB and the keepers.’ 4. On 20 January 1996 a landowner on a small estate which was actively running down its game management, phoned the RSPB to say that a dead eagle had been found in one of its crow cage traps. The bird had been ringed as anestling in 1995, inanest that had been subject to an intensive nest watch. The site had been previously plagued by egg collectors and a massive effort by some 100 local volunteers had resulted in this chick fledging, the first successful nesting at this site in 9 years. This bird’s death by starvation, due to the neglect of a legal trap, caused great concern. Crow cages could be officially monitored to prevent such incidents but they remain. unregulated. A police investigation found no evidence of deliberate wrongdoing. 5. Finally, in July 1982 a recently fledged juvenile eagle was found in Glen Tilt with a badly broken wing and the main primary feathers missing from the other wing. It is difficult to imagine how this damage might have occurred naturally and human interference was suspected. Considerable force is necessary to extract these feathers and the broken wing was consistent with a foot having been braced on it to give leverage to pull feathers from the other wing. Concluding remarks This list of incidents for Tayside could be repeated for other parts of Scotland, and incidents are listed and published as a matter of course by the RSPB. The main point to be 1997 made is that such occurrences are still frequent. No knowledgeable person doubts that illegal persecution of raptors continues, or that it is mainly associated with game preservation. It is, however, difficult to prove in court which individual is responsible in any one case. Appendix 2. Persecution at a Buzzard and Raven communal roostin Tayside Some raptors and Ravens gather to roost communally, and, undersuch circumstances, are particularly vulnerable to persecution. Persecution can kill many birds in ashorttime and these birds are drawn to the roost froma very wide area; thus the activities of one persecutor can affect a substantial part of the overall population of the species in question. It is thought that Red Kites and White-tailed Eagles were exterminated in Scotland more easily than some other predatory birds because, although they are wide ranging, they often congregate to roost or exploit local food abundance, where they become vulnerable to persecution. It is therefore of some concern that persecution at communal roosts still occurs; it is suspected at some Hen Harrier roosts and is documented here for a Buzzard and Raven roost. The roost accommodated up to 12 Buzzards and 70 Ravens at any one time and was occupied from atleast 1988. On 23 December Illegal persecution of raptors in Scotland 85 1990, shots were heard from the roost wood and regular inspection of the site from then onwards in January 1991 discovered a concealed dead Buzzard and 4 dead Ravens lyinginthe open. The roostwas only occupied intermittently for the next 4 years, perhaps because of the persistent shooting taking place there, but numbers built up in 1995 when intensive shooting started again; a Buzzard injured by shot was found in October 1995 and another dead one in October 1996. The overall impact of such shooting is impossible to quantify. The small glen where the roost is situated is very good Buzzard habitat and could support several breeding pairs but none have been found there. Appendix 3. Scientific names Red Kite Milvus milvus; White-tailed Eagle Haliaeetus albicilla; Marsh Harrier Circus aeruginosus; Hen Harrier Circus cyaneus; Goshawk Accipiter gentilis; Sparrowhawk Accipiter nisus; Buzzard Buteo buteo; Golden Eagle Aquila chrysaetos; Osprey Pandion haliaetus; Peregrine Falco peregrinus; Red Grouse Lagopus lagopus; Ptarmigan Lagopus mutus; Black-headed Gull Larus ridibundus; Raven Corvus corax, Mountain Hare Lepus timidus; Fox Vulpes vulpes; Wildcat Felis silvestris. Manuscript accepted September 1997 Publication of this paper has been accelerated as the information is of direct value to the current debate on raptors. We thank the George Lodge Fund for funding an extra Scottish Birds which will be published in March 1998. 86 Scottish Birds 19: 86-92 SB 19(2) Movements of Fulmars from the Firth of Forth JOHN C DAVIES Recoveries of Fulmars ringed as chicks in the Firth of Forth were concentrated on the shores of the Firth of Forth, on the East Coast and the Low Countries of the North Sea. Other birds were recovered in the English Channel, the Celtic and Irish Seas and the North Atlantic. One third of the birds were recovered in their first year. Immature birds appear to return to British waters in their fifth and sixth years. Introduction Fulmars Fulmarus glacialis first bred in the Firth of Forth on the cliffs at Tantallon in the 1930s. They now breed at many other sites including the Isle of May (first bred 1930) and Inchkeith (1948), (Harris et al 1987). The Fulmars on Eynhallow, Orkney have been the subject of a long term population study (Dunnet 1991) and recoveries of Fulmars ringed in Britain up to 1975 have been analysed by Macdonald (1977). Most of the Fulmars ringed in the Firth of Forth have been ringed on Inchkeith by the Lothian Ringing Group and its predecessor, the Edinburgh RG. Fulmars have also been ringed on the Isle of May by the Isle of May Bird Observatory andat Tantallon by individual ringers. This paper presents the results of this ringing and updates a previous report by Poxton (1981). Method of analysis This paper is based on 109 recoveries of birds ringed as chicks during the first fortnight of August reported up to the end of 1995. All but 8 of the recoveries refer to dead birds. The birds are assumed to have fledged by 1 September each year, although some will not have fledged as early as this, and the year of recovery is from 1 September to 31 August of the following year. The 4 seasons, autumn, winter, springand summer referto the months September to November, December to February, March to May and June to August respectively. Since the Fulmar is a pelagic bird, except when breeding, the place of recovery has been allocated to sea areas rather than countries or regions. East Coast refers to the east coasts of Scotland and England; the Low Countries are the North Sea coasts of France, Belgium, the Netherlands and Germany; Scandinavia is the North Sea coasts of Denmark and Norway and the Skagerrak. The NW Approaches are the Malin Sea, the Minches and the Sea of the Hebrides (Hardisty, 1990). North Atlantic includes the Faroes and Newfoundland, Canada. Results The temporal and spatial distributions of 1997, Movements of Fulmars from the Firth of Forth 87 recoveries are shown in Tables 1 and 2 and are mapped in Figure 1. Nearly a third (31%) of all recoveries were from the Firth of Forth and over a third (85%) of the birds were recovered in their first year, mostly within 2 months of fledging. Other recoveries in the first year came from the East Coast, the shores of North Sea and the North Atlantic. The number of recoveries reduced considerably in the second, third and fourth years, again with recoveries from the shores of the North Seaand North Atlantic. Inthe fifth and sixth years recoveries increased with birds from the English Channel, the Celtic and Irish Seas and the NW Approaches as well as the North Sea and North Atlantic. The majority of birds were recoveredin spring andsummer. Few were recovered in autumn or winter with the exception of the autumn of the first year. Figure 1a Recoveries of Firth of Forth Fulmars Years one to 4. Firth of Forth 20 birds Plus 2 birds at sea: 1 off Newfoundland & 1 off NW Spain 88 John C Davies SB 19(2) Figure 1b Recoveries of Firth of Forth Fulmars Year 5 onwards. 15W 12W 9W 6W 3W The finding details of recoveries were divided into 4 main categories shown in Table 3. Almost two thirds (65%) of the birds recovered were simply found dead on the shore. Sick and injured birds (11 %) included birds which subsequently died or were taken into care with unknown results. Oiled birds (8%) could have been involved in an oiling incident or become contaminated after dying. Fulmars suffer insignificant losses from chronic oil pollution except maybe in the southern North Sea (Stone et a/1995). Included inthe caught or shot category were 8 birds (out of 17) caught and released again by ringers at other colonies; 2 local movements (Tantallon to Firth of Forth 14 birds 0 3E 6E SE 12E Inchkeith and Inchkeith to Inchmickery); one to the Farne Islands; 3 to Hunstanton, Norfolk; one to lle de Rouzic, Brittany, France; and one to the Copeland Islands Northern Ireland. The median age of these birds was 11 (range 6 to 20 years ). There were also 11 recoveries of birds ringed as adults in the Firth of Forth. They were recovered between one and 14 years after ringing (median 4 years). Four of the recoveries were within the Firth of Forth, 5 within the North Sea, one in the Faeroes and one in the Barent’s Sea. There were 5 89 Movements of Fulmars from the Firth of Forth 1997 jseo9 jsez seat yjog ‘papnjoul jou Aia@a0de/ BUD *8\0U}004 %001 Pk KE KC RR ML NG *L RL HL HEL %*YI RE Kr wry GE abejueaed %001 801 L € rd € L G 8 8 8 Ck eZ € 14 v BE je}O] %9 9 L lL L I rA SUI} YUON WI lL l sayoroiddy AAN %G G L L L C B98S YS %v v lL re I B9S IID Sp v Lo ee, L jauueyd ysijbuz %G G L L L C BIABUIPUBIDS ALS €2 lL L L I L Cesc I lL L LL SalJUNOD MO7 ve 92 IL re L IL € ac L Gm ec re G jseOd \sey VLE ve re Ee OG 5G L lL re 02 YOY jo yy JUBDIBdg je}0 | MaAodeyY JO eal WISE UL EL USL ULE UOL UIE YS YZ YO WG Up PJE pug js} AW¥SAQ034 AO YHVSA "SILWIIN4 YO JO YL JO As9AOD9, JO Bale Puke JedA | ZIGeL SB 19(2) 90 John C Davies Jawuwns “eat yiog ‘papnjoul jou Ala@A0Ias BUC *34OU}OO 4 KOOL Mb VE %S VE Mk KG HL ML WL MEL %I RE wy Rr %GE obejusoled %00b 80L [> (Gre So © yk eg 8 G5 8 . Cla. Zw Cen Poe 1km from the nearest known nest site. Until mid April a few frequented ski ar parks and buildings at 620- 670m, outside B. A tape-recorder there at 0520 on 13 April 1975 revealed much song during a snowfall (A Archer-Lock in /itf). All birds seen in May were paired or single, except for small flocks on days of fresh snow in early May. Before nesting began, | saw no birds on A during many ski trips on days in May and early June when fresh snow covered all vegetation and soil there. Birds seen outside A were on land where snow cover was incomplete due to wind exposed terrain or lower altitude, mainly on B but also on Derry Cairngorm (Fig 1). Some on B ate picnic scraps at a cafe at 1090m, at summit cairns, and along footpaths and vehicle tracks much used by people (Table 1). Others foraged on reseeded patches, where ground bared by machines or trampling had been treated with grass seed and fertilizer to reduce soil erosion. They ate ungerminated seeds from new reseeding, as well as seeds produced by grass from earlier reseeding In May and June before nesting, most days were mild, with many snow free patches of vegetation and bare soil, and thawed upper soil horizons on most ground. Birds fed on scraps, reseeded patches and other habitats. Habitat use by Snow Buntings in Scotland spring to autumn 107 On warm days | saw some eating insects on snow. They often took cranefly larvae (Tipula spp) on thawing bare soil beside snow. In such conditions, many larvae came to the surface to escape drowning. The upper soil horizons had become waterlogged, due to thaw water being unable to percolate downwards because of frozen lower horizons. Foraging of adults with nestlings and fledglings Food taken by cocks to hens on eggs, and by both sexes to nestlings and fledglings was mainly craneflies, but also other flies, beetles, and moths. In amass influx of aphids | never saw adults get insects so easily (Watson & Stroyan 1984). Foraging adults often ate insects and grass seeds. Many ate cheese, bread, and cake, but seconds later gathered insects for their young. | saw no adult feed picnic scraps to yqung, except in 1996 and 1997 on Cairn Gorm with nestlings >8 days old and fledglings. | saw no dependent young find and eat scraps for themselves. Fledglings scattered within 100m on the day of nest departure, each in a hole in boulders, and came to the entrance to be fed. Later they came into the open, but ran or flew to boulders when disturbed, and sheltered under them in rain, sleetand snow. Upto 5days out, they often gave a food cheep even when no parent was present, but after a week they did this only when a parent was seen or heard nearby. Each fledgling churred when about to be fed, and up to 10-12 days when finding food for itself. At 15 days, fledglings occasionally gave the food cheep from boulder tops, but | saw none fed after this age. Most young foraging in the first week were at edges of boulder patches, but in the second week they went also to other habitats. SB 19 (2) 108 Adam Watson "L00°0 > d ‘}S9} jOexe Jaysi4 Aq SI g SA , jo uosuedwod *10000'0 > d ‘aUO SlWa1}xXe BJO B JO BNJEA paAJasgo oy) Hulule\qoO Jo SaiIqeqoud uOSsIOg eAeinuuNd Aq sisAjeue 0} SJ9jo1 es@WINU \duoS -Jodns yore ‘sedh} yeyiqey pesn-jsow ay} Jo Seale ajqeeA 00} BUSY PUL ‘JBAOD MOUS BI|GeLeA 00} JO esNedeq ajqissodwi Sem Siu} SJOYM Z|} PU O| ‘py ‘€ SMOJ Ul }d90x9 ‘pe}se} SEM MOJ YOO U! UOIWOdOud \sebbiq au, “paj}se} sem odAj yeyIqey Siu} JO esn UBYyM Bale au} JO %E> pesudwoo MOUS ‘eale Apnys ey} JO %|>> ‘peidnooo eae ul ‘pesiudwod yokes - yoyed pepeese! pue ‘abpea pjalj-sapjnog ‘ebpe mous - sedAj yeyIqey jeseAeS UO SdeJdS 1UdIq "SOA|OSWAU} JO} Hues a19mM a1ay ynq ‘HBunoA pey YydIuM sinpy ® ‘soenoaig Japjnog ye Maj e SEpNjouj, ‘yyed e Episeq WI UBYOO| puke ‘syed ‘syny ‘Sued WINS jy ‘\/eymous Ysa. OU ‘HuNSeU s1Ojoq oUNL-AeW\# ‘|!e/MOUS YSe.j ‘HuI}SoU a1Oyoq ouNr-AeE-WW ~ 0 0 0 0) 0 0 ziee Zt 0 Ov OL 8 1840}90 0 ot 0 0 0 i) Z mee = 0 lz el= Oc Jequia\dag 0 v 0 0 0) Z 92 o1/ec 61 Gt Ze ‘Bny pue Ajnr BunoA juapuedapu! ¥ s}jnpY 0) QE /) 0 0) Z 0 vl 0 6c = Se sBuij6pey; g SINpe syxdO}4 6 0 6 al 0 g9F 0 6 0 €? 0 co c 199M - € 0 0 0 0 Lv6 0) 0 0 € 0 ve | 480M - BHuibes0) s6ulj6paj4 0) 961 Ol G ) G G 6L G 6 vl -te@Bunok ym ‘Buyee - 0) s6c 6 € 0 lL ) ve OO LZ € 99 sburjhpal Buipes, - 0) r6l Ol 9 rd rAd 0 6s Gt €7 p 2G sbuljseu Bulpes; - 0 e9I 9 OL rd rAd ra ol 6 dO v2 vi. ¥S she jseu UjIM - Z vl 0 0 0 0 Ol vl tb Zt zvye 436 6e~SCs« Su BuO Jaq - 0 0 0 0 0 0 Le 0 0) GZ iby ~=6 @E_~—Ss« JMU - S}INPY 4904 los yyeoy gy =_ , 8B pa yyeay 9 yoyed MOUS uo uo puejssei6b ~— pyja- yoyed puejssei6 ysdeios WH episeg SSOW SSOW yjoowS Japinog paepsesey ysni4 mMouS Aepinog dudiq U "16-996. Ul sleliqey Juasayip uo sp4iq Bulbesos so sBuljybis jo abejuacsad | ajqeL sseouen an cieedcoss é z Sa ee caer = 1997 Parents gathered invertebrates on snow, flushes and other habitats (Table 1). Nearly all habitats used had many boulders. Smooth ground with few or no boulders and with continuous vegetation covered 41% of Awhen it was snow free, and >30% of it on 1 July each year, when much of A was under snow. However, |sawno bird with a nestor fledglings on smooth ground, and none foraging on it except at edges within 10m of a boulder patch. The most marked absences were on big tracts of grassland (eg in the centre of basins at Coire Raibeirt and at Cairn Lochan’s Feith Buidhe, where the nearest boulder patches were >200m away). However there can be too many boulders where they formed the entire surface, with no ground vegetation. Boulder fields occupied 41% of A in snow free conditions, and >50% of snow free land on 1 July, but! saw no birds with nests or fledglings foraging there except near the edge. Most adults with nests and fledglings foraged on intermediate ground with many boulders (18% of A when snow free). Adults foraged much on snow patches (Table l). For example, in 90 minutes a cock with fledglings that could fly up to 30m after him made 18 trips on to snow, 4 to beside snow, _ 3toa flush, and one to a boulder-field edge. However, during the nesting and fledgling _ periods | saw none foraging in the middle of | | | | snowfields >50m from boulders. Snow free land had a brief insect peak. Snowbeds delayed its start and prolonged its duration. Land beside snow was frozen or waterlogged, but in a few days dried and warmed, leading to cranefly emergence. More snow had melted by then, so emergence lasted for weeks if the snow lasted. Foragers _ exploited this. Ground beside snow was much _ used by adults eating for themselves, cocks Habitat use by Snow Buntings in Scotland spring to autumn 109 feeding brooding hens, cocks and hens feeding nestlings and fledglings, and fledglings feeding themselves. Forexample, on8 August 1979 a hen and 2 fledglings 3-4 days out of the nest were beside a snowbed. She spent ail of a 30-minute watch gathering food ina band 0.5-2 m out from the snow, mostly craneflies on moist soil. The chicks often followed her for up to 100m, which saved her time, but found food for themselves when temporarily on their own. A day or two after nest departure, parents and young occasionally stayed within 200m of the nest, but often moved to a new place up to 1km away, where craneflies were relatively abundant. Even though nest sites lay near snow patches, frequently the snow there had gone by the time of nest departure, whereas the new places were beside big snow patches or in hollows that had held snow within a week. Interesting exceptions were in 1996 and 1997 on Cairn Gorm summit dome; adults took several broods up to 500m to bouldery grassland far from snow but near spots with picnic scraps which they ate and often fed to young. Parents gathered most food near the young, but sometimes afar, eg a cock flew 600m thrice in succession for craneflies, and a pair gathered insects beside snow and together flew >1km to day-old fledglings. M Marquiss (pers comm) watched a cock gather a beakful of insects and fly >|.5km, but the stage of breeding was unknown. Adults and dependent young in a flock Once chicks were getting most food for themselves, they followed parents, and adults stopped making foraging trips out of sight of the young. In 1977, a year of late first nests and no seconds (Watson 1996), 3 families flocked together. On 29 July, 2 cocks and 3 hens had broods 9, 10 and 15 days out of the 110 Adam Watson nest, in a loose flock though each hen and brood formed a unit. They fed in a 4-ha area and then in only 0.1 ha of bouldery grass, before flying in gloomy light at 2030 into a boulder field to rest. On 31 July the hens and young were ina 1-ha area about 300m away, again ina loose flock but acting as 3 units, and hens still fed both the later broods. Adults and independent young in late July and August All sightings in this period were on A, mainly on flushes in bouldery hollows. Some flushes issued temporarily from snow, but most were permanent, including rock-controlled ones, some of which were on low wet corries not seen to be used for nesting. A few sightings were of single adult cocks eating scraps at summit cairns, and insects and seeds. Adults and independent young, singly andin groups, foraged on snow more than breeders. On 23 and 26 August 19781 saw 2 cocks, 3hens and 5 juveniles from 2 broods run up steep snow in line abreast cim apart, eating insects, and on reaching the top of the 60m wide snowfield fly down to start again, and repeat the sequence often. | watched independent young eating moss capsules and grass seeds as well as insects. September Though! saw some moulting adults at the end of August, the main annual moult occurred in September, as judged by the appearance of the plumage (Watson 1957). All sightings were onA. Reluctant to fly, most birds flushed within 5m, sometimes almost at my feet, and flew <15m to vanish among boulders, but sometimes an adult passed overhead, flying well. Most were single (Table 2). In most cases, singletons were in bouldery hollows 1-3m SB 19 (2) wide, with a small flush. | saw none in flushes >5m from boulders, except for birds in flocks. About a fifth of sightings were in flushes below clifftops, and a few beside snow. Birds fed mainly on insects and seeds. Some singletons and small groups frequented summit cairns and tors visited by many walkers, eating mainly scraps with some insects. This habit already occurred in 1946 with a cock at Ben Macdui cairn, and in 1956 a cock and hen on Cairn Lochan. October After the moult, many were flying when first seen or heard. As in September, singletons were reluctant to fly, but they skulk in winter too, so skulking is inverse to group size, irrespective of moulting. However, they tended to flush at greater range than in September, and flew further (>20 m and usually >50 m). Most birds were in flocks, with singletons less frequent than in September (Table 2). | saw none on smooth grassland or heath. As well as using A, birds used B much. The main foods were seeds of grass, sedge, rush and other herbs, with moss capsules. Occasionally birds ate insects, eg flies emerging on thawing new snow. Some ate picnic scraps at cairns, tors, footpaths, and the cafe on B. Ungerminated seeds of reseeded grass abounded after reseeding, eg all of acumulative total of 64 birds seen on Cairn Gorm in October 1996 were on patches reseeded in late July 1996, and birds were seen to eat ungerminated seeds exclusively. Discussion Adult presence and absence on the main habitat types Most breeding adults were on intermediate land with boulder patches near poorly drained ee aeons 1997 Table 2 Percentages of sightings of different group sizes in autumn 1966-97. Number September October in group 1 47 10 2 13 6 3 10 6 4 if 8 5 3 15 6 3 13 7 3 10 8 3 6 9 3 2 10 0 2 11 3 2 12 3 4. 16 0 2 17 0 2 20 0 2 21 0 2 28 0 2 30 0 2 33 0 2 n 30 48 An analysis comparing proportions of singletons and all other group sizes combined, in September vs October, gave chi? = 11 (Yates correction),df = 1, P < 0.001. or freely drained plant: communities or snowbeds, and at boulder field edges beside such land. None was seen on continuous grassland or heath with few or no boulders or Cliffs, oron Three-leaved Rush Juncus trifidus with grit patches and scattered boulders. In Baffin Island, Watson (1957, 1963) found up _to a cock per ha in wet or dry spots where boulder cavities offered nest-sites, but few or _ none where boulders were smooth, orisolated Habitat use by Snow Buntings in Scotland spring to autumn 117 (ie with cavities, but lacking nearby boulders as cover). Most boulders among Three-leaved Rush on A were tabular and embedded in soil, and so lacked holes, but interlocking boulders provided many holes. Foraging on and near snow patches, and insect abundance Many adults foraged on snow patches while breeding andshortly after. This supports ideas onsnowas an important insect trap for insect- eaters (Mani 1962; Swan 1963; Nethersole- Thompson 1966; Pattie & Verbeek 1966; Watson 1966). Hendricks (1987) did not find this for insectivorous birds, but Antor (1995) did for high-alpine ones. In some years | saw no foraging buntings on snow patches during the nesting and fledgling periods. As judged by visual impressions, craneflies abounded in 1969, 1971, 1972, 1973, 1975, 1977, 1979 and 1981 but not in 1968, 1970, 1974, 1976, 1978, 1996 and 1997 (I did not note this before 1968 and in 1982-95). In the first set of years | saw no foraging on snow patches in June and early July, when birds foraged mostly on grassland and flushes with abundant craneflies, but did see foraging on snow in late July and August, when visual impressions indicated few craneflies on the grassland and flushes used earlier. By contrast, in June and early July in the second set of years | saw more birds foraging on snow than on any other habitat except beside snow, which was favoured in every year. A sole adult foraging on snow may be at more risk from predation, and so may feed there only if insects are scarce off the snow and there are no other accessible foods such as picnic scraps. | saw no dependent fledglings on snow, and no adults or others on snow in September during the moult. Avoidance of 112 Adam Watson predation might explain this. However, | saw very few insects on snow in September, and the area of snow left then was minute. | observed far more foraging beside snow than on it. The role of frozen soil in delaying insect emergence and providing bird food for a longer period is not well recognised and deserves study. Craneflies were fewer in late July and August than earlier, even in peak years. Often the Only ones seen in late summer were on patches where foraging Snow Buntings and other passerines had concentrated. Adults feeding picnic scraps to nestlings and fledglings (as noted above) were seen only in late July and early August 1996 and 1997 on Cairn Gorm, a period and area with few craneflies as judged by visual impressions. Movements during the breeding season Above, | noted long flights by foraging adults, and movements with fledglings to new sites far from nests. Such movements might increase predation and may indicate poorer habitat. In a high density area in Baffin Island (Watson 1957,1963), | saw no such movements till young were independent The birds’ land use and protection measures Moutling birds were reluctant to fly and may be vulnerable to predators. R. Smith (in Gibbons et a/1993) noted that adults become very unobtrusive during the moult in August and September, and ‘a major loss of adults appears to occur at this vulnerable time’. Reseeded patches on the ski area and picnic scraps provide food near boulder cover then, and people there may keep predators away. | had too few data from before the 1960 ski SB 19 (2) development to test for differences since. My impression was that birds used B more since 1960, associated with reseeded patches, the cafe, and more scraps due to more people. Nevertheless, in May and early June snowfalls before nesting, | saw birds on B and on Derry Cairngorm in many pre development years in 1943-59. Hence their use of these areas was not due to development, even though the frequency of such use currently may be. Sites for protection should include all breeding grounds (area A), but also B, which birds use in autumn and in snowfalls before nesting. Existing and proposed Sites of Special Scientific Interest omit land north east of Cairn Gorm and the ski area’s upper parts north-west of Cairn Gorm, which form some of A and much of B. All of A and B should be considered for inclusion. One of the rarer regular breeders in the UK, the Snow Bunting did not nest in other EEC countries when the Birds Directive was announced in 1979 (79/409/EEC). Yetitis not in the Directive’s Annex |, though the far more abundant Peregrine Falco peregrinus, Capercaillie Tetrao urogallus, Golden Plover Pluvialis apricaria, Short-eared Owl Asio flammeus and others are. A proposed Cairngorms Special Protection Area for Annex | birds omits more of A and B than the SSSls. The Snow Bunting should be considered for Annex | and the SPA amended accordingly. Acknowledgements | thank Miss J Watson for drawing the map, and Drs M Marquiss, R D Smith andDBA Thompson for comments. References Antor RJ 1995. The importance of arthropod fallout on snow patches for the foraging of high-alpine birds. Journal of Avian Biology 26: 81-85. Gibbons DW, Reid JB & Chapman RA 1993. The New Atlas of Breeding Birds in Britain and Ireland: 1988-1991. Poyser. London. Hendricks P 1987. Habitat use by nesting water pipits (Anthus spinoletta): a test of the snowfield hypothesis. Arctic & Alpine Research 19: 313-320. Mani M S 1962. /ntroduction to High Altitude Entomology. Methuen, London. Nethersole-Thompson D 1966. The Snow Bunting. Oliver & Boyd. Edinburgh & London. Pattie DL & Verbeek N AM 1966. Alpine birds of the Beartooth Mountains. Condor 68: 167-176. Perrins. C M 1994. Birds of the Western Palearctic. Vol 9. Oxford University Press. Oxford. Smith R D 1994. Snow Buntings Breeding in the Cairngorms: Population Dynamics and the Influence of Recreation. Scottish Natural Heritage. Perth. Smith R D & Marquiss M 1994. Breeding seasons and nesting success of Snow Buntings in north east Scotland Scottish Birds 17:223-234. Swan L W 1963. Aeolian zone. Science 140: 77- 78. Habitat use by Snow Buntings in Scotland spring to autumn 113 Thomley A 1896. On some Coleoptera from the summit of Ben Nevis, collected by Mr. W.S. Bruce. Annals of Scottish Natural History 1896: 28-37. Watson A 1957. Birds in Cumberland Peninsula, Baffin Island. Canadian Field-Naturalist 71: 87- 109. Watson A 1963. Bird numbers on tundra in Baffin Island. Arctic 16: 101-108. Watson A 1966. Hill birds of the Cairngorms. Scottish Birds 4: |79-203. Watson A 1991. Increase of people on Cairn Gorm plateau following easier access. Scottish Georgraphical Magazine 107:99-105. Watson A 1996. Scottish Snow Bunting breeding and climate. Ornis Fennica 73: 137-140. Watson A, Davison RW & French D D 1994. Summer snow patches and climate in northeast Scotland. U.K. Arctic & Alpine Research 26: 141- 15: Watson A & Stroyan H L G 1984. Unusual concentration of aphids at high altitudes in the Cairngorms. Entomologist’s Monthly Magazine 120: 145-149. A.Watson, c/o Institute of Terrestrial Ecology, Banchory, AB31 4BY Revised manuscript accepted December 1996 Male Snow Bunting 114 Scottish Birds (1997) 19:114-127 SHORT NOTES Red-backed Shrike breeding records in Scotland With reference to previous successful breeding records of Red-backed Shrikes in Scotland (Scottish Birds 18:118-119; 190), a pair of Red-backed Shrikes also bred successfully in Grampian in 1981, producing 4 young of which 3 survived. R C Dickson, Lismore, New Luce, Newton Stewart DG8 0AJ The note describing a successful breeding attempt in Perthshire in 1994 clearly did not summarise all previous breeding records. We do not intend to publish more such records unless they are part of a comprehensive review. Accepted June 1996 < ammaciays gas e Red-backed Shrike SB 19 (2) Red-backed Shrike - summer records in Scotland An adult male was at Glen Doll, Angus, during the summers of 1989 and presumably the same bird, in the same area, during 1990. It was seen by several observers, butno females nor any form of display, or territorial behaviour was noted. During the summers of 1988 and 1989, both sexes were regularly noted near Braemar. They appeared to be paired and to be defending territory in a fragmented Hawthorn hedge. No nest nor fledglings were seen by me. In 1981 an adult male lived at Muir of Dinnet from 17 May till last seen on 3 July. This was a year, when Wryneck Jynx torquilla was singing throughout June elsewhere in Deeside suggesting a Scandinavian origin. On 3 June 1994 an adult female was at Milton, north of Blairgowrie but could well have been a displaced migrant. Remains of a male were found at a Sparrowhawk plucking post near Lochmaben, Dumfriessshire on 19 June 1969. It may well have been a migrant too. John Young, Scottish Natural Heritage, West Lodge, Airlie, Kirriemuir, Angus DD8 5NP Accepted June 1996 GIG. ov” 4 bslf, Zz — ~ Mike Ashley 1997 Ground nesting Treecreeper in Deeside On 26 May 1996 | found an unusual Treecreeper Certhia familiaris nest at Dinnet Nature Reserve, Deeside. The nest was located in grass and Ling, Calluna vulgaris, at the base ofa 15m Scots Pine Pinus sylvestris. The tree was in a mixed stand of Scots Pine and Birch Betula spp. The ground vegetation was approximately 0.15m tall and concealed asmall grass lined cup placed almost directly on the ground. There were 2 live, naked young, one dead, naked young and one Short Notes 115 unhatched egg. By the 30th one of the young had died and, by 1 June 1996, the remaining chick was ready to fledge. On the 5th the chick had gone, presumably having fledged successfully. The nest contained the 2 dead chicks and the egg. The only previous record of ground nesting | have found refers to anest in Derbyshire in May 1969. That nest too was at the base of a Scots Pine, though the ground vegetation was 0.3m high Bracken Pteridium aquilinum (Frost 1969, British Birds 62:446). The stand of trees in which the Dinnet nest was located was quite young, and probably lacking suitable nesting crevices, hence the unusual choice of nest site. Simon Gillings, British Trust for Ornithology, The Nunnery, Thetford, Norfolk, [P24 2PU. Accepted August 1996 Numbers of Snow Buntings on arable land in north east Scotland Wintering Snow Buntings Plectrophenax nivalis in north east Scotland have been recorded mostly on hills and coastal dunes, but with some flocks in fields, mostly in Buchan and Clatt (S T Buckland, M V Bell & N Picozzi, 1990, The Birds of North-East Scotland). This understates their occasional high numbers and wide distribution in fields. RR noted birds in 1984, 1986 and 1994-96 near Newburgh, and AW in 1988-96 in the course of studying Corn Buntings Miliaria calandrafrom Portsoy to Arbroath. Effort and locations searched were the same each year. 1983-84. January-February, big flocks in coastal fields in prolonged snow. 1985-86. January flocks of 600 at Collieston and 2,000 in uncut coats at Rashierieve. 1988-89. One at Towie in November, and 120 at Kinellar in December-January. In January, 225 at Inverurie, 108 at Corse, 84 at Crudie and 50, 80, 120 and 200 at 4 farms near Portsoy. At a fifth Portsoy farm 300 in January-20 February, 250 on 24 February, 30 on 26 February, 10 0n 11 March, and none on 26 March. At Whitehills 1,100 in January, 700-800 till 26 February and none on 11 March. 116 Short Notes 1989-90. Twoin8cmof snow at Kennethmont in January. 1990-91. In January, 50 at Boyndie. 1991-92. In February, 8 in 6cm of snow at Badenscoth, and one in 5cm at Inverurie. In November-March, 100 in uncut oats at Gannochy moor near Edzell, not seen after the oats were burnt in early March, and none in 1988-91 when the field was rough pasture. 1992-93. In November, 2 at Garvock and one near Buckie. In November-January at Gannochy, 100 in uncut oats. One in 4cm of snow at Barras in February. 1993-94. In December, 70 in 4cm of snow at Crawton; deep snow further inland. In November-February, 15 in uncut oats on Gannochy moor. 1994-95. Noneseen. Wild geese devastated uncut Gannochy oats in October and Red Deer at the rest of the standing oats by 31 October. 1995-96. None seen. Red deer devastated uncut Gannochy oats in October. Numbers seen in the winters of 1988-96, excluding repeated observations of the same flock, were 2,388, 2, 50, 109, 104, 85, 0 and 0, a big variation. Of the total of 2,739, 25% were 15-50km from the sea, 21% at 3-9km and 54% within 1km. Most were in north Banffshire and central or north Aberdeenshire, SB 19 (2) few in Kincardineshire, and none on study areas on and near the Angus coast. A few in 1989-94 were in fields with thin snow when higher land had deep snow; _ these birds vanished with the thaw. To assess foraging we used cases where flying flocks were first seen to alight, to avoid bias due to birds being more easily seen on some fields such as short grass than on others suchas stubble. Totals included some flocks observed more than once. Out of 8,286 cases, 49% were on wheat stubble, 26% on barley stubble, 13% and 4% on recently ploughed wheat and barley stubble, 4% on pasture, 3% in standing oats, 1% ona football pitch, 30 birds in rough grass by a farm road, and 2 each in uncut wheat, weedy set aside, and rough grass with herbs. As more stubble fields became ploughed, some foraged on newly or recently ploughed stubble in late February and early March, where grain was conspicuous on the dark earth. Birds husked oats and barley, but not wheat, which lacked husks. A few ate flies and other insects. Snow Buntings disturbed by a raptor or person flew to low electricity wires, and sometimes to telephone wires and fences. They often flew to wires to loaf, and a flock was seen twice in a 10m Beech Fagus sylvatica tree, loafing on branch tips. One flock of 300 weighed down electricity wires. Obvious cocks in flocks varied from 0% to 10%, averaging 5.1% out of 1,848 in different flocks checked. A Watson & R Rae, c/o Institute of Terrestrial Ecology, Banchory AB31 4BY Accepted September 1996 1997 Capercaillie and Black Grouse south of Banff in the 1940s A map in Catt et a/ (1994, Abundance and Distribution of Capercaillie in Scotland, 1992- 94, Institute of Terrestrial Ecology, Banchory) showed, incorrectly, that Capercaillie Tetrao urogallus had ‘never been present’ between Banff and Fyvie. Though their references list Pennie (1950, Scottish Naturalist 62:65-87), they did not use his good account for the area. Here | add my sightings of Capercaillie and Black Grouse Tetrao tetrix made in 1943-48, and information from others. All woods had mature/old Scots Pine Pinus sylvestris with some Birch Betula sp and Rowan Sorbus aucuparia. Capercaillie 1. Forglen, 250 ha. Each year | saw a cock and hen in 4 ha of heather and blaeberry with self sown Norway Spruce Pinus abies and Birch. They bred in big tracts of self sown Birch and Pine with Blaeberry Vaccinium myrtillus and Heather Calluna vulgaris, and in planted pine, Norway Spruce and Larch Larix decidua. Up to 25 were seen in a winter pack; not seen after clear-felling and planting. 2. Delgaty, 160 ha. Capercaillie bred in old pine with Birch, Heather and Blaeberry, and in mature planted pine, with up to 20 ina winter pack. | saw a hen eat Beech leaves in May. Birds not seen after clear felling and Forestry Commission (FC) planting. 3. Lescraigie. A hen was seen in May in 30 ha of pine and Norway Spruce, later clear felled. 4. Lendrum, 55 ha. One to two seen each spring in Heather with pine and Norway Spruce. Short Notes 117 5. Den of Woodhead, Balquhollie, 20 ha. One to 2 seen each spring in scattered pine and Norway Spruce with grass, Blaeberry and nearby Beeches. 6. Brownside, Mountblairy, 95 ha. This was cut by 1943, but the gamekeeper said it held many Capercaillie before cutting; not seen after FC planting. 7. Montcoffer, 120 ha. Capercaillie bred in old pine with Heather and Blaeberry, and nearby Beeches. Up to 16 were seen ina winter pack; not seen after clear felling and planting. 8. Den of Rothie. Pennie gave useful information on breeding numbers here, calling it ‘Den Wood’, and estimated that there were 500 acres of suitable habitat. Mr Main, Fyvie Castle gamekeeper, told me in March 1949 that ‘they shoot 8-9 every year and usually see 20-30 in a day’s drive. Occasional wanderers move into other wood on Fyvie (Cranna, Brownhill, Cockhilland Fyvie Castle). | haven’t noticed any difference in numbers over the years’ (he had been there for 37 years). 9. Hatton Castle. The owner Lt Col Duff told me in March 1949 that he shot a Capercaillie at Hatton about 1934, and ‘at one time | saw them fairly often and there were a nest or 2 at the east end of the Den of Balquholly, but those trees are now cut down. | saw one hen in the Wood of Colp 2 years ago, but | think any Caper at Hatton now are merely visitors’. Den of Balqhuolly is near Lescraigie (3 above). It had semi open patches with some Heather and Blaeberry. The Wood of Colp was of old planted pine and Norway Spruce, with grass below. 118 Short Notes SB 19 (2) Black Grouse 1. Forglen. Black Grouse bred each year. Up to 6 were seen ina winter pack; not seen after felling. 2. Delgaty. Birds bred each year with up to 12 in a winter pack; not seen after felling. 3. Waggle Hill, 175 ha of sedgy and rushy Heather. Breeding ocurred each year with up to 10 in a winter pack; not seen after FC planted nearly all of it. 4. Greenness Hill, 30 ha of sedgy and rushy Heather. One to 2 hens bred each year, only 1 km from nearest part of Waggle Hill; not seen after FC planted all of it. 5. Fyvie Castle and Den of Rothie. The laird (Forbes-Leith) told me there used to be ‘quite a number’, in both these places, the last in 1912 as planted woods thickened. Gamekeeper Mr Main confirmed this in 1949, adding that they had bred annually and he shot the last one in 1912. They were regularly on Minnonie and other Ythan braes to Gight Castle and east of it. Later the FC planted conifers densely on big parts of the braes. 6. Hatton Castle. Lt Col Duff told me in 1949 that they nested annually in the middle and north ends of the Den of Balqhuolly. Delgaty, Montcoffer and parts of Hatton and Den of Rothie resembled Caledonian pinewood and much of Forglen was developing into natural boreal woodland. Their destruction was tragic. Felling, burning and dense planting of mostly exotic conifers eliminated heath, save in narrow rides. This ended thriving stocks of both species, decades before recent concerns over their decline and years before Foxes Vulpes vulpes colonised the area in the late 1950s. Adam Watson, c/o Institute of Terrestrial Ecology, Banchory, AB31 4BY Accepted September 1996 ! CL ere i \" 4 \ | YW, UN ! NGG, f0 tps aR Ly! ANY ty os wea : YQ DMN ey Ai Black Grouse if i ; TRS OKA Ss DAG WA Hi pe At VPRE Lit iy) EPA ; ge nee ee WP Ky, ” 4 i x edie) Ms! we BN Pale ae ma pet, we he age af; HANAL cr i g' Sv yeh \ Pr) ip Wy AK Brent Hurley 1997 Reed Warblers breeding in south west Scotland In Scotland the Reed Warbler Acrocephalus scirpaceus occurs regularly in spring and autumn with only one breeding record in Shetland in 1973 (Thom V M 1986. Birds in Scotland). In England and Wales the overall breeding population is estimated between 40,000-80,000 pairs (BTO/JNCC: Britain’s Birds in 1990-91). On 17 September 1992, members of the _ North Solway Ringing Group were mist netting | in a Galloway reedbed when a juvenile Reed _ Warbler was trapped and ringed (Table 1). It could not be ascertained whether the bird was locally bred or was a migrant. This was the first Reed Warbler caught by the Group. Short Notes 119 On 3 visits to the reedbed during August 1994, 6 adult birds, 3 being female with well formed brood patches, were caught. Two of the adult birds had been ringed at the site in 1993 (Table 2). Three juvenile birds, all strong fliers, were also trapped and marked. Singing males had been heard in the reedbed every month between May and August. During July and August 1995, 18 Reed Warblers were caught at the site. Eight were juveniles with either remiges or rectrices still growing with large parts of the body still bare. Also trapped were 9 adults; 4 females, 5 males, with 3 of the birds having been ringed at the site in previous years. One of these birds had been ringed as a recently fledged juvenile on 5 September 1993. An adult bird was also seen in attendance with 2 newly fledged young. Table 1 Annual totals of Reed Warblers caught in a Galloway reed bed 1992-96. 1992 1993 | Juveniles 1 2 Adult (new) - 4 Adult (retrap) - - _ Totals 1 6 _ In1993 visits were made to the same reedbed | On several occasions when at least 2 singing | males were holding territory. By the end of the | summer mist netting had produced 4 adult birds, one of which was a female with a well | developed brood patch. On 5 September, 2 juvenile birds were caught. The birds, recently fledged, had traces of down, weak and loose body feathers with bare skin under the wings Still quite discernible and they were barely able to fly. 1994 1995 1996 3 9 20 4 6 3 2 3 6 9 18 29 During 1996, while only 5 males were caught, it is believed that 7 singing males were in the reedbed. Juvenile fledging was up on previous years with 20 birds trapped (Table 1). This would appear to be a notable range extension of the species into mainland Scotland. The nearest known breeding Reed Warblers are in Cumbria where 2, possibly 3 pairs, bred in 1993 (Birds in Cumbria: County Natural History Report 1993). In The Atlas of 120 Short Notes SB 19 (2) Table 2 Reed Warblers returning to reed bed in subsequent years. | Ringing year Year of retrap and No. of birds ringed 1992 1993 1992 1993 1994 1995 1996 Oana — i] No — Breeding Birds in Britain and Ireland (Sharrock 1976) there were no breeding dots for Cumbria between 1968-72 while The New Atlas ol Breeding Birds in Britain and Ireland: 1988- 1991 (Gibbons, Reid and Chapman 1993), gives 7 full dots. This obvious range expansion northwards can also be seen on the east coast of England, in North Yorkshire and Durham, where in the same 20 year period between the Atlases 4 dots compared to one dot is shown. During the summer of 1996 asecond breeding site in south west Scotland was found at Caerlaverock National Nature Reserve. Acknowledgement | am grateful to Dr Malcolm Ogilvie for helpful comments on an earlier draft. Ken Bruce, North Solway Ringing Group, Mallaig, Wellington Street, Glencaple, Dumfries. Accepted October 1996 1994 1995 1996 ; Nh ‘1 — N ! 'wMNw! DOLLA Lia Reed Warbler Steven Brown 1997 Inland nesting by Razorbills Razorbills Alca torda breed solitarily or colonially in rocky coastal areas, on mainland cliffs and offshore islands. The breeding site is usually in a boulder scree or on a small ledge or niche, on a rocky cliffface. Less commonly, birds nest in earthy burrows or on the nests of Kittiwakes Aissa tridactyla. As the chick leaves the nest site when only partly grown and still unable to fly, most colonies _ tend to be directly above the sea. Although _ there are published records of breeding sites _ upto 300 minlandin Greenland (Salomonsen 1950, Granlands Vogel), to our knowledge similar cases have not previously been ' described in Britain. es SSS Short Notes 121 During the last few decades, the numbers of several species of cliff nesting seabird, including Razorbills, have increased substantially on the Isle of May in the Firth of Forth, a trend which is in line with many other colonies around the North Sea (Lloyd et al, 1991, Status of Seabirds in Britain and Ireland). A notable feature of the increase on the Isle of May has been that several species have ‘spilled over’ from the seacliffs to breed up to 100 minland on low cliffs and steep vegetated slopes. The main area of overspill has been around a manmade freshwater loch, which is about 20m above sea level and separated from the sea by adam, a50 mslope of eroded . SSS SSS SS SNS SQV SS The loch on the Isle of May. The main group of nesting Razorbills on the south side was in the Kittiwake colony in the top left corner of the photograph. Mike Harris 122 Short Notes soil and loose rock and a boulder beach (see photograph). By the early 1990s, the slopes and low cliffs either side of this loch had been colonised by Kittiwakes, Fulmars Fulmarus glacialis and Puffins Fratercula arctica but Razorbills had never been recorded in the area, although large numbers were breeding on the cliffs at the seaward end of the loch. Annual counts of a monitoring plot on the seacliffs just north west of the loch showed that the number of Razorbills present increased by 6- 7% per annum between 1983 and 1995. It was also apparent that, each year, birds were progressively colonising the cliffs above the boulder beach and the eroded slope, and were moving towards the dam. In 1994, Razorbills were occasionally seen flying over the loch and we assumed that these birds were taking a short-cut to the east side of the island. However, in June 1995, birds were observed standing on, and among, Kittiwake nests on both sides of the loch. One pair of Razorbills on the north side definitely had an egg on 15 June, but it is not known if they fledged a chick (Harding & Thompson, 1996, Studies of breeding birds on the Isle of May in 1995, SNH Report). On the south side of the loch, a pair with a recently-hatched chick was recorded on the late date of 14 July. The chick was still present on 25 July, but SB 19 (2) | again, the breeding outcome was unknown. | In 1996, a pair bred at the same site on the north side of the loch and their chick lefton 23 | July. Up to 12 individuals were recorded on |} = the south side; four or five pairs laid, and at | least three of these reared a chick to leaving © age. 4 Although Razorbills appeared to have no ! difficulty rearing chicks at these inland sites, ° they faced a potential problem at fledging. | Under normal circumstances, the chickjumps | from the nest-site at dusk, and glides, or | plummets, to the sea, accompanied by its © male parent. All being well, the adult and A young meet up on the water and immediately * swim out to sea together. It is impossible for | Razorbill chicks from the lochside to reach the sea directly and it is almost inevitable that they will land on the loch. On the morning of 20 July 1996, there was an adult and a chick on the loch, but within an hour both had disappeared. To have reached the sea, the chick would have had to climb around the dam, and walk about 50 m through several Herring Gull Larus argentatus territories, a fairly perilous procedure. It will be interesting to see whether further colonisation of this inland habitat occurs and ifthere is any evidence of successful fledging. M P Harris & S Wanless, Institute of Terrestrial Ecology, Banchory AB31 4BY Accepted December 1996 2 | 1997 Hooded Warbler on St Kilda | Atmidday on 10 September 1992 J Vaughan ' and myself were at the army base at Village | Bay on Hirta, JV saw a bird that ‘looked like a | large phyllosc but with black on its head’ fly _ past him and disappear over the sea wall on | tothe beach. Together we almost immediately _ relocated the bird. The heavy rain and strong i i _ gusts of wind made viewing condition difficult. My firstimpressions were of an almost wagtail like passerine, basically olive green above and yellow below but with black on its crown, _ extending back to the nape and around the sides of its head to behind the ear coverts and _ with a black gorget. The legs were bright pink | | | and robust. The bird frequently cocked up or flicked its tail. It was very lively and would not permit a closer approach than 10m. The flight was fast and low, on shortish wings which served to exaggerate the length of the tail. We continued to watch it for some 4-5 minutes making mental notes as the poor weather made writing impossible. We adjourned to the Factor’s House to check reference books. Identification was straight forward. The large size, robust features and general colouration combined to eliminate all but 2 species: Hooded and Wilson’s Warblers. Thelatter was immediately eliminated due to our bird’s distinctive head Short Notes 123 pattern. We went out again and, after a short search, relocated the bird in the same area. The diagnostic presence of white patches on the 3 outer feathers of the upper tail was eventually confirmed, though this feature was only apparent when the bird flicked open its tail or hovered. The bird was watched for a further 2-3 minutes before we agreed not to push it around any further in the appalling weather. The bird was seen briefly later in the afternnon by JV, perched on a wire outside the army base. That night, continuous heavy rainfall was severe enough to cause flooding in the Village Bay area. The bird was never seen again. A series of fairly active Atlantic depressions had crossed the area during the preceding few days and a depression was centred over sea area Hebrides atthe time of the discovery. A vigorous westerly jetstream had apparently become established at upper altitudes several days before(| Robinson, Benbecula Met Office) which would have assisted trans Atlantic vagrancy. There has been one other occurrence of this species in Britain and Europe, a female from 20-23 September 1970 on St Agnes in the Scilly Isles. T J Dix, 2 Drimsdale, South Uist PA81 5RT Accepted December 1996 A full description is in the files of the British Birds Rarities Committee. This is the first Scottish record. Although it has been customary for such records to be published in Scottish Birds that practice originated when Scottish Birds was the SOC’s only publication. In future such records will be published in the Scottish Bird Report. 124 Short Notes Scavenging Merlins | read with interest the note (Thomas 1992, Scottish Birds 16:219-220) on a Merlin Falco columbarius feeding on Rabbit carrion in Tayside. Scavenging by Merlins has rarely been seen or reported. During intensive observations of Merlins between 1965-96, | have only recorded one apparent attempt at scavenging by wintering Merlins in over 800 observations. On 25 August 1977 at 1650hrs, in a sandy bay on the coast over a kilometre from a Merlin’s roostin Galloway, afemale or juvenile Merlin was watched standing on and apparently pecking ata juvenile Greater Black- backed Gull Larus marina. The Merlin was soon displaced by a Carrion Crow Corvus corone. \t landed a short distance away. Ten minutes later the Merlin chased a passing flock of Ringed Plovers Charadrius hiaticula enough to indicate that it was healthy and SB 19 (2) uninjured. The Merlin flew on out of sight.On | inspecting the dead gull, the feathers were found to be intact and the skin unbroken. | cannot say whether the Merlin was attempting to feed on the gull or using the carcase as a hunting perch, pecking at it inquisitively. Apart from the reference given in Thomas’s note | know of 2 others on scavenging by Merlins. Warkentin (1986 Canadian Journal of Zoology 64:262-264) recorded a radio tagged Merlin released 51 days after being found injured that was killed by a vehicle when feeding on a feral pigeon Columba livia killed on a road. Warkentin & Oliphant (1988, Wilson Bulletin100:137-139) recorded a radio tagged female Merlin apparently scavenging a feral pigeon frozen to a roof top. Although other species of raptors like Peregrine Falco peregrinus and Sparrowhawk Accipiter nisus have been illegally killed in Britain while scavenging at poisoned baits, Merlins have not (D Dick in /itt). R C Dickson, Lismore, New Luce, Newton Stewart, Wigtownshire DG8 0AJ Accepted December 1996 Buzzard robbing juvenile Peregrine of prey Peregrine Falco peregrinus occasionally steal from other raptors (Ratcliffe 1993, The Peregrine Falcon, London) but records of Buzzards Buteo buteo stealing from Peregrines are generally lacking in Britian. On 3 September 1995 a juvenile Peregrine was Standing on a dead, adult Black-headed Gull Larus ridibundus in a shallow muddy pool on flooded pasture near Stranraer, Galloway. As the falcon tried to lift its prey out of the mud it began to call as a Buzzard landed on a fence post 10-15m away. The falcon managed to free the carcase and, as it landed, it was immediately displaced by the Buzzard. The Peregrine flew up calling and swooped on the Buzzard 3 times, before landing 2-3m away. The Buzzard pecked at the carcase and once approached the i uw | i i f 1997 | Peregrine on foot in a threatening posture. Nine minutes later the Buzzard flew back to _the fence post while the Peregrine flew to the _carcase and fed on the gull for the next 25 minutes. | The Buzzard only made a half hearted attempt to eat the gull but was possibly intimidated by Short Notes 125 the presence of the Peregrine. In North America, Peregrines usually yield their prey to Buteos (Beebe 1960, Condor62:145-189) Inthis case the Peregrine regained possession of its prey without too much effort, from, | suspect, an inexperienced, juvenile Buzzard. Three Buzzards had been seen together on previous occasions in the same field, possibly juveniles from the same brood. R C Dickson, Lismore, New Luce, Newton Stewart, Wigtownshire DG8 OAJ Accepted December 1996 | Interaction between Sparrowhawk and Jackdaw near Newtonmore, Highland On 14 January, 1995 | was alerted to a ‘disturbance by the agitated calling of a large number of corvids, close to my house, near Newtonmore, Highland. | saw a female Sparrowhawk Accipiter nisus and, as | later discovered, an adult Jackdaw Corvus _monedula fighting on the ground. It was obvious that the confrontation may have turned into a lengthy one and | started the timer on my wristwatch. The ground was _Sloping and during the first few minutes of | struggle, both birds tumbled a couple of metres downhill. The Sparrowhawk had one talon locked into the breast of its prey but the Jackdaw was resisting fiercely, flapping its | wings, jumping up to 30cm into the air and | Stabbing with gaped beak as they tumbled. The Sparrowhawk arrested its fall with its free talon by grasping at the bank; the other remained locked firmly into the flailing Jackdaw. The mixed corvid flock of about 40 Jackdaws and Rooks Coruvs frugilegus continued to call from the surrounding trees. Several individuals swooped to branches just a few metres above the Sparrowhawk but none alighted on the ground or attempted to make physical contact. During the next tumble down the slope, the Sparrowhawk locked its other talon into the Jackdaw. The birds came to rest on a small level area, the hawk on its back, with the Jackdaw held by extended legs. Both birds rested for between 2 and3 minutes. Following another short tumble, the Sparrowhawk gained the upper position using its outstretched wings to balance and mantle its prey. Only at this stage did it seem to assess its Surroundings, glancing around atthe corvid flock, which caused them to increase the intensity of their calling. The struggle continued for 9 minutes and the Jackdaw made repeated attempts to escape. After 11 minutes the Jackdaw’s resistance had lessened and the corvid flock dispersed. One or 2 individuals settled in isolated trees about 200m away. i 126 Short Notes After 23 minutes, the Sparrowhawk attempted to pluck the Jackdaw which responded by weakly flapping its wings. After 29 minutes the Sparrowhawk plucked at feathers and began to rear at the Jackdaw’s flesh and no more movement was seen from the prey. Between the 39th and 43rd minute of the observation, the Sparrowhawk made 3 unsuccessful attempts to move the carcass. After resting for 7 minutes it resumed feeding for the next 2hours 15minutes. As it fed, the corvid flock reappeared landing in nearby trees, flying around and calling excitedly. After a minutes or so, they dispersed once again as quickly as they had arrived. Again, 2 individuals settled in isolated trees, in sight of the Sparrowhawk. Three hours and five minutes after first taking the Jackdaw to ground, the Sparrowhawk moved approximately 1m from the carcass and preened. The movement elicited a response SB 19 (2) from a single Rook which flew back to a tree | above the Sparrowhawk and called | intermittently. The Sparrowhawk left the area | after 3 hours 20 minutes and was not seen to . return, though the carcass disappearedwithin . 16 hours. Jackdaws have been recorded before as . Sparrowhawk prey items (Newton, The . Sparrowhawk, Poyser 1986) as has the fact that large prey may not be killed immediately, . but may die later during plucking and eating (Cramp & Simmons Birds of the Western Palearctic, Vol 2, October, 1980). This observation highlights the potential hazards to a predator which has preyed on a relatively large and social species. The rewards must be set against the energy required to subdue the prey and the risk of potential injury. | thank Steve Redpath for his thoughts on this observation. P R Moore, Scottish Natural Heritage, Achantoul, Aviemore, Inverness-shire PH22 1QD | Revised manuscript accepted January 1997 Sparrowhawk attacking a Jackdaw Mike Ashley ; t i | f i "| 1 | (Mobbing of Waxwing by \Chaffinches and Blue Tits ‘The short note in Scottish Birds 18:250 \coincided with a very similar experience we jnoted on 10 December 1996 on a gloomy, damp morning about 0900hrs, when from our /bedroom window we saw a single Waxwing, ‘\Bombycilla garrulus, eating the berries of a Rowan Sorbus vilmorinii. We noted that the jother birds in the same tree comprising Blue Short Notes 127 Tits Parus caeruleus, Chaffinches Fringilla coelebs, and Greenfinches Carduelis chloris were acting in a very agitated manner. Our fears that this was due to a predator in the form of a domestic cat proved groundless. It seems possible that itis the solitary Waxwing, rather than a flock, which arouses the hostility of other birds. On 18 December 1996 16 Waxwings held sway over the berries of the Rowan both in the tree and on the ground. We were not aware of any antagonistic behaviour from other birds. S & ME Shimeld, 1 Rtichie Place, Crieff PH7 3SL Accepted January 1997 | ) | | | ~ s 2 | Waxwings feeding on Cowberries British birdwatchers tend to associate | Waxwings with urban parks and gardens, }/'where they are most often seen feeding on the berries of small trees or tall shrubs notably dawthorn Crataegus Rowan Sorbus and Cotoneaster. On 28 January 1996, | saw 8 Waxwings Bombycilla garrulus feeding on the ground on the berries of Cowberry Vaccinium vitis-idaea on heather moorlanda ew miles north of Blair Atholl at 360 metres above sea level. | Joan Howie has told me that she used to see Waxwings feeding on Cowberry in Strathspey when she lived there. Birds of the Western Palearctic (vol V pp494- 496) records Waxwings feeding on berries of Vaccinium though the records are mainly from breeding areas. This behaviour may be under recorded in Britain since birdwatchers are most likely to see Waxwings in or near towns. Ron Youngman, 20 East Moulin Road, Pitlochry PH16 SHY Revised manuscript accepted January 1997 128 Advice to Contributors 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 will be reviewed by at least 2 members of the editorial panel and, in some cases, by anindependent 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; 2 copies are required and the author should also retain one. We are happy to accept SB 19 (2) papers on computer discs, however, please contact Sylvia Laing on 0131 556 6042 to discuss this. Headings should not be underlined, nor typed entirely in capitals. Scientific names in italics should follow the first text reference to each species. Names of birds should follow the official Scottish list (Scottish Birds Vol 17 : 146-159). 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 5 August VOGT eo tees but on the 5th (if the name of the month does not follow). Please note that papers shorter than c700 words will 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 keptas simple as possible, with footnotes used to provide extra details where necessary. Each table should be on a separate sheet. Maps and diagrams should be in Indian ink and be camera ready, but drawn so as to permit reduction to half their original size. NEOTROPICAL BIRD CLUB Neotropical bird club launched A club has been launched to promote the study and conservation of the birds of the Neotropics (South America, Central America and the Caribbean). It is currently seeking founder members to help reach the launch budget of £2000, which is required to get the club running and to publish the two first issues of its intended journal ‘Continga’. Founder members will be asked to pay a minimum of £25, and will be formally acknowledged in the first issue of ‘Continga’. ‘Continga’ will provide a colourful and much needed forum for exchange of information on the avifauna of this extremely rich and diverse area, and will contain papers and features on the birds and their conservation as well as news of recent observations and discoveries (at present, new species are still being discovered at the rate of more than two a year). It is hoped that in due course the club will be able to provide direct funding and support for practical conservation programmes. For further details and membership forms, please contact: Rob Williams, Publicity Officer, Neotropical Bird Club, c/o The Lodge, Sandy, Bedfordshire SG19 2DL Scottish Birds Volume 19 Part 2 December 1997 Contents Main papers The illegal persecution of raptors in Scotland A report from the Scottish Raptor Study Groups. 65 Movements of Fulmars from the Firth of Forth. John C Davies 86 A winter survey of sawbill ducks and Cormorants on the River Deveron, north east Scotland. P J Cosgrove 93 Declines in Turnstones and Purple Sandpipers wintering in south east Scotland. H E M Dott 101 ; Habitat use by Snow Buntings in Scotland from spring to autumn. i Adam Watson 105 «Gi Short notes | Red-backed Shrike breeding records in Scotland. RC Dickson 114 “i Red-backed Shrike - summer records in Scotland. John Young 114 | Ground nesting Treecreeper in Deeside. Simon Gillings 115 } Numbers of Snow Buntings on arable land in north east Scotland. \ A Watson & R Rae 115 i Capercaillie and Black Grouse south of Banff in the 1940s. Adam Watson 117 Sy Reed Warblers breeding in south west Scotland. Ken Bruce 119 | Inland nesting by Razorbills. M P Harris & S Wanless 121 | Hooded Warbler on St Kilda. T J Dix 123 | Scavenging Merlins. R C Dickson 124 | Buzzard robbing juvenile Peregrine of prey. R C Dickson 124 Interaction between Sparrowhawk and Jackdaw near Newtonmore, Highland. P R Moore 125 88 Mobbing of Waxwing by Chaffinches and Blue Tits. S & M E Shimeld 127 35 Waxwings feeding on Cowberries. Ron Youngman 12%, | | Advice to contributors 128 90 Published by the Scottish Ornithologists’ Club, 21 Regent Terrace, Edinburgh EH7 5BT. © 1997 Scottish Birds The Journal of the Scottish Ornithologists’ Club Editor: DrS da Prato Assisted by: Professor D Jenkins, Dr J B Nelson, Dr | Bainbridge, Dr M Marquiss and R Swann Business Editor: The Secretary SOC, 21 Regent Terrace Edinburgh EH7 5BT (tel 0131-556 6042, fax 0131 558 9947). Scottish Birds, the official journal of the Scottish Ornithologists’ Club, publishes original material relating to ornithology in Scotland. Papers and notes should be sent to The Editor, Scottish Birds, 21 Regent Terrace, Edinburgh EH7 5BT. Two issues of Scottish Birds are published each year, in June and in December. Scottish Birds is issued free to members of the Scottish Ornithologists’ Club, who also receive the quarterly newsletter Scottish Bird News, the annual Scottish Bird Report and the annual Raptor round up. These are available to non- members at a subscription rate (1997) of £36. 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 21 Regent Terrace, Edinburgh EH7 5BT is the most comprehensive ornithological library in Scotland, and is available for reference during office hours (Monday to Friday, 0930-1630 hrs). (But please phone beforehand). SOC annual membership subscription rates Adult £18.00 Family (2 adults and any children under 18) living at the same address £27.00 Junior (under 18, or student under 25) £7.00 Pensioner/unwaged £10.00 Pensioner Family (2 adults living at the same address) £14.50 Life £360.00 Life Family £540.00 A\l subscriptions may be paid by Banker’s Order and Covenanted. Subscriptions paid this way will greatly assist the Club. Published by the Scottish Ornithologists’ Club, 21 Regent Terrace, Edinburgh EH7 5BT Printed by Douglas Gauld & Co Limited, 22 Tannadice Street, Dundee DD3 7QH Scottish Birds (1998) 19: 129-133 Zo Observations at a winter Rook and Jackdaw roost in Aberdeenshire |S FRANCIS A large corvid roost near Alford, Aberdeenshire was counted during winter 1994-95. Up to 18,500 birds used the roost from early September to late March, with peak counts in late November. Around 68% of the birds were Rooks, the remainder Jackdaws. The roost was utilised by birds from perhaps as far as 20 km distant. Arrival times and behaviour are described. The roost is important in a national context, and reflects the very high densities of Rooks in Aberdeenshire. Introduction Large winter roosts of Rooks Corvus frugilegus and other corvids, including Jackdaws Corvus monedula, have been described from numerous parts of Scotland, particularly during an SOC inquiry from 1969-1975 (Munro 1975), _ when 164 such roosts were noted. During the | winter of 1994-95, | visited a mixed corvid zs = bes - —— roost from September to March, estimated numbers of roosting birds and made some observations of arrival times and behaviour. Site description Meiklemoss Wood, near Alford, Aberdeenshire (NJ 602161; 145 asl) is a mixed Scots Pine - Birch (Pinus sylvestris - Betula spp.) woodland of clO ha, on level _ ground with a thicker Birch fringe and a few | scattered spruces (Picea sitchensis & P abies). Canopy height is around 10-14 m. The understory is mostly grassy and dominated by the grasses Holcus mollis and Deschampsia caespitosa. There are some open areas with Heather Calluna vulgaris and Broom Cytisus scoparius. Within the roost area there are some dead trees, possibly killed by the birds’ droppings , beneath which lie dense patches of Chickweed Stellaria mediawith Nettles Urtica dioica. Feathers and pellets are scattered over an area of around 100 x 200m, indicating the approximate extent of the roost. In general, the impact of a large number of roosting corvids on the woodland appears minimal, though the Chickweed and Nettle patches are clear evidence of a fertilising effect. The introduction of Stellaria media to a rookery has been recorded elsewhere (Rieley and Shah 1984), though Meiklemoss Wood is not abreeding rookery. This corvid roost appears to have arisen in the last 15 years or so (J Latham pers comm); in 1971, Munro (1975) could find no Rooks in Strathdon in winter, the nearest roost being 5 km to the south west at Muir of Fowlis. Methods The roost area was visited 21 times, around 7-9 days apart, initially by car but later viewed from one vantage point, and estimates made of numbers of birds entering the roost. On most occasions, itwas necessary to estimate in 5-10 minutes large numbers of incoming 130) |S Francis birds, which were counted in blocks of 50 or 100. This inevitably led to inaccuracies. In order to assess the extent of this on one occasion, 3 independent observers estimated numbers which varied from 8,000 to 9,500. This indicates the probable degree of error, but also shows that the numbers recorded were probably not more than 20% under or overestimated. The proportions of Rooks and Jackdaws were usually extremely difficult to determine under count conditions. Therefore, on 5 different dates, 30 counts of large pre roosting aggregations were made by 3 observers. Proportions of Jackdaws varied from 25% to 38%, with no apparent trend, but most lay in the range 28-36%. A figure of 32% Jackdaws was used in all flock counts to estimate numbers of that species. However, much of the account below relates to total corvids. Carrion Crows Corvus corone were noted only very rarely. SB 19(3) Results Total numbers of birds and changes through the winter The roost peaked at 18,500 on 23 November 1994, comprising around 13,900 Rooks and 4,600 Jackdaws. Rook numbers rose from around 300 in early September to over 5,000 by mid October. Numbers were over 10,000 from mid November to early January, dropping to 2,800 by mid March then very rapidly to under 300 by the end of March. There was no indication that the proportion of Jackdaws changed greatly through the winter, and the roost held over 3,000 from late October to mid January. Figure 1 illustrates the trend in numbers for total corvids. There was no significant roost before early September or after the end of March, though in late summer, at least one other area nearby (Breda Wood, Figure 1 Combined numbers of Rooks and Jackdaws roosting at Meiklemoss Wood. (Thousands) 2 9 22 2 1624 2 1323 4 1425 2 1528 4 1226 9 2030 October September November December January February March Winter 1994-1995 1998 Winter Rook and Jackdaw roost in Aberdeenshire 131 5 km to the west) was used for roosting, prior to utilisation of Meiklemoss Wood. Roosting times On all occasions, arrival in the roost proper occurred shortly before almost total darkness and well after sunset. Weather conditions did not influence this. Arrival times in the roost proper, therefore, changed systematically, linked to time of darkness, through the winter, from 1930 GMT in early September to 1624 GMT in late December and 1938 GMT in late March. However, arrival in the pre roost gathering areas was much more variable, with the first birds usually arriving to feed in favoured fields up to 55 minutes before roosting, though 25-30 minutes was more usual. Pattern of arrival and catchment area Birds usually arrived in pulses of several hundred, presumably feeding flocks, separated by periods with few arrivals, from several consistently favoured directions, mostly to the south and west. They usually dropped into fieids (both grass and stubble) to the south of the wood and the A944, where they fed and undertook occasional pre roost flights. Birds were observed flying to the roost from over 5 km to the west and over 10 km to the south west. Other observations of flightlines also suggested that the roost was holding birds which fed much further west up Strathdon, perhaps 20 km distant. However, most birds appeared to come from the Alford to Tarland area. Behaviour Birds arrived in pre roost flocks in nearby fields, sometimes in nearby small woods and trees, and most fed, though many spent time sitting quietly on fences. Periodically groups would fly up in short display flights in tight flocks, before settling again. The flights involved circling over the wood with sudden dives by tight flocks, followed by rising higher again. Sometimes these flights would extend over the roost wood, some 500 m distant, and arriving birds would often tumble and undertake short display flights before landing in the fields. Occasionally, birds would fly over and look at, or tumble into the roost woodland on arrival, then fly out again to join birds in the fields. On 2 occasions, shortly before the actual roosting flight began, the entire flock flew up over the roost wood and performed complex flights, before returning to the fields (once) or dropping into the roost. However, the usual behaviour immediately before roosting was for some small flocks to take off and carry out display flights over the wood. This then seemed to prompt the rest of the birds to fly to the wood in large waves, usually dropping in fairly quickly. Once birds landed in the roost it was rare for them to take off again. Discussion Communal roosting in Rooks and Jackdaws has often been described, though there have been few recent detailed accounts of individual roosts, particularly in Scotland. Early work by McWilliam (1924) and Munro (1948) was later succeeded by studies in the Ythan valley (Patterson et a/ 1971; Feare et a/ 1974), which concentrated on wider issues of Rook ecology, though Feare eta/(1971) considered roosting behaviour in detail. North east Scotland seems notable for the number and size of corvid (particularly Rook) roosts. Munro (1975) described 11 known Rook roosts in Aberdeenshire and atleast another 3 possible sites (see also Watson 1967; Dunnet and Patterson 1968). Estimated numbers of birds 132 | S Francis ranged from 2,000 to 10,000 at several roosts, 21,500 at Ellon, 49,000 at Straloch and 65,000 at Hatton Castle, the largest knownin Scotland. Roosts quite close to Meiklemoss Wood were Dunnideer Hill, Insch (‘thousands’) and Inver Fowlis (Muir of Fowlis, 7-10,000). At least 6 were known from Banffshire (numbering up to 30,000), one from Kincardineshire (5-10,000) and 3 from Morayshire (numbering up to 20,000). In all these estimates, Jackdaws were not counted separately. Earlier in the century, and elsewhere in Scotland, McWilliam (1924) considered Jackdaws were very much a minority ina roost of 12,000 corvids on Bute. | also visited Meiklemoss Wood during the winter of 1995-96, on 3 December 1995, when 19,500 corvids were estimated, and during winter 1996-97, on 13 December 1996, when c13,000 corvids entered the roost. It was also counted in 1992, when a much larger figure of 37,000 Rooks was recorded (RJ and S J Aspinall, NE Scotland Bird Report 1992); there were also up to 12,000 Jackdaws. Clearly, the roost can hold larger numbers than those recorded in 1994-96. However, using population and mortality figures givenin Lack (1986) and Gibbons et a/ (1993), the Meiklemoss Wood roost may have held, in 1994-95, around 0.5% of Britain’s winter population of Rooks. Rook breeding densities in Aberdeenshire are the highest recorded for Britain, with the largest rookeries (Dunnet and Patterson 1968; Thom 1986) and Feare et a/ (1974) considered this to be attributable to small scale field patterns, abundant winter Stubble, late harvests and later spring sowings. These high densities are clearly reflected in winter, and make Aberdeenshire in general, and roosts such as Meiklemoss Wood in particular, of great significance for this species in national, and probably also European, terms. SB 19(3) Rooks will travel up to about 50km (30 miles) to communal roosts in the winter, with similar movements for Jackdaws (Holyoak 1971). The roosts are linked to the rookeries, even in winter, with birds flying from roost to rookery inthe morning (Dunnet and Patterson 1968). In Aberdeenshire, maximum distances to roost in the Ythan valley were nearer 10- 19 km (Patterson et a/ 1971). Buckland et a/ (1990) note that in the years prior to 1990, a large roost was present at Muir of Fowlis, which served upper Deeside and Donside. This roost is not now present, following felling of many trees at the site around 1989, and may have been replaced by Meiklemoss Wood, 5 km to the north east. Birds observed flying to roost at Meiklemoss Wood at up to 10 km away would accord with observations from elsewhere. Feare et al (1974) found that for most of the year the times of entering and leaving the roost were related to sunrise and sunset, this relationship being affected to only a small extent by the weather. This was clearly so at Meiklemoss Wood. Munro (1948) and Feare et al (1974) also described pre roosting and roosting behaviour which was seen in this study too. The latter authors concluded that the main function of roosting lay in the post roost behaviour. Assembling after roost, and the dispersal from such assemblies, enables Rooks to locate food in unfamiliar surroundings when conditions prevent them from using their usual feeding grounds; each individual has access, through the flock, toa larger store of knowledge of potential food resources. The erratic display flights before roosting were considered to be anti predator behaviour. 1998 Acknowledgements |am grateful to John Latham, Graham Taylor and Nicky Penford for help with counts, and to lan Patterson and lan Bainbridge for comments and provision of references. References Buckland S T, Bell M V and Picozzi N 1990. The Birds of North-east Scotland. North-east Scotland Bird Club, Aberdeen. Dunnet G M and Patterson | J 1968. The rook problem in North-east Scotland. pp.119-39 in: Murton R K and Wright E N (eds). The Problems of Birds as Pests. Academic Press, London. Feare C J, Dunnet GM and Patterson | J 1974. Ecological studies of the Rook (Corvus frugilegus L) in North-east Scotland: Food intake and feeding behaviour.Journal Applied Ecology 11:867-896. Gibbons D W, Reid JB and Chapman R A. 1993. The New Atlas of Breeding Birds in Britain and Ireland 1988-1991. T & AD Poyser, London. Winter Rook and Jackdaw roost c Aberdeenshire 133 Holyoak D 1971. Movements and mortality of Corvidae. Bird Study 18:97-106. Lack D 1986. The Atlas of Wintering Birds in Britain and Ireland. T & AD Poyser, Calton. McWilliam J M 1924. The roosting habits of the Rooks of Bute. Scottish Naturalist (1924):5-7. Munro J H B 1948. Rook roosts of the Lothians, winter 1946-47. Scottish Naturalist 60:20-29. Munro J H B 1975. Scottish winter Rook roost survey - central and northern Scotland. Scottish Birds 8:309-314. Patterson | J, Dunnet C M and Fordham R A 1971. ecological studies of the Rook Corvus frugilegus L in North-east Scotland. Dispersion. Journal of Applied Ecology 8: 815-833. Rieley J 0 and Shah S E 1984. Eutrophication of afforested basin mires in the Midlands of England. Proc. 7th International Peat Congress, Dublin: 1: 372-387. Thom V 1986. Birds in Scotland. T & AD Poyser, Calton. Watson A 1967. The Hatton Castle rookery and roost in Aberdeenshire. Bird Study 14:116-119. lan S Francis, RSPB, 10 Albyn Terrace, Aberdeen ABIO lYP Revised manuscript accepted January1997 Jackdaw William Paton 134 Scottish Birds 19: 134-140 SB 19(3) The diet and foraging behaviour of the Red Kite in northern Scotland L WILDMAN, L O’TOOLE & R W SUMMERS Red Kites have recently been successfully reintroduced to northern Scotland to an area with a presumed rich food supply. This study showed that the diet was composed mainly of Rabbits, both in winter and summer. Other important components of the diet were voles and terrestrial invertebrates in winter, and birds, particularly Corvids, in summer. The kites also scavenged on sheep and deer carrion, household refuse and fish. Red Kites foraged over pasture, stubble, woods, the foreshore and village gardens. They employed different foraging techniques: pouncing from the air, hawking insects, scavenging and kleptoparasitism. The scavenging of Rabbits killed on the road makes the kites vulnerable to collisions with traffic, whilst feeding on shot Rabbits makes them susceptible to lead poisoning Introduction Ninety three juvenile Red Kites Milvus milvus, translocated as nestlings from Sweden, were released in northern Scotland between 1989 and 1993 as part of a scheme to reintroduce Red Kites to Scotland, after a period of extinction of about 100 years (McGrady et al 1994). Successful breeding was first recorded in 1992 and there has been a steady growth in the population subsequently, reflecting the success of the scheme (Evans eta/in press). The release sites were selected in areas of lowland farmland and woodland similar to Sweden where there are sufficient feeding and nesting habitats. Now that a breeding population is becoming established, their use of the habitat can be assessed. This paper describes the diet and foraging behaviour of the Red Kites both in winter and summer. Methods Many of the Scottish kites roost communally in woodland throughout the winter and early spring. The main communal roost was visited during January to April 1994 and 10 fresh pellets were sub sampled from alarge number collected each month. The pellets were broken up for examination and larger items removed from the general matrix (Yalden 1977). The matrix was scanned under a microscope for earthworm chaetae and insect remains (eg chitinous jaws of caterpillars). Mammal hair identification was initially based on shape, size, texture and pigmentation. If identification was not achieved, microscopic analysis was then used. Casts of guard hairs were made onafilm of 5% gelatin solution ona microscope slide. After the gelatin hardened, the hairs were removed and the imprint of the cuticular scale pattern was used for identification. DL —— 1998 Finally, confirmation of identification was based on the pattern of medullary cells, viewed at a magnification of x400, of hairs mounted in 70% ethanol (Day 1966). Bird identification was based on the lower barbs of feathers which were mounted in 70% ethanol, allowing the characteristics of the barbules to be seen at magnifications of x200 to x400 (Day 1966). During the breeding seasons of 1994 and 1995, 8and 14 nests were visited respectively and prey remains within and below the nests were identified and counted. If prey was dismembered, the numbers of limbs, wings and other identifiable parts were taken into account when counting items and minimum numbers estimated. Diet and foraging behaviour of Red Kites in northern Scotland 135 Results Diet In winter, the most frequently identified prey in pellets was the Rabbit, followed by sheep, voles, worms and insects (Table 1). Identified insects included caterpillars and beetles. Passerines included thrushes and Robins Erithacus rubecula. Insummer, the most abundant and frequently found prey item was again the Rabbit (Table 2). Corvids were the next most important prey. Three Chaffinch nests were found at Red Kite nests. It seems likely that the kites had plucked the entire nest from its site and eaten the contents. Table 1 The percentage occurrence of prey items in 40 pellets of Red Kites in winter and early spring in northern Scotland. See Table 2 for scientific names. January February Sheep 40 30 Rabbit 90 80 Rat 0 10 Voles 20 30 Mice 0 0 Mole 20 30 Shrews 0 0 Ducks 0 10 Galliformes 0 20 Gulls 10 10 Pigeons 30 0 Corvids 10 10 Other Passerines 20 40 Worms 10:0 = 20 Insects 30 30 No of pellets 10 10 March April All months 30 20 30 50 60 70 10 0) 5 30 40 30 20 0 5 0 20 18 10 0 5) 0) 0 3 20 20 1S 20 30 18 10 0 10 10 30 1) 10 20 23 40 50 30 30 30 30 10 10 40 SB 19(3) 136 L Wildman, L O'Toole & RW Summers v0 v0 v0 v0 oC v0 v0 v0 v0 8 Sel 6 | ol v0 oC v0 GIS omouowyrToOonWonmMowmMoWM WH ww Ou + CO v ie LO Oo 990U9IINDOO JU9DI9d JU9DI98d sieak ylog Jequinu Aq 90U981JNDDO JUa0I0d Jaquinu Aq ]U9D19d S661 oa 80 oa 80 oa GV Ge sy cl 80 cl 80 cl 8°0 Ge Gab oa 80 oa 80 8E Le OO GIS smjeb snyje5 SIJQJOJSUC XCJODOIOBJEYd snduin; sniajdojoA9 snsayy SAYJYONE|d sIINpa sniAW snjoeideo snjoaideg snaedoina snedeuuy SAIC SIAQ snoibeAiou snjyjey eaedona eaje! sisuadeo snda7z sninoiund snbejojax1iC g90UdIINDDO Jaquunu Aq JU80J9q 4UddI0d ve6l uay d1Sswog Bbeus JEAW ISCO! JO 80d! DJe| JO BDaIq Sa}eIQELaAu| Jayonsdwiny Japuno}4 QesD ljassny\ 199q 90Y BoyebhpsyH squie|/daaus yey UMOIG a|OW\ Sa|O/ 12H UMOIG WOQEY SWALI ASdYd sou BY] JEAU JO JE PAlJUAP! SULA]! WOdJ GEG L PUL PEGL Ul PUB/IOIS UsJBYJIOU UI JAWILINS UI Saji pay Hulpaesg jo JaIp ay] Z aqQeL N © To} SS SS SS SS ee Ne ee a EE ee eee ee G 8 5) 6G2 Gel vel SW}! JO JOQUNN S o a € S CZ Sill LS QZ 0Ol yell SPIAIOD < val Gl Z 870 G2 Gal SOU YOUNeYD 0 G 70 : : EL 370 sqaje09 eBy/IbUL4 youlyjeyo S G 70 Z 80 - = sojawiojiyd snpiny ysniy| Buos ra G 7'0 wh 80 : : BOlSNs OPUNIIH MO||EMS or OF OL OF 08 SE 09 snquinjed equinjop uosbidpoom iS vi Gil i 8°0 GZ Gi BIAI] EQUIN|OD uoebid je1e- =| 6 C7? » = G2 GY SIND) ae) - c : = 6 Aa l 8:0 om Gk snjejuebie snie] INS) Guia} s G v0 - - El 80 SJIOPEM 9 vb 61 LZ Ov : : snoiyojoo snueiseyd jueseoud 2 > Le 8 Xo} i v 90U9IINIDDO Jaquinu Aq 90U8IINDDO §=saquinu AG = = 9 2.4% 0.8% 2 0.6% 9 2.4% 0.5% 4 1.3% - = - : 1 0.3% “ 1.3% 2 1 0.3% - 2 0.6% - 0.5% = 2 0.5% 1998 Tn ——————————————__...._ Te Fish prey in the diet of Great Skuas at St Kilda 163 Table 2 Numbers and percentages of otoliths of different fish species from pellets regurgitated by Great Skuas from the club site at Hirta, St Kilda. Fish species Poor Cod Trisopterus minutus/ 453 Norway Pout 7. esmarki Bib T. /uscus i, Whiting Merlangius merlangus 28 Haddock Melanogrammus aeglefinus 5 Saithe Pollachius virens 4 Sandeel Ammooaytes sp. 8 Scad Trachurus trachurus 1 Blue Whiting Micromesistius poutassou = Unidentified gadids 3 Silvery Pout Gadiculus argenteus 6 Argentine Argentina sphyraena 6 Lepidion eques 2 Tadpole fish Raniceps raninus 1 Blackspot Grenadier Coelorhynchus coelorhynchus to feed on Sandeels (Harris 1984, Leaper et al 1988), but it may well be that they are unavailable to Great Skuas except through kleptoparasitism. Furthermore, regurgitations by Great Skua chicks did not contain any Sandeel in 1996 (Phillips et al 1997b), nor were Sandeel recorded in either adult or chick regurgitated food in 1997 (S Bearhop pers comm). This lackis particularly interesting given the dramatic effect the reduction of Sandeel availability at Shetland had upon the breeding success of Great Skuas at Foula (Hamer et a/ 1991). Furness & Hislop (1981) noted that Sandeels were fed to Great Skua 1994 1995 1996 84.5% 457 80.5% 66 51.6% 3.2% 12 2.1% = > 5.2% 60 10.6% 19 14.8% 0.9% 9g 1.6% 2 1.6% 0.7% WZ 3.0% 9 7.0% 1.5% - : > 0.2% 4 0.7% = - = - 12 9.4% 0.6% = = 3 2.3% 1.1% = = 16 12.5% 1.1% 1 0.2% = - 0.4% 2 0.4% = = 0.2% - 1 0.8% chicks at Foula in preference to whitefish or other foods. Furthermore, Hamer et a/ (1991 ) described poor chick growth, a decrease in the attendance of adults at the colony and poor breeding success in Great Skuas at Foula in association with a reduction in the availability of Sandeels. In response to this adult skuas fed to an increased extent upon seabirds and spent more time foraging (Hamer et al 991). That the Great Skuas at St Kilda appear to breed as successfully as at sites where Sandeel is thought to be an important prey (Phillips et a/ 1997b) is particularly noteworthy. 164 DR Thompson, K C Hamer & RA Phillips The majority of the fish species recorded are likely to be derived from commercial fishing operations near to St Kilda. Trawling and long line fishing activity have been noted over the continental shelf edge, to the west of St Kilda (Leaper et a/ 1988, pers obs). With few exceptions, these fish can be considered mid water or bottom dwelling species (Whitehead et a/ 1989), and, therefore, most likely made available to skuas through discarding. Noteworthy in this category are Silvery Pout Gadiculus argenteus, Lepidion egues and _ Blackspot Grenadier Coelorhynchus coelorhynchus which are either bathypelagic or benthopelagic, occurring at depths generally greater than 200m and often in excess of 1000m (Whitehead et a/ 1989). Deep water fish have been recorded in the diet of seabirds previously (for example, Imber 1973, Hamer et al 1994, Thompson et a/ 1995), but to our knowledge this is the first time that meso and bathy pelagic fish have featured in the diet of Great Skuas. References Baber|A 1992. Breeding ecology of some seabirds on Handa Island, Sutherland, with particular reference to the use of seabirds as indicators of food availability. MSc Thesis, University of Glasgow. Barrett R T & Furness R W 1990. The prey and diving depths of seabirds on Hornoy, north Norway after a decrease in the Barents Sea capelin stocks. Ornis Scandinavica 21:179-186. Furness R W 1979. Foods of Great Skuas Catharacta skua atnorth Atlantic breeding localities. Ibis 121:86-92. Furness R W & Hislop J R G 1981. Diets and feeding ecology of Great Skuas Catharacta skua during the breeding season. Journal of Zoology 195:1-23. Hamer K C, Furness R W & Caldow R W G 1991. The effects of changes in food availability on the breeding ecology of Great Skuas Catharacta skua in Shetland. Journal of Zoology 223:175-188. SB 19 (3) Hamer K C, Thompson D R, Rundle A J, Lewis S A & Stewart F M 1994. Mesopelagic fish eaten by yellow-legged herring gulls in the Azores. Seabird 16:30-33. Hark6nen T 1986. Guide to the otoliths of the bony fishes of the north east Atlantic. Danbui Aps, Hellerup. Harris M P 1984. The Puffin. T & A D Poyser, Calton. Harris MP & Wanless S 1991. The importance of the lesser sandeel Ammodytes marinus in the diet of the shag Phalacrocorax aristotelis. Ornis Scandinavica 22:375-382. Imber M J 1973. The food of grey-faced petrels (Pterodroma macroptera gouldi (Hutton)), with special reference to diurnal vertical migration of their prey. Journal of Animal Ecology 42:645-662. Jobling M & Breiby A 1986. The use and abuse of fish otoliths in studies of feeding habits of marine piscivores. Sarsia 71:265-274. Johnstone | G, Harris M P, Wanless S & Graves J A 1990. The usefulness of pellets for assessing the diet of adult shags Phalacrocorax aristotelis. Bird Study 37:5-ll. Klomp N | & Furness R W 1990. Variations in numbers of nonbreeding great skuas attending a colony. Ornis Scandinavica 21:270-276. Leaper G M, Webb A, Benn S, Prendergast H D V, Tasker M L & Schofield R 1988. Seabird studies around St Kilda, June 1987. Nature Conservancy Council, CSD Report No.804. Mund MJ & Miller G D 1995. Diet of the south polar skua Catharacta maccormicki at Cape Bird, Ross Island, Antarctica. Polar Biology 15:453-455. Phillips R A, Catry P, Thompson D R, Hamer KC & Fumess R W 1997b. Inter- colony variation in diet and reproductive performance of great skuas Catharacta skua. Marine Ecology Progress Series 152:285-293. Phillips R A, Thompson D R & Hamer K C 1997a. The population and feeding ecology of great skuas Catharacta skua at St Kilda. Unpublished report to Scottish Natural Heritage Inverness. Pollock K 1963. Great skua breeding on St Kilda. Scottish Birds 2:427. Thompson D R, Furness R W & Lewis S A 1995. Diets and long-term changes in 615N and 613C values in northern fulmars Fu/marus glacialis from two northeast Atlantic colonies. Marine Ecology Progress Series 125:3-ll. 1998 Fish prey in the diet of Great Skuas at St Kilda 165 Whitehead P J P, Bauchot M-L, Hureau J-C, Nielsen and Movement Control (Benbecula) for J & Tortonese F 1989. Fishes of the north-eastern help and support with fieldwork. For expert Atlantic and the Mediterranean. UNESCO, Paris. help with otolith identification we thank Werner Schwarzhans and for help with Acknowledgements collection of pellets in 1994 and 1995 we thank Jim Vaughan and Paul Tyler. We thank John Love and Nigel Buxton (SNH) David Thompson, Applied Ornithology Unit, Division of Environmental and Evolutionary Biology, University of Glasgow, Glasgow G12 8QQ Keith Hamer, Department of Biological Sciences, Science Laboratories, South Road, Durham DH1 3LE Richard Phillips, Applied Ornithology Unit, Division of Environmental and Evolutionary Biology, University of Glasgow G12 8QQ. Present address, Department of Biological Sciences, Science Laboratories, South Road, Durham DRI 3LE Revised manuscript accepted September 1997 Great Skua chasing tern Barry Larking 166 Scottish Birds (1998) 19:166-182 SB 19 (3) SHORT NOTES Common Buzzards cartwheeling during a food pass Aerial activity in early spring between Buzzards Buteo buteo usually consists of soaring and diving displays between pairs but contact between birds is rare beyond touching of wings (Cramp & Simmons 1980, Birds of the Western Palearctic, Vol 2, Oxford). Buzzards will turn over and present talons in courtship and an extension of this behaviour occurred on 27 March 1996 when 2 Buzzards circled each other over moorland near Newton Stewart, Wigtownshire. The male had a reptile, either an Adder Vipera berus or a Slow Worm Anguis fragilis dangling from his talons above the female. The female tried to collect the prey by a food pass and 3 times turned over on her back trying to take the prey, talon to talon, uttering calls. On the fourth attempt they locked talons and cartwheeled several times. The reptile dropped, the female unlocked her talons and deftly caught the prey. She glided down to some trees followed by the male. The only other documented records of talon locking appear to be of a Red-shouldered Hawk Buteo lineatus which engaged in talon locking and cartwheeling when it attacked a conspecific in its breeding area in North America (Johnsgard 1990, Hawks, Eagles and Falcons of North America, London). R C Dickson, Lismore, New Luce, Newton Stewart, Wigtownshire DG8 OAJ Accepted February 1997 Jim Young 1998 Cannibalism in a Merlin brood Relations among siblings in broods of falcons are usually peaceful and deliberate fratricide is unknown (Cade 1982, The Falcons of the World, London). As there is no apparent aggression among Merlin Falco columbarius siblings, cannibalism is also rare and evidence of it hard to obtain. The only direct evidence of cannibalism obtained during long term studies in Galloway was on 18 July 1976 when | made a routine visit to a Merlin’s nest on a heathery ledge deep in dense conifer plantations. The nest contained 2 fully feathered young about 23- 25 days old and one dead young which was lying on its back. It was obvious that | had just disturbed the 2 live siblings eating their dead sibling because they were standing over it with their beaks freshly blooded and the 28s. =e oe EE _ Merlin Short Notes 167 breast freshly plucked. The dead sibling could have died naturally as there was no visible injuries, but Adder Vipera berus predation could not be ruled out as an Adder was lying sunning just below the nest site (see also Shaw 1994, Scottish Birds 17:162). Fresh prey remains at the nest site included a Skylark Alauda arvensis, 2 Meadow Pipits Anthus pratensis, and 2 Chaffinches Fringella coelebs. The 2 young eventually fledged. The only other evidence of cannibalism in a Merlin brood is that given by Brown (1976, British Birds of Prey, London), who gave details from ringing returns of 3 dead young in a nest where one had apparently been eaten by its nest mates. Cade considered it likely thatin severely food stressed broods of falcons a chick that dies will be eaten by its siblings, although starvation did not seem to be the cause at the nest in Galloway. R C Dickson, Lismore, New Luce, Newton Stewart, Wigtownshire DG8 OAuJ. Accepted February 1997 168 Short Notes Sys) IES) Mobbing behaviour by Barnacle Geese on a ground predator On 19 March 1996 at Caerlaverock WWT Reserve, Dumfries & Galloway, | was ob- serving a flock of around 2000 Barnacle Geese Branta leucopsis which were feeding on improved pasture. Suddenly, all the birds appeared very alert and uneasy. After about 5 minutes of vigilance by the goose flock a single Red Fox Vulpes vulpes appeared from the corner of the field and began walking very slowly towards the fence line bordering the neighbouring field (see diagram). MERSE FENCELINE a About 300 geese immediately took flight from the main flock and flew directly over the fox at a height of about 20m from the ground. These birds were continually calling. The flying geese then circled back over the re- maining flock on the ground but did not land to rejoin them. Instead, they flew back over the fox at the same height, calling constantly. Some birds even tried to hover over the mammal. This mobbing behaviour, which lasted a total of 3 minutes, was repeated 5 times in succession over the fox. It only stopped when the creature had reached the fence line of the neighbouring field. During FENCELINE A FLOCK OF MOBBING GEESE 1998 the mobbing some of the birds returned to the vigilant flock on the ground and at the last (fifth) mobbing attempt there were only around 200 birds left in the air. When the flying geese were mobbing the fox, the remaining geese on the ground were vigilant and calling con- stantly. | have been working on Barnacle Geese for 3 years at Caerlaverock and this Is the first time that | have witnessed geese taking any attacking action against a ground predator. The usual behavioural action taken by Bar- Short Notes 169 nacle Geese towards a fox at Caerlaverock is simply the alert posture, although some- times birds will walk towards and almost follow a fox until it is out of sight. This obser- vation of geese acting as a flock when mob- bing, and also defending, compares with other observations on Barnacle Geese seen at Caerlaverock in response to the presence of a Peregrine Falcon Falco peregrinus. (Patterson, Scottish Birds 18:101-102). How- ever | can find no reference to Barnacle Geese mobbing a ground predator. David J Patterson, Wildfowl & Wetlands Trust, Eastpark Farm, Caerlaverock, Dumfriesshire. DG1 4RS Accepted March 1997 _ Barnacle Geese Rodney Dawson 170 Short Notes Marsh Warblers breeding in Orkney in 1993: a first for Scotland On 14 July 1993, one of us (RGA) visited an area of reed swamp Phragmites australis in order to clear rides for mist netting As the work progressed, a single net was erected and an Acrocephalus warbler was caught. It was identified as a Marsh Warbler A. palustris on the basis of its general colouration and wing formula (Svensson L 1992, /dentification guide to European Passerines, 4th edition Stockholm). The bird was judged to be an adult and, as June is the normal month for the spring passage of this species in the northern isles, it was presumed to be a lingering migrant. However, on 20 July, asecond adult bird was trapped by RGA while the first was retrapped. The second bird was also identified as an adult Marsh Warbler although the wing formula was not as clear cut as on the first bird. On 22 July one of these birds was watched by RGA on 6 occasions as it carried large beakfulls of food and also called in a very agitated manner; a second bird was seen very briefly. RGA contacted ERM who visited the reed bed shortly afterwards. He too saw one, probably 2, birds one of which was seen with food, called repeatedly and sang frequently. This bird only gave brief views but appeared to be a Marsh Warbler with olive toned upperparts, distinctly pale tipped inner primaries and a song with an outstanding amount of mimicry. The activities of the birds were centred on an area of the Phragmites bed about 30m from its boundary with improved grassland and 80m from the open water of the nearby loch. Atthis point, running through the reeds, was a small ditch edged with Meadow Sweet Filipendula ulmaris, Marsh Marigold Cal/tha palustris, Forget-me- SB 19 (3) not Myosotis sp and Water Mint Mentha aquatica. Also close to the ditch is a substantial Willow Salix so bush from which the bird did most of its vocalising. On 24 July one of the ringed birds was again glimpsed by RGA and also heard and seen briefly by T R Dean who also considered it to be a Marsh Warbler. Thereafter, windier weather conditions made watching in the reed bed very difficult so that no further observations were made. On 11 August, RGA was mist netting for Sedge Warblers Acrocephalus schoenobaenus when he caught 2 unstreaked Acrocephalus warblers. The freshness of the plumage indicated that these were both immature birds while the lack of rufous in the upper parts and the length of the 2nd primary notch strongly suggested that they were Marsh rather than Reed Warblers A.scirpaceus (Svensson 1992). On 16 August, a third immature bird was caught and ringed by RGA and similarly considered to show characters of Marsh rather than Reed. None of the 5 birds was seen again after this date. The whole scenario suggested that a pair of Marsh Warblers had successfully fledged 3 young. In an attempt to verify the records ERM and Martin Gray independently examined the wing formula data for the 5 birds. The data were graphed using the methods described in (Walinder G, Karlsson L & Persson K 1988 Anew method for separating Marsh Warblers from Reed Warblers Ringing & Migration 9:55-62.). Using these methods the first adult was conclusively identified as Marsh Warbler but the second adult fell in the Marsh/ Reed overlap zone (Adam & Meek, 1994). The 3 immature birds were considered next. Using one of Walinder’s graphing methods (wing length v_ notch length) all 3 fell just within the range of Marsh. However, using his other method (wing length v notch position) 1998 produced a different picture with 2 birds falling within the range of Reed and the other falling close to the overlap zone. One explanation of these discrepancies between the 2 methods was, of course, that the immature birds, especially those measured on 11 August, did not have fully developed wings when trapped. Notch length is fixed once the feather has emerged from pin but its position in relation to the other primaries Is not set until the feather has completed growth. Because of the discrepancies we felt that it was not possible to claim, as certain, a first breeding record for Scotland although the evidence did strongly suggest this to be the case. A paper in the Orkney Bird Report 1993: 73-76, Did Marsh Warblers Acrocephalus palustris breed in Orkney in 1993? Adam RG & Meek ER _ 1994, reported this situation and the details were repeated in the report on Rare breeding birds in the United Kingdom in 1993, British Birds 89:2:87. Subsequently ERM and RGA were contacted by JH of Westphalia, Germany who has been working on the Marsh Warbler population of that area and who had developed an interest in the birds of Orkney on holiday in the islands in 1992. He confirmed that the description of the breeding territory fitted perfectly with the habitat utilised in Westphalia. Turning to the morphometric Short Notes 171 data, rather than the Walinder method, he applied the methods described in H Dorsch (1983 Bewertung verschiedener Merkmale zur sicheren Unterscheidung von Teichund Sumpfrohrsanger (Acrocephalus scirpaceus, A.palustris) mit einer praktischen Bestimmungshilfe Ber. Vogelwarte Hiddensee H.4: 111-120.) to the published data in the Orkney Bird Report. Using this method, the first adult and all 3 of the immature birds were clearly Marsh Warblers; only the second adult could not be certainly identified as Marsh. However, because the second adult was also identified as Marsh on general colouration and because the Walinder method placed it in the overlap zone and not in the Reed Warbler zone (see graphs in Adam & Meek, 1994), we are now confident that a pair of Marsh Warblers did indeed breed successfully in Orkney in 1993 and that this represents the first breeding record of the species in Scotland (Thom VM 1986 Birds in Scotland Poyser, Calton). The birds did not return to the site ialthough a pair did breed successfully ata site 15km to the south in SS. Acknowledgements We are especially grateful to Jorg Hadasch for his interest in this record and for his efforts in helping us to establish its validity. We also thank Martin Gray for his independent examination of the wing formula data. E R Meek, Smyril, Stenness, Orkney KW1 6 3JX R G Adam, Northolme, Old Scapa Rd., Kirkwall, Orkney KW1 5 1 SB J Hadasch, Uhlandstr 18, 32051 Herford, Germany Revised manuscript accepted April 1997 172 Short Notes An unusually open Wigeon nest At 0800 on 23 June 1996 near Ballater, Deeside we noticed a female Wigeon Anas penelope lying very flat and still in short grass approximately 10m from a small roadside pool. She was still in this position when we left 10 minutes later. In the evening we were Surprised to see the female Wigeon still inthe same place, again lying very low and still. SF began to walk towards the sitting bird 30metres away. She quickly flushed and flew away. TO our amazement she had been sitting on a nest which contained 5 eggs. The grass was no longer than a couple of centimetres with grazing cattle present in the field. The pool, which was approximately 20x8m and drying out fast, was in the centre of the field and 30m from any cover. Although Fe SB 19 (3) the Wigeon had made her nest in a slight hollow, the surrounding grass was so short that even when lying flat the top half of her body was visible. Wigeon nests are normally well hidden in vegetation. Of 16 nests found during a Wigeon study at another site in Deeside, all were fairly well hidden amongst Heather Calluna vulgaris, Rushes Juncus sp or rank grass. The Ballater nest was visited again on 29 June when it was found to contain 4 eggs. The female was still sitting on 9 July but there was no sign of her on 14 July when the nest contained 3 eggshells, from which chicks appeared to have successfully hatched, and an infertile egg. The laying date, when calculated back from the approximate date of hatching, and the clutch size suggested this was probably a repeat clutch. Raymond Duncan, 86 Broadfold Drive, Bridge of Don, Aberdeen Simon Foster, 17 Malcom Canmore, Tain, Ross-shire edd 7 7, 4g. Drake Wigeon 4 47, tl, yj, Montes tte Mag his Sige TA — as Accepted May 1997 ld David Mitchell 1998 Song Thrush apparently nesting on the ground On 12 April 1997, a member of the Royal Dornoch Golf Club informed me that on 2 recent occasions he had flushed a Song Thrush Turdus philomelos from 2 eggs ona bare patch on the ground below Gorse Ulex europaeus bordering one of the golf course fairways. Visiting the site later that day, | found that it had been predated and all that Short Notes 173 remained were several pieces of Song Thrush eggshells, which lay on apiece of bare ground below the overhanging Gorse. An inspection of the cover overhead failed to reveal any sign of anest from which the eggs might have fallen out. BWP Vol 5:998 describes various Song Thrushnestsites ending with the words, ‘and on the ground among thick vegetation’ which, | assume, refers to a normally constructed nest. D Macdonald, Elmbank, Dornoch, Sutherland 1V25 3SN Accepted May 1997 Andrew Dowell I ee 174 Short Notes Hunting times of Sparrowhawks in the non breeding season During studies of the hunting times of Merlins Falco columbarius and Hen Harriers Circus cyaneus (Scottish Birds 17:56-58, 18:182- 183) in winter in west Galloway, | also saw Sparrowhawks Accipiter nisus (adults and juveniles) hunting regularly in the same open country habitats between August and SB 19 (3) February 1970-97. Each year all hunts were timed in open country only, where | spent an equal amount of time at different hours of the day (see Dickson 1992, The Birds in Wigtownshire, Wigtown for details of sightings in various habitats). Two hundred and fifteen hunts were recorded by bothclasses: 131 by adults, 84 by juveniles (first years). Sixty nine percent of adult hunts 20 10 Se 0 0700 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 = Adults Juveniles (first years) Figure 1 Hunting times of Sparrowhawks in west Galloway, November - February 1970-97 12 0700 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 = Adults Figure 2 Hunting times of Sparrowhawks in west Galloway, August - October 1970-97 | ee Juveniles (first years) 1998 occurred in the period November-February (Fig 1) and 31% in August-October (Fig 2): 62% of juvenile hunts were in November- February (Fig 1) and 38% in August-October (Fig 2). Both classes showed pronounced peaks of activity at 1030 and 1500 hours in the period November-February. Inthe period August-October there was no pronounced peak of activity during the day by bothclasses but there was a definite peak by juveniles at 1830 hours (BST). Short Notes 175 The findings in this study of major peaks of activity in mornings and afternoons with lulls in activity around midday agree with the findings on the hunting times of other raptors in winter. The main peaks in Figure 1 by juveniles coincide with the winter peaks described by Newton (1986, The Sparrowhawk, Calton), but the more pronounced peak of activity in the afternoon shown by adults in Figure 1 was not recorded by Newton. RC Dickson, Lismore, New Luce, Newton Stewart, Wigtownshire DG8& OAJ Revised manuscript accepted June 1997 27 RT de Sant Sere As asa Se re Brent Hurley 176 Short Notes Merlin’s sunning behaviour in winter Sunning behaviour by Merlins Falco columbarius in winter has rarely been observed in the wild. In a list of species recorded sunning, Merlins were not mentioned (Kennedy 1969, British Birds 62:249-258) although sunning was subsequently recorded (in Summer?) in the North American subspecies (Cade 1982, The Falcons of the World, London). Informaton on sunning behaviour by wild Merlins in winter in Britain is thus generally lacking and probably under recorded. On 19 January 1997 at 1105 GMT, a relatively mild, cloudless day (9°C) with warm sun and very light winds, | watched a SB 19 (3) female or juvenile Merlin preening vigorously on a fence post on low ground in Wigtownshire. After preening it stood, back to the sun, with its wings held partly down and extended in the typical ‘ loose spread sun basking posture’ (see Simmons 1986, The Sunning Behaviour of Birds, Bristol). Eleven minutes later it stood with only its right wing partly extended sun basking before flying 5- 7m to another fence post. Here it stood for the next 4 minutes sun basking in the loose spread posture before flying away. Feather care used by Merlins in winter include preening, shaking and wing stretching. Cade considered that this loose spread posture is the typical sunning posture of falcons and likely to be shown by all species, although not previously documented in the wild by Merlins in mid winter. RC Dickson, Lismore, New Luce, Newton Stewart, Wigtownshire DG8 OAJ Accepted June 1997 Arthur Gilpin 1998 Territory sizes of Crested Tits at Abernethy Forest, Strathspey The Scottish Crested Tit Parus cristatus scoticus is confined to the Highlands of Scotland where it inhabits native pinewoods and old plantations, composed largely of Scots Pine Pinus sylvestris (Cook 1982, Breeding status of the Crested Tit. Scottish Birds 12:97-106). Itisan amber rated species in the list of species of conservation concern in the United Kingdom (Gibbons et a/ 1996, Bird species of conservation concern in the United Kingdom, Channel Islands and the Isle of Man: revising the Red Data List. RSPB Conservation Review 10:7-18). Elsewhere in its range, the Crested Tithas been the subject of detailed studies of individual birds (eg Short Notes 177 Ekman 1979, Coherence, composition and territories of winter social groups of the Willow Titand Crested Tit. Ornis Scandinavica 10:56- 68) but there has been little work in Scotland where the Crested Tit is generally regarded as being sedentary (Cramp & Perrins 1993, Birds of the Western Palearctic, Vol 7, Oxford). However, there have been few direct observations made of individual birds to confirm this. This note provides some detailed observations on a small group of birds. The study was carried out around Forest Lodge (57°14’N, 3°37’W) in Abernethy Forest where 6 adult Crested Tits were colour ringed with unique combinations. The study area comprised c.300ha of native pinewood and Scots Pine plantation, and set routes were NJO2 Figure 1 The study area at Forest Lodge. The habitat is native pinewood apart from | young (heavy shading) and old plantations (light shading). Dashed lines are roads. Grid } reference numbers are shown. 178 Short Notes SB 19 (3) NJO2 Figure 2 The locations of 4 Crested Tits in summer (open symbols) and winter (filled symbols). OR and YM were paired. walked during summer (24 May 1994-11 July 1994) and the following winter (2 November 1994-9 February 1995) to search for the marked birds (Fig 1). Once located, their positions were mapped to the nearest 10m. Observations one day apart were treated as independent of others and territories measured by joining the outermost points and measuring the size of the enclosed area (minimum area method; Mohr in Southwood 1966. Ecological Methods, Methuen). Because one requires a minimum of about 25 plots to estimate territory size, data for the 2 seasons had to be combined and, even then, this provided sufficient data for only 4 birds. Their territory sizes were (12.5, 13.1, 14.0 and 15.1ha (mean = 13.7ha). Birds OR and YM were a pair and had similar sized (12.5 and 13.1ha) and overlapping ranges (Fig 2). The territories of 2 neighbouring birds of this pair overlapped considerably, showing that territories were not mutually exclusive. 1998 There appeared to be changes in the sizes of territories between summer and winter but the results were not consistent, and insufficient observations were made of most birds to give realistic estimates of territories in each season. The pair (OR and YM), had smaller ranges in winter (7.1ha, n=17 and 6.8ha, n=18) compared with summer (10.4ha, n=24 and 10.3ha, n=16). In contrast, birds BR and SW had smaller ranges in summer (3.3ha, n=15 and 1.6ha, n=9) than in winter (15.1ha, n=30 and 12.6 ha, n=17). In order to find out if there were shifts in territories between summer and winter, the centre of each territory of all 6 birds was determined for the 2 seasons separately and the distances apart were measured. These shifts in centres were only 146m apart on average (range 81-256m). These distances were small in relation to overall territory size, indicating that these Crested Tits had fixed Short Notes 179 territories. The results of this limited study confirm that Crested Tits are sedentary, at least in a native pinewood in Scotland. Territory sizes of Crested Tits in other situations are likely to be different. The average size of 21 group territories in Swedish forests of Norway Spruce Picea abies, and Birch Betula spp. was 20ha (Ekman 1979). Other tit species tend to have smaller territories, though there is much variation depending on habitat and breeding density. For example, Great Tits Parus major have territories which range in size from 0.2 to 4ha (Gosler 1993, The Great Tit. Hamlyn). The birds were trapped and ringed by R Proctor and observations were made by M Adams, P Benstead, R Denny and C Jeffs. Territories were calculated by Dr M McGrady. The draft was commented on by Drs M | Avery and | P Bainbridge. R W Summers, RSPB North Scotland Regional Office, Etive House, Beechwood Park, Inverness IV2 3BW Accepted July 1997 . Crested Tit Steven Brown 180 Short Notes SB 19 (3) Baby Stoat in Peregrine nest On 8 June 1997 whilst ringing 2 small Peregrine Falco peregrinus chicks in an old Raven's Corvus corax nest in Strathcarron, Sutherland | was surprised to find a freshly dead Stoat Mustela erminea lying uneaten in the nest. It may have just been brought in by one of the adults as | had flushed the female from the nest as | approached. The Stoat was about 11cm in length from nose to tail with very short pink fur and looked as though it was still blind. It had obviously still been totally dependent on its parents. There were no other prey remains in the nest, apart from some feathers from a Grouse Lagopus sp and a Starling Sturnus vulgaris. Although this bird had laid 4 eggs there were only 2 chicks, one noticeably smaller than the other. Both also had empty crops indicating that prey may have been scarce. Ratcliffe (The Peregrine Falcon 1980) states that small mammals are very occasionally taken by Peregrines as is carrion. There is also a record of a Peregrine stealing a vole (species unknown) from a Kestrel Falco tinnunculus. One can only speculate whether the Peregrine obtained the Stoat from an adult Stoat carrying its youngster, from another raptor or by finding it freshly dead Presumably, if avian prey is scarce, the adults will resort to more unusual prey in order to feed their hungry young. RL Swann, 14 St Vincent Road, Tain, Ross-shire 1V19 1JR Accepted July 1997 > It hy Ie, et BT rr “o. ~- Peregrine at nest ms x A” oe Jim Young \ i] | = 1998 Short Notes 181 Oystercatcher incubating Lapwing clutch In 1995 at Colquhar, Leithen Water, Borders, a pair of Oystercatchers Haematopus ostralegus hatched a clutch of Lapwing Vanellus vanellus eggs, which produced 3 chicks, at least one of which survived for over 12 days (Dougall 1996, Scottish Birds 18:184). In 1996 at the same site a pair of Lapwings reared 3 chicks unaffected by Oystercatchers which were also present. In 1997, at the same site, | again found an Oystercatcher incubating a clutch of 3 Lapwing eggs on 3 May. On 21 April | had noted 2 Oystercatchers and 2 Lapwings on site, none of which was incubating. On 26 April an Oystercatcher was seen sitting ona nest with a Lapwing standing close by. The nest was visited on 3 May, when there was no danger of trampling by sheep. On 10 May the Oystercatcher was again sitting, with its mate and one Lapwing nearby. On 18 May there were no birds on the site and the nest was not relocated, having failed for an unknown reason. It seems probable that the same (infertile?) Oystercatcher was involved in both the 1995 and 1997 events. Oystercatchers were not seen again in 1997, but, on 31 May, 2 Lapwings were back in the area, although they were not seen to incubate on that or later dates. A single Lapwing with a brood of 2 was seen adjacent to the site on 10 and 15 June, but it is thought that this family had moved into the area from nearby. Again, | am grateful to Mr Templeton, the Colquhar shepherd, for his tolerance of my activities. Tom Dougall, 62 Leamington Terrace, Edinburgh EH10 4JL Accepted August 1997 Oystercatcher A D Johnson 182 Short Notes Song Thrush predation on marine gastropods Shell fragments belonging to 4 gastropod species were collected from an ‘anvil’ used by a Song Thrush Turdus philomelos on a rocky shore in south Harris during April 1997. Of 275 shells recovered, 244 were of the 4.00 8.00 12.00 L. littorea on shore. n = 92 4.00 8.00 L. littorea predated. n = 85 4.00 8.00 12.00 4.00 8.00 L. obtusata predated. n = 11 12.00 16.00 Shell height / mm 12.00 16.00 20.00 Shell height / mm 8 ews ") ih - ii i cee eae 16.00 16.00 Shell height / mm SB 19 (3) Edible Periwinkle Littorina littorea, 28 of the Flat Periwinkle Littorina obtusata, 2 of the Dog Welk Nucella lapillus and there was a single Littorina saxatilis. Observations of a Song Thrush breaking shells and examination of the discarded shell fragments revealed that the gastropods were held in the beak with the spire pointing to the left and struck 20.00 24.00 28.00 24.00 28.00 20.00 24.00 28.00 Shell height / mm 20.00 24.00 28.00 Figure 1 The shell heights of periwinkles Littorina littorea and Littorina obtusata found around the ‘anvil’ of Song Thrush and available on the shore. 1998 _ usually 3 to 4 times. The effect of this is either _ to break a hole in the side of the shell or to break the shell in half. Previous records _ suggested that such predation is largely associated with harsh weather conditions (Feare, C J 1966. British Birds 60:412-414). However, the quantities of shells to be found around anvils, and the regularity with which _ Song Thrushes were seen to behave in this way, suggests that this may not be the case _ in south Harris. Maximum height was measured for the _ predated shells of 85 L. /ittorea and 11 L. obtusata that were either holed or could be pieced together from broken fragments. , These were compared with the maximum shell height of 92 L. /ittoreaand 40 L. obtusata collected from the same shore at low tide (Fig | 4). Short Notes 183 Selection of gastropods by Song Thrushes was both species and size specific. L. obtusata was the most numerous gastropod on the shore and WN. /apillus was particularly conspicuous, being found on exposed rocks, yet many more L. /ittorea were taken than either of these species. The size of L. /ittorea taken in Harris corresponded to the size range of L. littoreain Wales that represent the adult cohort, those over approximately 4 years of age. (Williams E E 1964. The growth and distribution of Littorina littorea (L.) on a rocky shore in Wales. Journal of Animal Ecology 33:413-432). Selection of L. obtusata is also directed at the larger individuals in the population. Size preferences by Song Thrush presumably reflect a trade off between the amount of effort required to break the shell and the size of the gastropod obtained. Why Nucella lapillus are not eaten in greater numbers is not known. Alastair Pout, Flat 2, 60 Heslington Road, York YO1 5AU Accepted September 1997 _ Song Thrush / 7 7 Sr if a 5 AON fA CLE jae Bz Ze Shue, tg a perm le CEE epee ° a 7 David Mitchell 184 Short Notes 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 will be 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; 2 copies are required and the author should also retain one. We are happy to accept Erratum SB 19 (3) papers on computer discs, however, please contact Sylvia Laing on 0131 556 6042 to discuss this. 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 Vol 17:146-159). 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 5 August 19S oe Cu eee but on the 5th (if the name of the month does not follow). Please note that papers shorter than c700 words will 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 shouldbe kept as simple as possible, with footnotes used to provide extra details where necessary. Each table should be on a separate sheet. Maps and diagrams should be in black ink and be camera ready, but drawn so as to permit reduction from their original size. Scottish Birds 19:2 pp105-113 Habitat use by Snow Buntings in Scotland from spring to autumn Tablet Page 108 Flocks adults & fledglings - Beside snow should read 36° and not 36. Defective copies of Scottish Birds We were disappointed to learn that some copies of the December 1997 issue had faults. So far, we know of 7; unusually the printing faults differ between each. If your copy is defective please return it and we shall send a replacement. NEOTROPICAL BIRD CLUB Neotropical bird club launched A club has been launched to promote the study and conservation of the birds of the Neotropics (South America, Central America and the Caribbean). It is currently seeking founder members to help reach the launch budget of £2000, which is required to get the club running and to publish the two first issues of its intended journal ‘Continga’. Founder members will be asked to pay a minimum of £25, and will be formally acknowledged in the first issue of ‘Continga’. ‘Continga’ will provide a colourful and much needed forum for exchange of information on the avifauna of this extremely rich and diverse area, and will contain papers and features on the birds and their conservation as well as news of recent observations and discoveries (at present, new species are still being discovered at the rate of more than two a year). It is hoped that in due course the club will be able to provide direct funding and support for practical conservation programmes. For further details and membership forms, please contact: Rob Williams, Publicity Officer, Neotropical Bird Club, c/o The Lodge, Sandy, Bedfordshire SG19 2DL Scottish Birds Volume 19 Part 3 March 1998 Contents Main papers Observations at a winter Rook and Jackdaw roost in Aberdeenshire. | S Francis ) 129 The diet and foraging behaviour of the Red Kite in northern Scotland. L Wildman, L O'Toole & R W Summers 134 Breeding biology of Swallows in Easter Ross. D P Butterfield & A D K Ramsay 141 A survey of Storm Petrels on the Treshnish Isles in 1996. G Gilbert, D Hemsley & M Shepherd 145 The status of Storm Petrels on Mousa, Shetland. N Ratcliffe, D Vaughan, C Whyte & M Shepherd 154 Fish prey in the diet of Great Skuas at St Kilda. D R Thompson, K C Hamer & RA Phillips 160 Short notes Common Buzzards cartwheeling during a food pass. RC Dickson 166 | Cannibalism in a Merlin brood. RC Dickson 167 | Mobbing behaviour by Barnacle Geese on a ground predator. | David J Patterson 168 | Marsh Warblers breeding in Orkney in 1993: a first for Scotland. E R Meek, R G Adam & J Hadasch 170 An unusually open Wigeon nest. Raymond Duncan & Simon Foster 172 Song Thrush apparently nesting on the ground. D Macdonald WAS) Hunting times of Sparrowhawks in the non breeding season. R C Dickson 174 Merlin’s sunning behaviour in winter. RC Dickson 176 Territory sizes of Crested Tits at Abernethy Forest, Strathspey. R W Summers FE Baby Stoat in Peregrine nest. RL Swann 180 Oystercatcher incubating Lapwing clutch. Tom Dougall 181 Song Thrush predation on marine gastropods. Alastair Pout 182 Erratum Scottish Birds 19:2 p108 Habitat use by Snow Buntings in Scotland from spring to autumn. Adam Watson 184 Advice to contributors 184 Published by the Scottish Ornithologists’ Club, 21 Regent Terrace, Edinburgh EH7 5BT. © 1998 Scottish Birds The Journal of the Scottish Ornithologists’ Club Editor: Dr S da Prato Assisted by: Professor D Jenkins, Dr J B Nelson, Dr | Bainbridge, Dr M Marquiss and R Swann Business Editor: The Secretary SOC, 21 Regent Terrace Edinburgh EH7 5BT (tel 0131-556 6042, fax 0131 558 9947). Scoitish Birds, the official journal of the Scottish Ornithologists’ Club, publishes original material relating to ornithology in Scotland. Papers and notes should be sent to The Editor, Scottish Birds, 21 Regent Terrace, Edinburgh EH7 5BT. Two issues of Scottish Birds are published each year, in June and in December. Scottish Birds is issued free to members of the Scottish Ornithologists’ Club, who also receive the quarterly newsletter Scottish Bird News, the annual Scottish Bird Report and the annual Raptor round up. These are available to non- members at a subscription rate (1997) of £36. 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 21 Regent Terrace, Edinburgh EH7 5BT is the most comprehensive ornithological library in Scotland, and is available for reference during office hours (Monday to Friday, 0930-1630 hrs). (But please phone beforehand). SOC annual membership subscription rates Adult £18.00 Family (2 adults and any children under 18) living at the same address £27.00 Junior (under 18, or student under 25) £7.00 Pensioner/unwaged £10.00 Pensioner Family (2 adults living at the same address) £14.50 Life £360.00 Life Family £540.00 A\l subscriptions may be paid by Banker’s Order and Covenanted. Subscriptions paid this way will greatly assist the Club. Published by the Scottish Ornithologists’ Club, 21 Regent Terrace, Edinburgh EH7 5BT Printed by Douglas Gauld & Co Limited, 22 Tannadice Street, Dundee DD3 7QH Scottish Birds (1998) 19: 185-194 185 Minimal numbers and habitat of breeding Dunlin in north east Scotland R RAE & A WATSON National surveys consider breeding Dunlin scarce in north east Scotland. We estimated at least 436 pairs on only some potential habitat. Many were on alpine land, most on moorland blanket peat, some on glen peat and dune slacks. Numbers on 4 alpine and 4 high moorland areas showed no trend in 1967-95 but on some moorland may have fallen since 1980. We saw none at 8 low level sites since 1987 after habitat loss or deterioration. Rapid methods in transects and national surveys readily overlook this skulking breeder. Introduction and study area Recent books (Buckland et a/ 1990; Gibbons et al 1993) consider breeding Dunlin Calidris alpina scarce in north east Scotland. Here we give ahigher minimal estimate and assess habitat in the shires of Aberdeen, Kincardine and nearby parts of Angus, Perth, Banff, Moray and Inverness (Figure 1). Methods In 1943-86 we noted Dunlin during other fieldwork covering thousands of miles on land seldom visited by people. In 1987-97 we went specially to look for Dunlin. An egg, chick, rasping call or distraction display signified breeding and courting, nest scraping or a nest cup possible breeding. Method A involves birds with young<1 week being conspicuous on high points. On approach of aman, dog or Fox Vulpes vulpes, they flew from up to 100m to <10m to give distraction display, rasping calls and squeals. Method A misses birds on late nests and failed birds that have left. Method B is to scan for the cryptic, often motionless and silent birds while walking slowly with frequent stops, Zigzags and returns to previous stops. Although conspicuous when singing oncalm mornings and evenings, they seldom sang on windy or warm afternoons except soon after spring arrival. Nesting birds skulked, and were easily overlooked (Campbell & Ferguson-Lees 1972). Method C is that AW saw Dunlin adults, nests and young while counting Red Grouse Lagopus lagopus scoticus and Ptarmigan Lagopus mutus with pointing dogs (Jenkins et al 1963; Watson 1965). As in B, some Dunlin flushed far ahead, and were easily missed unless dogs were worked close. Method D is to hear the dawn/dusk chorus (Watson & O’Hare 1979a & b; Watson & Rae 1987; Parr 1990). From spring arrival to having small young, Dunlin sang for c1 hour from 10-15 minutes after sunset and from >1 hour before sunrise. This can miss isolated birds; R Parr (pers comm) heard none at dawns near Ballater but between dawns saw 186 RRae&A Watson SB 19(4) Figure 1 Minimum number of pairs seen in 1987-97. ? -- potential habitat viewed from afar with binoculars (0) -- no birds seen, apparently lacking potential habitat on farms, woods, freely drained moor and steep slopes, but only cursory inspection. Blank squares were outside our area. Most apparently lacked potential habitat, but 4 squares south east of Forres with breeders in 1956-70 (AW) were unvisited by us since. 00 10 20 30 is 50 [@0 70 30 90 NJ I ; lores shake | Rwledy leyinianl Fraserburgh SS bat i ~~ 2 Loch of e Strathbeg NY Forres 2a + / Be | | Peterhead nm {Nene albeit a 40 0 0 1. i aa, | __3 H | * Grantawn- ; (1) (13) | (1) 0 on-Spey | as ‘i ries A 20 yada ey | by Penal Cian, Aline2h ailnnO. cyl 0 ; Avismore | Strathdon [ce a 10 | | ABERDEEN ? {6} (10) (4) | (5) (5) | (0) (9) Cairngorms H Massif Ballater Banchary (4) 29 23 8 15 0 (2) 0 ’ e Braemar | weal 90 | i) | 48 | 2 | 7 | 41 | eo | @ | Ww | @) o bach Muck?) ain | ae i 80 Glas Maol 4 The our {5) (2) (4) 33 (18) (24) (1) ? (3 0 | ait NP SSS tC Te a reas zo ? | ? ? | ? ? ? ? (8) (6) ( ) = INCOMPLETE SURVEY | | | Fie pull a: i Montrose NO fittiernuir 4 | ACG pw | By | | 60 one oneggs. Method D requires prior daylight experience of Dunlin, walking on bogs in darkness and no strong wind or heavy rain. The number of pairs seenin daytime searches on Glas Maol by methods A and C was similar to the number heard singing by D (Table 1). While watching Foxes, J Robertson (pers comm) found up to 20 pairs by method A in the 1970s at Black Burn, Loch Muick, where RR saw 10-20 pairs by A+B. Data by A+B, C and D agreed well in a Caithness study (Parr 1990). Methods A, C and D may be accurate, and B sometimes so, but need to be tested against territorial spacing (Holmes 1970). We used A, B, C and D in that frequency. iin i Scottish Birds (1998) Numbers and habitat of breeding Dunlin in north east Scotland 187 Table 1 Consistency within years in number of pairs seen in similar weather and date by methods A, C and D on the same area of Glas Maol. Year A C no dogs dogs D, dawn/dusk watches 1967 1969 1970 19%2 1982 1984 1986 1987 1988 —_ —t ODDOAD OOO OD — ak iO) Co) SIO? SIO) OON ON We covered only some sites each year but assumed that each held similar numbers each year. This was valid when we visited sites in>1 year (Table 2). Of >100 sites with breeding in 1 year, all held birds when visited in a later year, bar those where there had been only 1 pair, or habitat loss. Results Alpine land About 15% of pairs seen were above 800m in a) Three-leaved Rush Juncus trifidus, b) grass, sedge or lichen heath, and c) blanket peat (Appendix 3). Groups (territorial pairs within 150m) and isolated pairs were at 800- 1100m, but above 1100m only the latter. Most usedb) andc), andafewa). Mostnests ona) and b) were in tufts of rush or Mat Grass Nardus stricta, and some in Crowberry Empetrum nigrum. Some birds near paths sat tight, apparently used to people. Most nests lay near flushes, but a few were up to 400m. Birds on a) and b) nested on freely drained land, buttook young to flushes, pools, lochans, or blanket peat. Flying young were on b) and c), and all Dunlin in alpine post breeding flocks of Golden Plover and Dotterel on b). Highest nests were at 1210monCairnLochan and 1230m by Wells of Dee which was a regular site in the 1930s (D Nethersole- Table 2 Number of pairs seen on the same parts of 4 alpine hills in Glas Maol area and 4 high moorland ones (3 in Ladder Hills, fourth north of Ballater). 6/5 O Saar Oi Valiec2 Glas Maol 6 Little Glas Maol 3 Cairn of Claise 2 Carn an Tuirc 6) OS 10 4 Lecht - - 5 Carn Mor - - 6) ee - Carn Liath Si vena oS Mona Gowan - She = S2Ny 1G4AEtCo Ov Foor 925 93 194595 Taw VOUdeT See We ate? snule 1e2 af Wine Wo 2leih ss UNiBeEh:> fish Gaeaipe Na ENERO CORR WORE SNES OWA G : ‘ : 2 3 L eghay! Sye PH RG BOP 1S Ell a4 ey ad Nee (Hd Ore hc-nia Wee ee IRHaS SAY Maingate C04 . i . , é e Ona gt Scors Burn in Ladder Hills, 3 pairs in 1992 and 4 in 1995. 188 RRae&A Watson Thompson, pers comm). AW saw young at the Wells and others at 1150m on Feith Buidhe, and at 1140m on Garbh Uisge Beag, Lochan Buidhe, and Cnap a’ Chleirich. At >1100m there was only an isolated pair in 1-2 km, suitable habitat being scanty. Moorland blanket peat Birds here formed c85% of the total seen, but we did not visit much of this habitat, so moorland may hold >90% of the population. Most were in groups (usually 3-8 pairs but up to 20). Most bred at 450-800m, the main altitude of blanket peat (Appendix 3). Some peat was uneroded, with pools and flushes on shoulders, saddles and basins. Most was eroded, with bare peat, hags, and few pools and flushes. Some of both types had meanders, oxbows and lochans. Glen peat This is Appendix 3’s ‘basin, valley and terrace peat’, mostly eroded. Basin peat had vegetation like moorland peat, but several lochside and riverside mires had more sedge, less Heather and some tall rushes. In the 1930s, Loch Morlich by Aviemore held 1-2 pairs, but none after 1984 when scrub had grown (Watson et a/ 1988). In each July in 1945-47, 1-2 in alarm at Loch Davan near Dinnet probably had big young; farm stock kept plants short, but since 1960 it was ungrazed, tall sedge and scrub grew, and no Dunlin were seen since 1970. Atnearby Loch Kinord, Prof D Jenkins (pers comm) often heard Dunlin in 1970-72 (probably 1-2 pairs), but not since. In the 1970s, 1-2 pairs bred at 300m by Dee near Braemar, but not since 1980, after high numbers of Red Deer Cervus elaphus overgrazed and trampled it. We saw none on Luibeg bog since 1986, after severe trampling, but they still nest on less trampled glen mires (eg Loch Tilt). SB 19(4) Dune slacks Atleast 5 pairs formerly bred on lightly grazed slacks with pools and ditches at the Loch of Strathbeg in the late 1960s, by the late 1970s there were only 1-2 pairs, and the last known bred after 1983. After drainage in the early 1980s for more cattle and sheep, cereal was grown on one part for 2 years. In the 1990s the land began to revert to slacks and would improve with less grazing. Two pairs displayed in 1992, a pair was not displaying in 1994 and 1995, and none was seen in 1996. In late 1977 a petrochemical plant destroyed the St Fergus site, where at least 2 pairs had bred. Main features in all sites: open water, gradient and plant height Alllay near open water, mostly alpine flushes and moorland pools. Most young were at pools, lochans, flushes, slow streams, and wet land nearby. Footprints and faeces showed much used pools. Pools used by young had shallow edges, and dried in some years such as 1994. We saw no birds at deep pools with vertical banks. Highest densities (up to 3 pairs/15 ha at570-590m on Monaltrie Moss and 760m on the Ladder Hills) were on uneroded peat with shallow pools, flushes and nearby vegetation with much Sphagnum. However, density is a dubious concept where birds occur linearly on flushes or clustered at pools, and often fly to other habitats to feed. Most were on slopes <5%, many on flats, and the few on 8-10% slopes were on flat shelves within the slope. Some in the Cairngorms bred on flats at 510m in Glen Geusachan and 860m by Loch nan Stuirteag 4km away, and not on steep land between. Another case was at 500m in Glen Derry and 850m on Moine Bhealaidh 3km away. Scottish Birds (1998) Numbers and habitat of breeding Dunlin in north east Scotland 189 D Nethersole-Thompson (in Watson 1966) noted this on Speyside. All these breeding sites were on peat >2m thick, with much Sphagnum and many pools or flushes (Appendix 3), features absent on steep slopes due to fast drainage. All 60 nests seen were in plants 10-20cm high, with short vegetation <10cm within 1m. We saw no Dunlin in vegetation >25cm high, and adults in <2cm were there only briefly in alarm. Estimates of numbers Work in 1987-97 showed at least 178 pairs breeding in 36 squares of 10 x 10km, and at least 258 possibly breeding. We searched all potential habitat in the 11 most studied squares. In 25 squares (Figure 1), not all potential habitat was visited; in only one did we visit >50%, and <25% in 21. Some unvisited land appeared from afar to be too freely drained or eroded to hold Dunlin, but the largest areas unvisited were blanket bogs with many pools in upper Feshie, Atholl- Gaick, Avon-Dorback, and northwest of Grantown. We saw some Dunlin there, but, to judge from such habitat searched more fully elsewhere, these 4 areas probably held well in excess of 100 extra pairs. Changes in numbers between years On 8 high areas, numbers were bigger at Glas Maol in 1970 and 1972, but changed little between decades and showed no trend (Table 2). We noticed no habitat loss or deterioration there. However, we saw no birds since 1987 on 8 low areas used by at least 15 pairs in earlier decades. Habitat was lost to industry (St Fergus) and tree planting (Abergeldie). Habitat deteriorated due to tall undergrazed plants (Lochs Davan, Kinord and Morlich), drains (Morven near Ballater), and deer overgrazing and trampling (River Dee by Braemar and Luibeg further up). Discussion The meaning of our current higher estimate Appendix 1 shows big numbers and wide distribution, bar recent surveys of all species by many observers using rapid methods that readily miss Dunlin. Such surveys lack the vigilance needed to count this small cryptic bird that skulks for most of the breeding season. Also, most watchers avoid remote pathless bogs and are on the hill at 1200- 1600 hours, a quiet time for Dunlin (Reed, Barrett et a/ 1983), especially inthe warm sun preferred by most people. Other skulkers such as Ptarmigan and Dotterel are readily overlooked (Watson 1965; Watson & Rae 1987), and Red Grouse skulk so much that some surveys omit them (Reed et a/ 1983). Transects for moorland birds gave densities too low and of fairly low repeatability within areas (Reed et a/ 1983), and cannot be used to calculate Dunlin density (Reed 1982). Estimates from nest finding plus ringing on 5 areas of South Uist machair on average considerably exceeded estimates on the same observers transects, but discrepancies on individual areas ranged from 44% fewer on the transects to 50% more (Jackson & Percival 1983). With such a wide range, using the average discrepancy elsewhere would be unwise. In any case, the percentage seen ontransects is likely to vary with density, season, hour, weather and whether birds have failed. A different method (point counts using supposed ‘constant search effort’) underestimated Dunlin densities to a varying extent on different areas (Brown & Shepherd 1993). The term ’census’ is often misused in surveys of bird numbers. 190 RRae & A Watson SB 19(4) Hill species cannot be counted accurately unless there is total enumeration with high repeatability. This has not been the case for breeding Dunlin, except perhaps for intensive study of individual territorial spacing as in Holmes (1970). Jackson & Percival (1983) obtained a steady cumulative total on each of 5 areas after a number of visits ranging from 5 to 13, but this is not the same as total enumeration with high repeatability on different visits. Sharrock (1976) gave British breeding numbers as 4000-8000 pairs, and Gibbons et al (1993) 2 estimates of 9150 and 9900 pairs. Our data raise the UK total. Good work in Monadh Liath, Wester Ross and other peatland would raise it by thousands of pairs (cf Whitfield 1997 recently on Lewis and Harris). Habitat Density in Alaska was high where pools abounded (Holmes 1970). Density on the Scottish flows was related to pools and other map features (Avery 1989), which indicated many of our sites. However, we saw birds where 1:10,000 maps showed no bog or pool, but where we saw flushes, pools and bogs too small to print on maps. Dunlin abundance is related to ground characteristics between pools as well as to total pool area (Lavers et a/ 1996). Also, flushes over base rich rock on Glas Maol held far more birds than at the same altitude and aspect on Cairngorms granite. Both types can have continuous vegetation, but Dunlin food may abound more on base rich sites. Nests were in low plants, unlike many Norwegian nests seen by RR in dwarf birch Betula nana and willow Salix spp up to 30cm. However, these were semi-open at the side, unlike tall heather, so physical structure be involved rather than height. Tall subalpine scrub is rare on UK peat due to overgrazing and burning, but we think that Dunlin would nest in it if it were here. Decline in numbers Since 1987 we have not seen birds at 8 low sites showing habitat loss or deterioration. The Braemar and Luibeg cases involved more Red Deer cropping plants and trampling pools. Although Dunlin numbers on 8 high areas in Table 2 changedlittle between years, the moorland ones had very few deer in summer and all 8 no deer in winter. The rise in deer density, including on grouse moors where deer were formerly rare, threatens Dunlin habitat. In the 1990s we saw 1000 deer atatime on Moine Mhor and the Mounth. They damage plants and peat, and trample bird nests (Thompson ef al 1996). Gamekeepers W Potts andJ Robertson (pers comm) noticed fewer Dunlin south of Loch Muick since a 1975-79 peak. Red deer increased two and a half times in 1966-86 in a block that includes Muick, faster than in Scotland generally (Youngson & Stewart 1996). The Dunlin decline may be more complex; JR (pers comm) said fewer returned after the hard 1981-82 winter, and many die on English estuaries in hard winters (Clark et a/ 1993a). Probably Scottish breeders winter further south, but evidence from ringing is scanty and some may die in hard weather. Also, some estuary habitat has been lost to industry (Clark et al 1993b). Conclusion We reject recent statements indicating Dunlin scarcity in north east Scotland. Though ea i Scottish Birds (1998) Numbers and habitat of breeding Dunlin in north east Scotland 191 localised, they are fairly common and our estimates are minimal. Dunlin are one of several hill species easily overlooked when breeding. These are unsuitable for rapid surveys of all species by many observers. Acknowledgements For information supplementing ours we thank gamekeepers D Calder, R Esson, P Fraser, J Oswald, W Potts, J Scott and J Robertson and colleagues E Duthie, J Edelsten, J Hardey, M Marquiss and S Rae. M Marquiss and an anonymous referee gave useful comments and R Duncan drew the map. References Avery M1 1989. The effects of upland afforestation on some birds of the adjacent moorland. Journal of Applied Ecology 26: 957-966. Bates M A, Galbraith C A & Tew T E 1993. An Ornithological Evaluation of Grampian Region, Scotland, with Special Reference to the Location of Future Commercial Afforestation. JNCC Report 136. Joint Nature Conservation Committee. Peterborough. Brown A F & Shepherd K B 1993. A method for censusing upland breeding waders. Bird Study 40: 189-195. Buckland S T, Bell M V & Picozzi N 1990. The Birds of North-East Scotland. North-East Scotland Bird Club. Aberdeen. Campbell B & Ferguson-Lees J 1972. A Field Guide to Birds’ Nests. Constable. London. Clark J A, Baillie S R, Clark NA & Langston R H W 1993a. Estuary Wader Capacity following Severe Weather Mortality. BTO Research Report 103. Thetford. Clark N A, Evans J, Rehfisch M M & Shepherd M 1993b. Variability in Waterfowl Distribution within Hypertidal Estuaries in relation to Sediments. BTO Research Report 109. Thetford. Cook M 1992. The Birds of Moray and Nairn. Mercat Press. Edinburgh. Cramp S & Simmonds KE L (eds) 1983. Handbook of the Birds of Europe, the Middle East and North Africa: the Birds of the Western Palearctic. 3. Oxford University Press. Oxford. Gibbons D W, ReidJB & Chapman RA1993. The New Atlas of Breeding Birds in Britain and Ireland: 1988-91. Poyser. London. Glentworth R & Muir J W. 1963. The Soils of the Country round Aberdeen, Inverurie and Fraserburgh. HMSO. Edinburgh. Gordon S 1912. The Charm of the Hills. Cassell. London. Gordon S 1925. The Cairngorm Hills of Scotland. Cassell. London. Harvie-BrownJA & Buckley TE 1985. A Vertebrate Fauna of the Moray Basin. Douglas. Edinburgh. Holmes R T 1970. Differences in population density, territoriality and food supply of dunlin on arctic and subarctic tundra. In: Animal Populations in relation to their Food Resources. Watson A (ed). pp 303-319. Blackwell. Oxford. Jackson D B & Percival S M 1993. The breeding waders of the Hebridean machair: a validation check of the census method. Wader Study Group Bulletin 39: 20-24. Jenkins D, WatsonA & Miller GR 1963. Population studies on red grouse, Lagopus lagopus scoticus (Lath.) in north-east Scotland. Journal of Animal Ecology 32: 317-376. Lavers C P, Haines-Young RH & Avery MI 1996. The habitat associations of dunlin (Calidris alpina) in the flow country of northern Scotland and an improved model for predicting habitat quality. Journal of Applied Ecology 33: 279-290. Nethersole-Thompson D 1966. The Snow Bunting. Oliver & Boyd. Edinburgh Nethersole-Thompson D 1973. Collins. Glasgow. Nethersole- Thompson D & Nethersole- Thompson M 1986. Waders: their Breeding, Haunts and Watchers. Poyser. Calton. Nethersole-Thompson D & Watson A 1974. The Cairngorms. Collins. London. Parr RA 1990. Moorland birds and their predators in relation to afforestation. \nstitute of Terrestrial Ecology. Banchory. Purvis O W, Coppins B J, Hawksworth DL, James PW &MooreD M 1992. The Lichen Flora of Great Britain and ireland. Natural History Museum Publications. London. Reed T M 1982. Transect Methods. Nature Conservancy Council. Peterborough. The Dotterel. 192 RRae&A Watson SB 19(4) Reed T M, Barrett J C, Barrett C & Langslow D R 1983. Diurnal variability in the detection of Dunlin Calidris alpina. Bird Study 30: 244-246. Reed T M, Langslow D R & Symonds F L 1983. Breeding waders of the Caithness flows. Scottish Birds 12: 180-186. Rodwell J S (ed) 1991 & 1992. British Plant Communities. 2 & 3. Cambridge University Press. Cambridge. Sharrock J T R 1976. The Atlas of Breeding Birds in Britain and Ireland. Poyser. Berkhamsted. Sim G 1903. A Vertebrate Fauna of “Dee”. Wyllie, Aberdeen. Thompson DB A, Watson A, Rae S & Boobyer G 1996. Recent changes in breeding bird populations in the Cairngorms. Botanical Journal of Scotland 48: 99-110. Watson A 1965. Research on Scottish Ptarmigan. Scottish Birds 3: 331-349. Watson A 1966. Hill birds of the Cairngorms. Scottish Birds 4: 179-203. Watson A, Nethersole-Thompson D, Duncan K, Galbraith H, Rae S, Smith R & Thomas C 1988. Decline of shore waders at Loch Morlich. Scottish Birds 15: 91-92. Watson A & O'Hare P J 1979a. Spacing behaviour of Red Grouse at low densities on Irish bog. Ornis Scandinavica 10: 252-261. Watson A & O’Hare P J 1979b. Bird and mammal numbers on untreated and experimentally treated Irish bog. Oikos 33: 97-105. Watson A & Rae R 1987. Dotterel numbers, habitat and breeding success in Scotland. Scottish Birds 14: 191-198. Whitfield D P 1997. Waders (Charadrii) on Scotland’s blanket bogs: recent changes in numbers of breeding birds. Peatland Conservation pp.103-111. CAB International. Wallingford Oxon & New York. Youngson R W & Stewart L 1996. Trends in red deer populations within the Cairngorms core area. Botanical Journal of Scotland 48: 111-116. Appendix 1 Literature account 1 Harvie-Brown & Buckley 1895. Hinxman, a few on hilltop peat Glen Livet, many 1893 Upper Cabrach/Blackwater hilltops and glen mosses Cairnbrallan (all north of Ladder Hills). 2 Sim 1903. A few breed St Fergus links, Loch of Strathbeg and other suitable localities. 3 Gordon 1912, 1915, 1925. Bred Forvie and Moine Mhor (detail SG pers comm) and 1 on Monadh Mor. 4 Nethersole-Thompson 1966. Isolated pairs or communities wherever there are high peat mosses on Mounth and Cairngorms. 5 Watson 1966, Cairngorms area. Small groups SW Cairngorms and Moine Bhealaidh, and many Clunie-Glen Ey hills. Odd pair or 2 Abernethy foothills (D Nethersole- Thompson). 6 Nethersole-Thompson 1973. Big groups Moine Mhor. High numbers Angus and Aberdeenshire Mounth (in field with AW). 7 Nethersole-Thompson & Watson 1974. Wide distribution on map (p258). DNT, formerly 1-2 pairs Loch Morlich, and AW saw them 1947. From AW, 1 or 2 Loch Davan, pairs 1400-1600 ft Luibeg, Loch Builg, Bynack (Mar), and elsewhere, scores of pairs Glas Maol and Carn an Tuirc to Glen Clova and Lochnagar (RR later saw 37-60 pairs), also good on Glen Ey hills, Moine Mhor and Moine Bhealaidh. On a few hills, eg Cairn of Claise and Moine Mhor, more than Dotterel, and locally more than Golden Plover. Nests sparsely high Invermark, Glen Muick, Balmoral and Atholl, and Ladder Hills. Though near the mark on much visited land, this underestimated numbers on land that was then less visited by AW (the ‘sparsely’ cases). 8 Sharrock 1976. BTO survey 1968-72. Bred 3 coastal and 2 inland 10km squares, possibly another 4 inland. 9 Buckland et a/ 1990. Regional survey 1981-84, scarce breeder Loch of Strathbeg and 3 inland 10km squares, probably bred 6, possibly 5. We found breeders in all these. 10 Cook 1992. 1980s Moray survey, scarce breeder, in our area bred in 4 tetrad squares, including 7 pairs Lecht/Carn Mor (Ladder Hills). Others near Forres outside our area. 11 Gibbons et a/1993. BTO survey 1988-91, Scottish Birds (1998) Numbers and habitat of breeding Dunlin in north east Scotland 193 in our area bred in 15 of the 10-km squares, plus seen in 2 inland and 1 coastal. 12 Bates et a/ 1993. 280 pairs Grampain Region and adjacent parts of Tayside and Highland, not based on Bates et a/ fieldwork, no detail on numbers or locations or sources, and coincided in time and in the 280 figure with local birdwatchers’ hearsay from RR earlier in our study. 1 and 4-7 show many (and 4-7 wide distribution), 8-11 scarcity, and 12 should be discounted. Appendix 2 Behaviour and breeding through the season The first birds usually arrived at the end of April, but later in years of much winter snow. The earliest were singles on 15 and 20 April near Glas Maol. They arrived after Golden Plover and often after Dotterel. By early May they were on most sites, and in brief snowstorms were seen sheltering in pairs close together, but after deep snowfall they vanished till the main thaw. First eggs hatched on 7-14 June (cf Nethersole-Thompson & Watson 1974), on the Ladder Hills <1 week earlier than on alpine Mounth land and about the same time in warm Mays, and the highest Cairngorms pairs nesting 1-2 weeks later. Snowfall delayed nesting (cf Nethersole-Thompson & Nethersole-Thompson 1986). In 1995 the Ladder Hills had far more new snow than at the same altitude by Loch Muick, and their Dunlin hatched a week later. In 1977 the main thaw on the high Cairngorms came on 18 June, and high pairs did not nesttill the last week of June. Strongly demonstrative display with young waned after 4 days and ended at about 7-8 days. Most hens vanished by the time young reached 10 days, and cocks took over (Cramps & Simmons 1983). Birds with big young >2 weeks could be seen by scanning from vantage points, at up to 200m. Work at this stage merely showed minimum adult numbers and whether young had been reared. As young 2-3 weeks old were usually apart (often 200-300m), and young from 2 broods can mingle, it was impossible to count separate broods and young per brood without ringing. Many young were fully plumaged by mid July apart from downy tufts, and most by the end of July (cf Nethersole-Thompson & Watson 1974). Usually, young in their last half week were unattended by adults. In late July we saw young singly (occasionally 2 together) without adults, eg 5 single young on 3 hills on 18 July 1993, which could only just fly. However, young were occasionally attended by an adult or less often by a pair, eg a full grown young with a pair on 20 July 1969 on Ladder Hills. At Glas Maol on 29 July 1969, a chick about two-thirds grown was with an adult pair in alarm, and a pair in alarm with a chick almost fully grown. Most Dunlin had gone by 1 August, and earlier in dry Summers such as 1994 when most attempts failed. The last seen with young was on 9 August in Glen Esk. We saw singletons in flocks of Dotterel and Golden Plover until mid August, but such birds might be migrants. Appendix 3 Main soils and plant communities in main breeding habitats Dunlin on alpine land bred on a) freely drained alpine sod podzols with a thin organic layer, and sparse vegetation dominated by Three-leaved Rush, grit and 194 RRae&A Watson SB 19(4) boulders (community U9 in the National Vegetation Classification (NVC) of Rodwell 1991, 1992), b) freely drained alpine podzols with a thicker organic layer, and vegetation dominated by Mat Grass (Rodwell’s U8), or Stiff Sedge Carex bigelowii(U10), or lichen heath (H13 & 19), and c) blanket peat with much Harestail Cotton Grass Eriophorum vaginatum (M19). Glentworth & Muir (1963) defined peat soils as having >30cm of organic matter above mineral soil. Many big subalpine peatlands lay at 800-900m on the Mounth and at Moine Mhor and Moine Bhealaidh inthe Cairngorms, and small alpine ones in the Mounth up to 1030m on Cairn of Claise by Glas Maol and to 1080m on White Mounth by Loch Muick. Each held breeding Dunlin. Alpine gleys due to waterlogging were frequent in a), b) and c) beside flushes. Most Dunlin seen bred at 450-800m, reflecting the altitude of a) ‘blanket hill peat’, and b) ‘flushed blanket hill peat’ (1: 63 360 Soil Survey of Scotland). Most of a) had Cross- leaved Heath Erica tetralix, Sohagnum moss and Harestail Cotton Grass, with some Heather Calluna vulgaris and Deer Sedge Scirpus cespitosus (Rodwell’s M15-17), and with less heath and more Sphagnum at pools (M2) and hollows. Most of b) had sedge Sphagnum mires (M6), or wet grassland with tall rushes Juncus spp (M23). Some on ‘basin, valley and terrace peat’ up to 550m bred in heath like a). We saw most Dunlin on hill peat >1m thick, and highest densities on peat >2m. Plants were short, especially on wet rarely burnt land. On a flat with high Dunlin density on Monaltrie Moss, the lichen Cetraria nivalis at 570m indicates severe conditions; usually it is at >900m (Purvis ef a/ 1992), and the lowest sites seen elsewhere on our area were at 800m on exposed ridges. Adults used other habitats, such as flying up to 1.5km to feed at lochans where we saw none breeding. On flushed peat, a few with young were on short grass and moss with some rush tufts by slow streams and pools (eg at 400m at Abergeldie by Ballater, where RR saw at least 3 pairs in 1971 and 1972), and on wet ground without pools (eg south of Morven). Robert Rae, 11 Millend, Newburgh, Ellon AB41 OAW Adam Watson c/o Institute of Terrestrial Ecology, Banchory AB31 4BY Revised manuscript accepted November 1997 Dunlin at nest E Pickard Scottish Birds (1998) 19:195-205 195 Changes in breeding bird populations in peatlands and young forestry in north east Sutherland and Caithness between 1988 and 1995 M HANCOCK & M AVERY As part of a long term study of breeding birds in north east Sutherland and Caithness, peatland areas were surveyed in 1988, 1991 and 1995. Nineteen plots (2.5km x 2.5km) were surveyed in all 3 years. Plots were surveyed once in May and once in June. Counts of birds recorded during transect surveys were used as an index of abundance. Skylark indices showed significant declines between 1988 and 1991, and between 1991 and 1995, with an average decline of 12% per year. Indices of Red Grouse, Golden Plover, Dunlin and Meadow Pipit abundance varied significantly between years and all were lower in 1995 than in 1988. There was no correlation between changes in indices and distance to recently planted forests and no evidence of an increase in Crow numbers. These surveys suggest dramatic declines in an upland breeding Skylark population, and, together with other recent work, suggest that nationally important breeding populations of some wader species on unafforested areas of peatland in Sutherland and Caithness may have declined since the late 1980s by amounts which are comparable to earlier losses due to afforestation. Point counts were made of birds in newly planted forests in the area in 1988 and 1991 and results are presented. Introduction Sutherland and Caithness contain the largest | expanses of peatland in Britain (Stroud et a/ 1987). The area is the stronghold for some upland bird species, holding, for example, an estimated 18% and 39% of Britain's breeding populations of Golden Plover Pluvialis ' apricaria and Dunlin Calidris alpina _ respectively (Stroud et a/ 1987), and areas of ) high Skylark Alauda arvensis abundance _ (Gibbons et a/ 1993). During the 1980s large | areas of peatland in the area were planted with exotic conifers (Avery and Leslie, 1990), | Causing considerable losses of breeding | waders (Stroud et a/ 1987, Bainbridge et a/ _ 1987). As well as direct effects through habitat loss, the new forestry plantations may have affected breeding birds of adjacent peatlands by increasing predator populations (Thompson et al 1988), or by affecting the local hydrology, although one 20 year Caithness study found that hydrological impacts were limited to within 20m of the forest edge (Pyatt et a/ 1992). Stroud et al (1990) showed that breeding densities of some wader species in the area are lower nearer to plantations. Other studies showed that this could be explained by the nature of areas selected for afforestation, rather than any ‘edge effect’ caused by the plantations themselves (Avery 1989, Avery et 196 M Hancock & M Avery al 1989). Parr (1993) found that some moorland birds suffer higher breeding failure rates nearer to plantations, and there is recent anecdotal evidence of increases of avian and mammalian predator populations in north Sutherland and Caithness, probably resulting from the favourable habitat for such species in the new plantations and a decline in keepering intensity (C Crooke pers comm). In 1988 peatland plots were surveyed to test predictions of bird numbers made using satellite imagery (Avery and Haines-Young 1990). Some of these plots were surveyed again in 1991 and 1995 and these plots formed the basis of this study. Methods - peatland plots Square plots of unafforested peatland in Sutherland and Caithness, 2.5km x 2.5km, were selected at random within 3 strata, defined by an index of infra-red reflectance derived from satellite imagery, such that plots with a wide range of reflectances were selected (Avery and Haines-Young 1990 and 1992). Five 2.5km straight line transects were drawn across each plot, separated by 500m, with the outermost transects 250m from the edge of the plot. Transects were orientated north south or east west, whichever was more convenient for the terrain. In 1988, 2 surveyors surveyed 37 plots. In 1991, 2 different surveyors surveyed 31 plots, permission to survey having been denied for 6 plots. In these 2 years each surveyor visited each plot once. In 1995, a different surveyor surveyed 19 plots (Figure 1), with assistance from another surveyor for 6 visits. The 19 plots chosen for survey in 1995 were all those plots surveyed in 1988 and 1991 which were either in the catchments of the Rivers Naver and Helmsdale, or further north and east, as SB 19(4) this was the area considered to be of greatest conservation interest (Evans 1994). Plots were surveyed once in May and once in June by a single observer walking the transects and mapping all birds within 100m of the transectline, including common species such as Meadow Pipit Anthus pratensis. Surveys did not take place in continuous rain or in winds of more than 35km/hr. Surveys took place between 0900 and 1700 BST to help prevent variation in bird detectability during visits (Reed et af 1985, Thirgood et al 1995). Registrations were totalled for each visit, and the peak count (May or June) for each species for each plot was used as an index of abundance. For each plot, the distance to the nearest area of forestry, and National Grid eastings, were taken from maps to compare with changes in indices of abundance. To determine whether counts differed between observers within years, counts of each species for each visit were compared for the 2 observers for 1988 and for 1991. Differences between observers between years could not be assessed. The visit dates for the survey years were compared for significant differences, and correlation coefficients between visit date differences and changes in indices of abundance were calculated. Counts made at the same sites on the 2 visits were compared for significant correlations, to see if counts were affected more by location than by visit date. Methods - point counts in plantations In 1988 and 1991, birds of new plantations in north east Sutherland and west Caithness were surveyed. Point counts were carried out | at 80 sites in plantations on peatland (Figure1), oe ll | Scottish Birds (1998) Bird populations changes in Sutherland & Caithness 1988-1995 197 Figure 1 Location of peatland plots surveyed in 1988, 1991 and 1995, and forestry point count sites surveyed in 1988 and 1991. [| Peatland plot following the methods of Bibby et a/ (1985). Point count sites were 1km grid intersections. Sites were visited once in late April - early May and once in late May, between 0700 and 0930, and counts lasted 5 minutes. Repeat records within one point count of abird singing from the same location were treated as records of the same bird. For each species at each point, the peak count (early or late visit) was used as an index of abundance. Planting dates were taken from forestry maps. An approximate visual estimate of the percentage cover of Heather Calluna vulgaris within 30m of point count sites was made in each year. « Forestry point count site Results - peatland plots For the 19 plots surveyed in all 3 years (Figure 1), bird abundance indices are shown in Table 1. All pair wise comparisons of years had more species showing decreasing indices than increasing indices. Differences between years were significant for Red Grouse Lagopus lagopus, Golden Plover, Dunlin, Skylark and Meadow Pipit. For all of these species, totals in 1995 were lower than those in 1988. For Skylark, but no other species, declines were significant for both 1988-1991 and 1991-1995 (Wilcoxon tests, Z=-3.8 and -3.6 respectively, p<0.01 in both cases), averaging 12% per year. 198 M Hancock & M Avery SB 19(4) Table 1 Peatland plots - sum of peak counts for the 19 plots surveyed in all 3 years for all species recorded at more than 10 plots over the 3 years. Peak number of registrations, Sum for all plots 1988 1991 1995 Red Grouse 80 130 Sih Golden Plover Sil 235 209 Dunlin 138 69 79 Snipe 23 We 11 Curlew 29 36 25 Greenshank 23 28 23 Common Sandpiper 12 5 10 Skylark 1235 565 238 Meadow Pipit 989 577 5338 Wren 28 26 2 Wheatear Sih 16 10 Crow 31 9 9 *Friedman 3-way ANOVA, N=19 For species showing significant differences in abundance indices between years (Red Grouse, Golden Plover, Dunlin, Skylark and Meadow Pipit), percentage change in indices of abundance from 1988-1995 was compared with distance to the nearest area of forestry, difference in dates of May and June visits, and eastings (Table 2). For Dunlin, the percentage decline between 1988 and 1995 was significantly correlated with eastings - the more easterly sites having the greatest percentage declines (Spearman correlation coefficient, r= 0.53, N=19, p<0.05). No other significant correlations were found. In neither 1988 or 1991 was there any significant difference between observers in indices of abundance, except for visit 2 in 1988, when one observer recorded significantly more Meadow Pipits than the other. On visit 1, the same observer also Percentage Significance of differences change between years 1988-1995 (7 —pd=,, “G0'0>d=,) junoo yeed ueal/| UOI}E}A09 JO JUeDIIUBIS 91S JUNOD JUIOd “@PIS 1UNOD jUIOd je abe uol}ejUe| 4 ye abe uolnejuRld (\UNOD ayNuIlW G 49d SpJiq) UZEMIJEq UOI]E|A0D JUNOO yead ueal/\ -(sieak Z ay} 4BAO SjUIOd Of UY] a4OW Je papsodas Saldads |e 404 ‘aBe19A09 JO Ssieak Z ay} JOJ SJUIOd je 40J SJUNOD yead Jo wins) abe uojejuejd yyM UuoHeleA — SJUNOD JUIOd fv 3|qeL Scottish Birds (1998) Bird populations changes in Sutherland & Caithness 1988-1995 203 suggests an average annual decline of 7% in Dunlin numbers in Sutherland and Caithness in the period 1979-1984 (D P Whitfield, pers comm). The British range of Dunlin expanded between 1968-72 and 1988-91 (Gibbons et al 1993), although declines in areas of high density such as Sutherland and Caithness could mean that the British population has declined overall. However, peatland breeding populations have recently increased slightly in Lewis and Harris (Whitfield 1996). The fact that the greatest percentage declines in indices of Dunlin abundance occurred in the more easterly sites is hard to explain. Many of these sites held high Dunlin numbers in 1988 and were remote from the main areas of afforestation. A model of Dunlin abundance innorth east Sutherland and Caithness, partly based on results from this survey, shows most of the areas of high predicted abundance to be east of Strath Halladale (Layers et a/ 1996). The analysis for this model showed that Dunlin have a preference for flatter areas with a high availability of pools, and where the ground between pool systems is dominated by short, sparse Trichophorum cespitosumand Eriophorum. Such areas are typical of the more easterly sites in this survey. Meadow Pipit densities increase in the early stages of afforestation, starting to decline when trees are about 5 years old (Moss 1978). Point count data from this survey are consistent with this pattern (Table 4). It is possible that numbers of Meadow Pipits on moorland plots in 1988 were swelled by additional recruitment from nearby populations in the new plantations, most of which would then have been less than 5 years old. However, the evidence of some observer bias in 1988 suggests that Meadow Pipit results should be treated with caution. Similarly, of allthe species showing significant between years variation, Meadow Pipit is the only one for which counts on different visits to the same site are not significantly correlated when all 3 years data are combined. This raises the possibility that the timing of visits, or the timing of breeding, could have had a significant effect on Meadow Pipit counts, possibly explaining the observed variation between years. Skylark numbers declined by more than 50% in Common Birds Census (CBC) plots since the early 1980s (Marchant et a/ 1990). Most of these plots are in lowland England. There has been little quantitative work on Skylarks in the uplands; indeed, for many early upland surveys, Skylarks were so abundant that they were not recorded. The abundance map for Skylark in Gibbons et al (1993) suggests that north Sutherland and Caithness may have held some of the highest densities of upland breeding Skylarks in Britain. This Survey suggests that breeding populations of Skylarks in some upland areas may be declining even more rapidly than those sampled by CBC plots. Little is known about the reasons for these declines in the uplands. Skylarks breeding in the peatlands of northern Scotland probably winter on farmland elsewhere in Scotland, and although little is known of the movements of Scottish birds (Thom 1986), there is little evidence of long distance migration (Dougall 1996). In Britain as a whole, the loss of winter stubbles as autumn sowing of cereals becomes more common is probably one of the main causes of decline relating to wintering habitat (Evans et al 1995). However, autumn sowing is uncommon in north Scotland and the Uists, where breeding Skylarks surveyed in this study may winter. A wide range of factors may be affecting populations of peatland breeding birds, including agricultural changes, climatic 204 M Hancock & M Avery factors, changes in moorland grazing and burning practices, moorland drainage and drainage for forestry, changes in predator populations following changes in keepering patterns and land use, fence strikes, acid deposition and habitat fragmentation caused by afforestation. Such factors may be acting over a wide area in ways not easily linked to adjacent forestry and farming practices. The fact that many species covered by this survey were recorded more frequently on June visits suggests that a repeat survey based on a single June visit might be worth considering. Asimilar conclusion was reached by Avery and Haines-Young (1992). However, such an approach would be highly sensitive to the timing of breeding (Thirgood et a/ 1995) sO more years’ data would be needed to show long term population changes. The results of the point count surveys in plantations are generally consistent with those of similar studies elsewhere (Avery and Leslie, 1990), showing a decline in open country species, such as Golden Plover and Skylark, and an increase in woodland species such as Willow Warbler and Redpoll. The correlation between Common Gull counts and plantation age, however, was unexpected and may be a coincidental result of the small sample size (Table 3). Alternatively, it could be explained by the fact that point count sites in the more mature plantations more often fell near flight lines used by gulls moving between lochs. The non significant positive correlation between plantation age and counts of Black- headed Gull Larus ridibundus (Table 4) is consistent with this explanation. SB 19(4) Conclusions This survey provides worrying evidence of declines in nationally important Golden Plover and Dunlin populations in areas of apparently unchanged peatland in Sutherland and Caithness. It also suggests dramatic declines in an upland breeding Skylark population. The reasons behind these declines are poorly understood and will need be addressed if these species are to remain key constituents of the bird community of this unique area. Acknowledgements This work was funded by RSPB, with assistance from the Forestry Commission in 1988 and the EU LIFE programme in 1995. The work was supervised by M Avery in 1988 and 1991, and by P Mayhew in 1995. The surveyors were P Akers and S Hayhow in 1988, M Hancock and V Egan in 1991, and J Fairweather, assisted by P Thompson, in 1995. We gratefully acknowledge the assistance of | Bainbridge, T Burns, C Crooke, J Fairweather, M Grant, P Mayhew, R Parr, R Summers, T Stowe, C Thomas, P Thompson, D P Whitfield and J Wilson during analysis and write up and the help of the landowners, estate staff, farmers and foresters of the area in permitting access to land for survey. References Avery M | 1989. Effects of upland afforestation on some birds of adjacent moorlands. Journal of Applied Ecology 26: 957-966. Avery M |, Winder F L R & Egan V M 1989. Predation on artificial nests adjacent to forestry plantations in northern Scotland. Oikos 55: 321 - S23k Avery M | & Haines-Young R H 1990. Population estimates for the dunlin (Calidris alpina) derived from remotely sensed satellite imagery of the Flow Scottish Birds (1998) Bird populations changes in Sutherland & Caithness 1988-1995 205. Country of northern Scotland. Nature 344 (6269): 860-862. Avery M | & Haines-Young R H 1992. Bird and vegetation surveys in the Flow Country. In: O M Bragg, P D Hulme, H A P Ingram & R A Robertson (eds) Peatland Ecosystems and Man: an Impact Assessment. University of Dundee. Avery M | & Leslie R 1990. Birds and Forestry. Poyser, London. Bainbridge | P, Housden SD, Minns D W & Lance A N 1987. Forestry in the Flows of Caithness and Sutherland. RSPB, Edinburgh. Bibby C J, Phillips BN & Seddon AJ E 1985. Birds of restocked conifer plantations in Wales. Journal of Applied Ecology 22: 619-633. Dougall T W 1996. The movement and mortality of British ringed Skylarks Alauda arvensis. Ringing and Migration 17: 81-92. Evans A, Appleby M, Dixon J, Newberry P & Swales V 1995. What future for lowland farmland birds inthe UK? RSPB Conservation Review 9: 32- 40. Evans R J 1994 Flow Country Action Plan. Unpublished RSPB report. RSPB, Inverness. Gibbons D W, Reid J B & Chapman R A 1993. The New Atlas of Breeding Birds in Britain and Ireland: 1988-1991. Poyser, London. Gibbons D W, Avery M | & Brown A F 1996. Population trends of breeding birds in the United Kingdom since 1800. British Birds 89: 291-305. Holloway S 1996. The Historical Atlas of Breeding Birds in Britain and Ireland: 1875-1900. Poyser, London. Lack P 1986. The Atlas of Wintering Birds in Britain and Ireland. Poyser, Carlton. Lavers C P, Haines-Young R H & Avery M1! 1996. The habitat associations of dunlin (Calidris alpina) in the Flow Country of northern Scotland and an improved model for predicting habitat quality. Journal of Applied Ecology 33: 279-290. Marchant J H, Hudson R, Carter S P & Whittington P 1990. Population Trends in British Breeding Birds. BTO, Tring. Moss D 1978. Song-bird populations in forestry plantations. Quarterly Journal of Forestry 72: 5-14. O’Connel M, Thomas C, Twiss S, Downie I, Evans P R & Whitfield D P (in preparation) Functional ecology of peatland animals in the Flow Country of north Scotland. 1: Habitat requirements of breeding waders. Scottish Natural Heritage, Battleby. Parr R 1993. Moorland birds and their predators in relation to afforestation. Unpublished PhD thesis, University of Aberdeen. Pyatt D G. John AL, Anderson A R & White IMS 1992. The drying of blanket peatland by 20 year old conifer plantations at Rumster Forest, Caithness. In: O M Bragg, P D Hulme, H A P Ingram and R A Robertson (eds) Peatland Ecosystems and Man: an Impact Assessment. University of Dundee. Reed T M, Barret C, Barret J, Hayhow S & Minshull B 1985. Diurnal variability in the detection of waders on their breeding grounds. Bird Study 32: 71-74. Stroud D A, Reed T M, Pienkowski M W & Lindsay RA 1987. Birds, Bogs and Forestry. The Peatlands of Caithness and Sutherland. Nature Conservancy Council, Peterborough. Stroud D A, Reed T M & Harding N J 1990. Do moorland breeding waders avoid plantation edges? Bird Study 37: 177-186. Thirgood S J, Leckie F M & Redpath S M 1995. Diurnal and seasonal variation in line transect counts of moorland passerines. Bird Study 42: 257-259. Thom V M 1986. Birds in Scotland. Poyser, Calton. Thompson DBA, Stroud D A & Pienkowski M W 1988. Effects of afforestation on upland birds: consequences for population ecology. In: Usher M B and Thompson D B A (eds) Ecological Change in the Uplands. Blackwell, Oxford. Whitfield D P 1996. Waders (Charaadrii) on Scotland’s blanket bogs: recent changes in numbers of breeding birds. In Parkyn L Stoneman R E and Ingram H A P (eds) Conserving Peatlands. Centre for Agriculture and Biosciences International, Wallingford, Oxon. Mark Hancock, RSPB, Etive House, Beechwood Park, Inverness, 1V2 3BW. Mark Avery, RSPB, The Lodge, Sandy, Bedfordshire, SG19 2BR. Revised manuscript accepted November 1997 206 Scottish Birds (1998) 19: 206-222 SB 19(4) Numbers of wintering seaducks, divers and grebes in the Moray Firth, 1977-1995 RJ EVANS Numbers and distribution of seaducks, divers and grebes in the Moray Firth were monitored for 18 consecutive winters. Most sections within the firth held nationally important concentrations of at least one species. The Moray Firth as a whole held internationally important numbers of Red-breasted Mergansers and Slavonian Grebes and nationally important numbers of Scaup, Eiders, Common and Velvet Scoters, Long-tailed Ducks, Goldeneyes, Goosanders and Red-throated and Black-throated Divers. Peak numbers for individual species occurred in different months from September to April. Numbers of Common and Velvet Scoters and Goosanders declined during the study period and numbers of Goldeneyes, Red-breasted Mergansers and Slavonian Grebes increased. There were changes in the distribution of Eiders and Scaup within the firth. Oil pollution had no obvious effect during the study period, but remains a potential threat. These nationally important concentrations of seaducks, divers and grebes have had no adequate statutory protection to date. Introduction The Moray Firth has been regarded as an important site for wintering seaducks since the nineteenth century (StJohn 1845, Harvie- Brown and Buckley 1895, Baxter and Rintoul 1953, Thom 1986) although counts exist only fromthe 1960s onwards (Milne and Campbell 1973). When exploration for oil started in the Moray Firth in the late 1970s, accurate and comprehensive data were needed on the numbers and distribution of waterfowl in order to assess any impact that oil industry developments might have. Surveys of wintering seaduck, divers and grebes were undertaken by RSPB in 1977-78 and 1978- 79 (Mudge and Allen 1980). Lower intensity monitoring was undertaken during the following 2 winters (RSPB unpublished data) and surveys throughout the winter resumed from 1981-82 until 1994-5 as part of the Beatrice Field environmental monitoring programme. The results of this monitoring for the period 1981-82 to 1983-84 were presented by Campbell, Barrett and Barrett (1986), with Supplementary papers by Barrett and Barrett (1985a) on Goldeneyes and divers (Barrett and Barrett 1985b) and by Aspinall and Dennis (1988) on Goosanders and Red-breasted Mergansers. This paper presents the results of monitoring for the whole period and discusses their implications for site safeguard purposes. Scottish Birds (1998) Study area The study area extended from Helmsdale in the north to Kingston-on-Spey in the south- east and included the inner Dornoch, Cromarty and Inverness/Beauly Firths (Figure 1). The combined area is known as the Moray Firth and is described in more detail by Mudge and Allen (1980). Numbers of seaduck, divers and grebes in the Moray Firth 207 Daytime land based counts gave adequate population estimates for most species except Long-tailed Duck and the divers, for which supplementary land based roost counts at dawn and dusk gave improved estimates. Land based counts were supplemented by aerial and boat surveys in the earlier years of the study (Mudge and Allen 1980, Campbell et al 1986), but they were regarded as Figure 1 The Moray Firth study area showing count sections: (1) Helmsdale to Brora; (2) Brora to Golspie; (3) Outer Dornoch Firth; (4) Inner Dornoch Firth; (5) East Ross Coast; (6) Cromarty Firth; (7) Inverness/Beauly Firth; (8) Riff Bank; (9) Culbin Sands; (10) Burghead Bay; (11) Burghead to Lossiemouth; (12) SpeyBay. Major Long-tailed Duck roost sites are shown by solid stars. CAITHNESS Helmsdale Golspie @) * Embo Domoch by (3) ’ — ~ e SUTHERLAND 4, Edderton @ 2 ‘ Tain EASTER ROSS Balintore . ) 6) Guillam Bank Invergordon e? ‘ ol Inverness Methods The frequency of counts and the period covered by each winter’s survey varied as shown in Table 1. Tarbat Ness 3 . ' e Findhorn ——4 10km ‘ t Lossiemouth ’ ARG) r4 © Burghead e Kingston-on-Spey MORAYSHIRE unnecessary in achieving adequate coverage provided daytime and dawn/dusk land based counts were made in suitable conditions. Locating the precise areas used by roosting Long-tailed Ducks could only be achieved by boat. 208 RJEvans SB 19(4) Table 1 Months in which complete surveys of the Moray Firth were carried out between 1977-78 and 1994-95. Month Winter Sept Oct Nov Dec Jan Feb Mar Apr 1977-78 F ‘ ra : 1978-79 : ‘ ’ 1979-80 : A 1980-81 ‘ 1981-82 : 3 i ‘i ; 1982-83 : ; rs * 3 Y 1983-84 5 ¥ ; 1984-85 a i : ‘ . 1985-86 ‘ i c ‘i 1986-87 ‘ ‘ ‘ 1987-88 * 1988-89 ; ‘ , 7 F 1 989-90 * * * * * * * 1990-91 7 : ; ' 4991-92 * * * r * * ri * 1992-93 r * * * * r * 1993-94 * * * * * * * 1994-95 * * * * * * * The study area was divided into 12 discrete sections (Figure 1), in which the birds could be counted by one person in a single day. Daytime land based counts were made from a number of points within each section, close enough together to ensure complete survey of the section, but far enough apart to avoid double counting. Counts were carried out from soon after first light to approximately 2 hours before sunset, the earliest time that roost movements could be expected to disrupt daytime distribution patterns. Counts were Carried out irrespective of the state of tide if weather conditions were Suitable, ie no wind or light offshore winds and calm seas. Coverage of the whole area was completed, weather permitting, in as short a period as possible and in geographical sequence in order to minimize the effect of any bird movements between sections. Although monthly differences in distribution were sometimes noticeable, short term changes on a large scale were not recorded. Birds were identified to species whenever possible. In practice, it was not always possible to separate Common and Velvet Scoters or the diver species. Unidentified scoters and divers were therefore categorised as ‘scoter sp’ and ‘diver sp’ Campbell et al (1986) presented results for Common and Velvet Scoters combined. However, an attempt was made to estimate numbers for the 2 species separately (Kirby et a/ 1993). Unidentified scoters were assumed to occur Scottish Birds (1998) in the same ratio as the numbers of Common and Velvet Scoter which were positively identified and the ‘scoter sp.’ total was split accordingly for each site to give an estimate for each species. Unidentified divers often outnumbered identified birds so unidentified birds were not apportioned. Counts of specifically identified divers were recorded for the years 1983-84 onwards for the Moray Firth as a whole and for individual sections from 1987-88 onwards The only seaduck species not counted reliably from the land during the day was the Long- tailed Duck. They are smaller, more cryptically coloured and tended to feed further offshore than the other seaducks, making daytime counts difficult. Therefore, counts were also made of Long-tailed Ducks flying to communal roosts at dusk, or ‘occasionally’ shortly after dawn. Daytime and roost counts were combined as follows: daytime counts for the 12 standard sections were made simultaneously with counts for other species. Roost counts were carried out aS soon as possible before or after the daytime counts for the relevant section. Observations were made from points between known roosis in order to detect large scale movements of birds overflying one roost in preference for another. This was only recorded in 1978-79 when birds feeding over the Riff Bank were found to be roosting in Burghead Bay. Counts of birds flying to a roost site in one count section from a particular direction were presumed to have been feeding in the adjacent count section in the direction from which they had flown. The monthly count for each section was taken as either the daytime total or the roost count total, whichever was the higher. The resulting section counts were summed to give a monthly total for the whole study area Numbers of seaduck, divers and grebes in the Moray Firth 209 Long-tailed Duck roosts tended to draw birds from areas which were larger than the standard count sections. Five main areas for Long-tailed Duck were identified within the Moray Firth and the results are expressed in terms of these larger areas: East Sutherland (sections 1-4), East Ross (sections 5 and 6), Riff Bank/Inverness Firth (sections 7 and 8), Burghead Bay/Culbin Sands (sections 9 -11) and Spey Bay (section 12). Eiders also used areas Somewhat larger than standard count sections and three main areas were identified: East Sutherland (Sections 1- 4), East Ross (Sections 5-6) and the Morayshire coast (Sections 9-12). For each species peak counts rather than indices of site usage (cf Mudge and Allen 1980) were used to express population levels for each section and are presented in the tables. In practice, peak counts occurred in different months in different winters. Each of the 12 sections was considered in its own right so that peak counts for different sections could be made in different months in the same winter. For the whole Moray Firth monthly totals were taken and the peak monthly count for each winter was used in the tables and for analysis. Trends for each species were tested by calculating Soearman Rank Correlation Coefficients (2-tailed) between numbers and years. The current status of seaducks, divers and grebes in the whole Moray Firth and at individual sites within the Moray Firth was expressed as the mean of peak numbers in the last 5 years 1990-91-1994-95 (Table 11). This is the standard way of assessing the importance of a site for a particular species (Pritchard et a/ 1992). 210 “vRREvans SB 19/4) Results Red-throated Diver Gavia stellata Red-throated Diver was the commonest diver. Numbers normally peaked at around 410, out of a total diver population of around 440 (Figure 2a). Red-throated Divers were recorded from all sections, though the largest numbers were found in sections also frequented by large numbers of seaducks, particularly the Outer Dornoch Firth, Culbin, Burghead Bay and Spey Bay. Large numbers were regularly observed flying out of the Inverness Firth during counts of roost-flighting Long-tailed Ducks (Table 2). Numbers between 1984-85 and 1994-95 showed no trend (r, =0.275, ns, n =11). Peak numbers tended to occur in October, occasionally involving large gatherings, including nearly 1000 off Culbin in October 1982 (Barrett and Barrett 1985b) and 600 in Spey Bay in October 1987. Individual sections generally matched this pattern, apart from the Inverness Firth where peak numbers tended to occur between December and February. No spring passage was detected. Black-throated Diver Gavia arctica Black-throated Divers accounted for about 6% of positively identified divers. The most favoured areas were the Outer Dornoch Firth and Burghead Bay (Table 2). Numbers between 1984-85 and 1994-95 showed no trend (r= 0.030, ns, n = 11) (Figure 2b). Peak numbers occurred in October, December and during February-March. The importance of individual sites followed the overall seasonal pattern. Great Northern Diver Gavia immer Great Northern Divers were the least common of the three regularly occurring divers in the Moray Firth, accounting for roughly 4% of positively identified divers. Great Northern Divers tend to occur further offshore than other divers (Parrack 1986) and probably did so in the Moray Firth (Barrett and Barrett 1985b). Therefore it is likely that a higher proportion of the unidentified divers were in fact Great Northern Divers. Great Northern Divers favoured the more open sections of the Moray Firth, particularly the Outer Dornoch Firth, Burghead Bay and Spey Bay (Table 2). There was no trend in numbers between 1984-85 and 1994-95 (r, = 0.418, ns, n=11). (Figure 2c). There was no clear seasonal pattern either for the whole Moray Firth or for individual sites. Peak winter numbers were recorded in October, December, February and March. Slavonian Grebe Podiceps auritus Slavonian Grebes occurred regularly in the Outer Dornoch Firth between Dornoch and Embo, the Cromarty Firth , the Inverness Firth and Burghead Bay (Table 2). There were occasional records from Culbin Sands. Numbers for the whole Moray Firth for the period 1984-85-1994-95 increased (r,=0.818, P<0.005, n=11) (Figure 2d), as did numbers for Burghead Bay for the same period (r, = 0.736, P<0.05, n = 11). It is possible that these trends were in part influenced by more Scottish Birds (1998) Numbers of seaduck, divers and grebes in the Moray Firth 211 Figure 2 Peak total numbers of divers and grebes in the Moray Firth 1984-85 - 1994-95. a) Red-throated Diver; b) Black-throated Diver; c) Great Northern Diver; d) Slavonian Grebe. a) Red-throated Diver b) Black-throated Diver NUMBER OF BIRDS NUMBER OF BIRDS g 64 8 86 87 88 8 90 81 92 93 94 WINTER (84 = 1984/5) Peder eaorgseee DEIN iN Tse ess 94 _C) Great Northern Diver d) Slavonian Grebe NUMBER OF BIROS NUMBER OF BIRDS 8 8 85 86 87 88 89 90 91 $2 93 94 ° WINTER (64 = 1984/5) 8 86 87 88 83 90 91 92 93 92 WINTER (85 = 1985/6) Table 2 Mean winter numbers of divers and Slavonian Grebes at the main sections in the Moray Firth. 1990-91 - 1994-95. Outer Cromary Inverness Culbin Burghhead Spey Whole Dornoch Firth Firth Sands Bay Bay Moray Firth Firth R-t Diver 80 12 154 68 74 225 459 B-t Diver 17 1 5 1 14 3 31 GN Diver 11 1 1 1 10 6 25 Slav Grebe 28 18 13 2 16 1 54 Ze. RJ Evans SB 19(4) frequent surveys during the latter part of the study period (Table 1) and also by improvements in the quality of telescopes used for counting. There was no clear seasonal pattern. Peak winter numbers were recorded in all months except September and March. Thisis perhaps not surprising considering that the number of birds counted depended on the weather conditions on the day of the count, in particular wind and sea state which affected the accuracy of counts of Slavonian Grebes more than other species. Red-necked Grebe Podiceps grisegena One or 2 Red-necked Grebes were recorded annually, with maxima of 6 in 1986-87 and 5 in 1991-92. Scaup Aythya marila Scaup were found almost exclusively in the Dornoch Firth, the Cromarty Firth and the Inverness Firth. Concentrations were associated with distillery and sewage outtfalls at these sites. Scaup were occasionally found elsewhere, including a flock of 600 in Spey Bay in 1978-79 (Mudge and Allen 1980). The Moray Firth wintering population peaked in most winters between 300 and 450 (Table 3) and there was no trend (Figure 3a). Numbers in the Dornoch Firth were higher (up to 650) during the early part of the study but declined latterly (r, = -0.709, P<0.002, n=18). Peak numbers in the Cromarty Firth gradually increased (r= 0.920, P<0.001, n = 18). Numbers in the Inverness Firth were more variable, peaking between 50 and 200. Numbers for the whole Moray Firth tended to peak between December and March, with most peak winter totals in February (8 out of 17 winters for which data are available). At individual sites, peaks occurred in every month from October to April with peak numbers occurring earlier in the winter (November to January) in the Cromarty Firth than in the Dornoch and Inverness Firths (December to March). Eider Somateria mollissima The most important areas were East Sutherland (Sections 1-4) and Morayshire (Sections 9-12) (Table 4). Flocks could be found off both rocky and sandy shores in these areas. No trend was detected in peak numbers for the Moray Firth as a whole (Figure 3b). However, there was an increase on the Morayshire coast (r, =0.732,P<0.001, n= 18) and a decrease on the East Sutherland coast (r, = -0.589,P<0.02, n =18). Peak numbers for the whole Moray Firth occurred during September - December and most frequently in October or November, the latter months accounting for peaks in 13 out of 18 winters. Numbers in the study area tended to drop as each winter progressed, implying that emigration to other parts of the Scottish coast occurred. Peak totals for East Sutherland occurred in October more frequently than in December but the reverse was the case for Morayshire. Peak numbers for East Ross occurred most frequently in January and February, probably reflecting dispersal of the large numbers in the north and east of the study area. King Eider Somateria spectabilis One to 3 drake King Eiders were recorded almost annually throughout the study period, i] Scottish Birds (1998) Numbers of seaduck, divers and grebes in the Moray Firth 213 Figure 3 Peak total numbers of seaducks in the Moray Firth 1978-78 - 1994-95. a) Scaup; b) Eider; c) Common Scoter; d) Velvet Scoter; e) Long-tailed Duck; ——— ———— — f) Goldeneye; g) Red-breasted Merganser; h) Goosander NUMBER OF BIRDS NUMBER OF B/RCS NUMBER OF BIRDS (Thousands) (Thousands) 85 AT BS 91 93 84 86 68 92 94 WINTER (77 = 1977/6) Qg) Red- breasted Merganser S00 i ae ae : 1500+ ie —H Baten Sud} - al Bates I 0 nme 77 7 7 a4 8 7A a0 $2 a) Scaup b) Eider S000 7 =] | ‘00 78 80 82 84 86 8a 4 90 92 94 8? WINTER (77= |977/8) AANTEENG ae Oa, a c) Common Scoter d) Velvet Scoter NUMBER OF BiROS = NEA oe aaa 76) f) Abate NUMBER OF BIADS £ S 87 9 78 80 82 ea 86 8B 9D WINTER (77 = 1977/8) h) Goosander S000 . , = xe — ae — -— = £5 é3 om -¢ ; ~ : cry <> a Es «co a § ae , ye > y SS pm «ch so & Scottish Birds (1998) more than 1500 in any other winter. It was the most important site in the Moray Firth for Common Scoter in 4 out of 18 winters. Burghead Bay held peak numbers in excess of 1000 more often than any other site in the Moray Firth (12 out of 18 winters) and was the most important site in 9 winters. Numbers declined at Burghead Bay (r= -0.703. P<0.002, n = 18). Despite large fluctuations in numbers elsewhere, no other section showed a trend. A decline in Common Scoter numbers for the whole Moray Firth was recorded (r, = -0.513, P<0.05, n = 18) (Figure 3c). During 1977-78 -1978-79 and 1981-82 - 1982-83, Common Scoter numbers peaked each winter between 7500 and 13500 (Mudge and Allen 1980, Campbell et al 1986). Peak numbers failed to exceed 4000 between 1983-84 and 1990-91, and in 1989/90 reached only 1175. During 1991-92 numbers were higher than during the mid 1980s, peaking at 4700 in April 1992. Winter peaks occurred in every month from October to April, but most frequently in February (6 out of 18 winters). There was no clear seasonal pattern at individual sites. Each site showed a winter peak as late as April and the Outer Dornoch Firth had peak winter counts in every month from September to April, with December the most frequent (6 out of 17 winters). Numbers at Spey Bay peaked only once before December in the 17 winters for which data are available and peak counts were fairly evenly spread between the remaining months. Peak counts at Culbin occurred most frequently between December and February, while peak numbers in Burghead Bay were spread mainly between October and February. Numbers of seaduck, divers and grebes in the Moray Firth 215 Velvet Scoter Melanitta fusca Velvet Scoter tended to occur in the sections also favoured by Common Scoter. As with Common Scoter, the section showing the greatest variation in numbers was Spey Bay and the section with the least variation was Burghead Bay. Numbers at Spey Bay ranged from 5000 during 1977-78 to only 8 in 1986-87; numbers increased again in subsequent winters. Itwas the mostimportant section in the Moray Firth for Velvet Scoters in 5 winters up to 1983-84 and none subsequently. Burghead Bay held at least 100 birds in every winter for which data were available except one. Itwas the most important section in 10 out of 18 winters, including every winter from 1988/-89 to 1994-95. The Outer Dornoch Firth held over 3000 in 1981- 82 and 1982-83, but numbers then declined sharply and peak numbers were fewer than 100 between 1986-87 and 1990-91. It was the most important section in 2 out of 18 winters. Culbin Sands never held more than 1000 birds andwas the mostimportant section only once. The Moray Firth total peaked at over 2500 in each of the winters 1977-88 1978-79 and 1981-82 - 1983-84, but failed to exceed 1000 in each winter from 1985/6 to 1991-92 (Table 5). The decline was not significant (r= - 0.335, ns, n = 18) (Figure 3d). Peak winter totals for the Moray Firth occurred in October and every month from December to April, with the highest frequency falling in April (5 out of 18 winters) followed by December and March (4 each). Peaknumbers at Spey Bay were spread evenly between January and April and the only section where the majority of peak winter counts occurred in December or earlier was Burghead Bay. 216 RdJEvans SB 19(4) Table 5 Five season (* = 3 season) means of Common Scoters (including appor- tioned unidentified scoters - see text) at the main sections in the Moray Firth, 1977- 78 - 1994-95 Winter Outer Culbin Dornoch Sands 1977-78 - 1979-80 *672 20750 1980-81 - 1984-85 2341 1566 1985-86 - 1989-90 446 683 1990-91 - 1994-95 634 749 Surf Scoter Melanitta perspicillata Surf Scoter were recorded annually and the highest number recorded in a single winter was 9 birds in 1978-79, followed by 5 in 1981- 82 and 1989-90. During 1990-91 and 1991- 92, 12 of 15 records of Surf Scoter were associated with Velvet Scoter and the other 3 were independent. Long-tailed Duck Clangula hyemalis Although the combination of day-time and roost counts was believed to give the best possible population estimate, the possibility Burghead Spey Whole Bay Bay Moray Firth PIS * 4030 * 6706 2205 1702 6715 974 202 1838 1075 1041 2996 remains that in some winters this method did not survey all sections adequately. In particular, the Easter Ross coast (Section 6) including the Guillam Bank in mid-firth was difficult to count. In some winters, birds using this area by day (and not counted) were regularly using the Burghead Bay roost, where they were counted. Also, it is possible, although unlikely, that some roost sites were undetected The most important sections for Long-tailed Duck were the East Sutherland coast (Sections 1-4), the Inverness Firth/Riff Bank (Sections 7-8) and Culbin Sands/Burghead Table 6 Five season (* = 3 season) means of Velvet Scoters (including apportioned unidentified scoters - see text) at the main sections in the Moray Firth, 1977-78-1994- 95 inter uter ulbin Dornoch Sands 1977-78 - 1979-80 * 169 * 499 1980-81 - 1984-85 2228 225 1985-86 - 1989-90 99 193 1990-91 - 1994-95 223 159 Burghea pey ole Bay Bay Moray Firth LT, * 2690 "2835 744 1537 SSM 359 S7/ 562 519 207 815 Scottish Birds (1998) Numbers of seaduck, divers and grebes in the Moray Firth 217 Table 7 Five season (* = 3 season) means of Long-tailed Ducks at the main sections in the Moray Firth, 1977-78-1994-95 Winter Sutherland Riff Bank/ Inverness Firth 1977-78 - 1979-80 * 1252 * 4602 1980-81 - 1984-85 3975 1225 1985-86 - 1989-90 1303 PaaS, 1990-91 - 1994-95 998 2588 Bay (Sections 9-11). Roosts for these areas were, respectively, between Brora and Golspie, over the Riff Bank and in Burghead Bay (Figure 1). Smaller numbers were found off the East Ross coast and in Spey Bay. Distribution between the 3 most important sections changed considerably over the study period. The Riff Bank, which held large numbers in the winters 1978-79 and 1990-91 had few in the intervening years. Large numbers were found along the East Sutherland coast in the early to mid-1980s, but many fewer during the early and later years of the study. The Burghead - Culbin area held larger numbers more frequently than the East Sutherland coast or the Riff Bank. Shifts of importance between areas occurred too frequently for any trends to be detected for individual sections. No trend in numbers was detected for the Moray Firth (r, = -0.119, ns, n =18) (Figure 3e) and numbers fluctuated considerably from winter to winter. Between 8000 and 12500 were found in most (11 out of 18) winters (Table 10). Much higher numbers (over 20000) were found in 1981/2 andlow numbers (fewer than 6000) in 5 out of 18 winters. Between 1984-85 and 1990-91 peak numbers alternated between average (4 winters) and low (3 winters). There was no clear pattern to Burghead Spey Whole /Culbin Bay Moray Firth “2719 1595 S723 9538 702 14741 4814 304 7281 4864 1257 8744 changes in the peak numbers of Long-tailed Ducks detected each winter. It is possible that between year differences in numbers were influenced by differences in distribution or survey intensity. Peak winter totals occurred between November and February, with an even spread between months. Individual sections reflected this pattern, although winter peaks at individual sections were also recorded in October, March and April. Goldeneye Bucephala clangula Prior to 1977-78, large numbers of Goldeneyes were regularly present at sewage and distillery outfalls at Invergordon and Burghead, but these declined after changes were made to the quality of the effluent from these outfalls (Barrett and Barrett 1985a). The Longman outfall at Inverness was the only site in the Moray Firth where large volumes of untreated sewage were regularly discharged up to 1994-95 and this site attracted a large proportion of the Moray Firth population (Table 8). Over the 18 years numbers of Goldeneyes increased in the Moray Firth (r, = 0.545, P20105n— 16), “due to changes” in’ the 218 RJEvans SB 19(4) Table 8 Five season (* = 3 season) means of Goldeneyes at the main sections in the Moray Firth, 1977-78 - 1994-95 Winter Cromarty Inverness Firth Firth 1977-78 - 1979-80 * 406 * 154 1980-81 - 1984-85 256 423 1985-86 - 1989-90 254 571 1990-91 - 1994-95 184 738 Inverness Firth (r, = 0.750, P<0.001, n= 18). Numbers decreased in the Cromarty Firth (r, = -0:68:.P<0.04,,. =e) Peak numbers occurred in all months between November and March, but mainly in January and February, when freshwater sites were most likely to be frozen Red-breasted Merganser Mergus serrator The Inverness/Beauly Firth and the Riff Bank were the main sections for Red-breasted Mergansers and large numbers were also found in the Outer Dornoch Firth, in the Cromarty Firth and off Culbin Sands (Table 9), Survey techniques for Red-breasted Mergansers were believed to have improved during the 1970s (Aspinall and Dennis 1988), so the upward trends for the whole Moray Firth (r, = 0.637, P<0.01, n = 18) (Figure 3g), Culbin (r, = 0.601, P<0.01, n = 18) and the Riff Bank (r, =0.477, P<0.05, n= 18) might be an artefact of improved coverage. No section showed a trend for the period 1981-82 - 1994-95, when surveys had constant effort. Peak numbers occurredin all months between October and March, but mainly during Burghead Whole Maltings Moray Firth V2i9 * 685 23 662 30 1002 38 1103 December - February (14 out of 18 winters). The Riff Bank, Inverness/Beauly and Cromarty Firths followed this pattern, whereas peak numbers at Culbin and in the Outer Dornoch Firth tended to occur either in October or during February-April. Goosander Mergus merganser Goosanders were found almost exclusively in the Beauly Firth and accurate count data exist from the mid 1960s onwards (Aspinall and Dennis 1988). Before 1980-81, peak numbers were generally lower than 1000, but between 1980- 81 and 1988-89 numbers peaked at more than 1200 in every winter (Table 10). Thereafter, numbers were much lower, peaking between 200 and 600. No significant trend was detected for the study period (r, = -0.426, ns, n=18) (Figure 3h), nor for the 31 year period for which good data exist (r,=0.151, ns, n = 31). Peak winter counts for the period 1981-82 -1994-95 showed astrongly significant downwardtrend (r, = -0.176, P<0.001, n =14). Peak numbers generally occurred during November-January. Scottish Birds (1998) Numbers of seaduck, divers and grebes in the Moray Firth 219 Table 9 Five season (* = 3 season) means of Red-breasted Mergansers at the main sections in the Moray Firth, 1977-78 - 1994-95 Inv’ness Riff Winter Outer Cromarty Dornoch Firth Firth 1977-78 - 1979-80 - old * 450 1980-81 - 1984-85 574 341 1985-86 - 1989-90 234 509 1990-91 - 1994-95 192 283 Table 10 Five season (* = 3 season) means of Goosander on the Beauly Firth, 1977-78 - 1994-95 Winter Beauly Firth 1977-78 - 1979-80 * 620 1980-81 - 1984-85 1800 1985-86 - 1989-90 1331 1990-91 - 1994-95 415 Discussion Although many sections within the Moray Firth qualify as nationally important in their own right, the arguments of Campbell et a/ (1986) that the Moray Firth should be treated as a single site because of changes in the relative importance of individual sites between years have been borne out by continued monitoring. Long-tailed Duck and Scaup illustrate this point. Winter numbers of both species in the Moray Firth have remained fairly stable at around 10000 and 350 respectively, but distribution within the firth Culbin Whole /Beauly Bank Sands Moray Firth Firth me mn 28 * 246 * 496 TATA, 559 47 1431 964 963 161 2498 als) 826 326 2242 has changed, gradually in the case of Scaup and more erratically in the case of Long-tailed Duck. Despite reductions in the numbers of some wintering species, the Moray Firth remains one of the prime seaduck sites in Britain and lreland, holding around 40% of British wintering Long-tailed Ducks and 30% of British wintering Red-breasted Mergansers (Kirby et al 1993). The most dramatic change has been the reduction in peak numbers of Common and Velvet Scoters. There is no apparent reason for this change; similarities in seaduck diet (eg Nilsson 1972) imply that altered food availability should cause changes in numbers of most seaduck species, not one or two. Scoter numbers fluctuate considerably, at other Scottish sites eg St. Andrews Bay (Milne and Campbell 1973). Inthe nineteenth century, St. John (1845) wrote of ‘considerable numbers’ of scoters in the Moray Firth, while 50 years later Harvie-Brown and Buckley (1895) described the status of scoter in the firth as ‘not particularly abundant’. More recently, Scottish Bird Reports show that between 1970-71 and 1976-77 peak winter numbers of Common Scoters in the Moray 220 RJEvans SB 19(4) Table 11 Sections of the Moray Firth which hold internationally (**) or nationally (*) important concentrations of individual species. SECTION SPECIES 1 2S 4 5 Sima hy 8 O34 100 a MARA Red-throated Diver : * . . % Black-throated Diver i “ 4 Slavonian Grebe * * . ne Scaup * Be t Eider oS P : : S Common Scoter x x “ A Velvet Scoter ‘ * x : “ Long-tailed Duck SON * * : P a Goldeneye a eae * * R-b Merganser ‘ ‘ * * ee Goosander * : Note: See text for fuller description of distribution of Eider and Long-tailed Duck. Firth ranged from 6000 to 14000. Such high numbers were not present during the 1960s (RH Dennis, pers comm). Campbell et a/ (1986) found that the Moray Firth held internationally important numbers of Long-tailed Ducks, as well as Common and Velvet Scoter. This is no longer the case for Long-tailed Ducks because of increased winter population estimates for north-west Europe, which raised the qualifying level for international importance from 5000 to 20000 (Laursen 1989). Numbers of Goosanders on the Beauly Firth declined during the study period. The increase during the 1980s was attributed to the introduction of a fisheries ban (Aspinall and Dennis 1988) whereas the decline during the latter years of the study coincided with reductions in numbers of other fish eating birds, notably Red-breasted Merganser and Cormorant Phalacrocorax carbo using the Beauly Firth. Red-breasted Mergansers were apparently able to redistribute to other parts of the Moray Firth, but Goosanders did not. There has been an overall decline in the number of Goosanders recorded winterirg in the UK (Waters and Cranswick 1993). This is of particular concern at a time when there is continuing pressure from fisheries interests in Scotland to shoot more Goosanders under licence (Carss 1994). Nationally important concentrations of Goldeneyes and Scaup were associated with sewage and distillery outfalls at 3 sites in the Moray Firth; Edderton Bay, the Cromarty Firth off Invergordon and at Inverness. Reductions in Goldeneye numbers at Invergordon and Burghead, which no longer holds nationally important numbers of | Scottish Birds (1998) Goldeneyes, coincided with changes to the quality of effluent at these sites (Barrett and Barrett 1985a). Similar changes are currently in train at Inverness, with further changes proposed. These are likely to have a further effect on the distribution of Goldeneyes and Scaup within the Moray Firth. The current study was initiated in order to monitor wintering seaduck populations and to determine whether oil industry developments in the Moray Firth affected them. There were no oil pollution incidents affecting large numbers of seaducks, divers and grebes in the Moray Firth during the study period and the Beatrice Field is now nearing the end of its planned production phase. However, with the difficult process of decommissioning yet to come and with the prospect of further inshore oil exploitation in the Moray Firth following the most recent licensing round, the waterfowl populations of the firth will still be at risk from oil pollution. Commercial fisheries, including shellfisheries, can affect seaducks and other waterfowl through direct competition for food and disturbance. A commercial mussel fishery already operates in the Dornoch Firth where there was a decline in Eider numbers during the study period. There have also been attempts at commercial cockle dredging at Culbin Sands and in the outer Dornoch Firth, though to date these have largely been confined to intertidal areas little used by seaduck: The provision of artificial reefs for commercial shellfish growing has also been mooted. Clearly, there could be conflicts of interest where commercial shellfisheries and large numbers of shellfish eating birds exist side by side (Galbraith 1992). Monitoring of waterfowl throughout the Moray Firth should be continued. In the case of seaduck, divers and grebes this requires additional coverage to that provided by the Wetland Bird Survey network. Numbers of seaduck, divers and grebes in the Moray Firth 221 The Moray Firth is a wetland of international importance and qualifies for Special Protection Area (SPA) and Ramsar status (Pritchard et a/ 1992). Current government guidance is that SPAs and Ramsar should be implemented through the SSSI network, which only allows for the protection of areas above the low water mark. While this is sufficient for some waterfowl, it is not for seaduck, divers and grebes, which are mainly to be found below the low water mark and thus would not be protected by SSSI designation. Provision for designation of Marine Nature Reserves under the Wildlife and Countryside Act 1981 has proved too unwieldy to deliver protection to important seaduck sites in the UK (Kirby et a/ 1993). Areas such as the Moray Firth which are important not only for seaducks but for other forms of marine life clearly require effective protection if the government is to honour its obligations under EC Directives. It is to be hoped that the opportunities for doing so under the EC Species and Habitats Directive are taken up. Acknowledgements The work was Carried out by the Research Department and the North Scotland Regional Office of the RSPB. but would not have been possible without funding from, chronologically, the Nature Conservancy Council, Britoil plc and their partners in the Beatrice field, BP Petroleum Development Ltd., and Scottish Natural Heritage. Dave Allen, Pete Akers, Simon Aspinall, John Barrett, Roger Broad, Brian Etheridge, Allan Mee, Greg Mudge, Sue Parsons and Jeff Stenning all gathered data while employed by the RSPB and 222 “wR Evans SB 19(4) were supervised by Len Campbell, Roy Dennis and Colin Crooke. Significant supplementary data were contributed by, among others, Martin Cooke, Roy Dennis, Dave Galloway, lan McCall, Tony Mainwood, Mick Marquiss, Fraser Symonds and Alan Vittery. Comments from Drs Ron Summers, lan Bainbridge, Greg Mudge and an anonymous referee greatly improved earlier drafts of this paper. References Aspinall S J and Dennis R H 1988. Goosanders and Red-breasted Mergansers in the Moray Firth Scottish Birds 15: 65-70 Barrett J and Barrett C F 1985a. Wintering Goldeneye in the Moray Firth. Scottish Birds 13: 241-249 Barrett J and Barrett C F 1985b. Wintering Divers in the Moray Firth Scottish Birds 13: 149-154. Baxter E V and Rintoul L J 1953. The Birds of Scotland. Oliver & Boyd, Edinburgh. Campbell L H, Barrett J and Barrett C F 1986. Seaducks in the Moray Firth: a review of their current status and distribution Proceedings of the Royal Society of Edinburgh 91B: 105-112. Carss D N 1994. Killing of piscivorous birds at Scottish fish farms 1984-1987 Biological Conservation 68: 181-188. Galbraith C 1992. Mussel Farms: management alongside Eider ducks. Their Scottish Natural Heritage, Edinburgh. Harvie-Brown J A and Buckley T E 1895. A Vertebrate Fauna of the Moray Basin. David Douglas, Edinburgh. Kirby JS, Evans R J and Fox A D 1993. Wintering seaducks in Britain and Ireland: populations, threats, research and conservation priorities. Aquatic Conservation: Marine and Freshwater Ecosystems 3: 105-137. Lack P (ed) 1986. The Atlas of Wintering Birds in Britain and Ireland. Poyser, Calton. Laursen K 1989. Estimates of Sea Duck Winter Populations of the Western Palearctic. Danish Rev. Game Biol. 13/6. Milne H and Campbell L H 1973. Wintering sea- ducks off the east coast of Scotland. Bird Study 20: 153-172. Mudge GP and Allen DS 1980. Wintering seaducks in the Moray and Dornoch. Wildfowl 31: 123-130. Nilsson L 1972. Habitat selection, food choice and feeding habits of diving ducks in coastal waters of south Sweden during the non-breeding season. Ornis Scandinavica 3: 55-78. Parrack J D 1986. ‘Great Northern Diver’ in Lack P (ed). The Atlas of Wintering Birds in Britain and Ireland. Poyser, Calton Pritchard D E, Housden S D, Mudge GD, Galbraith C A and Pienkowski MW 1992. Important Bird Areas in the UK including the Channel Islands and the Isle of Man. RSPB. St John C 1845. Wild Sport and Natural History of the Highlands. James Thin, Edinburgh Thom V M (1986. Birds in Scotland. Scottish Ornithologists’ Club/Poyser, Calton Waters R J and Cranswick P A 1993. The Wetland Bird Survey 1992-93: Wildfowl and Wader Counts. BTO/WWT/RSPB/JNCC, Slimbridge. Richard J Evans, Royal Society for the Protection of Birds, Unit 3.1, West of Scotland Science Park, Kelvin Campus, Glasgow G20 OSP. Revised manuscript accepted November 1997 Eider Bernard Zontrillo Scottish Birds (1998) 19: 223-230 223 Feeding studies of the Lesser Whitethroat in Strathclyde T BYARS & D J CURTIS The feeding behaviour of the Lesser Whitethroat was studied in Strathclyde at 2 separate territories, 43 kilometres apart. Invertebrate sampling was also carried out there and at 2 other control sites during May - July 1991 and 1992. Observations on territorial males indicated that a considerable proportion of their time was spent singing and feeding in the upper canopy level. Various feeding strategies of which pecking was the most common were identified and appear to be dependent on prey type and size. Video footage taken at the only accessible nest showed that caterpillars constituted 86% of prey items fed to fledglings. The predominant size range of caterpillars seen on video was between 6-10mm in length and that particular size range was proportionally significant in samples collected around that territory. Although our studies failed to establish a link between climate and Lesser Whitethroat distribution, climate may be an important factor in limiting Lesser Whitethroat distribution in Strathclyde through its effect on habitat. Introduction The Lesser Whitethroat Sy/via curruca is still a comparatively scarce breeding species in Scotland. In Strathclyde, the breeding population has remained relatively stable at 9-12 pairs since 1986, with no indication of further expansion into new areas. This has been attributed to the lack of suitable breeding habitat and climatic conditions (Byars et al 1991). We know of no other feeding studies solely on the Lesser Whitethroat in Britain. It has been suggested by Mason (1976) that range limitations may be related to diet. Because our study sites lie at the north westerly edge of the species’ breeding distribution, they provide an opportunity to study Lesser Whitethroat feeding ecology at the edge of its range. Study areas The majority of breeding territories which have been discovered in Strathclyde (Byars et al 1991) were evenly divided between 2 comparable sites, 43km apart. The Renfrewshire site, which covers a 2km? area south east of Paisley, contained 3 territories which were located at Dykebar and Brownside Braes. The Ayrshire site covered a slightly larger area of 2.5km? and is located just south of Ayr. Regular breeding territories were established at the Heads of Ayr, Burton Farm and Bracken Bay. All territories were located in areas of dense Hawthorn Crataegus monogynaand/or Blackthorn Prunus spinosa scrub with a dense mosaic understorey of Bramble Rubus sp, Gorse Ulex europaeus, Dog Rose Rosa canina and Goat Willow Salix caprea. 224 T Byars & DJ Curtis Study sites Insect sampling and observations of feeding behaviour commenced when territorial males arrived back on site. The criteria for locating suitable breeding habitat and assessing site fidelity have already been discussed (Byars et al 1991). The sampling sites for occupied territories were located in 2 Lesser Whitethroat territories, 48km apart. The 2 control sites used for invertebrate sampling were chosen on the basis of their similar habitat characteristics to the 2 breeding territories and were 23km apart. These sites were located in 3 of the 18 climatic zones found in Scotland calculated by Birse & Dry (1971) and are based on measurements of altitude, accumulated temperature and potential water deficits. Territory 1, at Heads of Ayr, was in Zone EE awarmdry lowland region; Territory 2 and Control 2 at Brownside Braes in Renfrewshire were in Zone EM, awarm moist lowland region. Control 1 at Dalry was within Zone ER, a warm rather wet lowland region. The vegetation profiles at these sites were as follows: Territory 1, Heads of Ayr: The territory contains large dense patches of Blackthorn scrub (2- 3m+) interspersed with mature Hawthorn (3- 4m+) and a patchy understorey containing Bramble, Dog Rose and Gorse (0-2m+) This site is situated on a NNE facing slope. Territory 2, Brownside Braes: The territory is located in a disused limestone quarry which is situated on anorth facing slope. The habitat consists of mature Hawthorn scrub (3-4m+) with a dense shrub layer of Bramble, Dog Rose and Gorse (0-2m+). Control 1, West Brownside Braes: This site contains an area of open mature Hawthorn scrub (3-4m+) with extensive Gorse and Bramble understorey (0-2m+), situated on a north facing slope. SB 19/4) Control 2, Dalry coal bing: A disused coal shale bing cloaked in Hawthorn scrub (2- 3m+) with mixed Bramble, Dog Rose and Gorse understorey (0-2m+) on a SE facing slope. Invertebrate sampling Insect samples were collected from each of the 4 sites using the beating tray method (Southwood 1 978). Within each territory and control site, 30 white plastic markers were randomly placed on the vegetation at 2 different height levels. Ten markers were placed in the scrub canopy at a 2 metre height level and 20 markers were placed in taller canopy at 4 metres height level. A telescopic metre stick was then used to beat a one metre square quadrat surrounding each marker 6 times in the scrub canopy. Invertebrates which fell were collected in a metre square catching tray. All items were then “pootered” into plastic vials containing 70% ethanol for analysis. Invertebrates were classified into taxonomic orders and all items were individually measured. All invertebrate sampling was carried out on warm sunny days. Samples were collected once a month during May, June and July in 1991 and 1992. Insect samples were also obtained using the sweep net method (Southwood 1978). A40cm diameter sweep net was quickly trawled back and forth 6 times over a one metre quadrat, which was randomly located above the scrub canopy. Four 1 metre quadrats per territory were analysed and captured invertebrates were then collected into plastic vials containing 70% ethanol for identification. This method was used to sample flying invertebrates located in mid air above the canopy in both study territories. Sweep sampling took place during 2 warm, but relatively calm days, during May 1991 and May 1992 to maximise capture of Diptera. Scottish Birds (1998) Observation of feeding birds Detailed observations on male spatial distribution were conducted over 2 x 10 hour periods on 2 selected dates in May 1991 and 1992 in both study territories. Ten hour periods were from0700to 1700 BST and observations were timed immediately after the male was located at the start of each songcycle. Specific activity, vegetation type and duration of time spent at each of the 4 height levels, ie 0-lm, |-2m, 2-3m, 3-4m were noted. Observations stopped when the male disappeared from the vicinity and could not be relocated by song after a 10 minute period. Feeding strategies utilised by the 2 territorial males were observed in May-June 1991-92 on a non quantitative basis within the study territories. Observations of feeding bouts started immediately after the male was located by song and all feeding methods were noted in detail. Adult males are more conspicuous and easy to observe early in the breeding season as they feed quite openly in the scrub canopy. After pair bonding, observing both sexes proved extremely difficult, as Lesser Whitethroats tended to search and feed more furtively within the dense scrub canopy. However, adults appear to give a soft contact tuc call when approaching and leaving a nest ' site, so that established routes can be identified and occasionally followed. _ AllLesser Whitethroatterritories were mapped using the minimum convex polygon method (Kenward 1987). The enclosed habitat was then systematically searched for any indication of nest building. Once found, nests were kept under daily surveillance and the contents checked periodically. Most food collections that were observed took place within 25 metres of the nest. Only one accessible nest suitable for video footage was located during the study period. Feeding studies of Lesser Whitethroat in Strathclyde 225 Video monitoring A tripod mounted VHS video camera was placed approximately 1.2 m away from the nest with magnification set to x8. The video camera was switched on manually and left to record until the battery power pack became drained after 2 hours operation. The equipment was then quickly retrieved with the minimum of disturbance to the nest. The video tape was then replayed back at Paisley University, where feeding visits to the nest could be viewed in detail and freeze framed when required. The majority of prey items were easily identified and, by using adult bill length as a reference, these items could also be individually measured. Video footage was taken on 10 June 1992 at the Heads of Ayr, when the fledglings were approximately 9 days old. To our knowledge, this video technique had never been used on Lesser Whitethroats before. Results Invertebrate abundance and seasonal variation * All 4 sites revealed a uniform trend in insect abundance and most taxa were represented throughout. Statistical analyses calculated from total invertebrate numbers per site showed no significant differences between territories or control sites, although there was a significant difference in mean invertebrate numbers between the 2 years. Out of the 7 taxa groups analysed, only Diptera and caterpillars varied significantly between territory and control sites, especially in 1991. There were notably more Dipterain territories than in control sites for both years, but there were significantly fewer caterpillars recorded in territories compared to that found in the control sites during 1991. 1992 showed similar 226 T Byars & DJ Curtis numbers of caterpillars for both territory and control sites. Those 2 groups exhibited distinct fluctuations in populations between the 2 years. Six different invertebrate taxa groups revealed distinct seasonal patterns of abundance in both territory and control sites. Lepidoptera larvae, Diptera and Coleoptera all showed peaks of abundance during May, followed by a decline in numbers during June and July. Hemiptera and Opiliones displayed a reverse of this trend, with peaks of abundance occurring during July and Psocopterashowed a peak of abundance in June. Results from the sweep sampling revealed small numbers of slow moving Diptera, notably Bibionidae and Stratiomyidae. |t appears that Bibionidae are favoured prey items, as male Lesser Whitethroats have been observed picking up large numbers from Goat Willow Salix caprea_ flowers. Densities of 12 Bibionidae per m were calculated from 4 random quadrat samples taken from the Salix canopy. Appendices containing details of these results have been lodged in the Waterston Library, 21 Regent Terrace, Edinburgh and may be consulted on request. Spatial distribution During the 2x10 hour periods, a total of 418 minutes was observed at the Heads of Ayr and 378 minutes at the Brownside Braes territory. Males appeared to spend a significant proportion of time singing along with feeding; 68% (285/418) and 60% (228/ 378) were recorded for this particular activity at the 2 sites. The predominant canopy at both sites differed and this was reflected in the results. At the Heads of Ayr, males spent 83% of their time in Blackthorn at the 2-3 SB 19(4) metre level (347/418), 16% in Hawthorn at the 3-4 metre level (68/418) and less than 1% inAsh at 5+ metre level (3/418). The utilisation of Ash trees within Lesser Whitethroat territories has been observed before (Hunt 1947). At Brownside Braes, males spent 66%, a substantial amount of time in Hawthorn at the 3-4 metre level (251/378). 23% was spent in Gorse at the 1-2 metre level (87/378) and only 11% in Hawthorn at the 2-3 metre level (40/378). Foraging behaviour Lesser Whitethroats can appear like Phyllocsopus warblers, especially when they forage in the scrub canopy. This distinctive feeding behaviour has been previously noted before (da Prato 1980). Males regularly patrol their territorial boundaries, singing and feeding as they move through the upper canopy. Lesser Whitethroats obtain their prey in 4 contrasting feeding methods 1.Pecking 2.Chase and snatch 3.Brief hover and 4.Flycatching. 1. Pecking was by far the most common type of feeding activity observed at both sites and accounted for 90% of all feeding observations. Pecking itself can be further split into two types: ASingle peck. This is apecking action towards individual food items (<5mm) dispersed on vegetation. B Rapid pecking. This occurs when multiple food items (<5mm) are obtained by rapidly pecking into a productive area of vegetation eg leaf clumps. Lesser Whitethroats appear to use their body weight in shaking leaf clumps when they hop on to adjoining branches. Hidden invertebrates attempt to scatter when disturbed and, are quickly taken by this rapid pecking action. — wt Oo ss fos Scottish Birds (1998) Food items which had been gleaned off the foliage by pecking were too small to identify visually in the field ie <5mm. Although beat sampling can reveal the diversity of invertebrates located in the canopy, we could not identify what the males were eating. 2.Chase and snatch. This strategy always involves a quick chase on moving prey, either by running along a horizontal branch in an attenuated posture, or ascending a trunk using flicked wings. Only 3 chase and snatch observations were made during the study period. The prey items caught were invariably large Tipulids (>10mm) which were dispatched by beating on nearby branches. Feeding studies of Lesser Whitethroat in Strathclyde VEE, 3.Brief hover. This feeding method is utilised when food items are located away from an available perch. Birds flutter very briefly for up to a second to pick at food items located on thin branch ends or in vertically placed cobwebs. This was observed 3 times during the study period 4.Flycatching. Two types of flycatching have been observed: A Flight. Seen twice, this involves short (5- 8m) one way twisting sallies from the upper canopy. Figure 1 Upper plot shows numbers of prey items availabe (as recorded in beating tray samples during June 1992 at Heads of Ayr) and taken (seen to be eaten in a video recording taken during same sampling date) as left and right bars respectively for various categories C = caterpillars, D = Diptera, M = Moths; 1-5 indicate size categories 0-5mm, 6-10mm, 11-15mm, 16-20mm and 21-25mm respectively. The lower plot uses a preference index where positive values indicate relatively more prey taken than in proportion to their availability, ie preferred, and negative values indicating relative avoidance. This is calculated as (proportion taken/proportion available - 1). 35 30 25 Number C1 C.2 C.3 C4 c.s a : D.i D.2 D.3 M1 M.2 Prey categories Pref, index value B preferred Oo §=6aavoided 228 T Byars & DJ Curtis B Static. This type was observed on 6 occasions. The method comprises short neck lunges from an openly perched position on top of the canopy, often accompanied with an audible bill snap. Only 114 minutes observation was obtained from one session in the field, during which a total of 47 food carrying visits were made: 31(66%) of such visits were made by the . female; while 16 (84%) were made by the male. Fifty two invertebrate items were brought to the nest, 45 (86%) were identified as Lepidoptera larvae, 4 (8%) were adult Tipulids, and 3 (6%) were adult moths. See (Figure 1). Asignificant 67% of the caterpillars were inthe 6-10mm size category. The female averaged 1.3 caterpillars per visit, while the male averaged 1.0. The female spent a mean of 208 seconds away (31 forays analysed), while the male spent 228 seconds away (16 forays analysed). The male was responsible for the only brooding of the young during the observation period. Post fledging observations Fledglings vacate the nest at around 10-11 days after hatching and hide in the dense vegetation layer (0-2m) within the territory, as their flight feathers are still in pin and not fully formed. The adults continue to feed them for a further 2 weeks, foraging in the canopy level. Food items collected by the adults could not be specifically identified during the fledgling period. Once fully fledged, juveniles roam further afield with the adults in a family party. Juveniles appear to loiter around the general vicinity of the territory for approximately 3-4 weeks before completely disappearing. Ringing studies in England have revealed that juvenile Lesser Whitethroats begin to leave their natal area once post juvenile moult has fully commenced SB 19(4) at around 30-40 days old (Norman 1992. Boddy 1994). Discussion Due to their dense habitat and skulking behaviour, prolonged observations on Lesser Whitethroats are difficult during the breeding season. During early May, however, observations on established males are easier as they patrol their territories. Data collected from timed observations within sampled territories indicate that males appear to spend a considerable amount of time patrolling as they sing and feed in the canopy. This roving behaviour of the territorial males could be territorial defence, mate searching, a response to low prey density, oracombination of these factors. When adults were feeding nestlings, they seemed to largely forage within <25 metres of the nest. As invertebrates and caterpillars in particular are relatively abundant in June, the adults we studied seemed to obtain prey within a small proportion of their territory when foraging to feed their young. Our study suggests that-in Strathclyde male Lesser Whitethroat foraging behaviour depends on habitat structure. At the Heads of Ayr site, where Blackthorn covered approximately 90% of the territorial area, males consistently fed at approximately 2- 3m in the Blackthorn canopy. At Brownside Braes, Hawthorn covered 60% of the occupied territory and the male Lesser Whitethroat spent more time feeding in the top (3-4m) canopy than at any other level. This suggests that invertebrate numbers were more abundant in the canopy, but our sampling data found no evidence from any of the 4 sites to confirm this theory. We need to carry out more detailed field work to investigate this. During May, males use various foraging Scottish Birds (1998) techniques depending on the prey item concerned. When males are hunting slow moving Diptera such as Bibionidae the prey items are easily observed and collected. However, the majority of food items taken appeared to be under 5mm in size and were not identified. Although both sexes hunt in thick cover when foraging for their young, video evidence showed that caterpillars are important prey. If caterpillar populations fluctuate between years - as our results suggest - then fledging success may depend upon caterpillar abundance during the breeding season as caterpillars are large and relatively easy to collect. In 1992, when caterpillar numbers were higher, a brood of 5 successfully fledged at the Heads of Ayr. Clearly more nesting attempts would need to be studied to confirm this link. The timing of the nestling period coinciding with peak caterpillar abundance has been well documented and variations in such timings have been shown to affect the breeding success in insectivorous birds such as tits (Perrins, 1979, 1991) and the Pied Flycatcher Ficedula hypoleuca (Lundberg & Alatalo, 1992). We previously suggested (Byars et al 1991) that the breeding population of Lesser Whitethroats in Strathclyde may be restricted by the climatic effects on prey populations. However, this study found that territories in different climatic zones did not differ in invertebrate diversity and overall numbers to any significant degree. The Brownside Braes control site had the highest total of invertebrate numbers and yet the established breeding territory was located 100 metres away outside the control area. Hawthorn scrub, which was sampled in both territories and control sites, appears to have very high invertebrate numbers and diversity and this is reflected by Feeding studies of Lesser Whitethroat in Strathclyde 229 the variety of other insectivorous bird species which breed within all 4 sites. Although our studies have failed to establish that climate affects Lesser Whitethroats through invertebrate distribution, it may have an important influence on habitat distribution within Strathclyde. We previously suggested that the distribution of Lesser Whitethroats in the Lothians is linked to the climatic zone defined by Birse & Dry (1971) as Zone EE, which is a warm dry lowland region below 200 metres. What appears to be ideal breeding habitat for Lesser Whitethroats can be found well over 100m in the Lothians (pers obs). Climatic conditions appear to limit the growth and diversity of Hawthorn scrub over 100m in Strathclyde and this seems to restrict the extent of Hawthorn scrub within the region, whereas more favourable climatic conditions in south east Scotland allows for more widespread Hawthorn scrub and therefore a larger breeding population of Lesser Whitethroats. Acknowledgements We are extremely grateful to everyone who helped us during this particular study, especially those who participated in the invertebrate sampling ie Stephen Cambell, Julie Figures, lain McDonald, Shona Quinn and Andrew Stevenson. We also thank; Professor C M MacDonald for supporting this work in the Biological Sciences Department at Paisley University, John Sweeney for constructive criticisms and Norman Tait for taking the excellent photograph of the Lesser Whitethroat. We would also like to thank Mr C Rankin of Laigh Kyleston farm for allowing us access onto his land at the Heads of Ayr. 230 T Byars & DJ Curtis SB 19(4) References Birse EL & Dry F T 1971. Assessment of climatic conditions in Scotland based on accumulated temperatures and potential water deficit. Macaulay Institute for Soil Research. Aberdeen. Boddy M 1994. Survival-return rates and juvenile dispersal in an increasing population of Lesser Whitethroats. Ringing & Migration 15: 65-78. Byars T, Curtis DJ, McDonald!1991. The breeding distribution and habitat requirements of the Lesser Whitethroat in Strathclyde. Scottish Birds 16: 66- 76. Hunt G H (ed) 1947. Leicestershire and Rutland Ornithological Society Report. Kenward R 1987. Wildlife Radio Tagging, Equipment, Field Techniques and Data Analysis. Academic Press, London. Lundberg A 1992. The Pied Flycatcher. London: T & A D Poyser. Mason C F 1976. Breeding biology of the Sylvia warblers. Bird Study 23: 2 13-232. Norman S C 1992. Dispersal and site fidelity in Lesser Whitethroats. Ringing & Migration 13: 167- 174. da Prato S R D 1980. How many Lesser Whitethroats breed in the Lothians? Scottish Birds 11: 108-112. Perrins C M 1979. British Tits. London: Collins. Perrins C M 1991. Tits and their caterpillar food supply. Ibis 133: 49-54. Southwood T R E 1978. Ecological methods with particular reference to the study of insect populations. 2nd edition. London. Chapman & Hall. Tom Byars, Department of Biological Sciences, University of Paisley, High Street, Paisley PA1 2BE. Dave J Curtis, Department of Biological Sciences, University of Paisley, High Street, Paisley PA1 2BE. Revised manuscript accepted November 1997 Male Lesser Whitethroat at Brownside Braes Norman Tait == na = =& 2 > E | Scottish Birds (1998) 19: 231-238 Za) The breeding bird community of upland Juniper scrub in eastern Scotland S GILLINGS & R J FULLER During spring 1997 3 sites containing Juniper scrub were censused to describe the breeding bird communities present. A combination of territory mapping and transect counts recorded 18 bird species. The community was dominated by Willow Warbler, Meadow Pipit and Chaffinch but also included locally scarce species such as Ring Ouzel, Whinchat and Black Grouse. The findings are discussed in relation to the likely effects of expansion of Juniper scrub on local avian diversity. | valuable in understanding the changes that might occur in bird communities with _ continued scrub expansion. Such knowledge Introduction It has been suggested that decreases in grazing pressure and reduction of traditional upland management techniques such as burning might promote rapid scrub and woodland regeneration in the Scottish uplands (eg Usher & Thompson 1993). On some _ estates andreserves inthe central and eastern _ Highlands concerted efforts are now being made to reduce numbers of Red Deer Cervus elaphus so it Is likely that scrub will develop in several parts of the Highlands in future decades. Where regeneration occurs on open moorland the new habitats will probably _ support markedly different assemblages of bird species but the exact community likely to _ develop is not known. The British Trust for Ornithology (BTO) has been undertaking ' censuses within existing upland scrub of various types in order to describe the range of bird species present and the densities at which they are found. This information will be is an important component in developing conservation and management policies as required (Hester 1995). This paper reports on work undertaken in 1997 aimed at describing the bird communities found in Juniper Juniperus communis scrub in the eastern Highlands of Scotland. Juniper scrub is largely restricted to the east central Highlands, particularly in the Cairngorm and Monadhliath ranges where it occurs between 350m and 650m (Rodwell 1991). It may be present as open, short scrub on moorland and also as understorey in birch Betula and Scots pine Pinus sylvestris woodland. Stands are typically even aged and of low productivity (Hester 1995) hence conservation of Juniper may be an issue itself. Scrub frequently exists as complex mosaics consisting of scrub patches differing in structure interspersed with moorland grass and Hseather Calluna vulgaris. Another aim of this study was to assess how this patchiness and structural variation affected bird community composition. 232 = SGillings & RJ Fuller Methods Study sites and field methods Juniperis relatively widespreadin the Scottish uplands but large tracts suitable for a study such as this are scarce. We selected 3 sites inthe eastern Highlands (Table 1) dominated by Juniper but differing in the height, ground cover and shape of bushes. The locations of these sites have been withheld at the request of the landowners. All fieldwork was undertaken in May and June 1997 on dry, clear, relatively calm mornings. Sites 1 and 2 formed discretely definable plots and were censused by territory mapping. Site 3 was extensive and more suited to census by transect counts. On a preliminary visit a hand held global positioning system (GPS) was used to define the plot boundaries of Sites 1 and 2 (accurate to nearest 10m) for drawing plot maps (scale 1:2500). Each of these plots was visited twice during which the locations of all birds were recorded on the plot map using standard territory mapping notation (Marchant et a/ 1990). Site 3 was surveyed using 4 transects (total 4.02km) set out on the first bird recording visit. These crossed extensive tracts of Juniper varying in height and ground cover and followed the hill contours in order to remain at the same altitude. The GPS was used to fix the start, changes of direction and end so that the length of each transect could be derived (to nearest 10m). Each transect was visited 3 times and the direction walked was switched to reduce bias arising from decreasing song activity after dawn. All birds encountered were placed into distance bands (0-50m and 50-100m) on basic transect maps using territory mapping notation. Basic vegetation descriptions were made at each site. The mapping plot at Site 2 was split SB 19(4) into3 sections (A-C, Table 1) on the basis of gross vegetation differences. On Site 3 the transects were split into sections of similar vegetation by fixing the location of major vegetation transitions using the GPS. In this way the length of transect in 4 vegetation types (I-IV, Table 1) was derived. Different grazing animal populations were present at each site and this may partly explain variation in vegetation structure. Rabbits Oryctolagus cuniculus and Mountain Hares Lepus timidus were present on all 3 sites but at varying densities (see Table 1). Roe Deer Capreolus capreolus were seen on Sites 1 and 3 and Red Deer were seen on Site 3. Sheep were recorded only on Site 1. The fact that we did not record one or both deer species or sheep at a particular site should not be taken as an indication of their absence. Estimation of densities The number of registrations of each species | on each visit was extracted from visit maps | forSites 1 and 2. OnSite 3 only birds recorded | within the 50m bands were used for density estimation (hence a 100m section is equivalent to 1ha). In all cases if anindividual | was recorded in more than one section (or | inside and outside a plot boundary) the proportion of registrations in each section counted towards the total. Excluding Black | Grouse’, species densities were calculated | using the mean number of registrations across | visits. This method provides a good approximation to densities derived by territory mapping and Distance Sampling procedures | (Gillings et a/ in press). Distance Sampling was not deemed necessary because of the large distances over which individuals were | | detectable. For male Black Grouse, which were often detected in groups, calculations used the number of males rather than the number of registrations. | | Scottish Birds (1998) Bird communities of upland Juniper scrub in eastern Scotland 233 "E} WL> “%06 - %09 1909 Jadiunr colt Al "WE} WL> ‘%6G - %9e 1909 Jadiunr oe} HI "(g’O Aljensn) {Je} WG2"0> “%GZ - %0|} 4aa09 sediunr L'6 Tl JE} WG'O> AAOD %O|> Jodiunr ‘usyyeEaH/SsesyH 100/y UBdO 9'rl | ‘yelqey JejIWIS JO Sdiujs JOSSUeJ] O]U! PEpIAIDgns SEM SIIS 9U] “NOYUHNOJY] JUsSeJd Ie ULeEJUNOW| “OleSOW JediuNfyeUuleaH/SseJy .ZOp SIOUM G6HS-8IS E=alS JBAOD %09-01 ‘Wel Wy E> sediunr (iqqey) pezeib Ajineay ‘sses6 ayi-lin paezesi6 Ayineay OUS ySl 9 ‘I9A09 %09-Or ‘I/e} W}.> sediunp Aysnq pezesbun “seyyeayH/sses6 pezesbun boat q ‘(49A09 %G>) Jodiunr ou Ajjenwia ‘ssess JOOH/\ Gili Vv “SeaIe € OJUI PAPIAIDGNs Sem OjIS BU, “OrESOW JediuNPaUJeaH/Ssesy L2p lOUM ZSP-96E ZeaIsS ‘aay UOWWOD Ayejnoed aey uleJUNOW\| °%09-G ebues Ul JAAOD ‘|\e} WG" } SeUSNG O} |je} WG'O> sjue|d oy!) WeEW WOJ) PaUeA JOdIuNr ‘olesoW sediunrjayyeaH/SssesH 9°61 SJOUM 99E-SEE | els (ey) (tu) uondiosap uonejabeaj = @ZIS_-«UON}IEG)9= ApPNINIY = AHS “QUI| JOBSUBJ] BY] JO APIS YOLA UIQS JO S}]/aq UO paseg SI SJuNOD JIaSUeI Aq PalaAOD ease AY, “SJUNOD JOaSUeI Aq F AIS Ppue Buiddew Asojiisay Aq pasnsuad 31am Z pue | sajig ‘spuesybipyy usajsea ay] ul sjojd sadiunr 404 sjiejap ayis yo Aseusruing | 31QeL 234 S§Gillings & RJ Fuller SB 19(4) "Scientific names of birds are given in Table 2 Results Eighteen species of bird were recorded across the 3 sites and estimated densities are presented in Table 2. Results are presented for whole sites and for the different habitats (A-B and I-IV). Willow Warbler was the commonest species in virtually all stages of Juniper scrub, followed by Meadow Pipit and Chaffinch There was much vanation in species density between sites and the different vegetation types represented by each section. Ourassessment of associations of each species with the extent of Juniper cover are given in Figure 1. Curlew, Snipe and Wheatear were mostly recorded from open moorland areas. At the other extreme, Goldcrest was only recorded where tall (c2m) Juniper bushes existed. Willow Warbler was recorded throughout but increased in density with increasing Juniper cover. Total density (all species combined) increased with increasing Juniper cover and height (Figure 2). Discussion The bird assemblage found in Juniper scrub was low in species richness and was dominated by Willow Warbler, Meadow Pipit and Chaffinch. The species recorded, and their densities, were very similar to those found in birch scrub in the eastern Highlands (Gillings et a/ in press). Both in birch and Juniper Willow Warbler densities were in the range 10-15 birds/10ha and comparable with densities in southern England and Scandinavia (Gillings et a/in press). Meadow Pipit densities in Juniper scrub were less than half previously published densities on open moorland (Thirgood et a/ 1995). Apart from the species mentioned above, there Figure 1. Graphical representation of the association of each species with the extent of Juniper cover. Bar width indicates relative habitat association: narrow bars = poor usage, wide bars = regular usage, based on density esti- mates and field observations. % Juniper cover So i uu YY i) —I On So con) a Red Grouse Black Grouse Snipe Curlew Skylark Meadow Pipit Pied Wagtail Wren Dunnock Robin Whinchat Wheatear Ring Ouzel Song Thrush Willow Warbler Goldcrest Chaffinch Redpoll were also notable densities of Dunnock, Song Thrush, Ring Ouzel, Whinchat and Black Grouse. Although densities of Black Grouse appear higher than those reported by Baines (1996), they are probably not representative of larger areas, and are not truly comparable because of methodological differences. Juniper scrub is not homogeneous and plots show wide variation in the extent of cover, growth form and size of bushes. Several factors may causes such variation including genetic differences, local climate (A MacDonald pers comm) and different grazing animals. Where abundant, Rabbits may graze the lower foliage, even within the base of the bush, producing stands that are much more open. This combined with the short, turf like Scottish Birds (1998) Bird communities of upland Juniper scrub in eastern Scotland 235 H : : . P ‘ f : ‘ G0 eawwely syjanpsed |\|jOdpay Cae owl |. O10 GiOpered re 30) 00 vk ee $qa/909 eI//BUII4 YOUulJyeYyo ¢ ‘ : . : SO OO OO re) 0’ sninbas sninbay \saJj9pjo5) MS MPSS SATS Cy cale Cente 710 v9 Gel SNIYIO4] SNAOISO||AYd 19\QVEM\ MO|||M\ COP Oke 80 EO ci SOO O10 Z0 8’ | sojawojiyd snping ysniy| buos Oe a Oe Oe “010m OO One oO: Zi0 60 A smjenbio} snpiny |ezno bury O00 0) = Ol CO ae. Ore ey 0) 00 20 . BYJUBUBO BYJUeUAC Jeayeay\\ HaOem 2OlO! = OF0m 6.0. aci0 O0m 6 0) 0:0 Lo) ; BHOQN BJOIIXES YEyOUlyM : . : A : FO 00) O10 ie) 60 Bindages snoeYFA UIGOY Vee tc O = Sie ci0 = OF Gin, OO) O10 6k OKs SLe/NPOW BjauNJd YOOUUNG Gi0les 20°0) Ee CO a0 FO 310) 0) 0 v0 : sajApoj6o1) sajApojbol, UasN S : ‘ : : Om 10.0) 010 LO : eqe eyjoejoyy \!e\Oe\\ Pald G Glee Chl & 6 CISC KOT Gime 20'S: eit 9°€ ime sisuajesd SNYJUY did Moped Oe, 200) 010F 0). oO : j : Z : sisuaAle epnely y1e\AYS O00) OO Soy Wy LO Gm 60) =8ic 80 : Byenbie sniuawinny MEIN ; ‘ : ‘ ; Ome 60 6250 20 0} oBeuyyje6 obeuyje adius Oe 100 ex aOiOneee sel Ciomee 0'O) 0.0 vik G0 X/4JO] ORAL ISNOIS YOR! Sa O°} C Camm OM Ol ‘ ‘ ? : : sndobe; sndobe7 asnoi5 pey A\ (II 1 | SIOUM 2) q VY SOU S|OUM sal9edS € 8S Z OS | OUS ‘IS B Je papsooas JOU saizeds ajeoipul sysuajsy *,.eYo, 4ad sayeul SI YOIYM asnosy yoe/g jdaoxa ‘,.eyoL 4ad suonessibas ase sayisuag ‘spuelybipy] usajsea ay] ul sjojd sadiunr ¢ uo spilg bulpseasg jo saijisuag Z a\qeL 236 SGillings & RJ Fuller SB 19(4) Figure 2. The total density of all species broken down into sites, sections and habitats. Total Density birds/10ha 4 _ nN N So uo i=) uo on Whole Whole A B Site 1 Site 2 grass sward produces a very different habitat from Juniper that is grazed only by Mountain Hares. Such spatial and structural diversity may provide many different niches for birds. Densities produced for gross vegetation types indicated that there were clear preferences by some species for different habitats on a continuum from open moorland to closed canopy Juniper thickets. Without comparable density estimates in open moorland adjacent to our study sites, it is difficult to determine how tolerant of scrub are species typically associated with open moorland. However, it is clear that Snipe, Curlew and Skylark persisted in areas where sparse, prostrate Juniper occurred. Moreover, Meadow Pipit, Ring Ouzel and Red Grouse persisted where Juniper occurred at up to 75% ground cover with some grass. A Ring Ouzel nest and several broods of Red Grouse were found within open canopy Juniper scrub on Site 3. C Whole 1 II It IV Site 3 Section C of Site 2 was Rabbit grazed grass with Juniper bushes and appeared particularly favourable for Ring Ouzels. Two pairs were recorded here, including one feeding a family of 4-5 fledglings on open lawn like turf. At the other extreme, species such as Willow Warbler, Chaffinch and Redpoll preferred more continuous tracts of tall bushes. When predicting the effects of possible future expansion of Juniper it is clearly important to use the appropriate measure of density. This paper presents several density estimates which are applicable at different spatial scales and to different types of Juniper structure. On asmall scale, Juniper scrub may be relatively homogeneous particularly as it often grows in even aged stands (Ward 1973). On alarger scale, however, the densities presented in Table 1 for specific stages will be inappropriate because scrub regeneration is likely to be Scottish Birds (1998) Bird communities of upland Juniper scrub in eastern Scotland 237 patchy and to produce mosaics of Juniper amongst Heather or grass. In these cases, the densities for whole sites might be more useful since they average out high and low density patches. Can these results be generalised to Juniper scrub in other areas of Scotland? The 3 sites used in this study exhibit arange of spatial structure, vegetation form and presence of 5 grazing animals (Rabbit, Mountain Hare, Red Deer, Roe Deer and sheep). Hence they are probably typical of Juniper scrub on moorland in the Scottish uplands but not necessarily in other habitats, for example understorey in birch woodland. The lack of study sites in the western Highlands may reduce the generality of conclusions because local bird species pools may differ, although this is unlikely to have a major effect in these simple habitats. These results suggest that colonisation of open moorland by Juniper scrub would probably Cause an increase in local species richness. Scrub regeneration might be at the expense of upland breeding wader populations but there are several reasons why this might not be the case. Firstly, Juniper prefers well drained soils and will only regenerate on bare soil or recently heavily grazed swards (Miles & Kinnaird 1979) where wader populations may typically be low. Furthermore, scrub regeneration is extremely slow or non existent at the high altitudes at which waders such as Golden Plover Pluvialis apricaria and Dunlin Calidris alpina breed in eastern Scotland (Brown & Stillman 1993). On the positive side, raptor species such as Kestrel Falco tinnunculus. Sparrowhawk Accipiter nisus, Merlin Falco columbarius and Hen Harrier Circus cyaneus may benefit from an increase in the local availability of songbirds. We encountered Kestrels and Hen Harriers on or around our study areas but cannot provide any information on their use of the scrub. Also, mosaics of Juniper scrub and grass adjacent to heather moorland may be particularly beneficial to Black Grouse populations. Males from a large lek at Site 3 were often seen within the scrub during fieldwork and this habitat probably provides important daytime cover and food resources (D Baines pers comm). In summary, we have found moderate to high densities of several species typical of open moorland and scrub breeding in Juniper scrub. Further expansion of Juniper scrub could be beneficial for many species, including some that are vulnerable, but possibly at the expense of strictly open moorland species More work is needed to identify the areas most likely to be colonised by Juniper and to describe the breeding bird population there so that any necessary areas of action can be identified. Future research could also focus on wide ranging species such as raptors to assess how they utilise scrub habitats. Acknowledgements This work was funded by the Joint Nature Conservation Committee on behalf of Scottish Natural Heritage, English Nature, the Countryside Council for Wales and the Environment and Heritage Service in Northern Ireland. We would like to thank Angus MacDonald, Jim Parkin and Phil Whitfield (SNH) for help during the study and for comments on an earlier draft. Thanks to Su Gough for producing the figures. We are grateful to the landowners for giving us permission to work on their land. References Barnes D 1996. The implications of grazing and predator management on the habitats and breeding success of black grouse Tetrao tetrix. Journal of Applied Ecology 33: 54-62. 238 = S Gillings & RJ Fuller SB 19(4) Brown A F and Stillman R A 1993. Bird-habitat associations in the eastern Highlands of Scotland. Journal of Applied Ecology 30: 31-42. Gillings S, Fuller R J & Henderson A C B 1997. Avian community composition and patterns of bird distribution within birch-heath mosaics in eastern Scotland. Ornis Fennica in press. Hester A J 1995. Scrub in the Scottish uplands Scottish Natural Heritage Review No 24 Scottish Natural Heritage. Edinburgh. Marchant J H, Hudson R, Carter S P & Whittington P 1990. Population Trends in British Breeding Birds. British Trust for Ornithology, Thetford, Norfolk. Miles J & Kinnaird J W 1979. The establishment and regeneration of birch, juniper and Scots pine in the Scottish Highlands. Scottish Forestry 33: 102- 119. Rodwell J S (ed) 1991, British plant communities, Volume 1: Woodlands and scrub. Cambridge University Press, Cambridge. Thirgood S J, Leckie F M & Redpath S M 1995. Diurnal and seasonal variation in line transect counts of moorland passerines. Bird Study 42: 257-259. Usher M B & Thompson D B A 1993. Variation in the upland heathlands of Great Britain: Conservation importance Biological Conservation 66: 69-81. Ward. LK 1973. Theconservation of juniper present status of juniper in southern England. Journal of Applied Ecology 10: 165-188. S Gillings, Terrestrial Ecology Unit, British Trust for Ornithology, The Nunnery, Thetford, Norfolk IP24 2PU R J Fuller British Trust for Ornithology, The Nunnery, Thetford, Norfolk IP24 2PU Revised manuscript accepted January 1998 Willow Warbler foe Sidney Clarke Scottish Birds (1998) 19: 239-243 239 Seasonal differences in pellet remains from Golden Eagles in the Isle of Harris A C POUT An analysis of prey remains in pellets collected from 4 pairs of Golden Eagles in the Isle of Harris revealed that ‘live prey’ remains were more frequently found in pellets from the prebreeding period than from later in the season, and in pellets collected from an active nest site than from other sites within the same range during the fledging period. It is argued that these differences arise from the feeding preferences of the adult Golden Eagles. Possible reasons for such a preference and the implications are discussed. Introduction Golden Eagle diet in Scotland can be usefully thought of as consisting of 2 distinct components: the small and medium sized mammals and birds that Watson et a/ (1992) termed ‘live prey’; and the remains of sheep and Red Deer Cervus elaphus that eagles scavenge as carrion. Live prey is, in the vast majority of cases, likely to have been killed by the eagles, will thus be fresh at the time of consumption, and is generally of a size that enables it to be transported by an adult Golden Eagle. In contrast, carrion is scavenged and not killed, therefore need not be fresh when it is consumed. Due to the size of the carcasses it is not readily transportable by the eagles themselves and consequently the vast majority of the feeding occurs at the site where the animal died. Watson etal (1992), inastudy encompassing 9 geographic areas across Scotland, founda positive correlation between the availability of live prey and Golden Eagle breeding success, and between the availability of carrion and Golden Eagle nesting density. A suggested explanation for the relationship between live prey and breeding success (Watson & Langslow 1989) was that, as carrion was not readily transported, the feeding of the young on the nest was dependent on the availability of live prey. In much of Scotland, and particularly the west, a substantial number of pairs of Golden Eagles is consistently recorded as being present within home ranges for which no evidence of breeding is subsequently found (Murray 1992-1996). While this may in part be due to observer effort, it also suggests that failure to lay eggs may be as important a factor as fledging success to overall breeding productivity. This study is based on pellets from Golden Eagles in South Harris, from which Red Deer are absent and where sheep are the only source of carrion, to examine seasonal changes in the live prey and carrion components of Golden Eagle diet. Methods Cast pellets were collected from the perches, roosts and the nest sites of 4 adjacent pairs e400 “AC Pout SB 19(4) of Golden Eagles in the Isle of Harris, Western Isles, during 3 periods in 1996: 25 March-11 April; 26 May-2 June; and 26 August-2 September. Collections were made from the same sites during each period and, therefore, the time interval within which pellets were cast could be determined, allowing pellet contents to be related to particular periods in the breeding season of the adult Golden Eagles. The pellets collected in March-April would have been cast in the period prior to egg laying, those collected in May-June will have been cast during the incubation period and the early fledging period, and those collected in August-September will have been cast during the late fledging period and the first 4 weeks of the post fledging period. In addition, pellets cast by a young eaglet were collected from a successful nest of one of these pairs during June and August, enabling a comparison between the remains of food consumed by the young on the nest and those from food eaten by the adults of that pair away from the nest over the same time period. Prey seen on the nest was also recorded, though was notusedin the analysis. Maximum length and width of pellets was recorded and hair and feather remains were examined under x 47, x 100 and x 400 magnifications and identified with reference to Day (1966), Brunner and Corman (1974), Brom (1986) and Teerink (1991), and also by comparison with a reference collection gathered in Harris. Skeletal remains recovered from the pellets were identified from skins in the Department of Zoology, University of Aberdeen and, in one instance, from Yalden (1977). Remains were recorded by their occurrence within a pellet, the presence or absence of live prey remains being compared. No attempt was made to relate the frequency of occurrence of remains in pellets to the frequency of occurrence of individual animals as food except when the presence of skeletal material enabled a minimum number of individuals to be inferred. Statistical tests followed Fowler & Cohen (1990). Results Remains of live prey were found to be present in 31 of the 51 pellets from the prebreeding period, 8 of the 23 pellets from the second collection period and 3 of the 10 pellets from the third collection period, the remaining pellets consisting entirely of carrion remains. The greater frequency of live prey remains in pellets collected during the prebreeding period in comparision with pellets from later in the season is statistically significant (X?= 6.12, d.f.=2,p<0.05). There was no evidence that the size of pellets changed over the period (Kruskal-Wallis test: K=4.12,d.f.= 2,n.s.), nor that the pellets became any more or less heterogeneous in the number of discernible prey remains a single pellet contained (X = 0.73, d.f.=2, n.s.). The live prey remains recorded in pellets included 49 bird and 29 mammal items (Table 1). Of the birds, Red Grouse Lagopus lagopus was the most common species with 24 occurrences. Passeriform remains were found in 11 pellets; of those that were identifiable to family, 2 were corvids and one was a thrush Turdus sp. Three passerines were identifiable from skeletal remains as being small, ie under 40g, but were not identifiable to family. Three pellets contained the remains of diver Gaviaspanda Woodcock Scolopax rusticola was identifiable as one of the Scolopacidea recorded. A single pellet contained Grey Heron Ardea cinerea feathers and another contained feather remains that were unidentifiable due to lack of downy barbules. | i2tsorcdit:=(sp< 0:01). Scottish Birds (1998) Of the mammalian live prey, Rabbit Oryctolagus cuniculus was the most common single item, remains being found in 8 pellets. The remains of small mammals, such as Wood Mouse Apodemus sylvaticus and Pygmy Shrew Sorex minutus, were found in 6 pellets and a single pellet contained the fur of a cat Felis sp. The 12 pellets recovered from the nest differed from 20 pellets collected from the adults of that pair over the same period in that nest pellets contained no sheep remains (X = Red Grouse was - presentin 11 of the pellets and Mink Mustela vison fur was present in 6 pellets. A total of 27 Mink claws were recovered from these pellets indicating that at least 2 Mink must of Harris in 1996. | Total pellets 25/3 to 26/5 to 11/4 2/6 Gavia 3 Ardea Lagopus i 4 _ Charadridae 3 Scolopacidae 5 Passeriform 5 3 _ Unidentified 1 Sorex 1 Mustela |. Felis 1 Sheep 33 18 | Apodemus 3 2 | Oryctolagus 5 2 _ Total occurrences 66 30 51 23 Pellet remains from Golden Eagles in the Isle of Harris 241 have been consumed. Prey items actually observed at the nest site included 4 Red Grouse, 1 Golden Plover Pluvialis apricaria chick, 1 Snipe Gallinago gallinago, 2 Rabbits, 2 lambs and a Wood Mouse. The surviving eaglet also consumed its sibling which died at between 2 and 3 weeks of age. Discussion Live prey predominated in the pellets from the young eaglet on the nest at a time when the pellets from the adult Golden Eagles of that pair contained a substantial proportion of sheep carrion. This suggests that the adult eagles are either constrained in, or discriminating between, the types of food they consume themselves and that which | Table 1 Frequency of indigestible remains recovered from Golden Eagle pellets, Isle 26/8 to Nest Total 2/9 3 1 1 2 ib 24 3 1 6 §} il 1 1 3) 6 1 9g 60 5 1 8 14 21 131 10 12 96 242 AC Pout they use to provision their young. It also suggests that studies that do not distinguish between pellets from adults, ie those cast away from the nest site, and pellets from the young are likely to misrepresent Golden Eagle diet. The frequency of live prey remains in pellets from adult eagles during the prebreeding period also runs counter to the seasonal abundance of the live prey species present in Harris. While this may in part be explained by adult eagles feeding their young on the live prey they catch in the fledging period rather than eating it themselves, it nevertheless suggests that live prey is favoured over carrion by the adult eagles during the prebreeding period. Indeed, based on the stocking levels, reported mortality rates and the numbers of carcasses found within the ranges (own data, unpublished BSc honours project, University of Aberdeen), the availability of sheep carrion in South Harris is such that it would be feasible for Golden Eagles to feed on nothing else for the majority of the year if it were their favoured food. Thatso, one explanation for favouring live prey might be that there is a need for female Golden Eagles to gain body condition prior to egg production and incubation, a requirement demonstrated in Sparrowhawks Accipiter nisus (Newton et al, 1983) and argued (Ferrer, 1994) for the Spanish Imperial Eagle Aquila adalberti, and that the carrion that is available is nutritionally inferior to live prey in this respect. Again, were it not so then, as Brown (1969) pointed out, one would expect breeding failure to be manifest as death of the young during the fledging period due to lack of live prey delivered to the nest and not as high levels of nonbreeding. Consideration of the circumstances under which sheep die on moorland grazings suggests that many of the carcasses available to scavenging birds, at least in Harris, will be those of sheep in poor condition. Carrion is SB 19(4) also likely to be much more variable in quality than live prey ina situation where consistency of diet over a prolonged period may be important. Barton and Houston (1993) have shown that sheep carrion has lower fat content and energy yield than Rabbit or Pheasant Phasianus colchicus. Carrion abundance correlates with high Golden Eagle nesting density (Watson et al 1992) hence smaller relative range sizes and, everything else being equal, fewer live prey per pair. In addition, heavily grazed areas support fewer live prey than, for example, areas with good heather cover. If there is arelationship between the availability of live prey and egg production then one might expect that areas with high carrion availability and high Golden Eagle nesting density would be characterised by the widespread incidence of nonbreeding by established pairs of Golden Eagles that are | able to subsist on the ample carrion that is available. This appears to be the case in © South Harris where Golden Eagle density is approaching 40 pairs per 1000 km?of suitable habitat (Green 1996), sheep carrion is abundant for the majority of the year and, of 3 pairs of Golden Eagles monitored over an 8 year period, 2 have consistently failed to lay eggs despite regular occupancy and with no evident signs of disturbance. Acknowledgements This work was completed in partial fulfilment of the requirements for a degree in Honours Zoology at the University of Aberdeen. | would like to thank my supervisor Professor Paul Racey and Dr Jeff Watson and Mr Dick Balharry for help andadvice. Specialthanks | to Anne Campbell who helped with the collection of pellets. The comments of an anonymous referee were also appreciated. Scottish Birds (1998) Pellet remains from Golden Eagles in the Isle of Harris 243 References Barton WH & Houston DC 1993. Acomparison of digestive efficiency of birds of prey /bis 135: 363-371. Brom T G 1986. Microscopic identification of feathers and feather fragments of Palaearctic birds. Bijdragen tot de Dierkunde 56: 181-204. Brown LH1968. Status and breeding success of golden eagles in North West Scotland in 1967. British Birds 62: 345-363. : Brunner H & Corman B J 1974. The identification of mammalian hair. \nkata Press, Melbourne. Day MG 1966. Identification of hair and feather remains inthe gut and faeces of stoats and weasels. Journal of Zoology 148: 201-217. Fowler J & Cohen L 1990. Practical statistics for field biology. John Wiley & Son, Chichester. Ferrer M 1994. Nutritional condition of Spanish Imperial Eagle nestlings Aquila adalberti. Bird Study 41: 120-123. Green RE 1996. The status of the golden eagle in Britain in 1992. Bird Study 43: 20-27. Murray RD 1992-96. (ed) Scottish Bird Reports 1990, 1991, 1992, 1993, 1994. Scottish Ornithologists’ Club, Edinburgh. Newton I, Marquiss M & Village A 1983. Weights, breeding and survival in European sparrowhawks Auk 100: 344-354. Teerink BJ 1991. Hair of West European mammals: atlas and identification key. Cambridge University Press, Cambridge WatsonJ &LangslowDR 1989. Can food supply explain variation in nesting density and breeding success amongst golden eagles Aquila chrysaetos? in Mayberg B -U & Chancellar R D (eds) Raptors in the modern world. Proceedings of the II World Conference on Birds of Prey and Wwils. WWGBP, Berlin. Watson J, Rae S E & Stillman R 1992. Nesting density and breeding success of Golden Eagles in relation to food supply in Scotland. Journal of Animal Ecology 61: 543-550. Yalden DW 1977. The identification of remains in owl pellets. Mammal Society, Reading. Alastair C Pout, c/o Department of Zoology, University of Aberdeen, Tillydrone Avenue, Aberdeen AB9 2TN Revised manuscript accepted January 1998 _ Ahr Zr Golden Eagle Mb Ulfir Wil naite . Yaw ty, ~~ a4 Steven Brown 244 Scottish Birds (1998) 19: 244-247 SHORT NOTES Inland Common Gull coloniesin north east Scotland Tasker et al (Scottish Birds 16: 106-112) reported very big gull colonies on north east Scotland moorland. Here | give past data that may throw some light on this. In 1943- 48 fairly big Common Gull Larus canus colonies (> 100 pairs) were on Monaughty near Pluscarden, Wishach Hill (also >100 pairs of Lesser Black-backed Gulls Larus fuscus), Aultmore, Ordiequhish, Hill of Towie, The Balloch (>1000 pairs), and Fetteresso, and small ones south of Cuminestown on Greenness Hill (20 pairs in 1945) and Waggle Hill. At Waggle’s south end in 1945, 14 pairs nested and a pair of Great Black-backed Gulls Larus major. Atits north end, hundreds of Black-headed Gulls Larus ridibundus nested, many Common and Lesser Black- backed and a few Great Black-backed. The Balloch held thousands, mostly Common Gulls but many Lesser Black-backed and some Herring Gulls. Ordiequish had a few Great Black-backed pairs also. These sites lay on heather moorland that was later destroyed by tree planting on Forestry Commission land or on private land using FC methods. Since 1970 this happened to much of the Corrennie and Suie colonies and all the SB 19(4) Cairn Mon Earn and Spyhill ones. Birds moving from these to the increasingly few moors near the main farmland tracts may have contributed to the current big colonies. At The Balloch, Edward Bruce (pers comm) said Common Gulls deserted the area within a few years of planting, but Lesser Black- backed and Herring Gulls continued to nest until tree branches met above the nests. Small areas remained unplanted at The Balloch and Hill of Towie, but gulls deserted them once the surrounding trees were >1m high. At a Mulben peat bog with stunted pines he found several Common Gulls nesting on top of the dense tree foliage. James Allan (pers comm) saw such a nest at Kellas moor by Elgin. Foxes Vulpes vulpes were absent in the 1940s in Buchan, lower Deveron and lower Don and colonised low farmland in Banff, Aberdeen and Kincardine counties only in the 1950s (Nethersole-Thompson & Watson 1974, The Cairngorms). This applied also to low Moray including Pluscarden. Small gull colonies are probably more vulnerable to Foxes than big ones. In 1945 crofters marked nests at Waggle with smallcairns. They saidthe moor was ‘moving with gulls’ in the 1930s but wartime shortages led to high egg prices and the colony declined. They took each new egg about every 3 days and fresh eggs were laid into July. Adam Watson, c/o Institute of Terrestrial Ecology, Banchory, AB31 4BY Accepted January 1998 i) Scottish Birds (1998) Hunting associations between Merlins and Hen Harriers Allthe reported cases of hunting associations between Merlins Falco columbarius and Hen Harriers Circus cyaneusin winter have usually involved single birds of either species (Dickson 1993 Scottish Birds 17:58-59). On 6 January 1996 at 1430 GMT on farmland in west Galloway, | watched a female or juvenile Merlin hunting 500+ Skylarks A/auda arvensis and 1000+ Linnets Carduelis cannabina without success. A ringtail Hen Harrier then hunted these same flocks but the Merlin did not take part. Thirty minutes later the harrier returned and again hunted Short Notes 245 the passerine flocks. This time the Merlin took part and, at the same time, another, smaller brown Merlin appeared. Both Merlins then attacked the passerines which were flushed by the harrier, swooping and stooping on them together in a quite spectacular manner for the next 4 minutes. The harrier landed and the Merlins began chasing each other and both flew out of sight, the larger Merlin leading. Eight minutes later the harrier resumed its hunting alone. On only 1.4% of hunts | have watched in Galloway have 2 Merlins hunted together (Dickson 1995, Scottish Birds 18:165-169). The occasion where 2 Merlins hunted together in association with a hunting harrier has not been recorded before. Fi C Dickson, Lismore, New Luce, Newton Stewart, Wigtownshire DG&8 OAJ Accepted January 1998 X\ a bay Skylark 4 AC Ft IRR ee = Andy Dowell 246 Short Notes Double brooding by Lapwings In 1997 apair of Lapwings Vanellus vanellus nested on an island in a freshwater pool at a nature reserve in East Kilbride, South Lanarkshire. Atleast 3 chicks hatched on 3 May. Between 11 and 14 May one of the chicks disappeared and by 21 May only one chick remained; however, this bird fledged successfully and was seen flying on 6 June. In late May and early June the male of the pair had been noted sitting in the same area of the Original nest, but was not thought to be incubating asecond clutch of eggs. However, on 13 June 4 chicks less than 24 hours old were observed. Of this second brood, only a single chick remained on 24 June, with no Lapwings at all present on 28 June - the adults either moved away with the chick, or abandoned the site following its death. BWP (vol 3) states the Lapwing is single brooded but replacement clutches can be laid after egg loss and that field experiments on the species have revealed that up to 4 replacement clutches can be laid following egg removal. Given a mean incubation SB 19(4) period of 21+/- 1.77 days (n= 96) from clutch completion in a Hampshire study (mean of 28.1 days (24.7 - 34.0), n= 41, including the egg-laying period, from a Scottish study) and clutch completion in 5 days - most females lay daily (BWP vol. 3), then the second clutch was almost certainly started between 17 and 20 May, when there was at least one chick still present from the first breeding event. In the Lapwing, laying of replacement clutches occurs 5-12 days after the loss of a previous clutch, but it can also be triggered by loss of chicks, at least early in the season (BWP vol. 3). The Lapwing nest site was approximately 30m from a frequently used public path, and the site was carefully monitored at least twice a week. No other adult Lapwings were observed anywhere in the area from 9 March to 28 July. The East Kilbride Lapwings are of interest, therefore, in that a second breeding attempt followed a successful first one when there was no complete loss of clutch or brood, and did not involve a replacement clutch, but rather a genuine second clutch. Andy Robinson, Seasonal Countryside Ranger, c/o South Lanarkshire Council Countryside Ranger Service, Calderglen Country Park, Strathaven Road, East Kilbride, G75 0QZ. Revised manuscript accepted February 1998 Lapwing Barry Larking Scottish Birds (1998) Cormorant calls at a winter roost Cormorants Phalacrocorax carbo on nesting territories are well known to show a loud and complex vocal repertoire (BWP Vol 1), yet away from this situation authorities consider that this species is usually silent. My experience is that this is certainly true of birds, singly or in groups, commuting to feeding grounds or actually fishing, but they can be very noisy at roost. For some years | have studied a large roost in the upper Forth estuary just above Alloa, the site being 20 pairs of piers of a dismantled railway bridge. Most of these piers consist of round brick pillars, about 10m high, and the birds mainly roost (a) on the flat tops of the pillars plus top connecting struts, (b) the middle horizontal struts plus the narrow ledges formed by metal bands around the brick pillars connected to these struts. My observations refer to watches made just before sunset in autumn and winter, when up to 200 birds assemble with 15 to 20 on the more favoured central piers. | cannnot recall a single watch which has not been enlivened by anumber of loud outbursts of calling. These are usually stimulated by an incoming bird flying onto a site already occupied by 5 or more birds, but occasionally when a squabble breaks out among some that are already settled. Often a bird, either an incomer or a ‘resident’, is displaced in such encounters and it then flies round to make an attempt to resettle, either immediately or after resting on the water. The disturbance is thus often intensified. Two main types of call can be distinguished, though many intermediates occur. The first is ashortseries of low pitched, hoarse bellows or roars, describable as ‘ gok-gok-gok’, that | Short Notes 247 would equate with the threat call described by van Tets and mentioned in BWP. The secondis amore rapid series of higher pitched barks, best rendered as ‘ag-ag-ag-ag-ag’; this could be similar (as a sound) to any of 4 of the intra pair calls in BWP, but | cannot see a clear relation. The roost encounters are also characterised by bill lunging and fencing (typical of territorial aggression at nest sites) (BWP) and also by the gaping display (bill up and tail up) figured in BWP and taken as a breeding season greeting ceremony that is accompanied by a loud call (not described in detail). There is no reason to suppose any greeting between the sexes when the gaping display occurs in winter roosts. In my observations the birds are some hundred of metres distant; this and the close grouping at a roost preclude a detailed study of any association between particular calls, displays and contexts. M Martin and E D Cameron have watched 2 inland roosts in trees in Perthshire, each sometimes holding 50 birds, and tell me that they have also heard loud calling in broadly similar situations during both daytime assembly (MM) and morning departure (EDC). Loud calling is a constant feature of aggressive interactions at the Alloa roost. However, the occasional midday roosts occurring on mud banks lower down the river are quite silent even though birds are often close together. The most likely explanation is that roosting for the night involves some aspect of territoriality. There may well be some individual preferences for particular roost sites but, without marked birds, my only evidence comes from a very distinctive juvenile (with pure white underparts and a blackish cap) that was seen twice and occupied the same site on 24 January and 14 February 1997. C J Henty, University of Stirling, Stirling FK9 4LA Accepted March 1998 248 Advice to contributors NN SS 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 will be 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 happy to accept papers on computer discs, however please state the type of word processing programme SB 19(4) used. Contact Sylvia Laing on 0131 556 6042 if you wish further information on this. 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 Vol 17:146- 159). 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 be treated as short notes, where all references should be incorporated into the text, and notlisted 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 should be on a separate sheet. Maps — and diagrams should be either good quality | computer print outs in black and white (please | do not use greyscale shading) or in black ink — and be camera ready, but drawn so as to | permit reduction from their original size. NEOTROPICAL BIRD CLUB Neotropical bird club launched A club has been launched to promote the study and conservation of the birds of the Neotropics (South America, Central America and the Caribbean). It is currently seeking founder members to help reach the launch budget of £2000, which is required to get the club running and to publish the two first issues of its intended journal ‘Continga’. Founder members will be asked to pay a minimum of £25, and will be formally acknowledged in the first issue of ‘Continga’. ‘Continga’ will provide a colourful and much needed forum for exchange of information on the avifauna of this extremely rich and diverse area, and will contain papers and features on the birds and their conservation as well as news of recent observations and discoveries (at present, new species are still being discovered at the rate of more than two a year). It is hoped that in due course the club will be able to provide direct funding and support for practical conservation programmes. For further details and membership forms, please contact: Rob Williams, Publicity Officer, Neotropical Bird Club, c/o The Lodge, Sandy, Bedfordshire SG19 2DL Scottish Birds June 1998 Part 4 Volume 19 Contents Main papers Minimal numbers and habitat of breeding Dunlin in north east Scotland. R Rae & A Watson 185 Changes in breeding bird populations in peatlands and young forestry in north east Sutherland and Caithness between 1988 and 1995. M Hancock & M Avery 195 Numbers of wintering seaducks, divers and grebes in the Moray Firth, 1977-1995. RJ Evans 206 Feeding studies of the Lesser Whitethroat in Strathclyde. T Byars & D J Curtis 223 The breeding bird community of upland Juniper scrub in eastern | Scotland. S Gillings & RJ Fuller 231 Seasonal differences in pellet remains from Golden Eagles in the Isle of Harris. A C Pout ; 239 Short notes : Inland Common Gull colonies in north east Scotland. Adam Watson 244 Hunting associations between Merlins and Hen Harriers. RC Dickson 245 Double brooding by Lapwings. Andy Robinson 246 Cormorant calls at a winter roost. C J Henty 247 Advice to contributors 248 Published by the Scottish Ornithologists’ Club, 21 Regent Terrace, Edinburgh EH7 5BT. © 1998 ASMA i | Nay it FEB OF iy ~L/8Raries Scottish Birds The Journal of the Scottish Ornithologists’ Club Editor: DrS da Prato Assisted by: Professor D Jenkins, Dr J B Nelson, Dr | Bainbridge, Dr M Marquiss and R Swann Business Editor: The Secretary SOC, 21 Regent Terrace Edinburgh EH7 5BT (tel 0131-556 6042, fax 0131 558 9947). Scottish Birds, the official journal of the Scottish Ornithologists’ Club, publishes original material relating to ornithology in Scotland. Papers and notes should be sent to The Editor, Scoitish Birds, 21 Regent Terrace, Edinburgh EH7 5BT. Two issues of Scottish Birds are published each year, in June and in December. Scottish Birds is issued free to members of the Scottish Ornithologists’ Club, who also receive the quarterly newsletter Scottish Bird News, the annual Scottish Bird Report and the annual Raptor round up. These are available to non- members at a subscription rate (1997) of £36. 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 21 Regent Terrace, Edinburgh EH7 5BT is the most comprehensive ornithological library in Scotland, and is available for reference during office hours (Monday to Friday, 0930-1630 hrs). (But please phone beforehand). SOC annual membership subscription rates Adult £18.00 Family (2 adults and any children under 18) tiving at the same address £27.00 Junior (under 18, or student under 25) £7.00 Pensioner/unwaged £10.00 Pensioner Family (2 adults living at the same address) £14.50 Life £360.00 Life Family £540.00 All subscriptions may be paid by Banker’s Order and Covenanted. Subscriptions paid this way will greatly assist the Club. Published by the Scottish Ornithologists’ Club, 21 Regent Terrace, Edinburgh EH7 5BT Printed by Douglas Gauld & Co (Tayside) Limited, Shanwell Road, Tayport, Fife DD6 9EA Scottish Birds (1998) 19: 249-258 249 North east Scottish counts of Goldeneye, Goosander, Red- breasted Merganser and Cormorant in 1944-50 compared with 1988-97 A WATSON, M MARQUISS & P J COSGROVE Unpublished counts of Goldeneyes, sawbill ducks and Cormorants are presented from part of the River Deveron and other sites in north east Scotland in 1944-50 and 1997, and compared with published data from 1988- 90 and 1996. The Deveron held many Goldeneyes in the 1940s, and also many other waterfowl, especially when lochs were frozen. The proportion of adult drake Goldeneyes on estuaries, at sea, and on the coastal Loch of Sirathbeg exceeded that on inland waters. The numbers of Goldeneyes and Goosanders on the Deveron were higher during periods of hard frost. Adult drake Goosanders formed a minority on the Deveron and on Haddo House lakes, but a majority on Loch of Strathbeg and on Lochs Davan and Kinord. Adult drake Mergansers were in a minority on all sites. Counts from the Don estuary in 1989-90 showed no material change in Goldeneye numbers or the proportion of adult drakes since 1944-50. Counts in 1996 and 1997 show more Goldeneyes and Goosanders than on the same parts of the River Deveron in 1944-50, fewer Mergansers, and few Cormorants in both periods. Many sawbills and Cormorants were culled on the Deveron in both periods. Introduction Counts of wintering ducks in Britain in 1966- 92 showed that most species increased over this period (Kirby eta/1995), but these counts were mainly at standing waters. Recent papers documenting counts in north east Scotland (Duncan & Marquiss 1993, Marquiss & Duncan 1994a, Cosgrove 1996. 1997) showed the importance of rivers for some birds such as Goldeneye Bucephala clangula, Goosander Mergus merganser, and Red- breasted Merganser Mergus serrator, yet few data are available for such sites. Here we document previously unpublished counts in 1944-50, to compare with 1988-97 data. Cosgrove (1996) reported his January 1996 count of Goldeneye on all of the main stem of the River Deveron, and noted that there appeared to be no previous published data on Goldeneye there. This led AW to inspect field notes for his many counts on certain stretches of the Deveron in 1944-50, and to visit these same stretches again in January 1997. Also, Cosgrove (1997) reported his 1996 count of Goosander, Red-breasted Merganser and Cormorant Phalacrocorax carbo done simultaneously with his Goldeneye count. Although PC’s summarised published data cannot be related to AW’s data because PC used different subdivisions of the river, PC had map references for each 250 A Watson, M Marquiss & P J Cosgrove 1996 sighting in his field notes. This enabled us to obtain his counts on the same stretches used earlier by AW. Similarly, much of MM’s data in 1988-90 has been published (Duncan & Marquiss 1993, Marquiss & Duncan 1993, 1994a and b), but we returned to the unpublished field notes to compare their counts on the same stretches covered by AW in earlier years. The aim of this paper was to find whether the numbers of these waterfowl changed between the 2 time periods, so we compared AW’s 1944-50 counts a) on certain stretches of the lower Deveron with PC’s single mid January 1996 count there and 2 counts by AW in late January 1997, and b) on other waters with MM’s data in 1987-94. Methods Nearly all Goldeneyes in north east Scotland are winter visitors arriving in October and leaving in March, but Goosanders. Mergansers and Cormorants occur all year round (Buckland ef a/ 1990). However, AW saw no Mergansers on the Deveron from April to August in 1944-50. AW’s counts from September till April were made using a telescope from concealed vantage points such as drystane dykes and shooters’ hides, and binoculars where birds were confiding (eg Deveron Bridge by Turriff, Banff Bridge and Bridge of Don). Most birds remained undisturbed. This minimised errors that arise when flushed birds alight unseen on water not yet scanned. He told adult drakes from young drakes and ducks by their different plumage. SB 19(5) Results Goldeneyes in 1944-50 Seasonal variation Early dates for AW were 3 at Meikle Loch of Slains on 28 September and 3 at Loch of Strathbeg on 5 October, but an occasional bird summered on loch and river, including one with a Scaup Aythya marila on Deveron above Banffin 1947. Birds on the Deveron by Turriff usually came in late October, increasing till late December irrespective of frost, eg in 1946 the first seen on a 1-km stretch were 2 on 25 October, rising to 5 on 28 October, and 9 on 5 November and 18 December. In 1947, however, none came until 10 November. He saw none there after 27 March, and a last flock of 12 on the Ythan estuary on 14 April 1947. Duncan & Marquiss (1993) found similar arrival and departure times for Goldeneye on 3 other rivers and on standing waters in north east Scotland. Counts during icy periods Numbers on the Deveron rose after ice sealed lochs and fell after thaws. On the Turriff stretch, numbers at 1.0-4.7/km rose to 18.4/ km during frost in January 1945 and 19.4/km in January-March 1947 (blggest flocks 15 and 28 respectively). After thaws in February 1945 and late March 1947, numbers fell to 4/ km. Many in a flock showed sexual display on 1 February 1947, a sunny frosty day. Flock size The biggest flocks were of 25 on the Ythan estuary, 28 at Turriff, 33 at Boyndie Bay, 43 at Don mouth, 60 in Aberdeen Bay, and once 400 at Loch of Strathbeg when flocks up to 50 combined temporarily after disturbance by Scottish Birds (1998) shooting, but AW saw a flock of 100 on a day at Strathbeg with no shooters. Proportion of adult drakes The mean percentage of adult drakes on the Deveron at Turriff was 28%. It fell from February to March 1946 (36% to 27% in 28 North east Scottish counts of Goldeneye, Goosander etc Zoi and 15 seen), and from January to February 1945 (50% to 25% in 16 and 12 seen) and 1948 (63% to 54% in 16 and 13 seen), but rose from January through February to March 1947 (21% through 24% to 45% in 24. 89, and 76 seen). Variation between years was too big and sample size too lacking in independence within and between months for statistical tests to be valid. Table 1 Percentage of adult drake Goldeneyes seen in 1944-50. River Deveron, Turriff Deveron, Eden-Ashogle Deveron, Alvah Don, Kinaldie Don, Br. of Dyce Loy Alvie of Strathbeg Meikle Loch of Slains Cotehill Loch Haddo, Upper & Kelly of Fyvie Davan & Kinord of Skene Spey Deveron Ugie Ythan Don Montrose Basin Findhorn Spey Bay Whitehills Boyndie Bay Foveran Links Aberdeen Bay Loch Estuary Sea N(day n)* % 405 (44) 28 108 (18) 21 35 (3) 40 10 (10) 20 34 (32) 29 7 (7) 0 4 (4) 50 c770 (c400) 60 96 (39) 36 16 (8) 63 18 (18) Wz 12 (3) 0 38 (31) 26 18 (18) 0 17 (8) S15) 24(25)) 58 12 (12) 25 94 (38) Si 213 (43) 82 13 (12) 23 8 (4) 50 10 (4) 40 38 (4) 82 66 (33) 100 72 (40) 46 123 (30) us * N is total seen, and (day n) is biggest number seen in a day at one place. “From a sample flock of 50. 252 A Watson, M Marquiss & P J Cosgrove Some seasonal variation may be confounded with frost hard enough to seal lochs and flood ponds. The percentage of adult drakes in January 1945 during frost (50% of 52 seen) exceeded that on ice free days before the frost (44% of 9 sightings) and after (21% of 14 seen). The percentages in hard frost in February-March 1947(24% of 89 seen and 31% of 103) exceeded those before the frost (22% of 27 seen), and after it in late March (25% of 8). Similarly, in frost during January 1948 the percentage exceeded that in ice free February but AW’s data from December 1947 were too few for a pre January check. However, the 1946 data came from months without hard frost. The percentage of adult drakes was high on estuaries (Table 1), especially on the Don. On one occasion AW saw birds fly from Donmouth on to the sea off the sandbar nearby, so they used both estuary and sea. The percentage SB 19(5) of adult drakes was very high at sea, up to 100% on Boyndie Bay, Banff. On lochs near the sea it exceeded the percentage on inland lochs, especially at Loch of Strathbeg which is hard by the sea. On inland lochs it broadly resembled that on rivers, but varied considerably between one inland loch and another. It would be invalid to apply statistical analysis and probabilities to the totals in Table 1, as they probably involved some birds seen more than once, especially at Turriff where AW did many counts on the same stretch even within a month, sometimes almost daily. However, itwas valid to test numbers at different places on the same day (he often saw birds at the first site as he returned from the second), and at widely different places visited within two days (Table 2). This showed that the main differences in the percentage of adult drakes noted above at different places were very unlikely to be random. Table 2 Percentage of adult drake Goldeneyes at pairs of places on the same date or similar dates Date Place 17 Feb 46 Haddo House lakes Foveran Links, sea 2 Mar 46 Don, Dyce/Kinaldie Aberdeen Bay, sea 1 Dec 46 Boyndie Bay, Banff, sea Deveron, Turriff 14-15 Dec 46 Don mouth Don, Kinaldie 19 Jan 47 Deveron estuary Whitehills, sea 25 Jan 48 Loch of Strathbeg Meikle Loch of Slains * number counted % n* Fisher exact P 17 18 Sih 30 0.014 28 65 M5 60 <0.0001 100 33 25 12 <0.0001 69 26 20 10 0.011 60 25 100 12 0.015 60 50 32 25 0.029 Scottish Birds (1998) Goosanders in 1944-50 The earliest were on the Deveron by Turriff on 7 October and Meikle Loch of Slains on 14 October, and the last small groups on 20 March by Turriff. AW saw an occasional pair as late as 26 April, so perhaps they bred there. In winter, he often saw Goosanders on Deveron at Turriff and on other parts of Deveron that he visited frequently between Laithers and Banff Bridge. Counts during icy periods As with Goldeneyes, numbers on the Deveron by Turriff were much higher when ice sealed lochs and ponds. During such periods, the total number seen was 98, with 4.4/km, whereas during mild ice free periods the total was 61, with only 1.6/km. As successive counts cannot be regarded as statistically independent, formal analysis with probabilities would be invalid. However, out of 45 counts in mild periods, 69% had no birds, 18% one bird, 7% 2, and 7% 3 or more (up to 15) birds. In contrast, out of 22 counts in periods of frost, only 14% had no birds, 18% one bird, 27% 2, and 41% 3 or more (up to 30) birds. On the Deveron by Turriff they sometimes flew in long strings far higher than usual (> 50 mup). Thiswas normal on days with shooting, but at times occurred on days without shooting, when ice had sealed lochs and flood ponds. For instance, AW saw 30 in strings flying up and down the Deveron on 23 January 1947, by which time lochs had 5 cm of ice. On 12 January 1947 he saw 52 ona3 km stretch (12 on the water and at least 40 flying up and down), when lochs and ponds in lowland north east Scotland were open but hard frost affected England and south Scotland. He noted this behaviour occasionally on days that were mild for the North east Scottish counts of Goldeneye, Goosander etc 253 whole UK, eg 15 on 29 February 1948. Shooting elsewhere on the Deveron can be ruled out as an explanation, as this was a Sunday and no shooting was done on Sundays in 1944-50. Proportion of adult drakes Adult drakes comprised a majority at Loch of Strathbeg (64% out of a total of 42, with 15/20 in acount selected at random), and at Lochs Davan and Kinord (67% out of a total of 24, with 6/7 ina count selected at random). They were in a Slight minority in most places. They comprised 42.5% out of a total of 254 seen on the lower Deveron from Laithers down to Alvah gorge (with 4/12 in a count selected at random from the Deveron stretch by Turriff), 18% in a total of 39 at Haddo House lakes (with 4/30 selected at random), and 30% out of a total of 37 at other places. A Fisher exact test showed that the difference in the proportion of adult drakes between the randomly selected Davan/Kinord count and the Haddo one was highly significant (P = 0.0001), andforthe randomly selected counts at Strathbeg and Deveron by Turriff was significant (P = 0.03). On the Deveron at Turriff, adult drakes comprised 45% of a total of 98 seen during frosty periods in 1945, 1946 and 1947 when ice sealed lochs and ponds. This included 45% out of 87 during the hard 1947 frost, when the largest flock of 30 birds contained 12 adultdrakes. The proportion of adult drakes there did not differ materially during mild periods, with 44% out of 61. The biggest flocks seen were of 30 on the Deveron at Turriff, 30 on the Upper Lake at Haddo House, and 20 at Loch of Strathbeg. AW twice saw a few fly downstream to rest on the lowest parts of the Ythan estuary including 254 A Watson, M Marquiss & P J Cosgrove the north bar and once a bird at sea just off Donmouth, but otherwise he saw none at sea. Some were in close pairs away from others nearby, in all winter months but especially in late winter. On 25 January 1947, 8 drakes and 4 ducks in a flock showed much chase and courtship at Alvah in hard frost. Red-breasted Mergansers in 1944-50 AW often saw these at sea, on estuaries, on lochs near the coast, and occasionally on the lower Deveron (up to 7 at a time on the estuary) and Haddo lakes (up to 5). The sighting furthest up Deveron was a duck below the bridge by Turriff. Flocks were seen at sea (up to 26 at Boyndie Bay), and largest numbers in September-October at sea and on the Ythan estuary. Many were in close pairs, as early as 6 November on Spey. The last seen were 2 pairs on 4 May at Loch of Strathbeg and 4 pairs on 7 May at Ythan estuary, maybe on passage. Adult drakes comprised a slight minority in December-May (46% out of 213). We regard the lower September-November value (13% out of 106) as false, due to moulting adult drakes being unrecognised then, especially when far out at sea. Dawn and dusk flights on the Deveron at Turriff in 1944-50 All duck species present flew upstream over Deveron Bridge at dusk, mostly after sunset, and particularly so on short midwinter days with poor light, as Marquiss & Duncan (1994a) noted for Goosanders arriving at their roosting site. The Bridge was a good watch point, as they flew close above it, within a few metres if one hid low. Goldeneyes, Goosanders and many others then turned south east in the SB 19(5) direction of Hatton Castle lochs and flood ponds on Turriff Haughs, where AW saw ducks fly in to roost on other dusks. However, a Goosander once came upstream on the dusk flight, landed on a quiet stretch above the Bridge, and stayed there until after dark. Ducks returned down the Deveron at dawn. Cormorants in 1944-50 During each winter in 1944-50, AW sawsingle birds on Deveron by Turriff (once 2 together), and up to 5 at atime at Alvah gorge. He often saw them flying but also fishing on the river, and basking on rocks and trees at the gorge. They were mainly adults, but some immature. On 18 January 1947 one swam on the Deveron above the A920 bridge at 150 m altitude, above Huntly. Shooting in 1944-50 In the 1940s, ducks other than sawbills were often shot for sport on the Deveron, occasionally including Goldeneyes (a shooter bagged 6 one evening at Forglen). Goosanders have long been shot. Every second Saturday during winter in 1944-50, many lairds and keepers had an onslaught in the afternoon, directed by the Deveron Fishery Board and aimed at killing sawbill ducks and Cormorants. Any that were missed at one spot might be shot further up or down and got no respite. The birds tended to fly directly above the river, and departed from the river line only across big bends, where shooters were placed. An immature drake Goosander injured by shooting could not fly on 29 February 1948, and PC saw an injured duck Goosander in 1996, so shooters did not adequately dispatch wounded birds. Scottish Birds (1998) Comparison of 1944-50 data with 1987-94 Duncan & Marquiss (1993) found that a high percentage of Goldeneyes on estuaries and at sea were adult drakes. AW’s data confirm this; clearly this feature is long standing. Figure 7 in Duncan & Marquiss (1993) shows a big variation in the percentage of adult drakes on different lochs, and AW’s data confirm this also. Marquiss & Duncan (1993) noted a September-October peak of Red-breasted Mergansers on the North Esk estuary but not on the river. AW’s data show a September- October peak also on the Ythan estuary and at sea. We compared AW’s data on the Don estuary with MM’s datain 1988-90 to check for change in numbers or proportions of adult drake Goldeneyes. The sole count in 1988 revealed an unusually high value of 118, but it was noted as atypical because it included many birds from off the nearby sea. The only counts in 1989 and 1990, one in each January, showed 17 and 20 Goldeneyes respectively, of which 14 and 13 were adult drakes (mean 73%). This signified no material change in numbers since 1944-50, when AW’s 7 counts ranged from 7 to 43 with a mean of 16. It indicated also no material change in the percentage of adult drakes which in AW’s counts had a mean of 82%. The proportion in a count selected randomly from AW’s data (18 adult drakes out of 24) did not differ significantly from that in a count selected randomly from MM’s data (13 out of 20, chir=0.33). On Lochs Davan and Kinord near Dinnet, AW’s 3counts in 1946, 1947 and 1948 showed 31, 7, and 0 Goldeneyes, little different from 5 counts run by MM in 1988, 1990, 1991, North east Scottish counts of Goldeneye, Goosander etc 255 1992, and 1994 with 33, 0,0,0 and 0. Adult drakes comprised 26% in AW’s counts and 36% in MM’s, and given the small sample sizes it is unsurprising that there was no significant difference (chi *=0.26). A preponderance of adult drake Goosanders was apparent in AW’s 3 counts at Lochs Davan and Kinord in 1946-48 (67% in a total of 18 birds). The larger number of counts in MM’s data at these 2 lochs confirms this preponderance, with 76% adult drakes in a total of 68 birds seen during single midwinter counts in each of 6 winters in 1987-94. The 2 sets of data suggest a possible increase (mean 1946-48 count 6, range 0-11, and 1987-94 mean 11.3, range 4-22), but there were insufficient counts for analysis. Marquiss & Duncan (1994b) found a preponderance of adult drakes on the lower parts of Dee also, unlike AW’s slight minority of adult drakes on the lower Deveron in 1944-50 Comparison of 1944-50 data with 1996 and 1997 As AW’s data from 1944-50 showed a bigger proportion of adult drake Goldeneyes on estuaries than inland, we checked this for the Deveron using PC’s 1996 field notes. These showed 61% adult drakes out of 28 birds on the tidal Deveron, but only 34% in the 273 further up (Fisher exact P = 0.006), so confirming the point. The 6I% was closely similar to AW’s 1944-50 mean of 58% there (Table 1) and AW’s 1997 value of 54%. PC’s proportion of adult drakes in January 1996 on the Turriff stretch covered by AW in 1944-50 was 29% out of 17, broadly similar to AW’s 1944-50 mean of 28%, and similar also to AW’s January 1997 value of 28% there. Hence the differences in proportion of adult drakes between sites were consistent and long standing. 256 A Watson, M Marquiss & P J Cosgrove Comparison was made with PC’s unpublished notes from the same stretches as AW covered (Table 4). This showed more Goldeneyes and Goosanders in 1996 than in 1944-50, fewer (no) Mergansers, and very few Cormorants. One must treat such comparisons with caution, as birds onastretch of a few km may not be restricted to it. This seems likely from the high variability in AW’s counts of Goldeneyes, Goosanders and Cormorants within months on the Turriff stretch (Table 3). Nevertheless, the higher values in 1996 on all 7 stretches covered in the 1940s are suggestive of an increase since 1950. The question arises whether this was a sustained increase or merely a short term peak, as a single recent winter count cannot distinguish between these alternatives To help elucidate this, AW did a count on 2 mild, ice free days in late January 1997. On 40 km covered in both years, Goldeneye numbers were very close to PC’s in 1996, with Goosanders slightly fewer; AW was told there SB 19(5) was some Goosander shooting in the week before his counts. On one of the 2 days chosen randomly, the proportion of adult drake Goldeneyes was higher at the estuary than inland (15/28 vs 20/92, Fisher exact P = 0.002) On every stretch that AW covered in 1944-50, Goldeneye numbers in 1997 exceeded those in 1944-50, in most cases exceeded AW’s highest count for the same stretches from the earlier years, and were closely similar to PC’s 1996 counts for the same stretches. Hence recent high Goldeneye counts on Deveron are probably due to a real increase. This fits the overall increase in Britain generally (Kirby eta/1995). Adult drakes comprised 47% of Goosanders seen on the entire Deveron in the 1996 count (Cosgrove 1997). This included a lower proportion on the section 10 km from the sea (Cosgrove’s Figure 2), thus fitting in broad terms the minority of adult drakes on the low sections near Turriff in AW’s 1944-50 data. AW’s 2 counts in 1997 again showed a Table 3 Day to day variation in counts on 2km of the Deveron by Turriff in 1944-48. Period Goldeneye Jan 44 0,0,2,6 Nov-Dec 44 0,2,0,0,0,0,0 Jan 45 mild OlOKS Jan 45 frost 16,23, 13 Feb 45 eC InZionl Jan 46 7-0)0;0/0)}05071 Feb 46 ASO A. 2, le Mar 46 9,0,6,4 Oct-Nov 46 2,5,3,9,0,0,0 Jan 47 Me ioy Piss) Jan-Mar 47 15,29,29,19,22, frost iepalaheciAclealors: Jan 48 frost 0,6,2,4,6 Feb 48 8,0,1,4,0;1 Goosander Cormorant O Oren 0,0,0,0 ON Z0+2,0,.0) 0,0,0,0,0,0,0 0,0,0,2 0,0,0,0 OF2N0 0,0,0 240):(0}(0).(0),(0)-(9). 0,0,0,0,0,0,0 2,1,1,0, 1,0, 1,0 - OHOLOG,OrGre (O)e22-(0)-{0)4(0)-c8)(6) 0:0, 0/0; 07050 00701 0,0,0,0 O04, 0;00)4 0,0,0,0,0,0,0 20)(0), 0,0,2,0 SOS 22.4: (OH(8){0} ah) (01, We Operorere 0,0,0,0,0 a aloteto 0,0,0,0,0 Vee eloLal(o 0,0,0,0,0,0 Scottish Birds (1998) North east Scottish counts of Goldeneye, Goosander etc 257 Table 4 January counts on ice free days on the Deveron in the 1940s, 1996 and 1997. Place Km Goldeneye 40s 96 97 Estuary IRON SeS) 26 120:5 Alvah gorge Oe 0's A ES) Bridge of Alvah 0.7 0.3 2 2 Eden-Ashogle 5.1 12.4 54 54 Turriff # ema Ore tae 25 Carnoustie OFSF uv ON7: 6) Si) Laithers 1 0.5 3 S) Total me ZOLo) NOs, sOs Goosander Merganser Cormorant 40 96 97 40s 40s 96 Ox 4 1.6 OO O18r eae 5 12 ISAO O50 1 0 ORI 5.70 GlSsigs 0 0 1 1 Arps 2 0.2 AO RO 04 2 1 0 ORO O-5gsr2 1 0 O30 O37e2459019 8). 2.8 1 1940s data are from counts on mild January days in 1944-48 (only one count in a given day at one place, number of counts ranged from 9 at the estuary to 26 at Turriff, and the mean number of counts at all 7 places was 12). The mean number of each bird species was calculated for each place in each year, and then an overall mean for that place for the five years, which is the value shown in the Table. 1996 and 1997 data cover several places in the same day. 1996 data from one count, 1997 data from 2 counts on 2 days, no Mergansers seen 1996 and 1997, and no Cormorants in 1997. # From 1km below Deveron Bridge to 0.7km above. For Goldeneye and Goosander separately, the probability of obtaining all 7 of the 1996 values greater than their 1940s paired mean values is 0.016 (sign test, on the hypothesis that one would expect as many increases as decreases in each 1996 value compared with its respective 1940s paired mean). The difference in the opposite direction for Merganser is not significant, as so few sites had any. minority of adult drakes on the lower Deveron centred on Turriff between Laithers and Alvah gorge, where a count selected randomly showed 44% out of 18, and at the estuary (25% out of 4). PC’s detailed field notes show 53% adult drakes out of 32 on these 2 stretches in 1996 but only 46% out of 39 when the stretch between the estuary and Alvah gorge was included. Cosgrove (1997) saw no Red-breasted Mergansers on Deveron, but in 1944-50 AW noted up to 7 at a time on the estuary, up to 4 at atime on deep pools at Alvah gorge, and once a duck by Turriff. In 1997 he saw no Merganser and no Cormorant on the same stretches as he covered in 1944-50. Discussion This paper’s main conclusions are that the River Deveron’s importance for wintering Goldeneyes and Goosanders is long standing, that Goldeneyes and Goosanders were more abundant in 1996 and 1997 than in 1944-50, and that the proportion of adult 258 A Watson, M Marquiss & P J Cosgrove SB 19(5) drakes in both species has not changed materially. The proportions of adult drake Goldeneyes at sea, at estuaries, and at the Loch of Strathbeg hard by the coast, exceed those on inland rivers and on most lochs. A possible explanation for the increase of Goldeneyes on the Deveron is eutrophication following more use of inorganic fertilizers on farmland, leading to more food for this duck. Another possibility is that coastal Goldeneyes declined after reduction of coastal sewage and industrial outfalls, as on coastal sites in the Moray Firth (Barrett & Barrett 1985), and may have moved to inland sites such as the Deveron. These and other possible explanations are speculative. John Edelsten’s many counts of Goldeneyes at sea between Banff Bridge and Whitehills in 1981-96 are of interest, as he found the main concentrations of them at the sewage outfalls at Banff harbour and from Ladysbridge Hospital. Numbers seen varied greatly within and between winters, the highest counts being 138 in January 1987, 222 in November 1993, and 121 in February 1994. On the Deveron in the 1940s, much shooting of sawbill ducks and Cormorants took place in winter, in addition to other shooting. Shooting of sawbill ducks is now under licence, and numbers reported to have been shot are known (Cosgrove 1997). We do not have a record of numbers shot in the 1940s. Despite all the killing, however, Goosanders were more numerous in 1996 and 1997 than in 1944-50, though Mergansers were scarcer, and Cormorants possibly so. Acknowledgements We thank J Edelsten for letting us use his unpublished counts of ducks, and K Duncan for help with counts in 1989-90. References Barrett J & Barrett C F 1985. Wintering Goldeneye in the Moray Firth. Scottish Birds 13: 241-249. Buckland S T, Bell M V & Picozzi N. 1990. The Birds of North-East Scotland. North-East Scotland Bird Club, Aberdeen. Cosgrove P J 1996. A winter survey of Goldeneyes on the River Deveron, north east Scotland. Scottish Birds 18: 242-246. Cosgrove P J 1997. A winter survey of sawbill ducks and Cormorants on the River Deveron, north east Scotland. Scottish Birds 19: 93-100. Duncan K & Marquiss M 1993. The sex/age ratio, diving behaviour and habitat use in Goldeneye Bucephala clangulawintering in northeast Scotland. Wildfowl 44: 111-120. Kirby J S, Salmon D G, Atkinson-Willes G | & Cranswick P A 1995. Index numbers for waterbird populations, Ill. Long-term trends in the abundance of wintering wildfowl in Great Britain, 1966/67- 1991/92. Journal of Applied Ecology 32: 536-551. 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 Diurnal activity patterns of Goosanders Mergus merganser on a Scottish river system. Wildfowl 45:209-221. Marquiss M & Duncan K 1994b. Seasonal switching between habitats and changes in abundance of Goosanders Mergus merganser within a Scottish river system. Wildfowl 45: 198-208 A Watson, M Marquiss, Institute of Terrestrial Ecology, Banchory, AB31 4BY P J Cosgrove, Culterty Field Station, Newburgh, Ellon AB4I OAA Revised manuscript accepted February 1998 Scottish Birds (1998) 19: 259-261 Amendments to the Scottish List Z5\9) R W FORRESTER for Scottish Birds Records Committee (incorporating an update to records of species recorded in Scotland on 5 or fewer occasions) The Scottish Birds Records Committee is responsible for maintaining the Scottish List first published in 1994 (Scottish Birds 17: 146-159). This is the second subsequent report of the Committee (the first being in Scottish Birds 18: 129-131), and contains 4 additions to and one deletion from the list. Also in 1996 (Scottish Birds 18: 132-143) SBRC detailed all acceptable records of species which had occurred in Scotland on 5 or fewer occasions. We also now update these records. Red-breasted Goose Brania ruficollis Adult Insh Marshes, Highland, 9-19 March 1994 (British Birds 89:485). Adult Vane Farm, then Tayside, now Perth & Kinross, 28 September 1994 (British Birds 90:461). 4". and 5" Scottish records. Lesser Scaup Aythya affinis Male, St John’s Loch and Loch Watten, Caithness, 1 February-10 March 1996 (British Birds 90:464). 4" Scottish record. Pallid Harrier Circus macrourus 2™ summer male Dunkadale area, Orkney, 18 April-27 June 1995, also 13 September (British Birds 89:496). 4" Scottish record. Black-winged Pratincole Gjlareola nordmanni Juvenile, Monikie, Angus & Dundee, 14-16 August, 1996. (British Birds 90:468). 3° Scottish record. Caspian Plover Charadrius asiaticus Female Skelberry, Shetland, 3-4 June, 1996 (British Birds 90:469). 2™ Scottish record. Terek Sandpiper Xenus cinereus Boddam, Shetland, 11-13 June 1995 (British Birds 89:503) | River Leven, Clyde, 22-30 September 1996 (British Birds 90:479) 5" and 6" Scottish records. Forster’s Tern Sterna forsteri 1st W Musselburgh area, Lothian intermittently from 16 December 1994 (British Birds 88:522) on coast from Portobello, Edinburgh to Aberlady Bay, but mainly Musselburgh, to 10 April 1995 (British Birds 89:507) and probably the same bird Ythan Estuary, Grampian 3 May-1 August, 1995 (British Birds 89:507). 2 Scottish record. 260 SBRC SB 19(5) Bridled Tern Sterna anaethetus Tiree, Strathclyde, 30 June-9 July 1994 (British Birds 89:507). 4 Scottish record. Eagle Owl Bubo bubo Remove from Category B Following an extensive review the BOU concluded that there was no evidence that Eagle Owl had occurred in a wild state in Britain and Ireland for over 200 years and removed the species from Category B of the British List (/bis 139:198). The species had been on the Scottish List as a result of 4 records during the period 1830-1883. No Scottish records are therefore now considered acceptable and the species is removed from Category B. Pallid Swift Apus pallidus Add to Category A Moribund, North Ronaldsay, Orkney 26 October 1996 (British Birds 90:490). 1*' Scottish record. There are now 5 species on the Scottish list. Only recorded as dead or dying. Blyth’s Pipit Anthus godlewskii Add to Category A ist W Fair Isle, Shetland, 31 October-4 November 1993 (British Birds 89:512). 1st Scottish record and 4" British record. Isabelline Wheatear Oenanthe isabellina The details for the 2° Scottish Record were shown as “20-21 September 1993, Whalsay, Shetland” (Scottish Birds 18:139). The year should read 1994. Hermit Thrush Catharus guttatus 1st W Fair Isle, Shetland, 19 October 1995 (British Birds 89:516). 2™ Scottish record. Swainson’s Thrush Catharus ustulatus South Uist, Outer Hebrides, 6 October 1996 (British Birds 90:497). 4'" Scottish record. Veery Catharus fuscescens Add to Category A Newton, North Uist, Outer Hebrides, 20-22 October, 1995 (British Birds 90:497). 1** Scottish record. Cetti’s Warbler Cettia cetti The details for the first Scottish record were shown as “5'" October 1993, freshly dead, Leith, Edinburgh”. (Scottish Birds 18:140). This should be changed to 4 October 1993. Olivaceous Warbler Hippolais pallida Fair Isle, Shetland, 5-13 June 1995 (trapped 5" June), (British Birds 89:517). 2™¢ Scottish record. Southern Grey Shrike Lanius meridionalis Add to Category A The BOU have now split Southern Grey Shrike from Great Grey Shrike Lanius excubitor (Ibis 139:199). The first acceptable record was of the subspecies L.m. pallidirostris (also known as Steppe Shrike), trapped Fair Isle, Shetland, 22 September 1964 (British Birds 66:401- 402). This is both the first Scottish and 1° British record. Tennessee Warbler Vermivora peregrina Hirta, St Kilda, Western Isles, 20 September 1995 (British Birds 89:524). 4" Scottish record (all 4 British records are from Scottish islands). Yellow-rumped Warbler Dendroica coronata 1st W North Ronaldsay, Orkney, 13 October, 1995 (British Birds 89:525). Scottish Birds (1998) 3 Scottish record. With the addition of Pallid Swift, Blyth’s Pipit, Veery and Southern Grey Shrike to Category A and removal of Eagle Owl from Category B, the totals for the various categories of the Scottish List now stand at: Category A 458 Category B 14 Category C rs) 478 Category D aal2 490 The British Ornithologists’ Union has recently separated Hume’s Warbler Phylloscopus humei from Yellow-browed Warbler Phylloscopus inornatus. Several records of birds resembling humei are currently being examined by the British Birds Rarities Committee and we await with interest to see whether they find that any Scottish records are acceptable. It now seems only a matter of time before the BOU separates Yellow- Amendments to Scottish List 261 legged Gull from Herring Gull Larus argentatus to enable yet another species to be added to the Scottish List. Also, we await acceptance by the British Birds Rarities Committee of the Blue-cheeked Bee-eater recorded in Shetland. A similar system of categories has been used for the Scottish List to that used for many years by the BOU for the British List. Early in 1998 the BOU announced the introduction of a revised categories to their British List. Although they have been working on the new system for some time, they have only now published the details. Consideration will be given to the merits of revising the categories used for the Scottish List, to keep it in line with the BOU’s British List. The Scottish Birds Records Committee, which is responsible for maintaining the Scottish List now consists of Peter Gordon, Eric Meek, Kevin Osborn, David Clugston, Bruce Forrester, lan Andrews, Colin Crooke and Ron Forrester (Secretary). Ronald W Forester, Secretary, Scottish Birds Records Committee, Se ; TST Se Caspian Plover 31 Argyle Terrace, Rothesay, Isle of Bute PA20 OBD Accepted April 1998 Steven Brown 262 Scottish Birds (1998) 19: 262-269 SB 19(5) Winter habitats of Twites in Scotland H CLARK & R M SELLERS This paper presents the results of a questionnaire survey of habitats used by Twites in Scotland in the winter months. A variety of habitats was identified, chief among them weedy turnip fields, rape and rape stubbles, other stubbles especially barley, pasture and other farmland, waste ground, saltmarsh and beaches and the strand line. These are important both in terms of the number of areas in which they are used and the numbers of birds using them. Other habitats including pasture and other farmland, dunes and machair were also used. These results contrast with the position in England where saltmarsh is the principal habitat used. Conservation aspects of the results are discussed and attention is drawn to the declines in the acreage of turnips planted in some areas in recent years and the apparent switch to rape and rape stubbles, a new crop in Scotland. Introduction Scotland is one of the main strongholds in Western Europe of the Twite Carduelis flavirostris. l\ts breeding range extends ina broad but patchy band from south west Scotland through Argyll, the Western Isles and north west Scotland to Caithness and the Northern Isles with smaller numbers in the Grampians (Jardine & Reid 1993). There is a complex redistribution of birds following breeding (Clark & Sellers 1998) with the main wintering areas including south west Scotland, Argyll, south-west Perthshire, the southern part of the Outer Hebrides, certain islands of the Inner Hebrides, Orkney, Caithness and low lying areas adjoining the Moray Firth, with lesser numbers in the Borders, Lothian, Central and elsewhere (Lack 1986). The habitats used in Scotland in the non-breeding season are not well known, but, in contrast to the position in England, it is clear from the general winter distribution that both inland and coastal habitats are used. In an earlier study of Twites wintering in Caithness we found that the principal habitats used were weedy turnip fields, fields of rape and rape stubbles and noted that these agricultural habitats might be used more generally in Scotland (Clark & Sellers 1997a). Materials and methods The material for this study was obtained primarily from a questionnaire circulated to all local bird recorders in Scotland except Caithness which we completed on the basis of our own studies and interested birdwatchers. This solicited information on the following: (i) Habitats used in the period November to February inclusive: Recorders were asked to assign one of the following status categories: 0 Habitats not used in winter or none of this habitat in the area Scottish Birds (1998) 1. Occurs irregularly in winter in small numbers (flocks <30 birds) 2 Occurs irregularly in winter in moderate numbers (flocks <100 birds) 3 Occurs regularly in winter in small numbers (flocks generally <30 birds) 4 Occurs regularly in winter in moderate numbers (flocks generally <100 birds) 5 Occurs regularly in winter in good numbers (flocks of >100 birds not uncommon) to each of the following habitat types: turnip field, rape and rape stubbles, other stubbles, pasture, other farmland, waste ground, birch woodland, machair, dunes, beaches and the strand line, saltmarsh and other. Compilers were asked to use only one abundance category per habitat, and in cases of doubt to use the highest appropriate category. For the other stubbles, other farmland and other categories, recorders were asked to provide additional information as appropriate. This list of habitat types was drawn up primarily on the basis of experience in Caithness and elsewhere in Scotland and various published sources, but was not exhaustive and the notes accompanying the survey form asked for details of other habitats where these were used. (li) Foods: Recorders were asked to supply any information they could on foods and, if possible, to differentiate between principal food sources and minor ones. (iii) Area: Area to which the habitats and foods related (usually a Region or District). Further details about the survey are given in Clark & Sellers (1997b), which includes sample copies of the survey forms. We supplemented the information from the Winter habitats of Twites in Scotland 263 questionnaire with data from fieldwork carried out by ourselves, covering especially low lying land adjoining the Moray Firth, and published sources including local avifaunas and local bird reports from all parts of Scotland. We obtained sufficient information to determine habitats for all parts of the Twite’s winter range in Scotland except Wester Ross and W Inverness. However, neither of these appear to be important areas for Twites in the non breeding season (cf Jardine & Reid 1993). For the purposes of this study winter is taken to be the period November to February inclusive. Results Habitats Table 1 shows the habitats identified in this study. They comprise 2 main types: agricultural habitats and the maritime habitats. The former category included weedy turnip fields, rape and rape stubbles, other stubbles (mainly barley), fields of potatoes, kale and other cultivated land, pasture (usually weedy pasture), crofting land, cattle feed troughs and silage rings, and set aside. The maritime habitats category covered saltmarsh, beaches and the strand line, weedy areas backing onto beaches, dunes and maritime grassland which might also be included under agricultural habitats. A number of other habitats were also recorded as being used, most importantly weedy areas such as waste ground and roadside verges (typically adjoining pasture or other agricultural land) and to alesser extent others such as moorland and moorland edge, rough grassland and birch woodland. The latter seems surprising for a species usually 264 H Clark & RM Sellers SB 19(5) Figure 1: Distribution of major Twite winter habitats in Scotland Rape and fields Nr = & Occurs irregulary rape stubbles s/ | +} Beaches and strand line By Occurs regulary in small flocks ey Occurs regulary in medium to large flocks associated with open country, but parallels some observations of Smart (1978) of birds wintering in woodland in England. The information presented here is not quantitative and the areas in which habitats were recorded were not of uniform size. However a rough indication of the relative importance of the habitats can be judged from the number of areas from which they were recorded. The results of this are shown in the lower part of Table 1. The most widely used habitat was weedy turnip fields, followed by other stubbles, beaches and the strand line, saltmarsh, rape and rape stubbles, pasture, waste ground, dunes and other farmland. The geographical distribution of the 6 most important habitat types is shown in Figure 1. Saltmarsh appears to be primarily a feature of birds wintering on the shores of the Moray Firth, Firth of Forth, Argyll, Ayrshire and Dumfries and Galloway. Agricultural habitats have amore northerly bias. Although we did not specifically collect information on the Scottish Birds (1998) Winter habitats of Twites in Scotland 265 altitude at which habitats were situated, it is clear from Figure 1 that they were located in low lying areas relatively close to coasts, presumably because these are the areas least likely to suffer from frosts and prolonged snow cover. One or 2 of the survey forms returned noted a tendency for birds to disappear from agricultural habitats during periods of hard weather for instance in Caithness and Strathallan in SW Perthshire and to appear on the coasts at such times, for instance in Fife. Birds usually reappeared, very shortly after the weather improved and evidently didnot move far. These fragmentary observations do suggest, however, that habitats only occasionally used may sometimes be important for the birds’ survival in bad weather. Foods Table 2 presents a summary of the principal foods seen to be taken by Twites in Scotland in the winter months. However, many of the questionnaire returns noted a lack of information on foods, so Table 2 is undoubtedly incomplete. Several forms made reference to plants such as grasses, sedges, seaweed (cf Jardine 1992), Nettle Utica, Chickweed Cerastium and Thrift Armeria maritima, used in the period April-October, and we think it likely that some of these may also be used in winter. The foodplants listed are, however, typical of those recorded elsewhere in the Twite’s Western European range (eg Cramp & Perrins 1994), and reflect a diet based primarily on small seeds, especially those of the weeds of cultivation, and of various coastal plants. Discussion This study confirms the importance of agricultural habitats, especially turnip fields, rape and rape stubbles, other stubbles, pasture and other farmland as wintering areas for Twites in Scotland. With the exception of rape and rape stubbles, the common factor in these habitats is the presence of weeds. Saltmarsh, beaches and the strand line are also important, however, and others such as machair, dunes and waste ground are also used but in fewer places and mostly by only small numbers of birds. These findings are in contrast to the situation in England where saltmarsh is by far the most important habitat used in winter (Davies 1988) and we believe that they have important implications for the development of proper conservation measures for Twites in Scotland. Weedy turnip fields, in particular, are important in Scotland for a variety of other species including all such declining farmland birds as Grey Partridge Perdix perdix, Skylark Alauda arvensis, Song Thrush Turdus philomelos, Tree Sparrow Passer montanus, Linnet Carduelis cannabina, Reed Bunting Emberiza schoeniclus and Corn Bunting Miliaria calandra (M Hancock pers comm based on observations in NE Scotland, and our own unpublished data based on observations in Caithness, Sutherland and Easter Ross). We have noted elsewhere declines by a factor of about 8 in the acreage of turnips in Caithness over the past half century, and similar trends in Sutherland (Clark & Sellers 1997a). It seems likely that the associated weed flora has decreased by at least the same amount and possibly more, given the widespread use of herbicides in recent decades. In Caithness, our observations suggest that only about a quarter of the current acreage of turnips contains enough weeds to be suitable for Twites. We suspect that the trends in turnip acreages in other parts of Scotland are similar, and this is also likely to be true of all the other weedy habitats. SB 19(5) 266 HClark & RM Sellers G 10 7 JO souepuNgYy c G v | ¢ € Cc v G v Ol UIM SBSJE “ON 6 cl Gl 6 v 6 8 6 Gl Ol Zt SEOJE ON 4 L v Z SIIUS-SOJWNG dc v G € G € v Kemojje ov c 4 L 14 uG v G CZ, G aiysiAy € € SOBIQWNY ‘aing ‘uBLY € € € apAjoyyeds |eJUsD ut G v v iv v v 4D iv (a3Ng [oxe) ||ABuy 1€ € L € uv € e€ c v pindaquag ‘S\siN L € € € SIMS] 8 SLIEH 6 | L L sJapl0g rt v | v 2 | v UBIYIO7 174 Z Yo Jeuu] v G G G SIIYSUUSd MS L L L L L S}I4 € i | p€ | | L v snbuy | | € | S 4 PUE}}OIS 4N € IL € UJIEN 9g Aeloyy v C € € v € G SSOUJBAU| F “SSOY J | | | L C PUBLSYINS M8 N 9G | L v € v Pue|IeyINS J c € Cc G G SSSUYIIED i i | € G G KauxJO q | a|s| 4Ie4 e€ € PUue}eYyS oul| sajqqnis Spal} usJewW pues 9 punos6 puejwey sajqqnis ede diuin} JaujoO yes sayoeeaq seunp i Jleyoew ajsem Jaujo ainjsed Jauy}o 9 ode Apaam Bay puejjoos ul Bursajuim sayimy Aq pasn sadA} yeyqeH | JTEVL (UOWWWOOUN JOU SPIIG OO} < JO SyOO}J) SIaquUNU POOB UI 4ajUIM UI A[eEINBey sunddO (SPulq OOL> Ajjes9uab syooy) suequinu ayesopow ul Ja}UIM UI AjsejnBas sind09 (Spulg OE> Ajjesoueb Syoo|J) Ssaquunu |yeLUS Ul Ja}UIM Ul AjejnBes sindoQ (SP4Iq OOL>SyOO}}) SJaQquUNU ayeJOpOW Ul! Ja}UIM UL AjZejnBei! Ssind0O (Sp4lq OE> SyOO}}) SYequuNu |jeWS Ul JaJUIM Ul ALejNBeu! sind9O (eave ul yeyqey Si} JO BUOU JO) 49}UIM UI P9sn jou sjeligey Aue ON ‘seuobeyeo soUepUNgY Winter habitats of Twites in Scotland 267 ~-N OT WO sbui abejis pue suBnou} pas ajeo puncuy d syed yoeaq ul sses6 yous O Spj9l} OFE}0q U (py) pue| Bulyos9 pue (7) puejsses6 jeyseoo ‘(7) pale Wu seale Hulpaes ajyyeo YyIM payeloosse UusyO {9S}! odes ay} UeY) JoYVes odes JaPPO} OU} YM Payeloosse spaem UO pee} AjqeuNseid sellMm, ‘pees s}es 11 Q10J}9q paezes6 S| PUL YIO}s JO} Paaj Ja]UIM Se UMOIH SI odes JOPPO4 “ades P|~aS |10 0} 19}9J ULUNJOD Si4} UI SaldJUS JOJO |;e ‘edes 1apPO04 A\uo Aejs| Spue|s! payeyiqeyulun pue sapispeol ‘pueOOY\ gjqqnis abpe puejooy\ pue|JoOOW puke apise jas puejssei6 ybnoy SpISe]9S a6pe puejioow pue pue|poom YoIg pue|Joow pue puejyJoOID PUB] PEJEAN|ND J9YJO pue SPjsl} O}e}Oq x = — -— SQOo0T OY DOL S9]0U]OO} =| a1qe] Scottish Birds (1998) 268 HClark & RM Sellers SB 19(5) Table 2 Principal foods taken by Twites in winter in Scotland Habitat Turnip fields Principal foodplants etc Charlock Sinapis arvensis, Turnip Brassica rapa, Dock Rumex, weed seeds Rape and Rape stubble Rape Brassica napus, Charlock Sinapis arvensis,Dock Rumex, weed seeds (cf footnote k in Table 1) Other stubbles Barley, Oats, Linseed, Rye, Hay (all presumed to be food Plants) plus associated weeds including Dock Rumex Pasture Dock Rumex, Thistle Carduus, Knapweed Centaurea, Annual Meadowgrass Poa annua Other farmland Waste ground Weeds in fields of sprouts, potatoes and set-aside Dock Rumex, Thistle Carduus, Dandelion Taraxacum and other blown seed, weed seeds Machair weed seeds Dune systems Seeds of maritime flowers (Elliott 1989), Thistle Carduus, Marram Ammophila arenaria, Chenopodium, Black Knapweed Centaurea nigra, weed seeds Beaches Strand line (seeds and other detritus washed up by tide),Orache Atriplex, Groundsel Senecio vulgaris, Chenopodium, weed seeds Saltmarsh Other The main mitigating factor in Caithness has been the recent introduction of oil seed rape as a crop and the strong preference that Twites have for it. The results presented here show that this is not unique to Caithness and that Twites have taken to using rape or rape stubbles in several parts of Scotland. To what extent rape has a role as a Substitute for the weeds of turnip fields etc. remains to be resolved. Inthe short term the availability of subsidies for growing rape is likely to be the main factor determining the amount of rape that is grown; any reform of the EU Common Agricultural Policy should take into account the consequences for conservation of any changes in subsidies onthis crop. The role of coastal habitats, especially saltmarsh and dunes, as a possible insurance against Glasswort Salicornia, Sea Aster Aster tripolium Birch Betula, weed seeds the non availability of agricultural habitats in hard weather merits further investigation. Protection of the remaining small areas of saltmarsh in Scotland is also a high priority. Acknowledgements This study would not have been possible without the patience and dedication of local bird recorders and other birdwatchers throughout Scotland and we are grateful to them all for their help. We thank also lan Bainbridge and Mark Hancock of RSPB Scotland for their advice and support during the course of this work and their comments on an earlier draft of this report. Scottish Birds (1998) Winter habitats of Twites in Scotland 269 References Clark H & Sellers R M 1997a. Distribution and abundance of Twites wintering in Caithness. Scottish Birds 19: 1-9. Clark H & Sellers R M 1997b. Twite Winter Habitats in Scotland: Results of the 1997 Questionnaire Survey [unpublished report]. Clark H & Sellers RM 1998. Movements of Twites in Scotland. Scottish Birds 19: 270-279 Cramp S & Perrins C M (eds) 1944. The Birds of the Western Palearctic, Vol. V\Il Oxford University Press, Oxford. Davies M 1988. The importance of Britain’s Twites. RSPB Conservation Review 2: 91-94 Elliott RE 1989. Birds of Islay, Christopher Helm, London. Jardine DC 1992. Twites eating seaweed. British Birds 85: 619. Jardine J D & Reid J 1993. Twite in The New Atlas of Breeding Birds in Britain and Ireland: 1988-1991 (ed D W Gibbons, J B Reid & R A Chapman) Poyser, London, p.418. Lack P 1986. The Atlas of Wintering Birds in Britain and Ireland. Poyser, Calton. Smart JH 1978. Twites wintering in woodland. British Birds 71: 86. H Clark, 3 Lindsay Place, Wick, Caithness KW1 4PF R M Sellers, Rose Cottage, Ragnall Lane, Walkley Wood, Nailsworth, Gloucestershire GL6 ORU Revised manuscript accepted August 1998 FE Lif = oe eS \\ \y ne \ Wintering Twite Mike Ashley 270 Scottish Birds (1998) 19: 270-279 Movements of Twites in Scotland SB 19(5) H CLARK & RM SELLERS Scottish Twites undergo a complex set of movements between breeding and wintering quarters. Three migration routes, probably best described as partial migrations, are tentatively identified: (i) from Shetland through Fair Isle to Orkney probably extending south to Caithness, the Moray Firth and NE Scotland, (ii) from the NW Highlands and the Western Isles to Caithness, (iii) from the W Highlands through Argyll to the north coast of lreland. The wintering areas of birds breeding in the Grampians, and the origins of birds wintering in southern Scotland remain unknown. The breeding populations of Caithness, and possibly those of E Sutherland, Easter Ross, Orkney, the southern islands of the Outer Hebrides and the southern islands of the Inner Hebrides appear to be resident throughout the year. There is no evidence that significant numbers of the Scandinavian race regularly reach Scotland. Conservation aspects of the results are discussed and the need for further studies highlighted. Introduction The Twite Carduelis flavirostris has a restricted breeding distribution in Western Europe, ranging discontinuously from the west and north coasts of Ireland through northern England and Scotland to western Norway and the Kola peninsula in northern Russia (Cramp & Perrins 1994, Hagemeijer & Blair 1997). The Norwegian population migrates south in the autumn mainly across the Skagerrak and through Denmark to winter along the coasts of northern France, the Low Countries and Germany (Bernhoft-Osa 1965). The movements of Twites in Britain are less well known. The small breeding population of the Peak District and southern Pennines moves to winter around southern North Sea coasts (Davies 1988); Scottish and Irish Twites, on the other hand, have been described as largely sedentary (eg Jardine & Reid 1993). This paper describes an analysis of what is known about the movements undertaken by Twites in Scotland. Materials and Methods The analysis draws on 4 main sources of information: a comparison of breeding and non breeding distributions, field observations of migrating birds, seasonal variations in numbers, and ringing recoveries. The distribution of Twites in Britain and Ireland has been described as a result of surveys in 1968-72 (Sharrock 1976) and 1988-91 (Jardine & Reid 1993) for breeding birds, and in 1981-84 for wintering birds (Lack 1986). We supplemented these surveys with information obtained from a questionnaire | Scottish Birds (1998) circulated to all local bird recorders in Scotland plus other knowledgeable birdwatchers with an interest in Twites (Clark & Sellers 1997a). The questionnaire asked for information on (i) timing of occurrence (especially where there was an appreciable difference in the numbers present between the breeding and non breeding seasons) and (ii) visible migration. The survey also solicited information on habitats, foods, numbers and changes in status; an analysis of these is published concurrently. Responses were received from all parts of Scotland except Wester Ross and west Inverness. We also checked relevant local avifaunas and bird reports. Information on Twites in Caithness came mainly from our earlier study (Clark & Sellers 1997b) and subsequent observations, and from records published in Caithness Birds, formerly the Caithness Bird Report. Ringing recoveries for Twites marked in Scotland were kindly made available by the British Trust for Ornithology. The total number of recoveries available was 26; 12 recovered at the site of ringing in the same season provided no information on movements and were excluded from the analysis. Results Distributions in the breeding and non breeding seasons When the distributions in the breeding and non breeding seasons are superimposed (Figure 1), it is evident that only a few areas hold birds throughout the year, chief amongst them being the Northern Isles (principally Orkney), Caithness, E Sutherland, SE Ross, E Inverness, the southern islands of the Outer Hebrides (Uists, Benbecula), the southern islands of the Inner Hebrides (Coll, Tiree, Colonsay, Oronsay and Islay) and W Movements of Twites in Scotland 271 Galloway. By contrast a large part of the breeding range including Lewis, mostof Skye, Mull, the NW Highlands from Cape Wrath to the Firth of Lorn, and N Perthshire is largely devoid of Twites in winter. Areas where birds appear only in the non breeding season include the coasts of Ayrshire, E Galloway, Lanarkshire, the inner Solway Firth, SW Perthshire, the shores of the Moray Firth south of the Dornoch Firth, Angus, Fife, inner Firth of Forth, coastal Lothian and the E Grampians. Other observations Asummary of observations on the occurrence of migrating Twites in Scotland is given in Table 1. These show that in several of the areas identified as holding birds in both the breeding and non breeding seasons there is a substantial winter influx of birds. Included in this category are Caithness, E Sutherland, E Inverness, Islay and E Galloway. By contrast, there is a reduction in the numbers of birds in winter in Shetland. It appears that most movements occur in October-November and March-April. The most detailed information on the passage of Twites comes from Fair Isle and North Ronaldsay, where regular observations have been made over many years. At the former, the main migration periods are August- October with a much smaller passage in April and May (Dymond 1991), whilst on North Ronaldsay there is a small autumn passage from mid August to early November but little evidence of aspring passage (K Woodbridge, pers comm). Migrant flocks have also been noted on the Isle of May, and what appear to be migrant flocks have been noted on both sides of the Pentland Firth, in E Sutherland, in several places on the west coast (Lewis, Eigg, Mull, mainland Argyll, Oronsay, Islay 272 HClark& RM Sellers SB 19(5) Figure 1 Distribution of the Twite in Scotland in the breeding and non breeding seasons i il! cgyllll||| yal “th i WY il f + Dd HUT present in the breeding season SESR present in the non-breeding season fs) present all year and Arran) and the Gargunnock Hills (Stirlingshire). In addition, Twites are occasionally noted on oil and gas platforms in the North Sea, with the majority of such sightings Coming from those in the Forties field off Aberdeen, although there are a few from the oil rigs off Shetland (Table 1). A number of the replies to our questionnaire made reference to hard weather movements. For instance, Twites were reported as appearing in hard weather on the coasts of Fife and Ayrshire and as disappearing from SW Perthshire, perhaps connected with their appearance in Fife. We have noted similar behaviour in Caithness, with birds disappearing with the onset of snow cover and reappearing more or less immediately once a thaw was underway. We do not know where the birds went but evidently it was not far; it was probably nearby coasts, which at such times are likely to be the only places where Twites can find food. Coastal movements noted near Aberdeen and in Lothian may also be related to hard weather. Scottish Birds (1998) Movements of Twites in Scotland 273 Table 1 Summary of sightings of migrating Twites in Scotland Area Principal observations North Sea OilOccasionally noted on spring and autumn passage; most records from oil Platforms Shetland Fair Isle Orkney Sule Skerry Caithness NW Suther -land E Suther -land E Ross, E Inverness Moray & Nairn NE Scotland rigs in the Forties field ca150km off Aberdeen, a few from those between Shetland and Norway (Bourne et a/ 1978, North Sea Bird Reports). Less common in winter than in breeding season (K Osborn, pers. comm) Large numbers on passage Unst, 26 September and 4 December 1911 (Baxter & Rintoul 1953). Fairly common as a breeding species; winters occasionally in small num- bers (formerly wintered more regularly); distinct spring and autumn passage, April-May and August-October (Dymond 1991). Present all year, in strength late October — early May. Regular autumn passage noted on North Ronaldsay, but little evidence of spring passage (K Woodbridge, pers comm). Regular passage suspected, Pentland Skerries, 1880s (eg Harvie-Brown 1886). Large numbers on passage, Pentland Skerries, 12 October, 6 November and 6 December 1911 (Baxter & Rintoul 1953); flock of 40 on Pentland Skerries on 18 September 1976 may also have been on passage (Booth, Cuthbert & Reynolds 1984). “...most commonly seen in April and May, sometimes in large flocks, which remain for a few day. Small numbers appear in August and September and occasionally in November” (Clark 1912). Much more common in winter than in the breeding season though varying somewhat between years (Clark & Sellers 1997b); present in strength in October-March. One record of flock of 250 birds coming in off sea at Duncansby Head in October 1995 and one record on Stroma probably relate to migrants. Post-breeding flocks of up to 100 birds present mid August-September; irregular in winter months. Present throughout the year but more common in late October-March when confined to low ground. Small parties of birds seen moving west through Clynelish Valley, Brora, late February and early March. More common in the winter months, when occurs on low ground near coast. Flock of 40 flying south, possibly migrating, Buachaille Etive Beag, Lochaber, 15 September 1990 (Highland Bird Report 1990). A century ago good breeding and wintering numbers with apparent move ment from high ground to low ground and coast in winter. Numbers now much smaller but still mainly present on high ground in breeding season and low ground October-April (Cook 1992). Wintering numbers variable, but far exceed the breeding population in most years (Buckland et a/ 1990); higher numbers present early November — end February. On 24 December 1995 at Aberdeen Beach several hundred moving “along coast and swinging inland’; this exceptional record was thought to be the result of a hard weather movement. 274 H Clark & RM Sellers SB 19(5) Fife Winter visitor in small numbers (Smout 1986); occurrence perhaps associ- ated with hard weather (D E Dickson, pers comm). Single birds or small flocks seen from time to time on Isle of May mostly in October and March/ April, often with Scandinavian migrants such as Brambling, Black Redstart, Lapland Bunting and Snow Bunting (I M Darling, pers comm). Thom (1986) says that “Some immigration is also believed to take place, as Twites are sometimes seen arriving on the east coast in autumn and parties of up to 50 have been recorded on the Isle of May in both autumn and spring”; we have not been able to corroborate this statement. SW Perth Winter visitor in good numbers; present mid-October — early May, in -shire strength early November — end February Stirlingshire Winter visitor, present mid-October — early March, in strength & Inner NovemberFebruary. Flock of 25 flying west, Stronend, Gargunnock Hills, 25 Firth of Forth October 1976 presumed to be migrants. Lothian Winter visitor (though breeds in very small numbers in the Pentland Hills), present early October — early May, in strength late October — early March. Small numbers noted occasionally moving west along the coast at Aberlady, Gosford and Musselburgh in October — December and March — April. Borders Winter visitor, present early December — late February. Outer Hebrides Present all year; irregular as a migrant (Cunningham, Dix & Snow 1995). - Lewis Large numbers noted on passage, Butt of Lewis, 24 September 1911 (Baxter & Rintoul 1953). - Uists, Benbecula Present all year; no obvious change in numbers. - Flannan ls Summer visitor (Clark 1912). Inner Hebrides - Rhum Influxes noted in April (Evans & Flower 1967). - Eigg Evidence of passage, eg flock of 150+ 18 August 1958 (Evans & Flower 1967). - Coll & Present throughout the year (Stroud 1989); on Coll present in strength Tiree August-April with some evidence of passage August/September. - lona Present in small numbers throughout the year, irregular in winter months; & Mull migrants occasionally seen (Madders & Snow 1987); post-breeding flocks of up to 60 birds present August-September. -Colonsay Present throughout the year (Jardine, Clarke & Clarke 1986). Possible & Oronsay northward passage noted on Oronsay in April 1987 (Jardine 1992). - Islay Breeds in small numbers; more common in winter; passage evident (Elliott 1989, Ogilvie 1992). Argyll Small numbers breed and overwinter; common in suitable habitat o (mainland) passage (Madders, Snow & Welstead 1991). C Strathclyde Winter visitor in small numbers. Arran, Bute, Occasional winter visitor in small numbers, present August-March. Some Cumbraes _ evidence of passage on Arran where birds have been seen arriving from the sea (T ap Rheinallt, pers comm). Scottish Birds (1998) Ayrshire Movements of Twites in Scotland 275 Winter visitor (though a few breed on Ailsa Craig), present end September — early May, in strength mid-November — late March; highest numbers possibly associated with hard weather when occurs with Brambling, Chaffinch and Linnet. Movements often occur at same time as visible migration of Skylarks believed to originate from N and NW Scotland, and possibly from the continent. Galloway Small numbers breeding in E Galloway; influx in non-breeding season, present September — April, in strength October — March. Some evidence of a passage of birds in Wigtownshire (Dickson 1992). Dumfriesshire Winter visitor. a_ Information from this study unless otherwise stated. Ringing recoveries Figure 2 shows the 10 recoveries of Scottish Twites involving movements of >40 km. There were 3 movements between Fair Isle (ringing dates September (2) and October) and Orkney where the birds were recovered in May, November and January respectively, and one movement between Shetland and Orkney. A bird ringed on Fair Isle in July 1990 was recovered near Ellon in NE Scotland in March 1992. Two movements were recorded between winter quarters in Caithness and breeding grounds in NW Scotland: one in Wester Ross and one in Lewis. There were 2 movements between W Scotland and Ireland, One in each direction, but involving birds breeding in Scotland and wintering in Ireland. A bird ringed on Fair Isle in July was found dying on a ship off the coast of Germany/ Denmark approximately 3 months later. These movements involved 4 males and 4 females of which 5 were first years and 3 adults, as well as 2 birds whose age or sex was not known. There were, in addition, 4 short distance movements which show evidence of fidelity to an area throughout the year. These included a bird ringed on Out Skerries, Shetland in September and controlled the following February on Whalsay, Shetland (13 km SW), a bird ringed on Foula, Shetland in October and recovered dead in the following January on Wester Skeld, Shetland (35 km E), a first year female ringed at Lynegar, Caithness in January and found dead the following April at Dunbeath, Caithness (29 km W) and a first year male ringed at Northfield, Caithness in December and found tangled in sheep’s wool at Hempriggs, Caithness, in June, 2.5 years later (2 km SSE). Discussion Movements Even though the information available is incomplete, itis clear that a substantial part of the Scottish Twite population moves away from the breeding grounds for the winter 276 HClark & RM Sellers SB 19(5) months, and, in general, there appears to be a shift from high to low ground. The pattern of movements is evidently complex and, on the basis of the evidence presented here, we believe that at least 3 distinct migratory routes can tentatively be identified as follows: (1) from Shetland through Fair Isle to Orkney, and probably extending southwards to Caithness and the low lying coastal strip around the Moray Firth, especially around the Dornoch, Cromarty and Beauly Firths, and NE Scotland. In the main this probably involves birds breeding in Shetland, though we do not exclude the possibility that some birds from Norway are also involved; (2) from the NW Highlands and the Western Isles to Caithness, and possibly to the coastal plain of E Sutherland, and by implication Easter Ross and E Inverness; and (3) from the W Highlands through Argyll to the north coast of Ireland. The wintering areas of birds breeding in the southern Grampians, and the breeding locations of birds wintering in southern Scotland remain unknown. Small numbers of Twites also winter in northern England and the Isle of Man, and we suspect that these are mostly of Scottish origin. In parts of nearly all the breeding areas just listed, some birds appear to remain during the non breeding season for instance, on Shetland and Orkney, and the Outer and Inner Hebrides, and it may be better to describe all 3 as partial migrations. Birds breeding in Caithness, and probably those in E Sutherland and Easter Ross also, appear to remain close to their breeding grounds throughout the year, undertaking perhaps short distance movements from high to low ground, though we do not exclude the possibility that some migrate. There is no direct evidence from ringing that birds from Norway winter in Britain. An extensive ringing programme was carried out in Norway in the 1960s and generated over 100 recoveries, mainly from the countries along the south eastern seaboard of the North Sea, but none from Scotland (Bernhoft- Osa 1965, Cramp & Perrins 1994). The occasional arrival of Twites at sites such as the Isle of May with known Scandinavian migrants suggests that some Norwegian birds do migrate into Britain and this is supported by the records of Twites from oil rigs in the North Sea, especially the northern ones. The sightings from the more southerly rigs are also probably birds from Norway, but it is possible that they are Scottish, probably Shetland, birds migrating to the continent; the bird ringed on Fair Isle and recovered off the German/Danish coast may be such a bird. We think it likely that small numbers of Scandinavian birds do appear in Scotland, but that there is no regular large scale movement across the northern North Sea. We had hoped to shed further light on the question of whether Scandinavian Twites C f flavirostris visit Scotland by examining skins inthe collection of the British Museum (Natural History) and some kindly obtained by them on our behalf from the University Zoological Museum, Oslo. In practice, we found the differences in plumage between the Scandinavian race C f flavirostris and that breeding in Britain (C f pipilans; the Outer Hebrides breeding population is sometimes treated as a separate sub species C f bensonorum) to be so slight that they could not be applied with confidence either to skins of wintering birds taken in Scotland or to birds caught in the field. The races also differ Scottish Birds (1998) Movements of Twites in Scotland UT. Figure 2 Ringing recoveries of Twites in Scotland. Lines and arrows connect move- ments from breeding to winter quarters; these may not be the actual routes taken. *Bird recovered on a ship off the coast of Denmark/Germany P Sites with a regular passage of Twites M Sites where migrant flocks have been recorded slightly in size with C f flavirostris the larger Conservation aspects (Clark & Sellers unpublished data based on skins), but it also proved impossible to The Twite is one of the few British breeding demonstrate conclusively on the wing _ landbirds whose population is of international measurements that any flavirostris occur in importance and its conservation is therefore Britain. of some importance. The prime conservation requirement is to provide protection to the N.B. Please insert this page into your Scottish Birds - Winter 1998 at page 277. Computer error on map. Rtg f + Seid i! ee arg "ho } ere Spears a “SACD: BSR: os 22ND BRB hid Mesiah ea oak ‘euaton seabiasstecebenill AP ee oe a au Liar ee: tf . f : Sie t tt ‘4 a : 4 a | 1) Bee j a eee | pte renee = ‘ yo? i . Bnv ahi rire eh Vi May 7 PTE) ne kG f one Mp in lee fa, Sion Sy ee ‘ 2 ¢ r ial 4 : SE ~ , Me a | ‘ rey ad rs P 14 ' ax BS as Bris 4) { hs i? OM * 3 | bi ol xree Pops nga ii ta e500 in the bated intr pm aes Ast mf i , een vitae Mi rest fi, @ Seo Gaetan ; pal Jy a * 1 ed Le, “i TCA (if ) ox .Heliidess |. eid og, ‘cole ys@n00 Lane ona | es Rosen abe 1 Bet He, s ha. ‘ ‘ ole erty read “ — ici. ioee fy bo er i ele gacctie Z Pi ie bi bral | iQ ple) rae a) iy its ; Nene ADI ats ae bank aocecl sn * o bad a ne ge aT ie Sue # ion Poaapagto wy og bids ano SOY abive 1Q OF Irion erlipxan "tae ee , we ct: om. Cc | 6 i aol eas — ae RET MD NN VANTIN © rs * A yagi don Scottish Birds (1998) Movements of Twites in Scotland CTE Figure 2 Ringing recoveries of Twites in Scotland. Lines and arrows connect move- ments from breeding to winter quarters; these may not be the actual routes taken. P Sites with a regular passage of Twites M Sites where migrant flocks have been recorded Slightly in size with C f flavirostris the larger Conservation aspects (Clark & Sellers unpublished data based on skins), but it also proved impossible to The Twite is one of the few British breeding demonstrate conclusively on the wing _ landbirds whose population is of international measurements that any flavirostris occur in importance and its conservation is therefore Britain. of some importance. The prime conservation requirement is to provide protection to the 278 HClark & RM Sellers SB 19(5) main breeding and wintering areas (Batten et al1990, Brown & Crockford 1997), but implicit in this is a sound knowledge of migration routes. The picture which emerges from this study is complex, but emphasises that any action to protect Twites in their principal wintering areas in Scotland (Orkney, Caithness, southern part of Western Isles, SW Perthshire, Argyll) must be balanced by equivalent action in the corresponding parts of the breeding range, namely Shetland, Orkney, Caithness, W and NW Scotland, Western Isles, and N Perthshire. Knowledge of the movements of Twites in Scotland is not complete and more information on this is needed to ensure the effective conservation of the species. This will probably be best achieved by a coordinated programme of ringing throughout Scotland. Acknowledgements This study would not have been possible without the help of a large number of people, many unknown to us, and to them all we extend our thanks. We owe a particular debt to the British Museum (Natural History), Tring and the University Zoological Museum, Oslo for permitting us to check Twite skins in their collections, to the Scottish Ornithologists’ Club for providing help with obtaining local Scottish bird reports and other publications, to Fair Isle Bird Observatory and North Ronaldsay Bird Observatory for supplying information on the passage of Twites at their sites, and to local bird recorders throughout Scotland and other birdwatchers for their patience in answering our questions. Ringing recoveries were 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), andthe ringers themselves. We would also like to thank Paul Collin, lan Darling, Angus Hogg, David Jardine, Kevin Osborn, Tristan ap Rheinallt, Alan Vittery, Audrey Walters and Kevin Woodbridge for their unstinting help and advice in our search for information on Twites in Scotland. Finally we are indebted to lan Bainbridge. Mark Hancock and Baz Hughes for their comments on an earlier draft of this paper. References Batten L A, Bibby C J, Clement P, Elliott GD & Porter R F 1990. Red Data Birds in Britain, Poyser, London. Baxter E V & Rintoul L J 1953. The Birds of Scotland, Their History, Distribution, and Migration. Oliver & Boyd, Edinburgh. Bernhoft-Osa A 1965. Om Birgiriksens Carduelis flavirostris trekk. Stavanger Museums Arbok. 35: 109-118. Booth C, Cuthbert M & Reynolds P 1984. The Birds of Orkney. The Orkney Press, Stromness. Bourne W RP, Knox A G, Merrie T D H & Morley A H 1978. The birds of the Forties Oilfield 1975- 1978. North-east Scotland Bird Report. 47-52. Brown A & Crockford N 1997. RSPB Report, Species Action Plan 1662. Twite Carduelis flavirostris, a red list species. Buckland ST, Bell M V & Picozzi N 1990. The Birds of North-east Scotland. North-east Scotland Bird Club, Aberdeen. Clark H & Sellers R M 1997a. Twite winter habitats in Scotland: results of the 1997 Questionnaire survey (unpublished report). Clark H & Sellers R M 1997b. Distribution and abundance of Twites wintering in Caithness. Scottish Birds. 19: 1-9. Clark W E 1912. Studies in Bird Migration. Gurney & Jackson, London. Cook M 1992. The Birds of Moray and Nairn. The Mercat Press, Edinburgh. Cramp S & Perrins C M (eds) 1994. The Birds of the Western Palearctic. vol.8 Oxford University Press, Oxford. Scottish Birds (1998) Movements of Twites in Scotland 279 Cunningham P, Dix T & Snow P 1995. Birdwatching in the Outer Hebrides. Saker Press. Carnduncan. Davies M 1988. The importance of Britain’s Twites. RSPB Conservation Review. 2: 91-94. Dickson R C 1992. Birds in Wigtownshire. G.C. Book Publishers Ltd, Wigtown. Dymond J N 1991. The Birds of Fair Isle. Elliott R E 1989. Birds of Islay. Christopher Helm, London. Evans P R & Flower W U 1967. The birds of the Small Isles. Scottish Birds. 4: 404. Gibbons D W, Avery MI, Baillie S R, Gregory R D, Kirby J, Porter RF, Tucker GM & Williams G 1996. Birds of conservation concern in the United Kingdom, Channel Islands and Isle of Man: revising the Red Data List RSPB Conservation Review 10: 7-8 Hagemeijer W J M & Blair M J (eds) 1997. The EBCC Atlas of European Breeding Birds; Their Distribution and Abundance. T & A D Poyser, London. Harvie-Brown J A, Cordeaux J, Barrington R M, More A G & Clark W E 1886. Report on the Migration of Birds in the Spring and Autumn of 1885, 7th Report. Jardine D C 1992. Twites eating seaweed. British Birds. 85: 619. Jardine D C, Clarke J & Clarke P M 1986. The Birds of Colonsay and Oronsay, Their History and Distribution. Jardine D C & Reid J B 1993. Twite in The New Atlas of Breeding Birds in Britain and Ireland: 1988- 1991, (ed. D W Gibbons, J B Reid & RA Chapman), T & A D Poyser, London, p.41 8-41 9. Lack P 1986. The Atlas of Wintering Birds in Britain and Ireland. Poyser, Calton. Madders M & Snow P 1987. The Birds of Mull. Saker Press, Lochdon. Madders M, Snow P & Welstead J 1992. Birds of Mid-Argyll Saker Press, Munro | 1988. The Birds of the Pentland Hills. Scottish Academic Press, Edinburgh. Ogilvie M 1992. The Birds of Islay. The Lochindaal Press, Glencairn. Sharrock J T R 1976. The Atlas of Breeding Birds in Britain and Ireland. British Trust for Ornithology/ Irish Wildbird Conservancy, Tring. Smout A-M 1986. The Birds of Fife. John Donald Publishers, Edinburgh. Stroud D A (ed) 1989. The Birds of Coll and Tiree: Status, Habitats and Conservation. Nature Conservancy Council/Scottish Ornithologists’ Club, Edinburgh. Thom V M 1986. Birds in Scotland. T & A D Poyser, London. Hugh Clark, 3 Lindsay Place, Wick, Caithness KW1 4PF Robin M Sellers, Rose Cotiage, Ragnall Lane, Walkley Wood, Nailsworth, Gloucestershire GL6 ORU Revised manuscript accepted August 1998 280 Scottish Birds (1998) 19: 280-289 SB 19(5) Distribution and foraging habitat preferences of Choughs on The Oa peninsula, Islay M MADDERS, F M LECKIE, J WATSON & C R MCKAY Foraging Choughs on The Oa peninsula, Islay, were studied in relation to the distribution of habitat. A total of 13-14 birds were present in winter 1995, and 6 pairs and one subadult in late summer 1996. In general, foraging Choughs selected grassland habitats. In Decmber-February, the order of Chough foraging preferences was acidic grassland > improved grassland > neutral grassland > other habitats. In July-September, when Choughs were distributed mostly around the coast, the order was neutral (dune) grassland > cliff and slope > acidic grassland and other habitats. Decline in the number of Choughs on The Oa does not appear to be associated with change in the availability of preferred foraging habitats. Other factors that may have had an adverse effect on Chough populations are discussed, and areas of further research identified. Introduction The Isle of Islay has traditionally supported a large number of breeding Choughs Pyrrhocorax pyrrhocorax (Thom 1986). A Survey in 1986 estimated that the Scottish Chough population comprised at least 105 pairs, of which 95 (90%) were on Islay (Monaghan et a/ 1989). Increases in the Islay Chough population between 1968-72 and 1988-91 have been attributed to stable land management and the conservation of nest sites (Gibbons et a/ 1993). However, evidence from The Oa peninsula on Islay suggests that numbers of chough have declined dramatically in recent years. Breeding surveys found 19 pairs and 22 non breeders on The Oain 1982 (Warnes 1983), and 23 pairs and 20 non breeders in 1986 (Monaghan et al 1989). These surveys indicated that The Oa held 31% and 24%, respectively, of the Islay breeding population. By 1992 only 13 pairs were present on The Oa (E. Bignal pers comm), representing 18% of the total number of pairs on Islay. A partial survey in 1994 indicated that The Oa breeding population had further declined to 7 pairs (Bignal 1994). Few data are available on the size of the wintering population on The Oa. Warnes (1983) reported that 42 Choughs were located during a coastal survey in December 1980, and considered that a further 15 birds found on the Laggan peninsula at the same time were part of The Oa population. More recently, observations of Choughs foraging inland indicated that a minimum of 23 birds were present during winter 1987 (C R McKay pers comm). In order to better understand the factors that influence Chough populations, we studied the distribution of Choughs on The Oa in relation to the availability of habitat. Definitions follow those of Mosher et a/(1987). Availability Scottish Birds (1998) of a habitat is its relative area within the study area. Habitat use is a measure of the quantity of a habitat utilised. Habitat selection occurs when a habitat is used disproportionately in relation to its availability. Preference for a habitat is a measure of the likelihood of that habitat being chosen if its availability is equal with that of other habitats. Study area The Oa is a broad peninsula covering approximately 47km2 in the southwest of Islay (Figure 1). A wide range of habitats are supported, including bog and dwarf shrub Figure 1 Location of The Oa, Islay Distribution & foraging of Choughs on Islay 281 heath, semi improved acid and neutral grasslands, andimproved and semi improved pastures and arable habitats. The coast comprises mainly steep rocky cliffs in the west and south, and lower profile cliffs and raised beaches in the east. An extensive dune system adjoins the area at Kintra, and extends northwards to the Laggan peninsula. The Oa is extremely exposed and generally treeless, although some of the eastern part was planted with conifers during the 1980s. Land use is geared mainly to the production of beef cattle and sheep. The inbye pastures and semi improved grasslands are used Port Askaig Bridgend Bruichfaddich Bowmore Port Charigtte Laggan peninsula Portnahaven Port Ellen 5km 282 M Madders et al SB 19(5) mostly for grazing in spring and silage production in summer. Most improved grass is left ungrazed by cattle during winter to avoid poaching, except at Kintra where the ground is better drained as a result of the sandy substrate. Elsewhere, the semi improved grasslands and heath/grassland mosaics are extensively grazed by sheep and small numbers of cattle throughout the year. The coastal cliffs are grazed by a population of c.100-150 feral goats (Newton 1984). Rabbits Oryctolagus cuniculus occur in several localities, and are particularly numerous in the dune grassland at Kintra. Methods Chough distribution and behaviour were recorded during 2 study periods: December-February 1995-6 and July- September 1996. Five survey routes were used to systematically search for Choughs on The Oa and in a buffer zone extending approximately 1km east. Routes were selected to pass within 0.5km of all areas of habitat considered likely to hold Choughs, ie sea cliffs and slopes, grassland, and cultivated ground (Figure Figure 2 Location of 5 routes used to systematically survey Choughs on The Oa, Islay. Kintra s 5 Upper Killeyan 1km Scottish Birds (1998) 2). Routes were allowed to meander so that Suitable ground could be covered as thoroughly as possible. Where routes followed the coast, headlands were used to carefully scan the seacliff and shoreline. The total area visible from all routes was estimated to be 40 km, ie 85% of the study area. Each route was walked 12 times (6 times each in December-February and July- September). The direction in which routes were undertaken was reversed between visits, to help control for diurnal variation in Chough activity. Possible bias due to differences in Chough detection between different observers was minimised by ensuring that each observer undertook each route an equal number of times. Observations were carried out only in conditions of good visibility, and in _ wind speeds <14 ms"! (Beaufort scale 6). _ All Choughs detected from the routes were | | | | recorded. To reduce dependency within the data, individuals considered to have been seen previously from the same route on the same day were excluded. These decisions were based on observations of Chough movements and, in a small number of cases, from observations of uniquely colour ringed individuals. Locations at which Choughs were initially detected were plotted onto 1:10,000 scale maps and the habitat at each location classified by reference to Phase 1 (NCC 1990) habitat maps drawn in 1995. For each observation, the number of Choughs were recorded, and the activity of each bird classified as foraging or not. Where possible, ' each bird was checked for colour rings and aged as either adult, sub adult (at least one calendar year old), or juvenile/immature. The area of each available habitat class was estimated using a point sampling technique. Points were selected by overlaying Phase 1 _ habitat maps with a grid, the cells of which Distribution & foraging of Choughs on Islay 283 represented 200x200m on the ground. Preliminary analysis indicated that, up to 200m from the route surveyed, the frequency with which Choughs were detected did not decline with distance. However, numbers declined rapidly beyond this distance, such that only 33% of the total observations related to Choughs located greater than 200m away (n = 24). Ideally, we would have liked to determine habitat selection using only data from points 0-200m from survey routes. However, this would have resulted in unacceptably small sample sizes, and would have reduced both the areas and range of the habitats sampled. Accordingly, habitat was classified at all grid intersections within 200m of the survey routes, and atarandom sample of 50% of the other available intersections within 1km. Intersections were used regardless of whether they had been classified for other routes. Using this approach, the distribution of habitat sampling points reflected bias inthe locations of the Choughs observed. Grid intersections that fell within areas of sea were ignored. The availability of each habitat was determined simply by counting the number of sample points within each habitat class. Foraging habitat preferences were investigated by comparing selection probability functions (Manly et al 1993) for habitats in which Choughs were observed to feed. These functions representthe proportion of the available units in a particular habitat class that are used. In this case, they can be estimated by dividing the proportion of Chough foraging observations in a particular habitat class by the proportion of that habitat class observed, ie Wie (UU ley ele) | + iad where w is the sample selection ratio, u, is the number of records of foraging Chough$ in habitat class |, u. is the total number of 284 M Madders et al SB 19(5) records of foraging Choughs, a. is the area of habitat class i observed, and is the total area of all habitat classes observéd. Selection functions were standardised so that they summed to one. These standardised ratios (B.) can be interpreted as being the estimated probability that a habitat would be the next one selected if each class could be made equally available to a foraging Chough. Possible values for B. can range from 0 (no selection) to 1 (always selected). Results Numbers and distribution Choughs were observed 68 times (29 times in December-February and 39 times in July- September). Each observation involved between one and 13 birds. The distribution of sightings differed markedly between seasons, with more birds encountered around the western coast during July-September, and greater use of inland pastures in December- February (Figure 3). This shift away from coastal habitats in winter was statistically highly significant (mean distance from High Water Mark in metres + Standard Error: July- September 325.6 + 62.7, n = 39; December- February 967.2 + 114.4, n = 29; t = -5.24, df 66, 2-tailed P < 0.001). Based on simultaneous sightings of Choughs in different localities, together with observations of 2 colour ringed individuals, The Oa wintering population in 1995/6 was estimated to comprise 13-14 birds, including at least 4 adult pairs. Excluding juveniles, a total of thirteen Choughs was present during July-September 1996, including 6 adult pairs and a sub adult. In July, 5 pairs were accompanied a total of 10 fledged young. However, one juvenile had disappeared by September. One family moved to the adjoining duneland north of Kintra shortly after fledging, and did not return to The Oa until September. The remaining birds mostly occupied the south western cliffs and adjacent pastures, where they often coalesced into a single foraging group. Habitat selection Fifty Chough observations involved one or more foraging individual (December- February: 24; July-September: 26). Foraging Choughs occupied 10 Phase 1 habitat classes, comprising 5 grassland and 2 heathland types, plus cliff and slope, arable, and inter tidal habitats (Table 1). A further 7 habitat classes available on The Oa were avoided by Choughs. These included woodland and scrub, marshy grassland, bracken, and mires and flushes. The only area of arable habitat selected was a stubble field at Upper Killeyan, which was used by varying numbers of Choughs after it was cut in late August. Choughs feeding in this field appeared to forage for invertebrates within the soil rather than on any residual grain. Overall, Choughs preferred semi improved to unimproved acidic grassland. Wintering Choughs foraged preferentially in acidic grassland and improved grassland (Figure 4). Neutral grasslands were less preferred, and other habitats were used little or not at all. In July-September, Choughs showed a strong preference for neutral grasslands, followed by maritime cliff/slope and acidic grasslands. Other habitats used by Choughs included arable and intertidal habitats (July-September | only), and dwarf shrub heaths. According to the Phase 1 vegetation maps Choughs foraged in wet and dry heaths during both study periods. However, field observations indicated that Choughs were in | fact using small patches of unimproved " ! Scottish Birds (1998) Distribution & foraging of Choughs on Islay 285 Figure 3 Sites where Choughs were initially located during systematic coverage of survey routes. Dot size represents the number of Choughs in each group located (including juveniles). Arrows show the direction and number of flying birds. (a) December-February (b) July -September Dot size represents: 1 bird e 2 birds @ 3-5birds @ 6-10birds Figure 3a Dot size represents: 1 bird @ 2 birds @ 3-5birds @ 6-13 birds I Figure 3b SB 19(5) 286 M Madders et al L ‘UQIeJ UOD9|AS pasipsepuels = g ‘UOTE UOOE|aS = mM -suBnoyuD BHuibeso} Jo sp109e/ JO equUNU |e}0} JO UOJOdoOJd e se} SseEjO yeYIGeY UI SUBNOUD Buibes0} 40 spiooel Jo Jaquinu = n/N ‘(9¢ = NZ 1deS-Inf ‘72 = NZ :qe4-0eq) syBNnoyD buibe19} Jo spuoded Jo Jequunu je}o} = 1-/ ssej yeygey U! SyBNoUD Buibes0} Jo spsoda1 jo Jaquinu = (7 ‘SOSSEIO JE}IGeY [e}0} JO UO!JOdoud e se/ ssejo JeIGQeY = &/e (GzG = EZ) Sjulod Huljdwes jo sequunu je}0} = eB {/ SsejO yeYIGeY Ul SjuIod Huljdwes Jo Jequunu = eB :Aey €0° GQ'l 8e0 €0° GO'l 8e0 OL" 8S°6 GLI’ G9" Ose GIle LO 19° Sl rAl) veel vSl €0° Loz 692 ZO" vol LAG GO" c6'c 8e0 L L fee g M nn 1das-ine ooooyrTrvroonn rT =) pa}9/9sS }eVIGeH qoe-4-00q OMOOrTrTNtoooneoo o>) o>) €c0° €c0° clO €00° 600° c8l Ofc GEL €c0° 890° 6c0° 600° e sIGe|IEAY JEVIQeH epi] 49}uU] PSljISSEOU/) PUe| 9/|Qely HD puejssei6 jejseoy Syo0o7 SOUSNI} 9 SOJIIN YJESY QN4IYS HEMP je uyesy qnius Yemp Aig usyoeIG puejssei6 Aysuey\ puejsses6 panosduy| peaosduiun puejssei6 snoaeieojey peaosduiun P@AOJAWI IWIAS purjssei6 jeujney peaosduiup PeAosdu! IWIaS pupjsseib6 o1pioy QnioS Y PU|POO/A ssejo yeyiqey | aseud ‘sy6noy) bulbesoy Aq uolNdajas 41ay] pue EC AYL UO Sasse/o jeyiqey | aseyd jo AjjigejieAy | a|qeL Scottish Birds (1998) Distribution & foraging of Choughs on Islay 287 Figure 4 Habitat preferences of foraging Choughs in Jul-Sept and Dec-Feb. The graph shows standardised selection indices for 5 Phase 1 habitat groupings (semi improved and improved acidic grasslands combined; coastal and semi improved neutral grasslands combined; other = wet and dry heaths, arable land and inter tidal habitats). Selection Acid grass. Neutral grass. grassland within heathland habitats. These were presumably too small to be discriminated by the Phase 1 mapping resolution. It was notable that many of these patches were located in areas overlying bands of limestone. Discussion Comparison with data from the 1980s (Warnes 1983; Monaghan et al 1989) indicates that The Oa breeding population has decreased by about 75% and the wintering population by about 60%, a serious long term decline. Warnes (1982) considered that approximately 30% of The Oa population wintered on the Laggan peninsula. We found no evidence to : | Winter | _] Summer Improved grass. Cliff Other suggest that this is currently the case, although one family group from The Oa took up temporary residence in duneland south of Laggan during late summer. Important differences in habitat selection by foraging Choughs were apparent between the 2 study periods. Choughs used a wider range of habitats in July-September, and were less reliant on improved grasslands and semi-improved acidic grasslands. This is consistent with McKay (1996), who found that Choughs on the Rinns of Islay used the narrowest range of habitats in late winter, when most birds fed in improved or semi- improved pasture. Chough nest sites on The 288 M Madders et al Oa are mostly coastal (Warnes 1982), and increased preference for maritime cliff/slope habitat in the breeding season (April-July) is perhaps unsurprising. However, examination of the data shows that this habitat was much used as late as September. This suggests that Choughs’ preference for cliff and slope habitat at this time may have been more than simply an artefact of nest site selection. Neutral grasslands were the most preferred habitat in July-September. By far the most intensively used area of this habitat was the dune grassland at Kintra, where 4-6 Choughs were more-or-less resident during September. It is probable that these birds found the short grassy swards, grazed by cattle and rabbits, preferable to the more structured vegetation characteristic of many other grasslands at this time (see Bullock ef a/ 1983). Most improved and enclosed semi improved pastures, for example, were effectively unavailable to Choughs much before August because they were used for silage crops. We can think of no dramatic change in the availability of preferred foraging habitats on The Oa over the past decade, although it is accepted that insidious change (eg the spread of bracken) is difficult to detect and that there is a clear need for a proper historical review, perhaps by mapping habitats from aerial photographs. Farming practices are likely to have had an important effect on Choughs. Several case studies (Bullock et a/ 1983) have demonstrated the importance of grazing herbivores to Chough feeding ecology. These maintain the sward at a height short enough to allow Choughs access to soil invertebrates (Bullock 1980), while dung beetles Aphodius found in the faeces of cattle and sheep represent an important additional source of prey (Bullock et a/ 1983). We do not know how numbers of grazing livestock have SB 19(5) changed on The Oa, and suggest that this would be auseful area for further investigation. Improved animal husbandry may have had an adverse effect on Choughs. An increasing number of cattle are wintered indoors on The Oa, and this will have inevitably led to a reduction in the availability of cow dung. At the same time, the trend toward the supplementary feeding of those animals wintered outdoors is likely to have concentrated dung in fewer areas. Finally, there is concern over the use of the anthelminthic drug, Ivermectin, which may prevent invasion of dung by invertebrates (McCracken & Foster 1992). Other factors which may have had an effect on Chough numbers, such as nest site availability and numbers of potential predators, are not known to have changed. The Oa holds a relatively strong population (10 pairs) of breeding Raven Corvus corax (Madders 1997). It is not known whether Raven numbers have increased in recent years, or if such an increase could have been detrimental to Choughs. The Chough’s decline on The Oa may be linked to changes in the population as a whole. A survey in 1992 (Bignal unpublished) indicated that numbers on Islay had declined by 35% since 1986, and that decline was greatest in the non breeding population. Our results indicate that a further widespread decline may have occurred, and the results of an island wide survey in 1998 are awaited with interest. Acknowledgements We gratefully acknowledge the co-operation of the landowners and farmers on The Oa. Lucy Arnold, Ciaran Cronin, Douglas Gilbert, Scottish Birds (1998) Ross Lilley, Fiona Rout and Anthony Williams assisted with various aspects of the study. | Wethank Paul Thompson and an anonymous referee for their valuable comments on the manuscript. Fieldwork was carried out under contract to Scottish Natural Heritage. References Bignal E 1994. Breeding survey of Chough on The ' Oa. Report to SNH. Scottish Chough Study Group. Bullock | D, Drewett D R, & Mickleburgh S P 1983. The Chough in Britain and Ireland. British Birds 76: 377-401. Bullock | D 1980. Some aspects of the ecology of | the Chough. MSc thesis. University College of North Wales. | Gibbons D W, Reid J B & Chapman R A 1993. The New Breeding Atlas of Breeding Birds in Britain and Ireland. T & A D Poyser, London. McKay C R 1996. Conservation and ecology of Red-billed Chough Pyrrhocorax pyrrhocorax. Unpublished PhD thesis, University of Glasgow. McCracken D | & Foster G N 1992. The effect of lvermectin on the invertebrate fauna associated _ with cow dung. Joint Nature Conservation Committee Report No.112, Peterborough. Distribution & foraging of Choughs on Islay 289 Madders M 1997. A population and behavioural study of ravens on Islay. Report to Scottish Office Agriculture, Environment & Fisheries Dept., Edinburgh. Contract no. MMA/001/96. Manly B F J, McDonald L & Thomas D 1993. Resource selection by animals. Chapman & Hall, London. Monaghan P, Bignal E, Bignal S, Easterbee N & McKay CR 1989. The distribution and status of the Chough in Scotland in 1986. Scottish Birds 15: 114-118. Mosher J A, Titus K & Fuller M R 1987. Habitat sampling, measurement and evaluation. Raptor Management Techniques Manual, NWF Sci. & Tech. Series no. 10 81-97. Nature Conservancy Council 1990. Handbook for Phase 1 Habitat Survey. A technique for environmental audit. Peterborough. Newton S (ed.) 1984. Observations of birds and mammals. Birds of Islay 1981-83. Brathay Field Studies Report no. 40. Thom V M 1986. Birds in Scotland. T & D Poyser. Calton. Warnes J M 1982. A study of the ecology of the Chough (Pyrrhocorax pyrrhocorax L.) on the Isle of Islay, Argyll, 1980-81. Unpublished report. University of Stirling. Warnes J M 1983. The status of the Chough in Scotland. Scottish Birds 12: 238-246. Address for correspondence: Mike Madders, Carnduncan, Bridgend, Isle of Islay, Argyll PA44 7PS \\ a a SEN s ‘ ‘3 in \ \ : x ae ‘ Choughs Steven Brown 290 Scottish Birds (1998) 19: 290-298 SB 19(5) Orkney Hen Harriers: a major population decline in the absence of persecution E R MEEK, G W REBECCA, B RIBBANDS & K FAIRCLOUGH The numbers of breeding male and female Hen Harriers in Orkney’s West Mainland in 1996 and 1997 were estimated to have declined by about two thirds since their peaks in the 1970s. For the whole of Orkney during 1996 and 1997, 25% of the females which built nests apparently did not lay eggs. Of the nests located with eggs, 60% did not produce any fledged young. A maximum of 39 young were fledged over the 2 years giving a productivity range of 0.4 young per female, 0.6 young per nest built or 0.9 young per nest with eggs. This represented a decline of about two thirds in breeding production since the peaks in the early and mid 1970s. It is suggested that reduced food availability, associated with land use change, and avian predation may have been the main reasons for the change in status. Introduction The Hen Harrier Circus cyaneus population of the West Mainland of Orkney (Figure 1) has been well documented (Balfour 1957, 1962, 1963, Balfour & Cadbury 1975, 1979, Picozzi 1980, 1984a,b). Orkney was one of the species’ few refuges from persecution in Britain during the late 19th and early 20th centuries and it is possible that the islands provided a source from which mainland Scotland was recolonised when gamekeeping pressure was relaxed during and immediately after World War II (1939-1945) (Watson 1977). More recently, it has been confirmed that Hen Harriers reared in Orkney have bred on mainland Scotland (Balfour & Cadbury 1975, Picozzi & Watson 1985). Hen Harriers have suffered severe persecution over much of mainland Britain for most of the 20th century (Watson 1977, Bibby & Etheridge 1993, Etheridge eta/1997, Redpath & Thirgood 1997, Scottish Raptor Study Groups 1997, Potts 1998, Stott 1998). The Orkney Islands are unusual in Britain in that human persecution of Hen Harriers is virtually non existent and there are no Red Foxes Vulpes vulpes, one of the species’ main natural predators (Watson 1977, Berry 1985, Redpath & Thirgood 1997). The Orkney Hen Harrier population increased between the late 1940s and the mid 1970s by which time polygynous breeding was widespread in the West Mainland (Balfour & Cadbury 1979). In 1978, an exceptional year, 95 females nested in the West Mainland and it was also estimated that there were 43 breeding males (Picozzi 1984a,b, Table 1). Breeding also occurred in the 1970s on Rousay (5 to 8 nests), Hoy (4 nests), East Mainland (3 to 4 nests) and Eday (1 to 2 nests) (Balfour & Cadbury 1975, MCockram, N Picozzi & E J Williams pers comm). The total number of breeding females for the whole of Orkney in 1978 may therefore have exceeded 110. Polygyny has also been recorded on Rousay and in the East Mainland but not on Hoy or Eday (RSPB unpublished, M Cockram, T Prescott & J Plowman pers comm). Thus, the total number of breeding males for the whole of Orkney in 1978 was probably between 51 and 55. a i | || | | i | _ north Scotland Scottish Birds (1998) Figure 1 Map of the Orkney Islands, 1 ES Ss, apa Fou, ie) i 9 6 North Ronaidsay fh Westray ag Si aE ci Shapinsay ree Sanday soe West Mainland 2g” sores Burray South Ronaldsay a eae NCO 7) r-) Wp, 70 NEEL In 1981, 57 female and 21 male Hen Harriers were found on West Mainland (Picozzi 1984a,b). Thereafter (apart from 1989) it became impractical to establish the annual number of harriers attempting to breed, mainly because of a lack of full time field study. However, it is believed that most, probably all, successful nests were found in the whole of Orkney for the period 1982-95 (RSPB unpublished). In 1989 a concentrated effort was made to locate all breeding attempts as part of anational Hen Harrier survey (Downing 1990, Bibby & Etheridge 1993) and a total of 71 breeding females was found, with 62 in the West Mainland (Table 1). The West Mainland figure was within the range found during Picozzi’s intensive study and did not cause immediate concern. However, by 1994, it Orkney Hen Harriers: population decline 291 was evident that the decline had continued, possibly as a result of increasing nest failure and declining clutch size during 1982-94 (Kalejta-Summers 1995). In 1996 and 1997 the Royal Society for the Protection of Birds attempted to census the Orkney Hen Harrier population and record breeding success, allowing comparisons to be made with the earlier studies of Balfour & Cadbury (1975, 1979), Picozzi (1984a,b) and Downing (1990). Methods In 1996 and 1997 all known nesting areas and other suitable moorland habitat in the West Mainland, Rousay, Hoy and Eday were visited and/or watched on several occasions between mid April and the end of May. The Table 1 The numbers of breeding Hen Harriers in West Mainland, Orkney in years when all breeding attempts were thought to have been known. Year Females Males Source 1970 42 16 Balfour & Cadbury 1979 1971 44 21 ‘ 1972 49 24 i 19%) 1 oll 36 i 1974 62 45 i 1975 47 Picozzi 1984 a,b 1976 56 : O77 385 44 : 1978" 95 43 5 1979 60 32 i 1980 80 if ‘ igs S7 21 : 1989 62 Dowing 1990 1996 24 14 present study 1997 26 14 i 292 ER Meek etal SB 19(5) East Mainland now has little suitable moorland (Bennett 1986) and was only visited once in each breeding season. The initial fieldwork coincided with the courtship, pairing, nest building and peak egg laying periods when Hen Harriers are at their most obvious on the breeding grounds (Balfour 1957, 1963, Watson 1977). In addition, 6 transects of 7- 8 km were each walked on 4 occasions during 15 April to 15 May in both years. Five were on West Mainland and one was on Rousay and they were designed so as to traverse blocks of moorland with recent records of breeding Hen Harriers. Further visits were made between mid May and early August to survey suitable habitat again, visit nests and record breeding success. A breeding female was defined as one which built at least a partial nest, whether or not eggs were proven to have been laid. A non breeding female was defined as one which was not observed nest building. For example, ifwe regularly saw the females of apolygynous group of 3 (see Balfour & Cadbury 1979 and Picozzi 1984a,b) between mid April and the end of May and only found or saw 2 nests being built, the third female was classed as a non breeder. As there was limited time available for lengthy observations we could have underestimated the number of females that built nests. Accurate assessment of the number of adult males was difficult as none were individually marked (see Balfour & Cadbury 1979 & Picozzi 1984a,b for Figure 2 The number of fledged or almost fledged young Hen Harriers in the West Mainland, Orkney between 1953 and 1997. Data from 1953-74 is from Picozzi (1980) and is a minimum count for some years as some broods were omitted from his analysis. Data from 1975-95 is the annual total of fledged young (Picozzi 1984a for 1975-81 and RSPB unpublished for 1982-95). Number of young Year 953 1956 1959 96 1965 968 1971 1974 3 year mean 28 26 44 “41 bse a52 73 9 year mean 33 46 Scottish Birds (1998) Orkney Hen Harriers: population decline 293 _ Figure 3 The number of broods of Hen Harriers with at least one fledged or almost | fledged young in the West Mainland, Orkney between 1970 and 1997, years in which all successful nests are likely to have been located. Data from 1970-73 from Kalejta-Summers (1995), 1974 from C J Cadbury (in litt), 1975-81 from Picozzi (1984a) and 1982-95 from RSPB (unpublished). 40 35 N wn Number of broods 8 =i wn 1970 1973 1976 1979 1982 descriptions of marked birds). In addition, the members of a male’s polygynous group of females may not necessarily have all been in the same valley, and therefore not all | simultaneously visible to an individual observer. Our estimated figures for males were quantified using plumage characteristics and behavioural traits, together with close | comparison of the timing of sightings in different areas by individual observers. To measure breeding success we aimed to visit all nests at least once during the incubation, nestling and fledging stages. If the number of fledged young actually counted was less than the brood size at ringing, at about 3 weeks old, (i.e. the maximum number) we searched the nest and its environs for the remains of chicks. Mortality in the late nestling 1985 1988 1991 1994 1997 period is known to be low for Orkney harriers (Picozzi 1980), and when it does occur, remains are found (ERM pers obs). Therefore if no remains were found we presumed that the complete brood had fledged and these maximum figures were used to calculate productivity. Results Numbers There were 29 and 31 breeding females respectively and 19 breeding males in Orkney in 1996 and 1997 (Table 2). Between the 2 years there was, however, amarked reduction in apparently non breeding females from 19 to 6, giving an overall total for females of 48 and 37 respectively (Table 2). In the West 294 ER Meek etal SB 19(5) Table 2 The estimated numbers of breeding Hen Harriers and non breeding females in Orkney in 1996 and 1997. Females breeding non breeding Island 1996 West Mainland 24 Eday 0 Rousay 1 Hoy 4 Totals 29 19 1997 West Mainland 26 5 Eday 0 0 Rousay 1 1 Hoy 4. 0 Totals 31 6 Mainland, the numbers of breeding females in 1996 and 1997 were 24 and 26 respectively, mean 25; males numbered 14 in each year (Tables 1, 2 & 4). Breeding success The breeding success in 1996 and 1997 is detailedin Table3. The percentage of females building, at least partial nests, but apparently not subsequently laying eggs, was 17% in 1996 and 32% in 1997. In total, 45 nests with eggs were located, 24 in 1996 and 21 in 1997. Of these, 9 failed during incubation in 1996 and 13 in 1997. Five complete broods were lost, 4 in 1996 and 1 in 1997. Overall, 40% of nests found with eggs fledged at least one young. The total numbers of young believed to have fledged were 22 in 1996 and 17 in 1997. The maximum productivity of both years combined was 0.9 young per nest located with eggs, 0.6 young per nest built or Males totals adults first year 40 2 2 - 0 0 2 | 0 6 3 1 48 16 3 31 13 1 - 0 0 2 1 0 4 4 0 37 18 | 0.4 young per female (Table 3). Inthe 4 study periods where realistic comparisons could be made (1970-74, 1975-81, 1989 and 1996- 97) there were significant differences between studies in the numbers of breeeding females and the production of young, ANOVA for breeding females, F, ,,=5.51, p<0.02 and for young per breeding female, F, ,,=5.12, p<0.02 Females averaged 68.6 during 1975-81 but had dropped to 25 by 1996-97. Average productivity was 1.5 young per breeding female in 1970-74 but had dropped to 0.6 by 1996-97 (Table 4). Interestingly, 1975-81 was the study period with the highest mean number of breeding females and was preceded by the highest period of mean productivity, 1970-74 (Table 4). The estimated numbers of Hen Harriers reared annually inthe West Mainland between 1953 and 1997 are shown in Figure 2. Since 1992, the numbers of fledged young have | | Scottish Birds (1998) Orkney Hen Harriers: population decline 295 | Table 3 Breeding statistics for Hen Harriers In Orkney in 1996 and 1997. | Female numbers from Table 2, NB = nest at least partially built, NE = nest with eggs. _ Island Number Nests Nests Nests Minimum Productivity of with with with & maximum maximum no. of nests eggs young fledged no. of fledged fledged young per located young young Female NB NE | 1996 West Mainland 24 19 12 9 13-16 Rousay 1 1 1 1 1 ~ Hoy 4 4 2 1 5 — Total 29 24 15 11 19-22 O4alOx%a 20:9 | 1997 West Mainland 26 16 5 9-13 Rousay 1 1 1 0 0 Hoy 4 4 2 4 — Total 31 21 8 7 13-17 0:4 ac0:5y20:8 1996 & 1997 combined 60 45 23 18 32-39 04 06 09 Table 4 Differences between studies in the numbers of breeding female Hen Harriers and in the production of young on West Mainland, Orkney (+/- = standard error). Studies Years Meanno Mean no of fledged young/ of females mean no of breeding females Balfour & Cadbury 1979 & Picozzi 1980 1970-74 49.6 (+/-6.3) 1.5 (+/-0.2) Picozzi 1984a 1975-81 68.6 (+/-5.3) 0.8 (+/-0.1) Downing 1990 & Kalejta-Summers 1995 1989 62 0.9 Present study 1996-97 75) 0.6 ~ been consistently low and in the worst year, | 3 shows the number of broods fledged in the 1993, only 4 were found. The 3 yearmeans West Mainland between 1970 and 1997, for 1992-94 and 1995-97 and the 9 year years in which all successful nests were mean for 1989-97 were the lowest since believed to have been located. The pattern records beganinthe 1950s (Figure2). Figure is similar to Figure 2. 296 ER Meek et al SB 19(5) Discussion Numbers and breeding success The surveys in 1996 and 1997 were considered to have thoroughly covered all Hen Harrier breeding habitat except in the East Mainland. The latter area did support at least 6 breeding females in the late 1960s (E J Williams pers comm) but has lost so much Suitable breeding habitat (Bennett 1986) that no more than one parr of harriers is believed to breed there now (ERM pers obs). Search effort in the West Mainland is believed to have been comparable in each of the 4 main study periods (Table 4). This being the case, there has clearly been a major decline since the 1970s and the population in 1996- 97 was very low. In order to avoid over emphasising the decline from the exceptional peak year of 1978, 3 year running means were calculated for the number of breeding females between 1970 and 1981. The peak 3 year running mean was in 1977-79 with 80 females, the 1996-97 mean of 25 representing a 69% decline. Three year running means cannot be calculated for males as data are lacking for 1975 and 1976. However, the mean for the 3 peak years for which data are available (1974, 1977 and 1978) was 44 (Table 1). The 1996-97 mean of 14 males represents a 68% decline . In our Study breeding success was poor, with a maximum of 39 young fledged for the whole of Orkney over the 2 years. Inthe West Mainland, the mean number of young fledged between 1992 and 1997 was only 12.3 +/- 4.6 s.d.; range 4to 17. Previous poor years in the West Mainland (eg 1979 which was particularly wet) were usually followed by a marked improvement, but the period 1992- 97 has been the least productive, in terms of young fledged, since detailed records began in 1953 (Figure 2). Three year running means of the number of young reared and the number of young reared per breeding female were calculated for 1970-81. The peak 3 year running mean for young reared was in 1973-75 at 80.7; the mean maximum figure for 1996-97 was 14.5, representing a decline of 82%. The 3 year running means for the number of young reared per breeding female peaked in 1971-73 and 1973-75 at 1.5; by 1996-97 productivity had fallen to 0.6, a 60% decline. These declines, for breeders and production, since the peaks in the 1970s, are in the order of two thirds. In contrast to many other areas of Britain (Etheridge et a/ 1997, Stott 1998) they have occurred in the virtual absence of human persecution, at least when the birds are in Orkney. Possible reasons for the declines There has been major moorland reclamation on Orkney since 1932 (Bennett 1986); for example there was at least 40% moorland loss on the Mainland between 1940 and 1985 (Gorman & Reynolds 1993). However, in the West Mainland, no Hen Harrier nest sites have been lost to reclamation since Picozzi’s study. Most of the core moorland areas, where the majority of the harriers nest, are Sites of Special Scientific Interest (SSSIs) where Scottish Natural Heritage have negotiated management agreements with landowners to restrict the stocking levels. Nevertheless, some of these areas are still grazed heavily enough to be degrading the areas of rank vegetation in which harriers prefer to nest. On the moorland fringes outwith the SSSls, there are no restrictions on stocking densities and it is these areas, which in the past were some of the preferred Scottish Birds (1998) hunting areas of harriers, which have changed mostin character between 1982-97. Changes have also occurred on lower ground where a considerable area of semi natural vegetation, previously used by hunting harriers, has been lost to more intensive farming (ERM pers obs). One major change in Orkney agriculture has been the increasing numbers of sheep. In 1981 there were 31,164 breeding ewes in the islands; by 1996 this had risen to 55,541, an increase of 78% (Orkney Islands Council 1982 & 1998). These changes may have resulted in reduced food availability and could be one reason why some females apparently do not nest (Newton 1986, Simmons ef al 1986) and others breed poorly. Apparent non breeding amongst female harriers in the West Mainland was known during Picozzi’s study and in 1981, for example, was estimated at 26% (Picozzi 1984b). Non breeding appeared to be particularly prevalent in 1996 (40%) and it is possible that such females may disrupt the breeding females. They certainly solicit males for food and copulation with some success (see also Simmons 1988 for Northern Harriers). Non breeding females may also attract predators, particularly Hooded Crows _ Corvus corone cornix, which were the main egg predators of Hen Harriers on Orkney during 1975-81 (Picozzi 1984a). Breeding corvids are abundant on Mainland and Hoy, and in addition flocks of up to 25 apparently non breeding Hooded Crows or Ravens Corvus corax were regularly seen on moorland during 1996 and 1997. These corvids were occasionally seen interacting with the apparently non breeding females in the vicinity of viable harrier nests. Another factor affecting the population may be an increased mortality of males. Picozzi (1984b) suggested that polygyny in the West Orkney Hen Harriers: population decline Zo Mainland was a result of an imbalance in the sex ratio, possibly caused by a higher mortality of immature males. Observations at a winter roost on Orkney during 1975-81 suggested that more males than females left the islands in winter, possibly exposing themselves to greater risks (Picozzi & Cuthbert 1982). The winter dispersal of young male harriers from Orkney to the mainland of Scotland is presumably because suitable prey is more readily available there at that time (Watson 1977, Picozzi & Cuthbert 1982). The risks of leaving the islands might include lack of familiarity with their new habitat and the possibility of being deliberately killed (Scottish Raptor Study Groups 1997). The extensive data sets have enabled us to quantify the alarming decline of the Hen Harrier in Orkney but detailed study is necessary to investigate its causes further. Acknowledgements We thank the many Orkney landowners who gave permission for access without which this work could not have been carried out. Mike Cockram, Nick Picozzi and Jim Williams gave us historical details for Eday, Rousay and East Mainland and Nick Picozzi and James Cadbury provided details of their West Mainland study areas. Dr lan Bainbridge devised the protocol for the transect work. Dr Tom Prescott and James Plowman provided information from Hoy, Tim Dean and Peter Carty did likewise for Rousay and Tim Dunn, Ann Humble, Jim and Stuart Williams and David Wood assisted in the West Mainland. Dr lan Bainbridge, Brian Etheridge, Dr Steve Redpath and Dr Ron Summers commented constructively on an earlier draft. Finally we thank Dr Mick Marquiss for statistical advice and for his helpful comments as referee. 298 ER Meek etal SB 19(5) References Balfour E 1957. Observations on the breeding biology of the Hen Harrier in Orkney. Bird Notes 27: 3-18. Balfour E 1962. The nest and eggs of the Hen Harrier. Bird Notes 30: 69-73. Balfour E 1963. The Hen Harrier in Orkney. Part 3 Courtship, display and sociability. Bird Notes 30: 145-153. Balfour E & Cadbury C J 1975. A population study of the Hen Harrier in Orkney. In Goodier R. (ed) The natural environment of Orkney 122-128. Nature Conservancy Council, Edinburgh. Balfour & Cadbury C J 1979 Polygyny, spacing and sex ratio among Hen Harriers in Orkney, Scotland. Ornis Scand. 10: 133-141. Bennet A 1986. An assessment of moorland loss in Orkney 1932-1985. Unpublished report to RSPB, Sandy. Berry R J 1985. The natural history of Orkney. Collins, London. Bibby C J & Etheridge B 1993. Status of the Hen Harrier Circus cyaneus in Scotland 1988-89. Bird Study 40: 1-11. Downing R 1990. Orkney Hen Harriers - 1989. In: Booth C, Cuthbert M & Meek E R (eds) Orkney Bird Report 1989: 57-58. Etheridge B, Summers R W & Green RE 1997. The effects of illegal killing and destruciton of nests by humans on the population dynamics of the Hen Harrier Circus cyaneus in Scotland. Journal of Applied Ecology 34: 1081-1105. Gorman M L & Reynolds P 1993. The impact of land-use change on voles and raptors. Mammal Review 23: 121-126. Kalejta-Summers B 1995. Abundance and breeding success of Hen Harriers Circus cyaneus in the Orkney Islands, Scotland between 1957 and 1994. Unpublished report to SNH. RSPB Inverness. Eric Meek, Brian Ribbands & Keith Fairclogh, RSPB Orkney Office, Smyril, Stenness, Orkney KW16 3JX | Graham Rebecca, RSPB East Scotland Office, 10 Albyn Terrace, Aberdeen AB10 1YP Newton | 1986. Population Ecology of Raptors. Poyser, Berkhamsted. Orkney Islands Council 1982. Orkney Economic Review, number 2. Kirkwall, Orkney. Orkney Island Council 1998. Orkney Economic Review, number 17. Kirkwall, Orkney. Picozzi N 1980. Food, growth, survival and sex ration of nestling Hen Harriers in Orkney. Ornis Scandinavica. 11: 1-11. Picozzi N 1984a. Breeding biology of polygynous y Hen Harriers in Orkney. Ornis Scandinavica. 15:1- ‘| lO: Picozzi N 1984b. Sex ratio, survival and territorial behaviour of polygynous Hen Harriers in Orkney. Ibis 126: 356-365. Picozzi N & Cuthbert M F 1982. Observations and food of Hen Harriers at a winter roost in Orkney. Scottish Birds 12: 73-80. Picozzi N & Watson J 1985. Breeding by an Orkney Hen Harrier on the Scottish mainland. Scottish Birds 13: 186-187. Potts G R 1998. Global dispersion of nesting Hen Harriers Circus cyaneus; implications for grouse moors in the UK. /bis 140: 76-88. Redpath S M & Thirgood S J 1997. Birds of Prey | and Red Grouse. London Stationery Office. Scottish Raptor Study Groups 1997. The illegal persecution of raptors in Scotland. Scottish Birds 19: 65-85. Simmons R E 1988. Food and _ the deceptive acquisition of mates by polygynous male harriers. Behaviour Ecology Sociobiology. 23: 83-92. Simmons RE, Barnard P, MacWhirter B & Hansen GL 1986. The influence of microtines on polygyny, productivity, age and provisioning of breeding Northern Harriers: a5 year study. Canadian Journal of Zoology 64: 2447-2456. Stott M 1998. Hen Harrier breeding success on English grouse moors. British Birds 91: 107-108. Watson D 1977. The Hen Harrier Poyser. | Berkhamsted. Revised manuscrip accepted October 1998 | | | Scottish Birds (1998) 19: 299-306 SHORT NOTES Raven nest on hydro-electric dam The Raven Corvus corax is known to nest on buildings, usually ruined, but there is only one record of anest onadam. This was on Glascarnoch Dam, Ross-shire, in the years 1970-73 (Ratcliffe D 1997, The Raven, Poyser. London). Inlate April 1996, | was told by the local landowner that a pair of Ravens had become very aggressive towards larger birds, such as Buzzards Buteo buteo and Common Gulls Larus canus flying over Errochty Dam, Perthshire and very vocal when people were walking the roadway surmounting the dam. Investigation of the down stream side of the dam showed the ends of afew sticks projecting from the recesses of a ledge close to the top of the dam, and just below the roadway. Apparently identical ledges are found between Short Notes 299 the buttresses across the face of the dam, apart from the central over spill facility. Each ledge is widest at the buttress, narrowing to virtually no ledge at the centre of the inter buttress space. The particular ledge used by the Ravens was not the highest. The height of the dam decreases markedly towards the ends, as the dammed valley is steep sided. The ledge was just above the level of the tops of conifers growing on the slope below the dam. The parent birds perched on the tops of these trees, or on the dam parapet, calling loudly, while | was in the vicinity. | estimated the ledge to be 25-30m above ground level, and 2-3m below the parapet. Itwas impossible to see the actual nest, due to the lie of the land below the dam. The Ravens raised 3 young in 1996. They used the site again in 1997, but the result is not known. | thank Wendy Mattingley and Dr Mick Marquiss for helpful comments on an earlier draft of this note. Euan D Cameron, 3 Stormont Place, Scone, Perth, PH2 6SR Accepted March 1997 Zs y Zee» ae Cfle-. -#f ee Lg Juvenile Raven poo shor David Mitchell 300 ~=Short Notes SB 19(5) Last breeding by native Red Kites in Scotland The Red Kite Milvus milvus was found throughout much of Scotland during the first half of the Nineteenth Century, being present in all of Buchanan White’s faunal districts except for the north and west coast archipelagos and east Dumbarton (Holloway 1995, Historical Atlas of Breeding Birds in Britain and Ireland 1875-1900, Poyser). It suffered heavily from persecution by both game preserving interests and collectors and, by the 1860s, it was confined to Inverness- shire, Perthshire and Aberdeenshire as a regular breeding bird (Gary 1871, The Birds of the West of Scotland, Murray), although still present in Ross-shire at this time, where Harvie-Brown and MacPherson (1904, A Fauna of North-West Highlands and Skye, David Douglas) had arecord ofa gamekeeper taking a clutch of eggs in 1883 and another record from 1881. The last proven breeding in Scotland was in Caithness in 1884; after this date it may have bred in Skye until 1886 although there is a very dubious record from Glen Garry (Inverness-shire) in 1917, where it had been seen in 1882 (Baxter & Rintoul, 1953, The Birds of Scotland Oliver & Boyd). Major William Stirling of Fairburn (1858-1914) put together a substantial egg collection between 1895 and 1910, the eggs coming mainly from his own estate in Easter Ross and from neighbouring districts. In 1983, Captain Roderick Stirling, the collector’s grandson, generously donated the substantial collection and documentation to Inverness Museum and Art Gallery. The collection has previously been referred to in McGhie 1994 (Discovery of the first British clutch of Slavonian Grebe eggs inamuseum collection, Scottish Birds 17: 166-167), and McGhie and Moran 1996 (Probable first breeding record of Brambling in Britain represented in a museum collection. Scottish Birds 18: 248- 249). In the course of work on the Stirling collection we came across a clutch of 2 eggs with a slip of paper which contained the following information: “Locality and Situation of the Fish tailed Glede On asummit in the Corry Done Craig, Kintail, Ross Shire. Date April 16" 1898”. The eggs measure 56.20 x 47.65mm and 54.45 x 46.60mm and agree well with descriptions of Red Kite eggs and other specimens inthe collections; Fish tailed Glede is an alternative name for the Red Kite. The name Corry Done Craig cannot be traced with certainty, but there is a Choire Dhomdain (NG9915) in Kintail which is spelt Coire Dhuinnid on a 1905 OS map. The clutch size and date are in agreement with observations from Wales (Birds of the Western Palearctic). The species was at the least extremely rare by the time of the 1904 Fauna and Harvie- Brown thought it extinct; this clutch of eggs therefore represent, the last proven breeding record of Red Kite in Scotland. The last records Harvie-Brown had for the west Highlands were from 1883: a note stating ‘3 eggs. Nest placed on arock called ....., Ross- shire 16'" May 1883’, and a note from Buckley of a pair supposedly nesting on an inaccessible rock in the Loch Carron district in 1881. Kintail is adjacent to the Loch Carron district, and either of these 2 records may even refer to ‘Corry Done Craig’. Nesting on crags was certainly more unusual than nesting in trees but Harvie-Brown and Buckley (1895, A Fauna of the Moray Basin, David Douglas) record 2 instances other than those already mentioned: Harvie-Brown supposed that they nested on a rock at the outflow of Loch Garry and records that Booth had a record of birds | Scottish Birds (1998) nesting on a crag in Rothiemurchus. Harvie- Brown, writing in 1895, did not know of any Red Kites anywhere in Scotland and was saddened by the relentless persecution which they had suffered, particularly at the hands of collectors. Short Notes 301 We thank Captain Roderick Stirling for reading a draft of this short note and for his donation of his grandfather's important collection to Inverness Museum and Art Gallery. Henry A McGhie, Nurse’s House, West Road, Muir of Ord, Ross-shire 1V6 7TD Stephen A Moran, Inverness Museum & Art Gallery, Castle Wynd, Inverness IV2 3ED. Accepted April 1998 Black Guillemot nesting on an operational car ferry Black Guillemots Cepphus grylle breed in a range of natural sites. In addition, they have taken to a wide variety of artificial sites, including holes in breakwaters, harbour walls, piers, wharf buildings, under fish boxes and other debris (Harris & Birkhead 1985, Breeding Ecology of the Atlantic Alcidae, in The Atlantic Alcidae, Academic Press, Cramp _ 1985, Birds of the Western Palearctic Vol. 4). Harris and Birkhead consider that they “virtually always accept artificial sites if those are offered”. ~ On 5 June 1992 while on the relatively short ferry crossing from Tayinloan, Kintyre to the Isle of Gigha in Argyll, | observed a Black Guillemot flying close to the car ferry MV Bruernish. The bird completed one circuit of _ the ferry, then approached the left side from the rear and landed on the moving ship, through a hole that carries the anchor chain, close to the bow. The bird remained on board until the ship approached the slipway on Gigha; when about 300m away, it flew off and to the rear, landing on the sea some 500m offshore. While the ferry was returning to Tayinloan, the Black Guillemot was observed flying around the ferry once it had left the slipway and had reached c300m offshore. As before, once the bird had flown around the ferry it landed through the same hole as previously. On 2 further occasions, when the ferry approached and left the Gigha slipway, this pattern of behaviour was repeated. The continued attraction of the ferry to an individual Black Guillemot raised the possibility of nesting on board. On the return crossing, | examined the area close to the anchor chain hole and discovered a nest with 2 eggs under a green pvc cover amongst a range of winching machinery. Before disembarking, | alerted the crew to the presence of the nest. They were already aware of the nest and told me that the birds had nested in the same place for the previous 2 or 3 seasons but had always been unsuccessful in hatching the eggs. | learned 302 Short Notes SB 19(5) later that the 1992 nesting attempt was also unsuccessful. During its normal operating hours (07.20-18.00hrs), the ferry undertakes 59 return trips each week. It is probable that in any one 20 minute sailing only 15 minutes were available for incubation, and that during the ferry’s turn around period of c15 min the birds were unable to incubate. On the 6 operational days per week, this would limit incubation toamaximum of 4.5 hours between 07.20-18.00hrs, losing 25% of available incubating time during the entire incubation period ie incubation would be limited to a maximum of 75% of the available time. This compares with the incubation taking 84% of available time (Preston 1968 cited in Cramp 1985). However, in this case, either the limitation of time or the frequency of disturbance, oracombination of both, appears to have prevented the eggs from hatching. At the end of the 1992 summer season the Operators replaced the Bruernish with another, larger ferry of a different design. In subsequent years, the birds have not attempted to nest on the new ferry. While this car ferry represents an extraordinary nesting site, it does not appear to be unique. Thom (1986, Birds in Scotland, Poyser) records “a niche on an inter island ferry in Shetland” and Tulloch (1992, A Guide to Shetland’s Birds. Shetland Times, Lerwick). comments “It has even been recorded nesting under a life raft on the stem of a regularly used ferry boat” [The comment was made by Bobby Tulloch| | would be interested to see further information on these records. Chris M Waltho, 73 Stewart Street. Carluke, Lanarkshire, ML8 5BY Revised manuscript accepted May 1998 Hen Harriers cacheing and retrieving prey Many raptors regularly and purposively store avian prey for later use in the breeding season (Cramp & Simmons 1980, The Birds of the Western Palearctic Vol 2 Oxford). Information on Hen Harriers Circus cyaneus cacheing prey is generally lacking and, despite intensive Studies of harriers since 1963, | have only 2 instances of it. On 14 July 1974 at 0957 hours | watched a female Hen Harrier return to its breeding area in Wigtownshire carrying a young Curlew Numenius arquata. She landed on a heathery knowe 100-200m from her nest, and emerged without the prey. She landed briefly at the nest, which contained 3 young about 18-24 days old, before taking off and landing on the moor well below the nest site. Thirty nine minutes later, the female harrier flew back to where she had deposited her prey. She circled there with lowered talons, flew back to her — = | Scottish Birds (1998) _ nest and hovered above it. She flew back to _ the cached prey, retrieved it and brought it _ back to the nest at 1040hrs, leaving with the remains 24 minutes later. _ On 17 July 1974 at 1800hrs the same female _ stood on the moor below the nest. She flew to _ the heather knowe, dropped her talons and again retrieved small to medium sized prey _ from the heather and flew back to the nest _ with it. Despite several later visits this behaviour was not seen again. Short Notes 303 Purposeful storage of ‘surplus’ prey by raptors is usually of prey brought by the male to the female who usually stores it. Interestingly, it was a large prey item, apparently caught by the female herself and then purposely stored by her. Cacheing behaviour becomes more advantageous if the food is less likely to spoil (Heinrich 1990, Ravens in winter, London). In one of the instances the prey was used quickly (within 43 minutes of storage) and in the other it may have been used quickly too. RC Dickson, Lismore, New Luce, Newton Stewart, Wigtownshire DG8 OAJ Manuscript accepted May 1998 Ravens nesting successfully on a specially provided platform In June 1996 | received a report that a pair of Ravens Corvus corax had attempted to nest in the barn of a ruined croft in the West Mainland of Orkney. The croft was situated in an area of grassland used for cattle and sheep grazing. | visited the barn and found the semblance of a nest lying across a roof beam at a height of about 4 metres. The structure consisted of sticks, lengths of barbed wire, part of a lamb skeleton and pieces of wool. The beam was only 5cms wide and the materials were draped over it. It would have been impossible for such a narrow beam to have supported a complete Raven's nest. It then occurred to me that a platform placed across 2 of the roof beams, which were 1.5 metres apart, would make a Suitable base for a nest. | contacted the farmer who owned the barn suggesting that a platform could be put in the roof and he was delighted with the idea and of possibly having a pair of Ravens nesting successfully on his land. The covering of the roof consisted of sheets of rather rusty corrugated iron, many of which were missing and several others were loose. Itwas decided to delay putting the platform in position until after the winter, in case more sheets were lost or the platform blown away. Unfortunately, it was not actually in place until 8 March, 1997, which was rather late for the Raven’s nesting season. However, on returning to the 304 Short Notes SB 19(5) barn 2 months later it was found that the foundation of a nest had been built on the platform but the lining had not been completed. | next visited the barn on 9 April 1998 and, as | entered the building, flushed a Raven from a fairly substantial nest that was on the platform. The bird flew around calling agitatedly so | left as quickly as possible without examining the nest any further. On 5 May 1998 there were 4 full grown young standing on the nest and a week later the young were seen perched on the roof of the barn. Ravens had been observed in the vicinity of this croft several times in the breeding season over the past 10 years and | had been to the site on 2 previous occasions but had not found any evidence of nesting. A pair of Ravens have nested successfully for a number of years, including 1998, on a ruined building 3.5 kms from this croft. | am very grateful to Bob Adam for his help with the building and placing of the platform in position and to Steven Harvey for allowing access to his barn. C J Booth, 34 High Street, Kirkwall, Orkney, KW15 1AZ \ Manuscript accepted September 1998 Proposals to extend the range of the Crested Tit in Scotland, 1945-1955 Historical research has revealed some interesting observations on early proposed nature conservation practices in relation to the Crested Tit Parus cristatus scoticus in Scotland, more particularly in Strathspey. It is apparent from the archives of the Nature Conservancy (NC) Scottish Committee which begin in April 1949, that an application was put before the committee on 7 December 1954 to transplant Crested Tits out of the Highlands to other parts of Scotland. The proposal was conceived by JMD ‘nestbox’ Mackenzie, a Forestry Commission employee, who was known for his nestbox work with ducks and flycatchers (P Sandeman pers comm). He sought to transplant the birds caught in Strathspey to ‘suitable pine forests in Perthshire, Angus and Fife’. Mackenzie’s declared aim was to extend the bird’s range as anature conservation strategy. The NC Scottish Committee seemed confused by this request, and decided to relay it to the Advisory Committee on the | Protection of Birds (SRO:SNH 1-1, Minutes — of meeting of Nature Conservancy —Scottish — Committee, Edinburgh, 7 December 1954). However, the issue remained alive, and the ‘ NC Scottish Committee returned to it ata | later meeting in March 1955. This was the ° first committee meeting attended by JPGrant |) of Rothiemurchus, and he was keenly } interested in a proposal that could see the | transplantation of some of the tits from his |} own estate. The laird was against ° transplantation as a nature conservation practice, but saw some merit in the ‘laudable’ scheme ‘of spreading the risk’. He praised Mackenzie’s vision that could have guaranteed ‘more areas that this rare bird frequented’, thus ensuring there was less Scottish Birds (1998) ‘chance of it being wiped out by localised disturbances’. The feeling of the Scottish Committee was that this application could not be supported, but no firm reason was given (SRO:SNH 1-1, Minutes of meeting of Nature Conservation — Scottish Committee, _ Edinburgh, 3 March 1955). This episode is made all the more interesting by the fact the Original application was probably drafted in _ Autumn 1954, just a matter of weeks before the new Protection of Birds Act passed into _ lawon1 December 1954. This was a powerful _ piece of new bird protection legislation that | repealed 15 previous Acts and offered full protection to all wild birds, their eggs and nests, with some limited exception relating to game and vermin species. This discussion also came just after the declaration of the Cairngorms NNR on 9 July 1954, within which Rothiemurchus Estate, and thus its Crested Tits, was a key component. Information taken from other archives regarding the Crested Tit | in the 1940s provides some background to ' this 1950s debate. The second half of the 1940s saw an intensification of the ' Cairngorms National Park and/or Nature ' Reserve debate, and the Crested Tit had a | Committee, | Conservation Committee (SWLCC) of 1946- | role to play in that. Baxter and Rintoul wrote in December 1945 to Joseph Westwood, the _ Secretary of State for Scotland, urging him to Support the Nature Reserve ideal, adding | that the Crested Tit demanded ‘special care’ | because it was ‘so typically Scottish and so _ very local in their breeding area’ (SRO: FC9/ | 3, Letter, dated 12 December 1945). The body that preceded the NC Scottish the Scottish Wild Life Short Notes 305 1949, confronted issues surrounding the protection of the Crested Tit on a number of occasions. The most controversial event took place in 1946, when the SWLCC became angered on hearing that PA Clancey had been collecting Crested Tits in Strathspey, and was then taking them away ‘for research purposes’ (SRO:FC9/2, Minutes of meeting of SWLCC, Edinburgh, 30 September 1946). This was Philip Clancey, a worker in the museum service in Scotland and later in Durban, South Africa, who was known to have done taxonomic work on the races of some passerines (RH Dennis pers comm). This perhaps explains his interest in the Crested Tit, which we must assume he was taking away to skin and/or stuff. Clancey later built up a large collection of African birds in Durban museum. A year later, James Ritchie reported to the SWLCC that some felling of pinewoods at Rothiemurchus was continuing, and that ‘adequate provision must be made for the protection of the Crested Tit’ (SRO:FC9/ 3, Minutes of meeting of SWLCC, Edinburgh, 29 September 1947). The harsh winter of 1947 then adversely affected the Scottish tit population, although in the 1950s the Crested Tit did expand its range into Culbin Forest. Primary sources SRO: SNH are the archives of the Nature Conservancy and Nature Conservancy Council in Scotland, held in the Scottish Record Office, Edinburgh. SRO: FC are the archives of the Forestry Commission in Scotland, held in the Scottish Record Office, Edinburgh. Robert A Lambert, Institute for Environmental History, University of St Andrews, St Johns House, South Street, St Andrews, Fife KY16 9QW. Revised manuscript accepted August 1998 306 __—Letters to the editor SB 19(5) LETTERS Numbers of Turnstones and Purple Sandpipers on the East Lothian Coast In H E M Dott’s paper on declines in Turnstones and Purple Sandpipers in S E Scotland (Scottish Birds 19:101-14), there is a table with the maximum counts of the latter Frank Hamilton, 23 Campbell Road, Longniddry, East Lothian EH32 ONP species at Gullane Point - Hummel Rocks in EastLothian. Forthe record, Keith Macgregor and | had peak counts at this site of 20 in 1952-53; 24 in 1953-54; and 20 in 1954-55. These counts are much in line with the last 8 winters on Table 2 (p 103), made between 1987-88 and 1994-95. It is nice to know we were not just enjoying ourselves inthe 1950s. Received February 1998 Comment by H E M Dott On 31 January 1998, a further low tide count was made of all shore birds from Gullane Point to the Scotland/England border; the same coast with the same methods and by mainly the same people described in Dott 1997 (Scottish Birds 19: 101-104). Results obtained for the species concerned were: H E M Dott, 114 Comiston Road, Edinburgh EH10 5QL ‘ Turnstone 264 (East Lothian - 233, Borders - 31) Purple Sandpiper 81 (East Lothian - 66, Borders - 15) This again shows continuing very low numbers in the 1990s compared to the much | higher numbers in the 1970s and 1980s, as : previously described. Received July 1998 ik | n ” Scottish Birds (1998) 19: 307-311 OBITUARIES Valerie MacLaren Thom 1929-1998 - Valerie Thom made a massive contribution to natural history in Scotland. She will be best remembered, probably, for her Birds in _ Scotland; recognition tends to go to major _ publications of that kind, as it also does to _ large personalities and high profile projects _ and research. But the remarkable extent to which the amateur (and professional) study of wildlife in Scotland has expanded and flourished ' since the War could simply not have happened without the largely unseen and wholly unsung _ infrastructure of committees, councils, record keeping, policy making, financial accounting, | journal editing, collating of information, and organising of volunteers. Itis in all of this that she had an outstanding role, and made _ perhaps her largest contribution. | Bornin Tynemouth, Northumberland in 1929, _ Valerie graduated BSc in Agriculture at | Edinburgh University in 1949, taking the first year medal. She went to work in the East of Scotland College of Agriculture and earned ‘high professional respect for her work | connected with the quality of milk. In 1969 she changed direction and joined the | newly established Countryside Commission | for Scotland where she worked for a dozen | years before taking early retirement. Here | she was involved in pioneering work in countryside interpretation in a team led by | Don Aldridge who defined it as ‘the art of | explaining the significance of a site to the | people who visit it, with a view to pointing a _ conservation message’. 307 Though taken for granted nowadays, this was then a novel concept, originating in America, combining information, enjoyment and a homily. It involved educating not only the public but also the agencies, local authorities and voluntary bodies providing much of that education. Valerie worked tirelessly to set and raise standards of presentation in texts and pictures, and especially in the spoken word of the new ranger services, ata time when visitor centres and other facilities were springing up all over the country, not always ideally located or focussed. The experience she gained in this was invaluable to the fledgling Scottish Wildlife Trust of which she chaired the Perth branch, and which was setting up interpretive services at its reserves at the Falls of Clyde and at the Loch of Lowes with its second most famous pair of Ospreys. Not that she ignored the most famous pair, lending a hand during a number of summers at the Royal Society for the Protection of Birds Loch Garten reserve, relishing the sights and sounds of the neighbouring ancient pine forests. She helped the RSPB in other ways too, serving on its Scottish Committee with advice which her colleagues remember as always wise and carefully thought through. This thoroughness also served well the bureaucratic complexities of the Secretary of State for Scotland’s Advisory Committee on Birds, and the wide-ranging collaborative research projects overseen by the Council of the British Trust for Ornithology, on both of which she sat in the 1970s. Earlier, she had been Secretary and Chairman of the Ornithological Section of the Perthshire Society of Natural Science from 1965 to 1972 and had found, or made, time to be the 308 Obituaries SB 19(5) Wildfowl! Count organiser for Scotland for the Wildfowl and Wetlands Trust from 1963 to 1970. This involved the pleasure, keenly felt, of counting the geese on a frosty, clear winter day at her local Dupplin Loch near Perth, but also carrying through the counts in foul weather as well, and meticulously correlating the efforts of many others. They soon learned that she was stickler for detail. Everything had to be right. As it had to be, too, in her work for the Fair Isle Bird Observatory Trust. To use her own word she became ‘addicted’ to the island when she worked there for a season in 1955 as the observatory’s first assistant warden. For 23 years from 1970 she was one of its Trustees and for six of them, from 1986 to 1992, its Honorary Secretary. What a deal of work there was in that, and what a valuable job she did in acting as a link between a scattered Board of Trustees and the observatory staff, with her great knowledge of the island she loved and its community. She was able to make good use of this knowledge in drawing together the work started by George and lrene Waterston in gathering material for a history of Fair Isle and which she completed in Fair Isle, An Island Saga, published in 1989. Throughout all these years, Valerie was an extremely active member of the Scottish Ornithologists’ Club, serving on and chairing several of its committees, and in 1978 she was elected its President. These were good years for the SOC with flourishing branches, revenue surpluses, bookshop sales rising by 25% or more a year, and no hint yet of the maintenance problems at 21 Regent Terrace. In her crisp and business-like way Valerie chaired Council meetings which galvanised the Scottish Bird Report (running three years late when she took over) and pondered such issues and threats to bird life as the Torness power lines and the Lurchers Gully skiing | development. In her last year as President, Council took the decision to go ahead with a new book onthe | status of the birds of Scotland, which she volunteered to write. Thirty years after The Birds of Scotland by Dr Baxter and Miss Rintoul (to this day scarcely anyone refers to their first names) the ‘good ladies’ mantle > would be handed down to Valerie Thom ina wholly appropriate way. The new book was |. partly financed by the entire remaining funds | in the Baxter and Rintoul Trusts. 4 It was an enormous task to undertake and it 5 took five years. During almost all of this time ; Valerie also edited both Scottish Birds and | Scottish Bird News, a formidable and unenviable combination. The new book drew on the willing assistance of many members of | the Club, both in providing and dealing with material; some 150 of them are listed in the acknowledgements. So itwas acollaborative | venture from the start, but at centre it was Valerie's achievement. Birds in Scotland — was published in 1986 and is her enduring memorial. In that same year she was elected Honorary President of the SOC. These were some of her public achievements, but Valerie was a very private person, reserved and not easy to get to know. She | was impatient of small talk and could be nippit with people whose standards of | commitment and performance were less | exacting than those she set for herself. Some | people found her slightly daunting. She was, though, a deeply compassionate and caring person, drawing strength and conviction from her Christian faith. Her minister in Perth recalls her saying to him: — — KF — Exe = Scottish Birds (1998) Obituaries 309 “I’ve been given this gift of faith. | know | have certain talents. | have the energy. | want to do things for my Lord.” Ordained as an elder inthe Church of Scotland, she took a two year theological course and cared faithfully and patiently for the people in her district and also cared, in different ways, for fellow believers in the former Soviet bloc countries. She devoted a great deal of time to Manna House in Perth, a coffee house and meeting place whose Christian witness proclaims the Gospel in the market place, and to establishing a pastoral care group. She had a great love for the Scottish countryside, its wild places and its birds. Perhaps both her joy in them and her concern for them are summed up in two verses from her poem published in Birds in Scotland: Valerie Thom pictured cutting the An eagle soaring high on a thermal, A peregrine swooping swift to the kill, Divers wailing on lonely lochans — Such sights and sounds have the power to thrill. The “everyday” birds such as Wren and Robin And the rarities blown here from distant lands All have their place in God's great pattern — But the future for many will lie in man’s hands. John Arnott cake at the launch of Birds in Scotland in Glasgow, 1986 Op eS EE SSS Ss eae EER eeeeeeESEEeaeaeresee_or_a 310 Obituaries SB 19(5) at John Morton Boyd 1925-1998 Morton Boyd, who died on 25 August 1998 at the age of 73, was one of Scotland’s best known ecologists and wildlife conservationists. He was born in Darvel, Ayrshire on 31 January 1925. He attended Kilmarnock Academy and then went to Glasgow University to study engineering. However, after one year he read Frank Fraser Darling’s book A Naturalist on Rona (1 939), and as aresult he changed from engineering to zoology. Service in the RAF interrupted his education but he graduated in 1953, obtained a PhD degree in 1957 and a DSc in 1964, followed eventually about 30 years later, again at Glasgow University, by a D Litt, a very rare recognition for a scientist. He was also awarded the Neill prize of the Royal Society of Edinburgh. Morton was a broadly based naturalist and, though his main interests were birds andmammals, he studied earthworms on Tiree for his PhD, the start of a long affection for this island. The great diversity of his interests established a fervour for ecology and land use of the Western Highlands and Islands. In all these topics he was a great communicator; to the end, however, Morton hadn’t quite mastered the art of sharing conversation and seldom used one word when 10 would do. He was along standing member of the Scottish Ornithologists’ Club. He made a substantial contribution to our knowledge of the birds of Scotland, notably by his studies on St Kilda. The report Birds of St Kilda (Harris & Murray 1977) lists 11 papers of which Morton was author, including 2 important papers on seabirds. These were Distribution and numbers of kittiwakes and guillemots (British Birds 1960) and The gannetry of St Kilda (Journal of Animal Ecology 1961). Fieldwork i for it began in May 1955 when Morton, in a motor drifter, sailed close by Stac Lee and Boreray, allowing him his first impression of the immense scale of the cliffs and the great numbers of birds there. He returned the following year with an expedition from Glasgow University. On 13 May 1959 the RAF photographed the entire gannetry from the air, with Morton together with Derek Ratcliffe and David Boddington watching from the top of Stac an Armin. Thereafter, Morton's task was to count the Gannets on the photographs, and he did this 4 times! Then he divided the gannetry into 672 sections, assessed the overlaps between pictures and concluded that the population was 44,500 pairs. Diagrams of the main counting areas and the details of the actual counts are given | in the paper which became a benchmark. It } was almost 30 years before the St Kilda — Gannets were counted for a second time. Morton was one of the first scientists to be recruited to the Nature Conservancy in Scotland and he was appointed in 1957-68 | as regional officer in charge of wildlife | conservation in the west of Scotland. He was instrumental in establishing several national | nature reserves and will long be associated | with the Grey Seals of North Rona as well as | the sheep and Gannets of St Kilda. The 1950s and 1960s were happy days for ecologists in Scotland; those on the staff of the Nature Conservancy were able to indulge their interests virtually unhindered. Morton led many expeditions to his favourite islands }) at home and abroad, and initiated long term) studies on the sheep on Soay and Boreray and the seals of the Hebridean Islands, often in very difficult field conditions. With his family, he spent many idyllic holidays on Tiree where, | from his deep seated interest in wildlife, land use and social customs, he gained the’ impetus to write 4 books on the Hebrides, one co authored with his son, lan. Morton, i | i Scottish Birds (1998) Obituaries 311 also worked with Ken Williamson with whom he co authored St Kilda Summer (1960). In Morton’s obituary in the West Highland Free Press (11 September 1998), Brian Wilson records how Morton and Williamson prevented the army, set on making a road, from demolishing the remains of the row of houses on Hirta. This would have been ‘an act of almost unbelievable vandalism’. Morton was supportive of a range of investigations in the north west and notably encouraged Jim Lockie in early work on Pine Martens, Niall Campbell with a study of Sticklebacks and Dick Balharry’s studies of Golden Eagle and Pine Martens. lan Newton remembers with gratitude Morton’s encouragement of his early work on Greylag Geese at Loch Druidibeg. In his appreciation in The Scotsman (10 September 1998) John Francis, Morton’s successor, mentions the reintroduction of the Sea Eagle as a part of Morton Boyd’s legacy. Francis recalls Morton’s stories of the early days of the Loch Ewe partnership, and in particular those about a Red Deer stag called lain which ‘are among the richest brand of humour imaginable’. This partnership culminated with the first successful deer management group. Morton was Assistant Directory (Conservation) of the Nature Conservancy _ (Scotland) in 1968-70 and became Director of the Nature Conservancy (later Nature Conservancy Council Scotland) from 1971 to 1985. He took over in a period of relative calm but shortly the peace was to be shattered. Suddenly the Nature Conservancy as designed by Max Nicholson, with national headquarters in London, Edinburgh and _ Bangor and research stations at Monkswod, Merlewood, Furzebrook, Norwich and Banchory, was divided. Hitherto, conservation managers and research workers had shared Office accommodation and this synergy stimulated dialogue and worked well. In 1972, following the Rothschild report on the organisation of government science, the Nature Conservancy Council was formed. The research arm was removed from the Conservancy to the Institute of Terrestrial Ecology and, as aresult, Morton lost authority over research staff in Edinburgh. Shortly afterwards, decentralisation from Edinburgh took place with regional officers outposted to local offices. The reorganisation demanded considerable resilience and Morton thrived on it. The implementation of The Wildlife and Countryside Act (1981) closely monitored from NCC’s Peterborough HQ and reinforced by voluntary organisations and parliamentary scrutiny, disrupted the non confrontational style that had characterised NCC’s relationships in Scotland with the powerful lobbies of landowners, foresters and farmers since the 1950s. In this new climate, some of Morton’s dedicated local staff, supported directly by NCC’s largely English based Chief Scientists’s Team, were able to achieve significant protection for important wildlife sites in the face of the kind of development pressures before which the NC had caved in during previous decades. The first half of the 1980s saw notable successes for conservation across Scotland: in Islay (with its goose problems); in the Northern Isles; in Speyside (Creag Meagaidh, the Cairngorms northern corries and Abernethy); and in the bird rich Flow Country in Caithness and Sutherland. Many of these victories were achieved in the face of opposition from more traditional sections of the Scottish land owning community and perhaps from the Scottish Office itself. Nothing demonstrates better Morton’s character and his commitment to nature conservation than his reluctance to bow to these pressures. His steadfastness and that of his staff in circumstances in which 312 Obituaries SB 19(5) they were occasionally subjected to personal abuse safeguarded numerous important wildlife sites and secured long term and positive shifts in attitudes to conservation in Scotland. However, there was a price to pay for the period of conflict that was necessary to secure this legacy. Shortly after Morton’s retirement in 1985 and the subsequent 5 year period under the leadership of John Francis, NCC was dismembered as a Great Britain wide body leading, with its amalgamation with the Countryside Commission for Scotland, to the creation of Scottish Natural Heritage which was under Scottish control. SNH is a very different kind of organisation; it espouses access and partnerships, not the strenuous defence of nature conservation interests, and has been known to attack the scientific emphasis of its predecessor. Nonetheless Morton, said by Brian Wilson to be a devolutionist, accepted these political changes and worked positively with his successors for wildlife conservation in Scotland. Morton and John Francis were the last scientific directors of nature conservation in the Berry/Eggeling tradition and, although the organisation which Morton led for 14 years scored major successes in a hostile political environment, it paid a heavy price for its achievements. Morton travelled widely, both as a professional ecologist in the 1960s and also more recently leading safaris and lecturing on cruises, notably those organised by the National Trust for Scotland to the outer isles and by Swan (Hellenics) Ltd and Serenissima Travel Ltd to the Mediterranean Sea, Indian Ocean, Indonesia and Africa. After retiring, when his contributions were acknowledged by his CBE, Morton seemed to live anew. He became ecological consultant to bodies such as North Scotland Hydro Electric and to the Forestry Commission, and worked with the National Trust for Scotland, as councillor for the Royal Zoological Society of Scotland, editorial consultant for Edinburgh University Press and for Mirror Publications Ltd, and became a vice president of the Scottish Wildlife Trust and the Scottish Conservation Projects Trust. He was committed to the Saltire Society and an active committee member of the Royal Society of Edinburgh, and a trustee of the Hebridean Whale and Dolphin Trust. He was accomplished in water colours and, by coincidence, he completed the design of the Colinton Kirk Christmas card on the Saturday before he died. In spite of all his outside interests, Morton had at heart 2 outstanding passions, his | family and the community of Colinton parish kirk. For 40 years he was an elder. He was | devoted to his wife, Winifred, who suffers | greatly from arthritis in her hips, and he was | very proud of his sons, theirwives andchildren | and also activities. They were a constant | source of support. | In 1997 Morton agreed to be Patron of | Dunbar’s John Muir Association. Muir and | Morton, both sons of Scotland, were 2 of a kind. They were free spirits and expressed with frisson the causes they chose to | champion, a course that had its dangers. | Few who knew Morton will forget his | contributions and his energy, charm and | affection, an enduring and endearing ; testimony. | David Jenkins |) The author thanks for their contributions John \y Forster, Mike Harris, Fred Last, Jim McCarthy, Sarah Wanless and the authors of published|, obituaries cited in the text. | 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 will be 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 cn 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 happy to accept papers on computer discs; however, please State the type of word processing programme used. Contact Sylvia Laing on 0131 556 6042 if you wish further information on this. 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 Vol 17:146- 159). 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 oth (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 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 should be on a separate sheet. Maps and diagrams should be either good quality computer print outs in black and white (please do not use greyscale shading) or in black ink and be camera ready, but drawn so as to permit reduction from their original size. Scottish Birds Volume 19 Part 5 December 1998 Contents Main papers North east Scottish counts of Goldeneye, Goosander, Red-breasted Merganser and Cormorant in 1944-50 compared with 1988-97. A Watson, M Marquiss & P J Cosgrove 249 Amendments to the Scottish List. R W Forrester, SBRC 259 Winter habitats of Twites in Scotland. H Clark & RM Sellers 262 Movements of Twites in Scotland. H Clark & R M Sellers 270 Distribution and foraging habitat preferences of Choughs on The Oa peninsula, Islay. M Madders, F M Leckie, J Watson & C R McKay 280 Orkney Hen Harriers: a major population decline in the absence of persecution. E R Meek, G W Rebecca, B Ribbands & K Fairclough 290 Short notes Last breeding by native Red Kites in Scotland. H A McGhie & S A Moran 300 Black Guillemot nesting on an operational car ferry. C M Waltho 301 Hen Harriers cacheing and retrieving prey. R C Dickson 302 Ravens nesting successfully on a specially provided platform C J Booth 303 Proposals to extend the range of the Crested Tit in Scotland, 1945-1955. R A Lambert 304 Letters Numbers of Turnstones and Purple Sandpipers on the East Lothian Coast. F Hamilton 306 Comment by H E M Dott 306 Obituaries Valerie M Thom. J Arnott 307 J Morton Boyd D Jenkins 310 Advice to contributors Inside back cover Published by the Scottish Ornithologists’ Club 21 Regent Terrace, Edinburgh EH7 5BT. © 1998 HECKMAN BSN GD ME “RAY: TENEG: Bound-To-Please® JUNE 00 N. MANCHESTER, INDIANA 46962 * - ers TENE APE Ra eM Mite TN Sen DIR Ee ie 4 = PERIIAANI a nat ag t Sat geen eae RL at ge Beta NCeMSIING re ot es Lehi ve . . : Bee Ae ype ews ee re ern eee) Pte pm were tng DES PARES win mln we CY a ee We MEN © Se = ae eee SPP eran : : x ~ CTP Sep ape Seles meas sey RYN PEE eee erererr Ce PPP PS AE oe Un Ce eT Whee ss TEAR ORERS EERSTE Bn PG eS NEB Ra PPPOE ER OLR ait ey fase ea ~oe — : ’ OR RN me ely RAR ea SN tp hte et eee tht Tale ae th Sete yokes Oo MeN m eee i ee Ce ae ee ALA EME ERIBL A. gods nea ee hetete WWOd AR 2 z ‘Ow PE Sethe p an a ARTI IRV AIF okiipvare ton oy whe Hips kee hy Oe oe ee a ee ewe ~ ee ee - ne Sn Oro crt ce 4 . ~ nee, = 5 \ avon ye Saes Peta as FOUN RB ete ee bi A ~ Ss eg ete ale . y Neu naa ons SOO TORE ISRO ADL OP ESO led dk UOE ee PROP MEAA STEM OE ARMM SBD cp vip get iad er an ty CN Cie Tae ae kee OY ey ee ee mags Senta ee mea ape ny, Lp Nae PIL ALN SERN ILE BP MTIS BERS OF RO eed . Peary , : Astisny LLLP OEE AA EEA GE FV OYE BREE MED AEE Ma Newey DUP MN AFR ET UAE SEOs ape MOLE Ws, NAF 4 I DBY ve ee a toe = hese eae Ste et “ae VS REN CRS Aone a ea - en Oe Seep ae 2 Aw SPR AMON DH et ete hey : ee A Ng ab N ae rey eSE 2 Soxtheny » WEV ae sryhar er arvas SEP Ue “? _ 72 “ Tt ee ee TOP DIMI 8 6 PAP RTI DED BARRA freer ere MARS “ut Ad Pome VOOMES Nw ew ay CNSR eS gre ere gen el age Re age wae : a ee Wa 40% ta : WI OF e0 feyn ne \aoheeey a 5 Seb hapete see AND we ay reaees Orpen ee ere AS Rae Ore age ae Ry ae x eee Rinee evant Sen erg Rete She seteensedcenee we-qhanegetectent se pe nee ae halite heii RDO a), WEED ena ea eegapeieie WAL Oe oN oP rete ee see ~~ e , “ee aan ee Set Ae ae ee So Se8 Se qQay PANO Alene D eens AAEM PON AGE eS AD MpR ee ay SVEN EE WAS iy Vp ee Ly pap nN, ieonan tw Ptah Ae Ahr Peers PER GD bP L AEM awe PREP CUA ee ste apy oy é SAT ee ry, ae eee | PASAT ASS SAME RS Qk am “s rer o SN ENTE AIUD thy id OEE ROSOR ae Teed ey SAP oe Ny SATIS apes Som km ater 2 ‘ ; Sa NMRA IS Ngee Se 8 FEARS Bw EL Oey eS” er i Phra SoS ET ER TPP ES ONES ASRS Retr Rr YS WO/ED se. Ueean oe ne Soy WTO APS Smee Te ace nye the et di seuena or Senor tan anes cae BNE C BREESE PATS OSEMCSE NO PLP Tomy RBAIW 1X a PONCE ne wk ae COUPE ACR, Wak tm, AAA AO aes Sy teB ie eae Hy batt et ae lgo pam rant ms Meena aca a NE RZAUS TAB OBES wx PETA EDR PANES ee Apmis eORP Ree Sah er AAS MOP» a ; ADIN TP PAN A me Oty ay DEN ” SSEER IA a em Comm ema She ghey PNW S? OE AeN FAY Ta AgD aE Ht YER OR Os enn Lac et MNO R SOFAS peercewe ae = eee nLee Peed 4 7 ” POPPE I LO ey eee hays teeta wee SP SAME AWN P awa per ces Yh eA MD CORE Lh el es pee TANCeQeNy= sane. tee a tn mirinoren tnd otha het CPi a AS ACAR SCAT LINBYS LR RABTALA UA RCA ne vo Paya i hes wety PAPE CASS AA 22 Vb Oe ae APN Arapeta POS eras ENPRA A APRS SAHRA K OTROS a peg WANE AEM E EURO NICE KAGAN LRIYAR OD OM roy SON we eee the auhepaess eh) DP APR ETE AG eee Mow ey nit 4 Ee Hee peiricinirt y GLa w Errore La ee ae oreerye PAPE a he VTE ADNAN R DD ove" Ae we ASS ate pe tbewee ag ee bey SOO eee ey Seen PAPE ee ee Gaps VeRMP ay Wet NY Ure ts Te eae eee See eee er ENDED DRE REO RSAT PAGO VERSUS PNPM CPN Doky A SS URama carer Losen (rer paras ~ Ue es Gase Tee SETS nas ortho COPE Teay LAP Geet ete LEEPISSEMEND ECW at Sesh nah 8g ie CO RAD INE #9 Mt CaP LAP LIAY TP wey NAAM eSeAW aU OoNRES PID age aera sacapac’ oonaranee . SONGS A, ESSER Oe RGOTONSTE Se, “ue NB BCID Sa FP me Pe Pas Naa gne h VIAPN Fwy Pav abyeen *yecese 2 ERE NESTENE NG 882 UR NP QION TED Vt MINES UE SYNE Vey & HERRERA RG MR SE) emp RSQ We ehey eS egeas meen Rewer ed keene Pa se ps, So SME FE WA Ry npeay erm HO ese rieeh ter ap inthe ed alaay ee fie Svar SUELO VENEER ND Ves TP OD RASRAD ENE AY Or 8 Gy San NUEO ORY Co Non ‘ MENSA PADS SAA RP 8 OBS SAD WMO OPES EHS ORM Qh Ob ere NORTON NS Went qe te Ra poe) Pe hy a PUP APW ADO ee . Aut nee SN SSPRPUPADAS OAPUD ADE STW AEN ENTE NEN LAIRD EE A EDET ND ERR TD PASE SRNL OE rae se yeeee, Ad ee a WOM TAN Sp dy g SAP BSS Or USO) dt UR Vie eS ey VAP aA Aer POPPA “O89 8 wea pes. PO Ne PPS de tal edicts tena. ht hed beatetcorteniad o'\VO? "pag arreseqptenns. ca... har (hn lo EEC) » eta ei min oe SOND bee Beer Pyne 20 8)* 2 koe Per Parheve ub wus <0