AUSTRALIAN INSTITUTE OF MARINE SCIENCE MONOGRAPH SERIES Volume 2 A REVIEW OF THE PHYSICAL OCEANOGRAPHY OF THE GREAT BARRIER REEF AND WESTERN CORAL SEA by G. L, Pickard F. Rougerie i a ee tel Tt a - AUSTRALIAN INSTITUTE OF MARINE SCIENCE MONOGRAPH SERIES Volume 2 A REVIEW OF THE PHYSICAL OCEANOGRAPHY OF THE GREAT BARRIER REEF AND WESTERN CORAL SEA G. L. Pickard Institute of Oceanography University of British Columbia Vancouver, Canada with J. R. Donguy C. Henin F. Rougerie Section d’Occunographie O.R.S.T.O.M., Centre de Noumea New Caledonia Australian Government Publishing Service Canberra 1977 Australian Institute of Marine Science Private Mail Bag No, 3. M.S.O. “Townsville, Qld 4810 ISBN 0 612 02635 1 © Commonwealth of Australia 1977 Printed by The Dominion Press, Joseph Street, Nth. Blackburn, Victoria. Contents PREFACE . N vil ACKNOWLEDGMENTS . Vili PART 1—THE GREAT BARRIER REEF G. L. Pickard I INTRODUCTION : . 1 II TOPOGRAPHY y Bo, 2 III CLIMATE 6 Winds 6 Cyclones : 7 6 Air temperature ia 8 Rainfall . ' 10 Mean annual rainfall : 11 Rainfall variability 11 Comparison of rainfall on land and on islands i 12 River runoff 13 Cloud and humidity —, oe 16 Cloud amount . ; ; 16 Humidity... ; : 17 IV TIDES Ys. = 19 Tide levels . a” . ‘ 19 North Queensland zone ihe 19 Mackay zone A gh y . . 21 Torres Strait .. iste ae — : oR Mean sealevel .. tee 23 Annual cycle... Pe 23 Mean sea level and latitude : 2 h 23 The age of the semi-diurnal tide a ks . 24 Tidal currents .. .. ae : . 24 North Queensland and Mackay zones... , 3 24 Torres Strait oom oe . 24 Vv WATER PROPERTIES IN THE LAGOON ...__.. + es a 25 Temperature . oe 7 aE. 25 Surface temperature at specific locations... . ’ : 25 Surface temperatures along the lagoon .. F 29 Sub-surface temperatures ... Co er . 31 Salinity Surface salinity at a specific location Surface salinity along the lagoon Surface salinity gradients across the lagoon Sub-surface salinities Dissolved oxygen Other chemical properties ‘Transparency Density Properties in other areas New Caledonia barrier reef lagoon Tropical estuaries VI TS CHARACTERISTICS OF SURFACE W A'TERS T,S diagrams T,S, time diagrams VII WATER PROPERTIES OUTSIDE THE REEF Individual properties outside the Reef T,S, time characteristics outside the Reef ‘Temperature and salinity variations across the lagoon— Further remarks VITE Currents Great Barrier Reef Lagoon Other Lagoons Bikini and Rongelap Atolls (Marshall Is., approx. 11.5 N, 166.0 E) Ouotoa Atoll (Gilbert Is., 1.8 S, 175.5 E) Addu Atoll (Maldive Is., 0.6 S, 73.2 E) Rangiroa Atoll (Tuamotu Is., 15.0 S, 147.5 W) Fanning Atoll (Line Is., 3.8 N, 159.3 W) The barrier reef lagoon off Nouméa (New Caledonia, 22.3 S, 166.4 E) IX SUMMARY 40 Ill IV PART 2—THE WESTERN CORAL SEA G. L. Pickard with FR. Donguy, C. Henin & F. Rougerie INTRODUCTION TOPOGRAPHY CLIMATE Winds Cyclones Air temperature Rainfall Cloud Willis Island WATER MASSES, PROPERTIES AND DEDUCED FLow PATHS Introduction—-salient characteristics Surface water characteristics Surface temperature—mean distribution Surface salinity—mean distribution Mean seasonal variations of surface temperature and salinity by areas Sources of surface waters in the Coral Sea~—Rochford Surface waters—Scully-Power Other surface data in seasonal sequence Speed of movement of water masses—Rochford Mixed layer depth Subsurface water masses and flow patterns Introduction Isentropic analysis Rochford. The 27.20, surface The 25.00, surface Core layer analysis—Wyrtki Continuity of water masses along the western boundary of ake’ Tasman and Coral Seas—Rochford Core analysis—Scully-Power Subtropical Lower water (upper salinity miaximitn) Antarctic Intermediate water (intermediate salinity minimum) Upper oxygen minimum Oxygen maximum Isentropic analysis—Rougerie & Donguy ‘Gorgone 1’ cruise 1972, late winter ‘Gorgone 2’ cruise 1975, late summer Consistency of subsurface water characteristics ~~ Wm UI to Vv CIRCULATION Introduction Surface circulation W'yrtki, 1960, 1962b— Current atlas and geostrophic circulation Rotschi, 1958-61—Earlier O.R.S.1.0.M. cruises Takahashi, 1959, 1960—LEddy in Coral Sea Donguy, Oudot & Rougerie, 1970—Review of 1956-68 northern Coral Sea data : Rougerie & Donguy, 1975—‘Gorgone 1’ 1972 cruise, northern Coral Sea Donguy & Henin, 1975b—Review of flow at 158 and 163 E . Scully-Power ef a/., 1969, 1973a, b—Western Coral Sea Subsurface circulation Hamon, 1958—Inflow to the northern Coral Sea from the east Wyrtki, 1962b—Geostrophic circulation in the Coral Sea Rotschi, 1958-61—O.R.S.T.O.M. earlier cruises, volume transports Donguy, Oudot & Rougerie, 1970—Review of 1956-68 northern Coral Sea data . Rougerie & Donguy, 1975—‘Gorgone 1’ 1972, late winter Rougerie & Donguy (unpubl.)—‘Gorgone 2’ 1975, late summer Scully-Power e7 a/., 1969, 1973a, b, c— Geostrophic circulation May-July 1968 and volume transport budgets, Western Coral Sea Flushing characteristics of the Western Coral Sea Upwelling off the East Australian Coast VI SUMMARY Water masses Introduction Surface waters Characteristic diagrams es Geographic distributions and seasonal variations Subsurface waters Characteristic diagrams Geographic distribution “Coral Sea water’ Circulation .. APPENDIX— Units, conversion factors and glossary REFERENCES ae, NN LP to WBNONes es P = ee arr rere NR bh Bee Nw WwW bl bv DOAoB A _ bo CO 130 Preface The object of this study is to review what is known of the physical oceanography of the waters of the Great Barrier Reef and Western Coral Sea. Because the two regions are sharply different topographically (the first being a reef-studded strip of shallow coastal water, generally well mixed vertically, while the second is chiefly 1000 to 5000 m deep with marked water structure in the vertical), they can best be treated separately with references to theit interactions where appropriate (or, more exactly, where known), In neither case is the available information sufficient for a full description. In the presentations for both areas, the topography is described briefly and available climatic information is summarised as background. Then descriptions are given in turn of the surface distributions of water properties, the subsurface characteristics and their distributions, and the circulation, referring to the sources of information in each case. Finally, a short summary is presented. For Part 1 on the Great Barrier Reef lagoon, information is limited, and some suggestions are offered for possible future investigations of the physical oceanography of this region. In this connection, it should be noted that there are very few data on the physical oceanography of amy barrier reef lagoons to which one can refer for guidance. An opportunity is available here for pioneering work. (A project for the study of the New Caledonia barrier reef lagoon is already under way.) For Part 2 on the Western Coral Sea, there is more information; here the salient papers are reviewed specifically, and then an attempt is made to summarise the present state of knowledge of the region, For some readers, the summary may be sufficient for their needs, while it is hoped that the review as a whole will form a useful introduction to this region for physical oceanographers prior to study of the original papers for a fuller understanding of the development of the knowledge and of the limited extent of the data available at present, Much of the review is a presentation of the analyses of previous workers, although the mode of presentation of results is sometimes different from theirs. However, some of the material in this review is derived from new analyses of existing data. If the original author is not quoted it may be assumed that the results presented are novel in this review. A common bibliography is attached. Most of the items are referred to in the text but some, such as data records used by some of the authors, have been included for completeness though not specifically quoted in this text. G. L. Pickard June 1976 vil Acknowledgments Having acquired an interest in the water circulation patterns in reef and lagoon areas during many visits to Pacific islands, the senior author was pleased to accept Dr M. Gilmartin’s suggestion that he review existing knowledge of the physical oceanography of the Great Barrier Reef and adjacent Coral Sea as background for further studies of this area. He is grateful to Dr Gilmartin, as Director, for making available to him facilities to work at the Australian Institute of Marine Science; to Dr K. R. Allen and Messrs B. V. Hamon, D. J. Rochford, and R. J. Edwards of the CSIRO Division of Fisheries and Oceanography, Cronulla, and to P. D. Scully-Power of the Royal Australian Naval Research Laboratory, Sydney, for assistance in locating and evaluating information; to many members of the Section d’Océanographie, O.R.S.1'.0.M. Centre de Nouméa, for suggestions and assistance related to both parts of this review; to Mr J. Ngai for his excellent drafting of the figures; and to his collaborators in Part 2. He wishes to thank the Director-General of O.R.S.T.O.M. and M. Legand, Director of the Centre de Nouméa, for permission to use the facilities of the centre at Nouméa and for other courtesies during his stay. He is particularly grateful to M. Henri Rotschi, until recently head of the Section d’Océanographie, who first roused his interest in the south- west Pacific and fostered it during two previous visits to Nouméa including participation in two major cruises in the western equatorial Pacific in N.O. Coriolis. Permission to reproduce figures as indicated in the legends is gratefully acknowledged. vill PART 1 The Great Barrier Reef by G. L. Pickard —-— TT ™— , 150°E L SBS ° 7 BRAMBLE CAY GUINEA Tay 7) 100 1's 4 Log TORRES STR. QTHURSDAY 1 to as © oy CAPE YORK’, 200-km ae a ds L 4 % oo CORAL Ln PRINCESS COENG \cHARLOTTE 7 BAY _>\CAPE MELVILLE 18° 184 COOKTOWN 2 SEA +PAPUAN PASSAGE : WILLIS IS. ] LOW IS, = TRINITY OPENING vO PORT DOUGLAS “S36 L CAIRNS® GREEN o Al HARVEY CREEK INNISFAILS, L N.BARNARD IS 4 & can , r 4 9CAPE CLEVELAND... 0 7 TOWNSVILLE Fe f 20° BOWEN Xn, D,WHITSUNDAY 15, rae 74 20044m | fe) QUEENSLAND MACKAY SPINE IS L METEOROLOGICAL ‘BROND 4 STATIONS oo 4 o LAND e REEF ST. LAWRENCE? beige feat 200m. ISOBATH SHOALWATER BAY [ewer : a: : ROCKHAMPTON, CAPE CAPRICORN, oat GLADSTONE, % : sea | BUSTARD HEAD’ s LADY ECLION IS; NOWNSVILLE “SANDY CAPE ONT | Or. : a ° 4 ° i loueeeeans L 2s BUNDABERG : 25 | HERVEY BAY ones fret Pay [ee : EY BA’ SANDY IS. | I BRISBANE® ‘oO : L DOUBLE 4 IS. PT. MORETON BAY AUSTRALIA ‘ao 4g. BRISBANE 150° —_—o 1 1 L - i ae Fig. 1 Australian Great Barrier Reef region. I Introduction The Great Barrier Reef off Queensland, Australia, extends more than 2000 km from about 9.1 S, just south of Papua New Guinea, to about 24 S (Fig. 1). Although a considerable number of biological investigations have been carried out in the region, relatively few studies have been made of the physical oceanography (water properties, currents, etc.). The main ones are those on water properties by Orr (1933 a,b) and Moorhouse (1933) (both at Low Isles during the Great Barrier Reef Expedition) and by Brandon (1973), and on the tides by Easton (1970). The Australia Pilot, Vol. IV (British Admiralty, 1962) gives some information on currents and the annual Australian National Tide Tables (e.g. 1976) on tides and tidal currents. Maxwell (1968) has described the geographical, geological and other features at length. There is also information on water properties and climate in Weather on the Australia Station (RAAF, 1942), Sea areas around Australia (Roy. Neth. Met. Inst., 1949), and the Monthly Oceanographic Charts, Tasman and Coral Seas (CSIRO, 1974). The present paper will review current knowledge of the physical oceanography of the Great Barrier Reef region for the use of marine scientists working in the area, and will suggest directions for future study. II Topography The Great Barrier Reef is an assembly of coral reefs and lagoons off the cast coast of Queensland, The coral reefs are found as far north as Bramble Cay atabout 9.1 S, close to the coast of Papua New Guinea; this cay will be taken as the northern limit of the Barrier Reef region for physical oceanographic consideration. A southern limit of 25 S is arbitrarily selected; this latitude is in the vicinity of a marked feature of the coast, Great Sandy Is., and is just south of the southernmost coral reefs. The western boundary is taken as 142 E in Torres Strait and the Queensland coast to the south. The eastern boundary is taken as the position of the 100 m depth contour just outside the outer reefs, a depth greater than those found between the shore and the outer reefs, and close to the drop-off to the continental slope. The seaward limit will not be adhered to too rigidly and, in the south, the Capricorn Channel will be included west of a line from outside the Swain Reefs to Sandy Cape, For this review, the term ‘(Great) Barrier Reef’ will be used to refer to the region as a whole from west to east boundaries, The term ‘outer reef will be used for the main boundary structure on the eastern side, and the term ‘lagoon’ to the main body of deeper water between the land and the outer reef, The word ‘reef alone will be used to refer to individual features of the coral structures of the region; these may enclose their local (shallow) ‘lagoons’. However, attention will be devoted chicfly to the main lagoon rather than to the local lagoons. Despite its name the Great Barrier Reef is not continuous and certainly does not present an impermeable barrier to the passage of water. Fig. 2 shows schematically the general character of the outer reef. The linear distance scale is taken along the eastern face of the outer reef, and the intersections with parallels of latitude are indicated; the latter scale is non-linear because of the changes in the orientation of the reef face. The northern 40", of the reef, from 9.5 to 16.5 S, is one of the more continuous stretches, with reefs coming close to the surface for about 90", of the distance. Then from 16.5 to 20 5 shallow reefs occupy only about 10", of the distance along the reef. There is a short stretch of dense reef from 20 to 21 S, followed by the Swain Reefs to about 22.5 S, in which the reefs are scattered but the east-west width is greater than elsewhere, The Capricorn Channel is open to the south and provides a wide and deep entrance to the southern end of the lagoon, The southern 60", of the outer reef length is thus only about 20", occupied by actual coral reefs. The above figures are approximate but do indicate that the lagoon waters inside the outer reef are by no means isolated from the Coral Sea outside. In addition, it should be noted that in the more continuous stretches of reef, water from the sea may enter not only through the gaps between the coral reefs (‘passages’ etc.) but also over the reefs as swells break on them. The Reef as a whole extends roughly ina NW-SE direction, with an outer reef length of about 2300 km; the width from shore to outer reef varies from 23 km at 14 S to 260 km at 22.5 S and avcrages 100km. The area between land and outer reef is about 230 000 km?. A more significant dimension oceanographically is the depth of water, An estimate of the distribution of depth in 20 m intervals has been made from Maxwell’s (1968) Figs. 16A-E. The areas between 0-10, 10-20 fm etc. were estimated by+ degree square units for each of Maxwell’s five regions and are shown in Fig. 3A as percentages of the area in that 2 COAST REEF CONTINUITY DISTANCE PAPUA NEW GUINEA 9°S_ 5 + BRAMBLE CAY Lo r 10° 4 I CAPE YORK at | b | L 7) IL 4 wn WwW ‘= Ww L J loa 2 L- 500 °o ° for) CAPE MELVILLE + | | r COOKS PASSAGE = 15° 4 | - COOKTOWN 4 L PAPUAN PASSAGE L = be 4 E Low 1s. + L TRINITY OPENING ~ [ GRAFTON PASSAGE P) - 1000 Te CAIRNS +1 ] Ww WwW INNISFAIL 4 ee r (od L 4 LL a Ww L Ww w # co & 3 °o o i ‘i © 2 fe 1 L a TOWNSVILLE 4 < 1 + 1500 WW z + 20° 4 r = wn WHITSUNDAY IS.4 a 7 o 1 Lo} ° [ I | r a MACKAY . 7 1 F oe a | I _| SWAIN a REEFS -— | 2000 : + CAPRICORN CHANNEL L 4 . on 2 HERON IS. o - i 1 SANDY CAPE—24-7°4 Fig. 2 Schematic representation of outer reef continuity, Great Barrier Reef. region (shore to 50 fm). There is also a component of uncertain depth in most regions because of the incompleteness of the hydrographic surveys. In Fig. 3B are presented the areas within the depth zones for the Reef area as a whole. From this graph the values for zones 0-20 m, 20-40 m etc. were interpolated (extrapolated for the 100 m value) and are presented in Table 1. From this it is estimated that the mean depth of the surveyed area of the lagoon is about 35 m, and depths over 60 m are uncommon. The unsurveyed areas are probably mostly in the shallow end of the depth range. It should also be noted that aerial photographs are available for about 80", of the Reef region (some sources are given at the end of the References). Fig. 3 (A) Depth distribution in Great Barrier Reef lagoon by 10 fathom intervals in each of Maxwell’s (1968) regions, expressed as percentage of area of each region, (B) Hypsometric curve for Great Barrier Reef lagoon, derived from diagram A above. PERCENT 3 ; 0) 10 20 30 40 50 60 70 80 30 100 Oo 9-3 Ss L 1 L 1 | 1 L 1 1 4 a vB a . 3 25500 % 13-2°S 7 NoT aa w WAT = SOUNDED 72000 ~ 9 17°s < my = = uw = MoH -54500 -100 504 a +80 — = Ww B40 4 Ww x = < 39 | ae w = x x & 20 4 p40 e WwW W [=) [=) 10 4 +20 0 T T T T T T T T T 130 () 20 40 60 80 100 PERCENTAGE OF TOTAL REEF AREA Table 1. Percentage of Great Barrier Reef area in various depth zones Not Depths (m) 0 20 40 60 80 100 surveyed Areas (")) 23 25.5 225 11.5 2.5 15 Mean depth 35 m; total area about 230 000 km? Ill Climate Information on some aspects of the climate particularly relevant to the physical oceanography of the Barrier Reef has been extracted from Weather on the Australia Station (RAAF, 1942), the Australia Pilot (1962), Sea areas around Australia (Roy, Neth. Met. Inst., 1949) etc., and are summarised in this section. One important factor is that most of the regular observing stations are on land; in this regard, two points must be noted, First, the local topography is likely to affect the meteorological factors, particularly wind direction and rainfall. Second, the degree to which land station measurements can be applied to the neighbouring sea areas, particularly at distances as great as 250 km, is always uncertain. However, there are some island stations which afford an opportunity to evaluate this factor to some extent. WINDS South of 15 S, the SE trade winds prevail and the wind direction is generally from between east and south. North of 15 S the NW monsoon invades the region. In the north, the NW winds start in December, develop through January and February and die out in March to be replaced by SE winds. Their effect is felt as far south as 15 to16 Sinlessened degree. Increased rainfall is associated with this wind direction. The most common wind directions and typical speeds are shown in Fig. 4 for a selection of stations near the Reef. Willis Is. is included in this and other climate tables, although it is 460 km east of the reef, because it is the only continuous long-period reporting station representative of open sea conditions with little topographic interference (the island is only about 8 m above sea level). The above information is taken from Weather on the Australia Station (RAAF, 1942), in which the frequency of occurrence of winds is given for eight directions (and calm) on a monthly mean basis. The speed is given as a single monthly mean value for 0900 hr and 1500 hr. The monthly sheets of the General Air Circulation in Sea areas around Australia (Roy. Neth. Met. Inst., 1949) show only a single arrow in each 1 square to represent the general direction and speed; this information agrees well with that in Fig. 4. The effect of the NW monsoon is chiefly to decrease stability and to introduce westerly to northerly components of wind to 15 to 16 S. South of this latitude, an easterly component of wind prevails from August to January or February, with southerly winds prevailing for the rest of the year. A feature in the Cairns—Townsville area (17 -19 S) particularly, and to some extent for other southern areas, is for the wind to back by 45 to90 (from SE toward NE) between 0900 and 1500 hr during much of the year. CYCLONES On the average, two cyclones per year affect the Queensland coast, the associated heavy rainfall in short periods of time probably being the most important factor oceanographi- cally. Cyclones originate in the Intertropical Convergence Zone between 8 and 18 Sin the northern Coral Sea and are most common (about 76",,) in January to March, with 6 another 18°,, occurring in December or April. They initially move to the west. Those north of about 12 S tend to continue west across the Cape York peninsula, while the more southerly ones tend to curve SE and parallel the coast. On the average there are the same number of lesser disturbances each year, 80°,, of which occur between January to July inclusive. Fig. 4 Preponderant wind directions in the Great Barrier Reef region. Wind blows toward the dot, numbers represent the mean wind speed (m s) for 0900 and 1500 hrs (Daru 0900 hr only). For stations from Daru to Willis Island the two wind vectors indicate two predominant directions during the month (associated with the NW monsoon), while from Cairns to Sandy Cape the two vectors indicate that the wind backs from the 0900 hr observation (the more clockwise vector) to the 1500 hr observation. Circles indicate 10",, or more calms in the month. JAN FEB MAR APR MAY JUN JUL AUG SEP Oct NOV DEC ow sw WS WS Wo & < << < £ @ ¥ POW SSN hs MONSOON Se RA S. —. TRADE WINDS COOKTOWN eo a ny < A NKR RS < a ES RRR 1rs 1) wns is A aa aA OY ae a al i ORS es SRS SE a Det < a . & —N— @ 10 PERCENT OR OVER CALMS oe eS FREQUENCY OF DIRECTION PERCENT AIR TEMPERATURE The air temperature cycles for monthly mean maximum and minimum temperatures show a maximum in December or January and a minimum in August. Curves for selected stations at the north (Cape York) and the south (Sandy Cape) are shown in Fig. 5. The curves for ‘lownsville and Cape Cleveland are shown partly to represent an intermediate latitude and partly to show how much difference there may be between neighbouring stations. These two are only 20 km apart and at the samc latitude. It is concluded that for any heat budget studies, it would be essential to use actual observations for the locality being investigated. Sea areas around Australia (Roy. Neth. Met. Inst., 1949) has monthly charts of mean air temperature averaged by 1 squares with whole number isotherms interpolated on the plots. For the Barrier Reef region temperature time curves for 11 S, 19 S and 25 S have been compared with the maximum and minimum plots of Fig. 5. The Dutch data values fall half way between the maximum and minimum values for Cape York (10.7 S), ‘Townsville (19.2 S) and Sandy Cape (24.7 S) (within + 0.5C ), except that the Dutch values fall about 1.5C lower near the minimum. Fig. 3 Seasonal variation of air temperature for selected typical stations along the Great Barrier Reef coast. (A) Monthly mean maximum, (B) Monthly mean minimum. Cape York— 10.7 S, Cape Cleveland— 19.2 S, Townsville—-19.2 S, Sandy Cape—24.7 S. w a ——! 2 ay ugg ON Se SS 5 feet pa ee ET ee 4 a 4 = 2 «254 a an : NN. (A) cm opt MONTHLY MEAN MAXIMUM PtH ae ee a 20-4 T T T T T T T T T T T 1 JAN FEB MAR APR MAY JUN JUL AUG SEP oct NOV DEC JAN MONTH 25 nN ° AIR TEMPERATURE °C uw (B) MONTHLY MEAN MINIMUM 10 Tr T T T T T T =Ie T T T 7 JAN FEB MAR APR MAY JUN JUL AUG SEP oct NOV DEC JAN -M E R woitieon T E— R S U M- An indication of the geographical scatter of mean air temperatures is given in Fig. 6 in which the monthly means of daily maximum and minimum temperatures (RAAF, 1942) and the monthly mean temperatures, are plotted against latitude. Mean lines have been visually drawn through the points, with reduced weight being given to stations more than 10 km inland from the coast and to Willis Is. offshore. It will be seen that the maximum is substantially constant from 9 to 20 S and then decreases southward, while the minimum decreases steadily to the south. The annual mean air temperatures derived from the Dutch data above lie about 0.4C_ higher than those in Fig. 5. Fig. 6 Minimum and maximum values of monthly mean air temperatures, and annual mean air temperature, for stations along the Great Barrier Reef coast. Means are of at least 20 years observations to 1939 except for Daru (16), C. Cleveland (8). (.) = more than 10 km inland, [ ] = Willis Is., about 460 km east of outer reef. DARU al . I Lio°s : THURSDAY IS. _| Ja . C.YORK |. = “| ° y >| = > x iS) L S Wl > S 3 = x ny ~| x S x = r y Sy] @ = = > =! S (COEN) +- (0) S al = (*) & ! x rs > | yi Lis’s S | RS x > COOKTOWN 4 ° im S]- L | = [wittts 18] [8 (*] CAIRNS + Ww ; ° INNISFAIL © ay e | C.CLEVELAND |. Es TOWNSVILLE | i L20°s ! | | Mackay + ° a] | L | (ST LAWRENCE) 4 (0) 4] | ry I (ROCKHAMPTON) — (0) ja GLADSTONE | B BUSTARO HEAD 24S 3 “y SANDY CAPE 4 Ia BUNDABERG 4+ al | 4| ! DOUBLE IS. PT. 1 _" ——* = 10 20 rs AIR TEMPERATURE C RAINEALL. In coastal seas particularly, rainfall and river runoff are important factors in their effect on salinity. Australia is a continent of tremendous variation in rainfall and the Queensland coast is the wettest part of the country. ‘Phis point is often overemphasised as the mean annual rainfall along the coast is not high by world standards (about 1500 mm, the same as Vancouver, one of the drier parts of the British Columbia coast). What may be more Important occanographically (and socially economically) are the very large variations in space and time of the rainfall, As this section of the review is intended primarily for orientation, not as a source of climatological statistics, tabulations of detailed data will not be presented. ‘Vhese are available in Weather on the Australia Station (RAAT, 1942), dlustratia Pilot (1962), or from the Australian Bureau of Meteorology as the basic source of climatological data for this country. Rather, the rainfall will be described graphically as are the other climatological factors. For any analysis of specific oceanographic field data, it will be mandatory to use actual weather information, rather than climatological averages, because of the large variations with time (day to day or year to year) which are Characteristic of this coast. Data from standard observing stations may be available, but in Hig. 7 Mean annual rainfall along the Great Barrier Reef coast and range from recorded minimum to maximum annul rafal ] MEAN ANNUAL RAINFALL | © LAND STATION 6000 © JSLAND STATION 5000 HARVEY CREEK Q' N 4000 \ \ as | é I ol £ fies HN = 1) |\ 4 ] \ 3000 | \ z \ b=; + MAX. | 7 a l \ \ HS ARU GREEN (PNG) 18.4 ss ~ Low ‘ ° 2000 -} ~\_ 15. 9 CAIRNS S\}rHurspay 1s COOKTOWN / 1 \ ° 7 MACKAY C.YORKTN L \ it Wey 7 \ iN SANDY Rel Z| WILLIS CAPE, - >< it. 1S NGTOWNSVILLE |] NORTH ee ~~" e | dn al reer | tom” |? 1000 Oy Km GLADSTONE t eeiners qd MERON IS I MIN. | 0 ge to Sa i ° os 5 LATITUDE zo 2 10 view of the significant spatial variations it may be necessary in some cases to schedule weather recording for the area of oceanographic study. Rainfall values have been rounded off to the nearest ten millimetres as the precision implied in the published figures to the nearest millimetre seems inappropriate when year to year variations may be measured in merrey (e.g. Fig. 7). Mean annual rainfall Fig. 7 shows the long-term mean annual rainfall as a function of latitude, with a few names added for convenience and the land and island stations distinguished (Sources: RAAF, 1942; Brandon, 1973). It will be secn that the mean annual precipitation is between about 1000 and 2000 mm except for the notorious but restricted zone near Innisfail (17.5 S$). Also it is noted that the island stations lie on the same curve as the land stations (except for Willis Is. far offshore). This point will be discussed later. Rainfall variability Fig. 7 also has bars indicating the range between recorded minimum and maximum annual rainfalls (Source: RAAF, 1942). This figure shows one aspect of the variability of rainfall on this coast, that the variation is often greater than the mean. A plot (not shown) of the differences between recorded maximum and minimum against long-term mean annual values shows a roughly linear relationship from 1000 to 4000 mm yr. A few values are: 000 3000 4000 500 3300 4100 Long-term mean annual rainfall (mm) 1000 2 Difference, recorded max.-min. (mm) 1700 2 This table gives an indication of the long-term variability, although it must be noted that the differences are from records of 50 to 72 years duration and such large variations need not be expected very frequently. In records of maxima and minima for individual months, the variability is even greater than that shown by the annual maxima and minima. A few examples from tabulations in Weather on the Australia Station (RAAF, 1942) will suffice to indicate the tremendous ranges of rainfall to be expected. For Innisfail (17.5 S), January rainfalls from 30 to 1580 mm (mean 310 mm) and August falls from 0 to 450 mm (mean 130 mm) have been recorded; for Bowen (20 S), the January range is 5 to 1180 mmand for April 0 to 640 mm. Ranges such as these are typical, not rare. On a still smaller time scale of days, the variability of rainfall can be very large, particularly during the wet season when the heaviest falls are associated with cyclonic systems moving through the region. Practically every coastal station south of 15 S has received at least one fall exceeding 250 mm in 24 hours (RAAF, 1942). The recorded 24- hour maximum rainfall values in the climatological tables (RAAF, 1942), when plotted against mean monthly values on a scatter diagram (not shown), can be summarised as follows: Monthly average (Rui) 24-hr maxima Dry season up to 150 mm 0.8 to 5 times Ry Wet season 100 to 650 mm 0.5 to 2 times Ruy ll Although the dry season 24-hour maximum factor may be larger (up to 5 times), it is the wet season 24-hour maxima which are likely to be more significant oceanographically because of their greater volume. A generalised picture of the seasonal variations of rainfall during the year is presented in Fig. 8. Mean monthly rainfall values for each station were normalised by dividing by the mean annual rainfall for that station. Curves drawn of normalised rainfall against month show the well-known features of (a) distinct wet and dry seasons and (b) more marked distinction between the seasons in the north and centre areas than in the south. Mean curves were presented for these three zones in order to avoid the confusing appearance of a large number of curves (for all the stations) on one diagram, (Note that there are only three stations, widely spaced at 10.6 , 10.7 and 14.0 S, for the north zone.) Because Fig. 8 is a simplification from the original diagram, which shows considerable scatter about these mean curves, it should not be used to scale off values for individual stations (from their mean rainfall values) except for very approximate values. Tables of climatic records should be used for this purpose, ¢.g. RAAF (1942), Australia Pilot (1962), Brandon (1973) for immediate reference, or the Bureau of Meteorology as the basic source, especially for recent data. However, Fig. 8 does show the two basic features of the rainfall distribution noted above. The concentration of rainfall (about 70°,, of the annual amount) in the January to March period for the north and centre zones is well shown. Note that Innisfail, in the centre maximum rainfall zone, does not fall clearly on any of the curves (it is shown by the symbol ‘TP’ on the diagram). It has an extended wet season similar to the south zone but falls between the centre and south zones in the dry season. This is attributed (RAAF, 1942) to orographic showers occurring in the nominal dry scason. Comparison of rainfall on land and on islands Earlier it was noted in reference to Fig. 7 that the rainfall statistics for the islands of the Reef were not much different from those on the land stations nearby. Brandon (1973) examined this aspect by comparing the rainfall at islands and land stations for the same group of years (the data in Fig. 7 are for different numbers of years for the different stations). The ratios which he obtained were, for the same years for each comparison but not the identical vears for all: (a) Low Is./Port Douglas =1.02 (b) Green Is./Cairns =1.09 (c) Pine Is./Mackay =0.51 (d) Pine Is.; Rockhampton =1.09 (c) Heron Is./C. Capricorn* =1.60 *Eight years only ‘Yo these are added, before discussion: (f) Pine Is.,St Lawrence =0,83 (g) Heron Is./C. Capricorn** =1.23 **AIl data (h) Heron Is.'’Gladstone =0.98 (i) Heron Is.’Bustard Hd = 0.84 Note: the latter are not for the same groups of years in each case. wma For Heron Is., Brandon concluded from the ratio 1.60 (¢) that rainfall at the island was significantly higher than at the neighbouring coast station. He did not point out that the rainfall at C. Capricorn for the 8 years for which Heron Is. data are available was substantially below its long term average (600 mm against 810 mm) and that usc of this latter figure would yield the lower ratio of 1,23 as in (g) above. Also, the ratios for Heron Is. rainfall to two other stations at the same distance as C. Capricorn are (h) 0.98 and (i) 0.84, Brandon regarded the Pine Is./Rockhampton ratio of 1.09 (d) as more significant than that to Mackay of 0.51 (c), which station is nearer but in a local high rainfall area. Note the Pine Is./St Lawrence ratio of 0,83 (f), St Lawrence being nearer than Rockhampton and the same distance as Mackay. Overall, the evidence above suggests that the rainfall over the Reef area is probably not very different (+15°,,) from that on land in the vicinity (same latitude), and that for climatological calculations the mean rainfall curve of Fig. 7 could be used for the lagoon. However, it would be desirable to have more data for the Reef area itself to verify this opinion, particularly from positions further from land, and it is not expected that the very high coastal rainfall near 17 S, related to orographic effects, extends very far across the lagoon. (Ref. Chapter X, Reef Climate.) RIVER RUNOFF Rainfall on the lagoon is not the sole cause of reduced salinity; river runoff from precipitation on land is also a factor in coastal areas. Previous descriptions of salinity distributions in the Barrier Reef lagoon have attributed reduced salinity in part to river runoff, but curiously only one paper (Endean et a/., 1956a) has given any values for river Fig. 8 Seasonal variation of rainfall along Great Barricr Reef coast, expressed as fraction of mean annual value: North zone: 10,5 to 15 S, mean annual rainfall 1750 mm, Jan.-Mar. — 70", of annual; Centre zone: 15.5 to 21.2 S, mean annual rainfall 2000 mm, Jan.—Mar. = 70", of annual; South zone: 21.5 to 25 S, mean annual rainfall 1000 mm, Jan.—Mar. = 45", of annual. 0-255 0-2-5 0-15 on 0:05 MONTHLY RAINFALL AS FRACTION OF MEAN ANNUAL VALUE o T T T DEC JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV 13 DEC runolf and no one has presented any quantitative comparison of river runoff and rainfall effects, ‘Vherefore an attempt was made to evaluate the relationship between these two fuctars. Only limited information on river runoff is available, the two prime sources being Stream Gauging Information, Australia, December 1969 (Aust. Water Res, Council, 1971), abbreviated to $.G.1. hereafter, and ‘Surface Water Resources’ in the Arlas of Austrahan Resources (Aust. Dept. of Nat. Dev., 1967), hereafter S.W.R. S.G.I. lists all river basins with their catchment areas and, in some cases, annual discharge figures. The North-east Coast Division is the one related to the Great Barrier Reef. Runoff information is available for about 318.000 km?, i.e. about 76", of the total drainage area of 417 000 km? from Cape York to 25.5 S, Untortunately, much of the catchment area for which runoff is measured is in low rainfall areas inland (less than 50 mm yr) which do not contribute much to the total runoff, Many of the short rivers in the high rainfall regions near the coast are not metered at all. In fact, only one-third of the estimated annual rainfall of 8,3 x 10'° m* (S.WLR. Commentary Text) is actually measured. When calculating river runoff from rainfall data in a catchment area which is not generally saturated with moisture, it is necessary to allow for water which is retained in the soil and for that which is lost by evaporation (the actual river runoff is as low as 10", of rainfall in much of Australia including the inland areas of Queensland). It should therefore be noted that rainfall maps, such as Fig. 2 in Brandon (1973) or Fig. 53 in Maxwell (1968) can be misleading as far as river runoff is concerned. Because of increased losses inland, the runoff decreases much more rapidly than does rainfall as one proceeds inland from the coast. As an example, in the same distance inland at 18 S the rainfall decreases by a factor of 3 but the runoff by a factor of 60. Fortunately, S.W.R. gives isopleths of runoff per unit area superimposed on the drainage basins; the runoff from each basin was estimated by numerical integration as suggested in the S.W.R. Commentary. For the basins for which the measured runoff is given in S.G,1, and for which the gauged area includes most of the catchment area, there was a reasonable correlation between runoff estimated from the S.W.R. chart and the actual gauged values in S.G.1., with the exception of the Burdekin River. For this, the estimate was only 40", of the gauged figure, The total runoff for the North-east Drainage Basin was estimated as 80",, of the average annual runoff quoted in the S.W.R. Commentary Text, presumably obtained in a similar manner but with larger size charts. ‘The discrepancy is attributed to the small scale of the published $.W.R. chart which makes it very cramped in the coastal high runoff area. Also, isopleths for more than 1500mm yr rainfall are not given in the published chart (because of crowding). Accordingly, estimates for individual basins were increased by 100 80 (with the Burdekin River value increased to the measured value) and presented in Fig. 9A. As these figures are for annual runoff vole, they cannot be compared directly with rainfall figures which represent volume per unit surface area, i.e. depti of water. To provide some idea of the relative importance of river runoff and rain components of fresh water input, the river flow value has been compared with rainfall volume over a selected area of the lagoon, ‘The latter was divided into 1 latitude zones and for each the fresh water volume for the annual mean rainfall (Fig. 7) was calculated for an area from the shore to the outer reef or to 60 km east from the shore, whichever was the smaller distance. (The narrow part of the lagoon, in the north, has an average width from shore to outer reef of close to 60 km, hence the use of this figure where the actual lagoon is wider.) The river runoff into each of the same 1 latitude zones was determined and plotted (dashed line) with the rainfall value (full line) in Fig, 9B. On this basis the river runoff is about 50", of the rainfall volume. Several criticisms may be offered. The difficulty of integrating the published chart has been mentioned, but the errors are not likely to be more than +25",, from this source (atter normalising to the published total runoff figure). However, the information on 1+ which the S.W.R. is based is admitted to be very limited and more extensive data would certainly improve the quality of the estimate. For the present, the value of trying to improve the runoff estimates is doubtful because of the extreme variability in time of river runoff. The variations from year to year are very large. The Burdekin River (19.7 S) has the largest recorded mean annual flow for the coast, 0.97 x 10! m3, and its minimum and maximum recorded flows are 4", and 300",, respectively of this mean value. For the Fitzroy River (23.6 S) the figures are 5.1 x 10!° m3, 3", and 560", respectively. The rivers north of 19 S generally do not show such extreme ranges of flows (e.g. average 40", to 210°,, of mean annual) as those south of that latitude (e.g. average 4", and 390",,). It should be noted that the mean seasonal variation of river flow, from the shorter rivers at least, is very similar to the mean seasonal rainfall pattern. S.W.R. gives figures for the Barron River at about 16.8 S, close to the coast, and for the Nogoa River at 23.6 S, about 110 km inland. The seasonal variations of runoff from these are very similar to those for rainfall for the centre and south zones respectively (Fig. 8). In shorter time intervals, the coastal river flow follows rainfall within a day or two but the smaller inland component may be delayed. Fig. 9 (A) Mean annual river runoff volumes estimated from Surface Warer Resources (Aust. Dept. of Nat, Dev., 1967) and Stream Gauging Information (Aust. Wat. Res. Council, 1971), (B) Comparison of river runoff volume with rainfall volume on lagoon to reef or to 60 km from shore, whichever is least. A 101 m3yyr. B 10° m3/yr/deg lat. ‘: te) 0-5 1 5 ie) 1 2 tos - 10° ++ 4+ + 1 1 4+ tt st RIVERS RAIN (MEAN) CAPE YORK 4 NORMANBY 15 rey it WW a @CAIRNS wo | =) fe INNISFAIL a - at J 4 s TOWNSVILLE | BURDEKIN 20° 20° 4 ra «MACKAY 1 4 4 — i 4 FITZROY ROCKHAMPTON @SANDY CAPE °. ° #5 25°4 15 Another criticism of the rainfall versus runolf comparison of Fig. 9B concerns the calculuted ruintall volume. Mor instance, the high rainfall component in the 17 to 18 S$ yvone occurs because the local coastal rainfall figure was used; however, the substantial orographic effect here may overstate the rainfall volume calculated for the lagoon in this zone. Second, the 60 km width of zone for the southern half of the lagoon was chosen rather arbitrarily; if the full width of the lagoon were used, the rainfall component would be inercused south of 18S. “Vhird, the tacit assumption in this comparison is that whatever width of zone is chosen, the river and rainfall components are distributed equally and unilormly over tt. In fact, this is not likely to be the case for the runoff component because the rivers are essentially point sources at the coastline and their effects will be emphasised near their mouths. ‘Vhis is another reason for not calculating the rainfall volume over too wide an area of the lagoon for comparison, It would be desirable lo make some measurements of runoff distribution as a function of time after a heavy discharge. Salinity depth measurements over an areca would be desirable but a first order estimate might be made by using the turbidity of the river water as a visual tracer, Another way to compare river runoff with rainfall is to assume the river component to be uniformly spread over an area of the lagoon as for the rainfall. [f this is done, the river water layer thickness would vary from about 15 cm to 255 cm (compared with rainfall depths ranging from about 90 to over 400 cm). Expressed this way, itis convenient to note that 100 cm of water added toa column of 35m depth (mean depth of the lagoon) would decrease the mean salinity by about 1 if mixed uniformly through the column. While rainfall is distributed over an area, river runoff is initially concentrated near the river mouths and therefore may have a greater effect locally in reducing the salinity. A frequently quoted example is Hedley’s (1925a) deseription of the destruction of a coral reef by an exceptionally large river flood, It is concluded from this discussion that river runoff does introduce a component of fresh water to the lagoon which is quantitatively comparable to the rainfall component. In the main, it will be contributed in phase with the rainfall but, being introduced at specific points along the coast, it could have greater effect in reducing salinity locally. CLOUD AND HUMIDITY ‘Two other meteorological factors relevant to heat budget studies are cloud amount and the humidity of the atmosphere above the sea, According to the data available, neither shows much variation. Cloud amount Vig. LOA shows that there is a small variation of cloud amount with latitude, the lowest values occurring between about 19 and 24 S, There is also some seasonal variation (Fig. 11A) with rather more cloud during the summer wet season than during the winter. The lower mean cloud amount in the south appears to be a result of less cloud all year round as is seen when curves a and b are compared in Fig. 11A, although the range of scatter of points about each line is 1 to 1.5 oktas (eighths) or rather more than the separation of the lines. The mean amount of cloud, 4.5 oktas of sky covered in the north in the monsoon season, seems to be low but it should be noted that for the three stations {rom 10 to 14 § the values are for 0900 hr observation only, which does not take account of any build-up during the day. Most of the others are for the mean of 0900 and 1500 hr observations. The summary in the Australia Pilot (1962) of reports from ships at sca shows much the same cloud conditions as described above, The above remarks apply to all cloud observed from the stations. 16 Humidity The relative humidity (Fig. 10B) shows a slight decrease southward, with lower values in the afternoon than in the morning, which is usual in coastal regions. The humidity also has slightly lower values in late winter (Fig. 11B) and 10", lower values through the year (curve d against curve c) for the same southern group of stations which show lower cloud amount (Fig. 11A). Fig. 10 Variation with latitude along Great Barrier Reef coast of: (A) Cloud amount (1 okta = one-eighth of sky covered), (B) Relative humidity. (A) CLOUD AMOUNT (B)RELATIVE HUMIDITY #1 Plo g S la myl g Rls ea) is) fo) i) eee S| 8 9S Ss = F15°4 = ° 3 e | Wi ° Pee O.-2] ) = |e < od Yy L 20° | 4 r e fo} o fe} ° e 2S. aie elas 1 pes T *s 6 7 8 50 60 70 80 90 100 PERCENT (A) CLOUD AMOUNT Hw 6 a9 < 4 fo) Zz ee 5 — MEAN is ou= SCATTER > = tt a od ro) va 2% : T T T T aN T T T T T 1 DEC JAN FEB MAR APR MAY JUN JUL AUG SEP Oct NOV DEC MONTH 8) RELATIVE HUMIDITY - 0900 HOURS MEAN SCATTER RELATIVE HUMIDITY °/o Qo to} T T DEC JAN FEB MAR APR MAY JUN JUL AUG SEP oct NOV OEC —-M ER WINTER Ss UM- Fig. 11 Seasonal variation along the Great Barrier Reef coast of: (A) Cloud amount: (a) 10 -18 S plus 24 —25 S, (b) 19 —23.8 S, (B) Relative humidity: (c) 10 -18 S plus 24 25 S,(d) 19 23.8 S. 18 IV Tides Easton (1970) has described the tides of Australia comprchensively and most of this section is based on his work, with references to the Australia Pilot and the Australian National Tide Tables. The Barrier Reef lies in two of Easton’s zones, the North Queensland and the Mackay zones. TIDE LEVELS North Queensland zone / Cape York, 10.7 S, to 19.5 S, north of Mackay } Cairns is taken as the standard port because its records are regarded as very reliable. The-tidal rise and fall is semi-diurnal in character with significant variation from springs to neaps (3.3 to 0.3 m at the equinoxes, March and September); sec Fig. 12. There is considerable diurnal inequality, particularly in the height of successive high waters, amounting to over 1 mat times (Fig. 12). This should be borne in mind if one runs a ship Fig. 12. Typical tide height records for three stations along the Great Barrier Reef coast (from Maxwell, 1968). “aettdddda ae: { {| | an | ‘| I fafbe ANAM rad Hl AM : ai [I nn | | haf ne hilntatatan An P ahs yA Na i ia Mn HNO A A A a rE “dU Mldduuuthdddblibiens “ul Mi Hi it i i i it SMH Ale aground. Spring tides occur | or 2 days before full or new moon. At neap tides there are usually one or two days of essentially diurnal tides. ‘The tidal wave progresses eastward from the Coral Sea toward the shore and then north-westward north of Cairns, south-castward south of Cairns (Maxwell, 1968, Fig. 41). Fig. 13A shows the time of high water along the shore (full line) relative to that at Cairns. ‘Vhe tides occur earlier on the reef (¢.g. 15 min earlier at Green Is. compared to Cairns, 30 km to the west). he dashed line in Fig. 13A shows the time of high water at the outer reef, estimated from Maxwell’s co-tidal chart (1968, Fig. 41). Hig. 13B, from Maxwell’s co-range chart (1968, Fig. 42) shows that the tidal range increases somewhat north of Cairns at the coast (full line) and at the reef (dashed line). Because of the very limited amount of tidal data other than at the shore stations, the reef values for phase and range must be regarded as only approximate. Vig. 14A from Easton (1970) shows the percentage frequency of occurrence of hourly tide heights (full linc) and of low and high tide levels (dashed lines) at Cairns. The full line shows the frequency along the ordinate, in terms of the tide level at hourly intervals, for which the sea level is at or above the corresponding height on the abscissa, e.g. the hourly Fig. 130 (A) “Vide phase relative to Cairns for Great Barrier Reef region, (B) ‘Vide range at springs for Great Barrier Reef region. (Fullline shore, dashed line — outer reef.) eo RANGE + 10 8 L qs5° 4 fe T CAIRNS N.BARNARDIS. \ P L TOWNSVILLE L 20 4 \ “b MACKAY b + BROAD SOUND T T t 25° 4 T T T T T T T T T 1 0) 1 2 3 0 5 10 EARLIER LATER (HOURS) METRES THAN AT CAIRNS 20 sea level is at 1.0m (above datum) or higher for about 76", of the time, at 1.5 m or higher for 43", of the time, at 2.0 m or higher for 16",, of the time, etc. The high tide line shows that the high water level occurs at 1,0 m or higher onalmost 100", of occasions, at 2.0 mor higher on 55", of occasions, at 2.5 m or higher on 15", of occasions, etc. The low tide curve shows that low water level is at 0.5 m or higher on 72", of occasions, at 1.0 m or higher on 22", of occasions, and at 1.5m or higher almost never, This information may be useful in estimating recf exposures in shallow waters. A minor point to note is that a few (of the lower) high tides occur below mean sea level, because of the pronounced diurnal inequality of the high tides. Mackay zone (about 19.5 to 25 S; The salient characteristic here is a very large tidal range, with maximal values in Broad Sound (22 S) south of Mackay (21.2 S), ‘The tides are semi-diurnal with marked inequality of the high tides (up to 1.8 m at the solstices) but little or none in the low tides (Fig. 12C). Easton stated ‘categorically’ that the higher tides occur during the night in winter and the day in summer. As the spring tide range is greater at the equinoxes than at the solstices (6.4 m compared to 5.8 m) and the mean sea level is lowest in September, the lowest water levels occur in September—October and afford the maximum exposure of corals. The spring tide reaches a maximum of over 9 m in Broad Sound (Fig. 13B). Fig. 14B, from Easton, shows the percentage frequency of occurrence of hourly tide heights and of low and high tide levels for Mackay. Fig. 13A shows that the delay in the tide, relative to that at Cairns, reaches a maximum of about 2.25 hours between 20.5 and 22.5 S and then decreases, while Fig. 13B shows the marked increase of tidal range to a maximum in Broad Sound and then a decrease to the south. At the outer reef the delay is less or negative (leading the tide at Cairns) and the amplitude is much less than at the shore. Easton attributed these features (delay and increase in range at the shore) to ‘the offshore break in the Barrier Reef and the presence of Broad Sound’. As the outer reef directly offshore from Mackay is one of the more continuous stretches of the whole outer reef, presumably the ‘break’ to which he refers is the deep Capricorn Channel opening SE of Mackay. Other than this he attempts no explanation of the tidal features. Presumably the topography is the main cause, i.c. the existence of a basin to be filled (Broad Sound and Shoalwater Bay) together with the reduction in depth in the approaches, as in other estuaries. However, there are other regions on this coast which have similar topography, e.g. Princess Charlotte Bay (14 S) and Hervey Bay (25 S), which do not exhibit such extreme tidal features. Possibly the deep Capricorn Channel opening is the critical feature. Hervey Bay is open to the Coral Sea with no reefs but does not have as deep a channel as Capricorn Channel, while Princess Charlotte Bay has extensive inner reefs and a fairly continuous outer reef between it and the Coral Sea. (The Australia Pilot (1962) does mention radial flood and ebb flows near Princess Charlotte Bay, similar to those for Broad Sound to be described below, but less noticeable.) An alternative possibility is that the extensive shoals of the Swain Reefs area slow the tidal wave and give rise to refraction and consequent focusing to Broad Sound. Torres Strait Technically this may not be a part of the Barrier Reef, but it is in the northern extremity of the region under review and, because of the unusual nature of its tides and currents, it warrants a brief description, based on Easton (1970) and the Australian National Tide Tables (1976). The Strait lies between the Barricr Reef on the east with predominantly semi-diurnal tides, and the Gulf of Carpentaria on the west with chiefly diurnal tides. The largest semi- diurnal components in the east occur 2 days before full moon, whereas in the west they 21 PERCENTAGE OF OCCURRENCES PERCENTAGE OF OCCURRENCES 0 0-5 1 1-5 2 TIDE HEIGHT (METRES) 00 80-4 60-7 40 L. uw > Ww i < WwW wn 2074 zi Ww = a B — O-+ a T Ly T ie} 1 2 5 6 3 4 TIDE HEIGHT (METRES) fig. Jd Annual distributions of recorded hourly tide heights and predicted high and low tides (from Easton, 1973): (A) Cairns, (B) Mackay. occur 2 days after full moon; in the Strait they occur from 7 days before to 7 days after full moon. Thus it can be spring tides at one end of the Strait and neaps at the other, and neap tides can occur at full moon, The diurnal tide is the same through the Strait. he character of the rise and fall is indicated in Fig. 12A. The information on tides is available for the southern quarter only of the gap between Australia (Cape York) and Papua New Guinea, as the northern part is shallow and not much used. MEAN SEA LEVEL Annual cycle The monthly mean sea level (m.s.].) shows an annual cycle, and three sets of measurements have been published. Values for selected locations are presented in Table 2. On the north-east coast of Australia, monthly mean values for Cairns, Townsville and Gladstone in the International Geophysical Year, July 1957 December 1958, showed a sharp peak in the autumn and a flatter minimum in spring (Hamon & Stacey, 1960). Easton’s (1970) data for 1966-68 show less difference between highest and lowest m.s.l, The Australian National Tide Tables (1976) point out that monthly m.s.1. may vary by 0.07 m from year to year and that it can stand +0.3 m relative to the average for as long as a month. It may be concluded that relative to the annual mean sca level, the monthly mean 1s likely to be about 0,15 m high in March, April and 0.1 m low in September—October. In the North Queensland zone, Easton (1970) stated that the mean sea level is not related directly to the wind, in contrast to the Mackay zone where the mean sea level is attained under the influence of a5 m’s southeasterly wind, which is typical for most of the year (Fig. 4). He presented interesting examples (his Fig. 5.8.9) showing the response of mean sea level to wind changes. In particular, an increase of SE wind from the 5 ms normal to about 15 m/s resulted in an immediate increase in mean sea level of 0.3 m, Mean sea level and latitude Hamon & Grieg (1972) drew attention to the fact that mean sea level apparently rises relative to geodetic levelling by 1.7 m from 30 Sto 11 S along the east coast. This change is opposite in direction to that in North America and Europe where sea level falls Table 2. Monthly mean sea level relative to annual mean sea level Variation tn mean sea level (it) Location Highest Lowest Annual Range Cairns 17S April + 0.23 Oct-Nov. —0.09 0.32 Townsville 198 April + 0.27 Oct-Nov. —0.10 0.37 Gladstone 248 April + 0.26 Oct. —0.10 0,36 Easton (1970): Cairns 17S March + 0.13 Sept. —0.10 0,23 Townsville 19S March + 0.13 Sept. —0.11 0.24 Mackay 228 March + 0.14 Sept. —0.14 0.28 Australian National Tide Tables (1976): Cairns 178 Feb.Apr. + 0.12 Sept. —().12 0,24 Townsville/ Mackay 19-22 S) Mar.—Apr. + 0.12 Aug.-Oct. —0.10 0,22 cqualorward. although in South America there isan cquatorward rise by 0.4m from 23 S tok S.’Vhe rise off Australia was not explained by the change in sea water density, and no satisfactory explaination for the equatorward rise was available. The ave of the semi-diurnal tide This is the time interval between the time of new or full moon and the time of the local spring tide. Webb (1973b) has pointed out that the area from North Queensland to the Samoan Islands is one of the few regions in the world where the age 1s negative, ic. the spring tide occurs before new or full moon (by 43 hours at ‘Vownsville), Webb (1973a) sugvested that this is due to a resonance in the Coral Sea with a period of about 1.97 cycles day (from tidal data for Cairns). TIDAL CURRENTS North Queensland and Mackay zones Haston (1970) stated, without qualification, that in the North Queensland zone the flood is to the north and the ebb to the south. However, the Australia Pilot (1962) states that the flood is generally to the north or north-west when north of North Barnard Is. (17.7 S) and to the south when south of that point, Exceptions are in the vicinity of Princess Charlotre Bay (14 S$) and Broad Sound (22 S), The latter description (in the Pilot) would appear to be the more likely in view of the relative phases of the tide along the coast (ig. 13A). In the vicinity of the outer reef, the directions may be modified by the flows through the passes which are into the lagoon (W or SW) on. the flood and out (E or NE) on the ebb, Speeds are greater in the narrow passes than in the wider ones, with values up to 1.5 ms during springs. Within about 1 of latude (110 km) of Broad Sound, the flood sets radially toward the Sound and Shoalwater Bay and the ebb radially away. A similar pattern is observed over a smaller area around Princess Charlotte Bay (/listra/ia Pilot, 1962). In the vicinity of the Capricorn Group of islands and reefs (23 to 26 8) the charts and P7/or note that the flood direction is tothe west, [tis not clear how much of the flow to Broad Sound comes by this route and how much through the clear opening of Capricorn Channel (22.5 $8) or through the Swain Reefs (21 to 22S). The tidal streams are generally semi-diurnal in character, flood and ebb being approximately opposite in direction. North ofabout 21 S, the diurnal inequality becomes marked with the stronger streams (flood or ebb) some 50", greater than the average, and the weaker streams 50", less than average. Away from constrictions to the flow, the speed isabout 0.5 ms in the southern half of the lagoon and 0.25 m/s in the north. Speeds of up to 2as are found between islands and in the entrances to coastal inlets. It should be noted that when the SE ‘Prades are effective, the current due to these winds in the open areas is stated to be greater than the tidal currents, and a continuing northerly drift of the water may be expected with a tidal perturbation in speed, This matter is discussed further is Section VIEL. In the north, between about 11 and 15 S, the NW monsoon wind drift (in a southerly direetion) may be greater than the tidal flood to the north, but the lighter winds of the monsoon season generally result in the tidal current being more important then than during the SE trades season. Torres Strait Although the tidal rise and fall has a marked diurnal character, the tidal currents are predominantly semi-diurnal and equal in either direction with maximum values up to 3.8 m/s (dlustralian National Tide Tables, 1976, Hammond Rock). There are considerable variations in speed in different locations and further observations are being made (B, V. Hamon, personal communication), a4 Vv Water Properties in the Lagoon The chief water propertics used by physical occanographers and for which there are some measurements for the Barricr Reef lagoon are temperature and salinity (independently as tracers and together for the calculation of density), dissolved oxygen and water transparency. There are some measurements of other chemical characteristics (phos- phate, nitrate, etc.). As the water column in the lagoon is generally well mixed, the surface values for the first three properties (T, S and O,) give a good idea of the overall distribution and will be described first; comments on the vertical distributions will follow. Variations in the vertical are chiefly in salinity, associated with rainfall and river runoff, and occur mainly near the coast. TEMPERATURE Surface temperature at specific locations Orr (1933a) described the data obtained at a ‘fixed’ station in the main lagoon channel 5.5 km east of Low Isles (station at 16.35 S, 145.6 E) in 32 m of water, This station will be referred to hereafter as ‘Low Is.’ Temperature (and other properties to be described later) was routinely measured at six depths from 0,5 m (‘surtace’ value) to 28 m, on 47 occasions at approximately weekly intervals for a year from 30 July 1938. The surface water temperatures are plotted in Fig. 15A from Orr’s data. The minimum value was 21.5 Cin mid-August and the maximum 29.9 C in Mid-lFebruary. Short term fluctuations were generally larger (ca. 1C ) in summer than in winter. During the warming period from September to December, the water temperature was gencrally about 2C less than the daily mean air temperature. During the summer maximum the two were about the same, and during the cooling period the water was about 1C warmer than the mean air temperature. According to Orr, the wet and dry bulb thermometer readings indicated ‘a certain amount of evaporation all the year but the difference in vapour pressure between sea and air showed no seasonal variation.’ He gave no calculation of the volume of water evaporated or of the heat loss. Moorhouse (1933) presented data and a graph of the temperature taken at 0900 and 1500 hr daily for almost a year at the surface and at 1 m depth at the Anchorage on the north side of Low Is., but he did not present any analysis of the data. Day to day variations in temperature at the Anchorage at 0900 hr averaged 0,3C with a maximum of 1.5C , while at 1730 hr the average variation was 0.5C witha maximum of 4.2C . Diurnally, the 1730 hr temperature averaged about 1C_ higher than the 0900 hr value, with a maximum of 4.5C higher. The larger differences both diurnally and from day to day usually occutred when the winds were light. The Anchorage temperatures at 0900 hr were close to the Low Is. station values, being on the average 0.2C lower, while the 1730 hr values at the Anchorage were generally higher by about IC on the average and up to 3.5C . Endean er a/. (1956b) gave temperatures for August 1952 to August 1954 obtained from the Marine Branch, Commonwealth Department of Shipping and Transport, for 25 30°C 7 TEMPERATURE AUG SEP oct NOV DEC JAN FEB MAR APR MAY JUN JUL 36% + 1 41 1 L 4 1 1 1 1 n 4 35 3475 > = z 5 q nn 33 AG = 10 atT = 25° S = 35% loo # RAINFALL (10 DAY TOTAL ) 100 31 1 I a | o | AUG SEP Oct NOV DEC JAN FEB MAR APR MAY JUN JUL Fig. 15 At Low Is. station, Great Barrier Reef (16.4 S, 145.6 E), one year’s records (data from Orr, 1933a) of: (A) Water temperature: full line=surface, dashed line = 28 m depth, dotted line =smoothed air temperature, (B) Salinity: full line = surface, dashed line = 28 m depth; and rainfall by 10-day periods. four islands. Unfortunately only the monthly range of values (from 15 or more observations per month) was given and not the mean value. If on¢ takes the mean of the range bars shown, the Low Is. values fur 1952-54 average 1.0C lower (2.5C maximum) than Orr’s values, while the annual range of the means is 6.4C against Orr's 8.0C , Although the location and procedure for obtaining the Marine Branch values were not stated, the differences suggest that year to year variations may be expected. In February, at stations 130 to 180 km north of Low Is., the temperatures were within 0.5C of those at Low Is., while in Papuan Pass (16.8 S) and in Cook’s Passage (14.5 S the temperature was slightly lower than at Low Is. Between August and November ata station in Trinity Opening (east of Low Ls.) the water was LC: warmer than at Low Is. At these times, then, temperature did not change much over about 2 of latitude (220 km) along the lagoon in this region. Orr bricfly summarised temperature measurements (presumably surface values) made by the Commonwealth Navigation and T.ighthouse Service at positions along the lagoon northward from 25 S for several years and quoted a maximum recorded temperature of 28.9 C, the same as the maximum at Low Is., and a minimum of 17.8 C at the south. He also noted that the maximum range of temperature during one month from south to north was 7C and the minimum range was less than 1C (this must have been during the summer; see Fig. 17). Orr gave no further data from this source. Kenny (1974) measured the sea temperature in the top 25 cm of water, and the air temperature aboye the water on some occasions, at the seaward end of the Townsville Harbour eastern breakwater (19.25 S$, 146.83 E) at irregular intervals and varying times of day during 1961-62, 1966-67, 1968-69 and at weekly intervals at 1400 hr EST during 1970-71, making a total of 212 observations. Fig. 16 (upper curve) presents Kenny's overall mean values by month (joined by a cutve) and the ranges of values observed in each of the years (vertical bars), (‘he individual values were not tabulated in Kenny’s paper.) The mean values show a minimum of 21.8 Cin July anda maximum of 31,2 Cin January (compared with 21.5 C and 29.9 C in a single year at Low Is.). ‘The individual temperature readings were plotted but not identified by year; the monthly ranges of temperature for the four years vary trom 2.0C to 4.4C , which is not negligible compared to the annual range of means of 9.4C . Kenny pointed out that the range of values in one year may lie completely outside that for another year for the same month, c.g. March and April. Kenny stated that water temperature (T\,) was linearly rclated to air Lemperature ('T,) at 1 m above the water surface (in the 1970-71 series) by the relation: Ty, =0.83 T, | 3.31. (The simpler relation T,,=0.95 T, fits the data just as well, considering the considerable scatter in the T,,, T, correlation plot.) Both formulae give a sea temperature less than air temperature (over the range of values recorded) all year round, although Kenny did not remark on this. This result contrasts with Orr’s (1933a) note that water temperature was less than air temperature during the spring warming season, the same during the period of maximum temperatures, and greater during the cooling season (see Fig. 15.4). Possibly Kenny’s result of constantly cooler water than air is an artifact arising from the fact that in deriving the water, air temperature relation he lumped together all his 1970-71 data (his Fig. 3), not attempting to distinguish seasons. However, more than 90", of the points on his Fig. 3 plot of T,, versus T, show the water temperature lower than the air temperature which does suggest predominantly cooler water than air (as his data points were taken at uniform intervals of time). Hedley (1925c) presented a series of daily (0900 hr) measurements of temperature of a bucket sample drawn from 1.8 m depth below the surtace at the Pile Lighthouse (27.3 S, 153.2 E) in Moreton Bay for a year from 1 June 1924, Although this location is a little 27 south of the present area under review, it is included here because of the paucity of data for the Reef area, and because it is one of only two series of dai/y observations available. he monthly means and ranges of values are presented in Fig. 16 (ower curve). The cycle is seen to be similar to that at Low Is. and ‘Vownsville, although the maximum at Townsville occurs a month earlier than at the other two locations. ‘Mhe monthly ranges at Townsville (0.5C to4.9C over four years) and Moreton Bay (1.3C to 4C_) are greater than at Low Is.(0.6C to 2G ). Inthe Moreton Bay serics the change between succ sive days averages 0.5C with 85°,, of changes being 1.1C. or less and the maximum being 2.2C . Trip, SURFACE WATER TEMPERATURE °C 16 Seasonal variation of surface water temperature at ‘Vownsville Harbour Breakwater (19.25 S, 146.83 19: fullline overall mean, vertical bars — monthly range for cach year (from Kenny, 1974); and at Moreton Bay (27.3 8S, 150.2 Ej: dashed line — mean, vertical bars = monthly ranges for June 192-Fto May 1925 (from Hedley, 1925b). ANNUAL RANGE 1970 — 71 1968-69 1966 —67 | 1961 — 62 = y | TOWNSVILLE HARBOUR BREAKWATER | fo Eee RY y= wel | 25° aut bare | | ly | | ° AN | ol | \ | | Pte l \| l | IN, y a ] | [X95] | I Nea | 4 1 NI 1 i, MORETON | | \BAY 1 1h | | \ i | “ | | \ | ’ 20 \ Fa l | N | / [MONTHLY RANGE | :% T peel | I | |/ \ | T \7 \ ee 4 Nl lao © OVERALL MEAN VALUE asta | | J | \ I I 1 15° T T T T T — ss oC OT T#ff.Vao OEC JAN FEB MAR APR MAY JUN JUL AUG SEP oct NOV DEC MONTH Surface temperatures along the lagoon Aurosseau (1938) transcribed some records of surface water temperature and specific gravity made by Capt, Ogura en route between Nagasaki and Sydney and return. Values in the Barrier Reef lagoon are presented in ‘Tabic 3. hese temperatures are similar to those given by Brandon, discussed below (see Fig. 17). Table 3. Surface water temperature, air temperature, and specific gravity in the Barrier Reef Lagoon. (Data from Aurosseau, 1938) Tome Date = Temperature (C) Sp. Gr. (hr) (1928) Lat. Water Air (15 4, 1751 28 July 9.68 25.0 AT =3.8 24.6 1,0253 0630 3 Aug. 24.18 21.2 et 71 1.0264 0636 30 Aug. 9.758 25.0 AT =3.4 25.0 1.0259 0600 25 Aug. 24.08 21.6 ~~ 21.4 1.0264 Brandon (1973) wrote a review of some of the water properties of the Great Barrier Reef province, using (a) his own observations (temperature measured with bathythermo- graph and reversing thermometers) collected during four cruises between September 1967 and August 1968, (b) those of Orr (1933a) and Moorhouse (1933), and (c) data from the CSIRO Division of Fisheries and Oceanography, described in his text as unpublished, and not referenced in his bibliography. (These data are probably from the CSIRO data bank from which their ‘Oceanographic Charts, Tasman and Coral Seas’ were prepared—present reference CSIRO, 1974.) Brandon generally did not distinguish the data sources in his review and unfortunately gave inadequate information to judge the statistical significance of his mean values and ranges. (Sce note at end of Chapter X.) Brandon (1973, Fig. 3) presented graphs of mean temperature against latitude by 1 intervals for January to November (December being omitted because of inadequate data although December values were referred to in his text). These graphs are mostly very irregular and lines for individual months cross over each other apparently at random. He gave no indication of numbers of observations but it is difficult to believe that the long term average values would be as irregular as those in his plots. He summarised the surface temperature extremes as: Temperature North Centre South Maximum January January February Minimum June-July to July-August These observations agree with the maxima shown in Fig. 16, but the Moreton Bay minimum in the extreme south is earlier than Brandon suggested. To show the seasonal cycle and the latitude variation, Brandon’s (Fig. 3) temperature values were averaged by month for zones 10.5 to 14.5 8, 15,5 to 19.5 S, and 20.5 to 24,5 S in an endeavour to smooth the data; temperature/time plots are presented in Fig. 17. Even with this averaging there is still considerable scatter remaining, particularly in the January to May period. Sea areas around Australia (Roy. Neth. Met. Inst., 1949) has monthly charts for sea surface temperature averaged by 1 squares with whole number isotherms drawn on the plots. For the Barrier Reef region, temperature/time plots drawn from these data show variations similar to those in Fig. 17, but the absolute values average about 0.7C lower 29 SURFACE WATER TEMPERATURE than those in Fig. 17. There are much larger numbers of values (30 to 150", per 1 square) than Brandon can have had, but despite this there is considerable scatter (+0.5C ) in some areas of the Dutch data and the positioning of the isotherms in the charts is by no means unequivocal. In addition, the original data will have been taken with a variety of instruments, probably not calibrated frequently, so the scatter is not unexpected. Thus, Brandon’s data and the Dutch data are probably reasonably consistent. The Marine Branch values quoted by Endean et a/. (1956b) for August 1952 to August 1954 compare with Fig. 17 as follows: Station Marine Branch values average: Booby Is. 10.7 S 0.1C lower Jan.-June than Fig. 17, NORTH; 1.0C higher July—Dec. than Fig. 17, NORTH, Low Is. 16.6 S 1.0C lower for the year with minimum difference in winter and maximum of 2.5C lower in Oct., Trent Is. 20.5 S | ss than 0.1C diff ok \ eI Pine Is, 21.5'S | Less than 0, ifference from Fig. 17, SOUTH. The range of temperature between annual maximum and minimum, irrespective of month, for each degree of latitude from Brandon’s Fig. 3 is shown in Fig. 18. Also included are the annual ranges for Low Is. (Orr, 1933a), Townsville (Kenny, 1974) and Moreton Bay (Hedley, 1925c). The first and last are close to the mean of Brandon’s data but the Townsville value is considerably higher. Kenny (1974) attributed the larger Townsville range (than at Low Is.) to the ‘more oceanic locality’ of the Low Is., and also noted that the range between extreme temperatures at Townsville (12.4C ) compared more nearly with the extreme range at the Low Is. Anchorage (12.7C ). However, the latter range is between the lowest 0900 hr reading for the year at the Anchorage and a Fig. 17 Seasonal variation of surface water temperatures along the Great Barrier Reef, inshore stations (data from Brandon, 1973). North zone—10.5 to 14.5 S, Centre zone—14.6 to 19.5 S, South zone—20.5' to 24.5 S. 20 . =f. “Let, ok TT T_ “T T JAN FEB MAR APR MAY JUN JUL AUG SEP oct NOV DEC JAN MONTH pronounced peak in 1730 hr readings which are generally higher than at 0900 hr. A more realistic figure for the Anchorage would be 10.0C._ for the 0900 hr range, Therefore the higher than average range for Townsville may be attributable to the location being very close inshore where surface water property ranges are often greater than those offshore due to the presence of river runoff giving a more stable surface layer. It may be noted that the fairly detailed Low Is. records (Orr, 1933a) showed the highest water temperature in December through March but there were four subsidiary maxima during that period, while the daily Low Is. Anchorage data (Moorhouse, 1933) showed a series of subsidiary maxima and minima of about 25 days period during December through March. The warming and cooling periods were less irregular. The fluctuation in surface water temperature in the north in summer might be due to more frequent periods of light winds during the NW monsoon, allowing greater surface layer heating than in the south where mixing induced by the SE trade winds may be expected throughout the year. Brandon (1973) commented on two periods of lower temperatures between 14 and 15 S and remarked (p. 198) that ‘This area of the shelf is unique in another respect as it contains the large, shallow Princess Charlotte Bay indentation of the coast.’ Itis not clear why Princess Charlotte Bay was considered topographically unique as there are other similar arcas along the coast, such as Halifax Bay (19 S) and Broad Sound (22 S), He continued, ‘It is felt that the water temperatures here are a combination of shelf conditions, wind and weather. This area is somewhat more independent than other shelf areas immediately north or south in that a larger number of factors can influence the water column here.’ The statement is grammatically faulty and very vague, and its meaning is not clear. Sub-surface temperatures At the Low Is. station (Orr, 1933a) the difference between the surface and the 28 m depth values was generally small, 38", of the differences being less than 0.1C , 60"), less than 0.2C and only 15", were greater than 0.5C . As may be seen in Fig, 15, the large values occurred during the summer when the surface layer had a low salinity and the upper part of the water column was more stable than usual. The surface water was generally warmer than the 28m water during spring and summer. For 28", of the stations, the surface water was cooler during the autumn when the water was losing heat, but the water column was stable or neutral within the probable accuracy of measurement. Fig. 18 Annual range of surface water temperature versus latitude for Great Barrier Reef. Full line and circles = mean for inshore stations (from Brandon, 1973), dashed line = Coral Sea (from Roy, Neth. Met. Inst., 1949), dots =zone averages outside reef (data from CSIRO, 1974). 2 10 7] 4 wi TOWNSVILLE & (KENNY) = MORE TON Bays im HEDLE Q ° 45 © oe EOE Q sb rd ° a OINSHORE STATION w o z | ~ 208i 54 7 a a ®ZONE AVERAGES .- = 5s OUTSIDE REEF wi = T 3 oh e. - Zz - z CY $ L—Ta ete Fal F ; "dae 10° ys° 20) 25°S LATITUDE 31 Brandon (1973) described the vertical temperature distribution for the lagoon, although only qualitatively, essentially as summarised above for Low Is. He also stated that ‘in the Capricorn Channel, vertical temperature variation 1s only 1 to 2C on the average’ but did not say over what depth range nor which way the temperature changed with depth. (‘Vhis channcl has depths to 120 m close to the reef line and increases to over 400 m just outside.) Brandon also stated that his bathythermograph measurements over the shelf revealed essentially isothermal water with a few tenths of a Celsius degree warming in the upper few metres, but he did not state which depth range bathythermograph he was using (standard depth ranges are to 60, 135 or 270 m). SALINITY Surface salinity at a specific location The annual cycle of values for 1928-29 at Low Is. is shown in Fig. 15B from Orr’s (1933a) data. Orr commented that the maximum values (35.40 at the surface and 33.47 | at 15 min October) are ‘a little lower than would be expected for this latitude’ but he did not say on what he based his expectation, He attributed the low values to ‘the proximity of the coast and the considerable land drainage’ without giving any data for the latter, although he did quote a few rainfall figures for the coast. He attributed the variations in salinity chiefly to reduction by rainfall, with subsequent return to higher values due ‘to wind mixing more saline water upward.’ He did not mention the possibility of increase due to evaporation, having previously noted that the ‘difference between the vapour pressure of water and air shows no important seasonal variation.’ Orr was probably in error in dismissing evaporation as unimportant because the rate of loss of water vapour (and hence increase in salinity as well as loss of heat) is also a function of wind speed, increase of which increases the eddy diffusive transport from the sea surface. In the Low Is. area the winds are steadiest and strongest in the dry season (e.g. Fig. 4, Cooktown) and evaporation could be a significant factor. (Brandon did suggest evaporation as being a significant factor as will be mentioned later.) Orr noted the minimum surface salinity of 31.3 |, remarked that lower values might have occurred in the intervals between occupations of the station, and warned that minimum values must be expected to fluctuate considerably from year to year. He commented that on 28 February the Daintree River (17 km NW of Low Is.) was in flood and its turbid water was visible at Low Is. This river is in a moderately high runoff area but no figure for its flow is given in the Australian Water Resources Council’s Stream Gauging Information (1971). (My estimate, 0.15 x 10!" m3/yr, is a little less than the coastal average of 0.2 x 10'° m*/yr). Orr also remarked on the rapid increase in salinity after the reduction due to rainfall, e.g. in late February and in mid-April. The vertical dashed bars in Fig. 15B show the 10-day accumulated rainfall from Orr’s daily values, as an indication of the relationship between precipitation and surface salinity decrease, e.g. a short term effect in November and a longer term and more marked salinity decrease in January to March, presumably associated with the increased rainfall. If these events are directly correlated, the response of salinity is rapid, within a few days at most, However, the heavy rainfall in the last week of February is associated with a sharp rise of salinity, which is difficult to understand. One possibility is that some rainfalls are very localised and this, combined with currents, may move a lowered salinity patch of water into an area where observations were being made but which did not itself receive much rain. Alternatively, a lowered salinity patch after rainfall might be moved away from the site before the next observation was made. 32 Surface salinity along the lagoon Brandon (1973, Figs 4, 5 and 6) presented graphs of monthly mean and of maximum and minimum values of surface salinity for restricted latitude bands from 11. to 12 S, 16 to 17S and 22 to 23 S. The average values are presented here in a single figure to facilitate comparison (Fig. 19) and smoothed curves are drawn through them. The maximum and minimum value curves presented by Brandon are omitted from Fi ig. 19 to avoid contusion. Because the number of observations on which they are based is not given, their significance is uncertain but the range between maximum and minimum values, varying from about 0.06 to 3.0 |, 1s very wide. The greatest range, for April in the centre zone, is almost as large as the mean annual range for that region {3.7}. Itis seen that the minimum value occurs in the first part of the year (February to May) for all zones, with the maximum in the last third. The lowest values occur in the centre zone (in association with the coastal rainfall maximum, Fig. 7), and this minimum dissipates more rapidly than that in the north zone. The difference in behaviour between the north and centre zones is puzzling because the distribution in time of the rainfall is the same in these two zones (Fig. 8) and the centre zone has rather more rain on the average. One possibility is that Brandon’s figures were based on limited data and the difference in behaviour in Fig, 19 is associated with the variable nature, in both time and space, of the rainfall in this region. On the other hand, it is seen that the Low Is. data, for a different period, compare well with Brandon’s centre zone curve (apart from the April value). Another possible explanation is that Brandon’s centre zone curve is for 16 to 17 S only, and he did state that this salinity/time curve is the ‘most extreme for the centre zone’ ‘ although the highest rainfalls occur between 17 and 18 S. Brandon himself attributed ‘the great annual range of salinity .. . not only to the high rainfall and runoff (no values given) during the monkoon season but alse ta evaporation which can create localised zones of high-salinity water.’ This latter statement is surprising because the great range in the centre zone is due chiefly to the very low wet season value, not toa high dry season salinity, i.e. to evaporation. In fact, the centre zone has the /owest maximum salinity of the three. Brandon attributed the high winter salinities in the north to evaporation but produced neither quantitative nor qualitative evidence to support this, although he did make meteorological measurements including wind speed and wet and dry bulb air temperatures. Itis agreed that evaporation is likely to be significant during the dry months but it would have been more convincing to have given some actual estimates from the meteorological data. Figs. 11 and 4 together do support the thesis that evaporation is likely to be significant in the spring and early summer (September to November) because the relative humidity decreases during this period, while the winds are steadiest and relatively strong at this time. Evaporation is usually calculated from a relation of the form: Evap. (gm)cm?,sec) = Constant x wind speed x (Vap. press. at water surface—Vap, press. in air at 10 m height) and rough calculations suggest that the potential for evaporation in October-November is 1.5 to 2.5 times that for January-February for the north zone, 1.3 times in the centre zone and 1,1 times in the south zone. Wyrtki (1961b) made some estimates of evaporation values for the Arafura Sea and Timor Sea, where evaporation exceeded precipitation from April to November and June to November respectively. He obtained net rates of loss of about 25 em/month from the water column. This rate in a column of depth 30 m would give rise to an increase in salinity of about 0.25 /| per month or about 0.75 | in three months of relatively low humidity. This is about the same as the amount noted in the north zone in the second half of the year (Fig. 19). These rough calculations support the suggestions that evaporation is a significant factor in increasing the salinity in the dry season in the lagoon, 33 SURFACE WATER SALINITY “/oo ‘The small annual change in salinity in the south (22 to 23 S near Capricorn Channel) was attributed by Brandon to the area being relatively open to mixing with the Coral Sea waters (ignoring the fact that the Coral Sea waters have an annual salinity cycle) and to the predominance of the SE Trades throughout the year (presumably to their mixing effect). He could also have added that the south zone rainfall is more uniformly distributed through the year than that further north, and the annual amounts are only about one-half of those in the centre and north zones, and the lagoon is much wider here than further north. Unfortunately, Brandon did not specify where the salinity values used for the above described graphs were obtained in the lagoon. The inference is that they were obtained in the ship channel which is generally close to the land (on the average about one-quarter of the distance from shore to outer reef). This point is relevant to the next section. Surface salinity gradients across the lagoon The salinity variations shown in Fig. 19 are for locations near the land, and the marked reductions in summer are assumed to be due both to direct rainfall on the sea and to river runoff from the land. Further offshore, in the lagoon, the latter contribution of fresh water will not apply directly. In addition, as the reef is approached, the proximity of the Coral Sea waters may be expected to moderate the seasonal variations (although the Coral Sea waters show significant seasonal variation, described later). In consequence, one may expect that during the wet season the salinity will increase from the shore to the reef, while in the dry scason, when the inshore salinities increase, a decrease of salinity from shore to reef may be expected. A few scattered tests of this thesis are available from Orr’s (1933a) and Brandon’s (1973) data, as presented in Table 4. These observations are reasonably consistent with the thesis but are few. Fig. 19 Seasonal variation of surface water salinity for one degree wide latitude strips. North zone—11 to 12 S, Centre zone—16 to17 S, South zone 22 to 23 S$ (data from Brandon, 1973), and for Low Is., 16.4 S (data from Orr, 1933a), we a T T T T TTT T T ¥ JAN FEB MAR APR MAY JUN JUL AUG SEP oct NOV DEC JAN MONTH It will be noted later that less consistent results are obtained when the inshore salinities of Fig. 19 are compared with Coral Sea values near the reef (Fig, 24) and ina short section near Low Is., Orr’s (1933a) data show an increase in salinity to seaward in August when one might expect a decrease or isohaline conditions (see the next section). Sub-surface salinities At Low Is. the salinity at 28 m (deepest measurement) had its seasonal maximum in October and minimum in March (Fig. 15B), following the surface salinity change, and was always equal to or greater than the surface value within the probable precision of measurement (not stated by Orr, 1933a). In winter the difference between surface and deep values was less than 0.2 (average (0.05. ) except for two occasions in November following moderate rain. In summer, January to April, differences ranged from 0.05 to 3.35 (average 0.8), the two values over 2 being associated with heavy rain (February, Fig. 15B). The greater part of the salinity difference occurred in the top 3 to 10 m. Orr drew attention to the abrupt creases in deep salinity associated with decreases at the surface on two occasions, 13 February and 5 April. He considered it unlikely that this was due to advection from the south, where the rainfall is also generally heavy, but then said that in the south there are many openings in the reef (through which, presumably, more saline water might enter the lagoon). He also referred to udal currents through Trinity Opening east of Low Is. In other words, he did not really attempt any explanation of the phenomenon. One possibility is that the large input of fresh water may generate a temporary estuarine circulation with outflow (to and through reef openings) in the surface layer and consequent inflow below (c.g. Pickard, 1975, Ch. 8). Same observations on other tropical areas which appear to be relevant to this phenomenon are described later in the section on Tropical Estuaries. At the same time it may be noted that there were occasions when the deep salinity decreased with decrease of surface salinity, e.g. early November, from mid-December generally to early February, carly March. Uherefore, although there may be a phenomenon to be explained it would be best to have more substantial data with which to work before spending much time on it. At stations 120 to 200 km north of Low Is. at the end of February, deep salinities were within about 0.2 of those at Low Is. at the same time, but surface values were much higher, the range of values in the column being only 0,4 to 0.6, . At Trinity Opening, 40 km east of Low IJs., the salinity in the column was within 0.1 of that at Low Is. in August, September and November. Table 4. Salinity gradients across the lagoon (selected examples). O=Orr (1933a); B= Brandon (1973) Charge : Laritude Salinity Zone (S) Month f ) Direction Temp, (C) Source North 14.5 February 0.2 increase toreef (isothermal) O 11.7 Oct. or Nov. 0.6 to 0.8 decrease to reef B Centre 15.8 March 0,1 increase toreef (0.3 decr.) O 14.7 Mar. or Apr. 2.0 increase toreef (isothermal) B 17 April 1.1 increase to reef B 17-18 = April 1.9to 2.9 increase to reef B 19 April O.1 increase to reef B 14.5-16.5 July isohaline (2.6 iner.) B 16,4 October 0.15 decrease to reef (0.3 incr.) Oo 16.4 November 0.1 decrease toreef (0.4 decr.) @) At a group of stations along a line 50 km long from the mouth of the Daintree River and past Low Is. to the inside end of Trinity Opening in August, the salinity change at all depths was within 0.4 and the temperature change within 1.75C , the highest values being in Trinity Opening for both properties. Brandon (1973) referred to making Nansen casts (for water samples and temperatures) but he presented no data on vertical profiles of salinity. He stated ‘The vertical salinity profile on the Queensland shelf is, for the most part, isohaline for the majority of the year. The only exception to this is during the rainy season when the surface salinity may be greatly reduced’ and also ‘From May to the commencement of the rainy season, salinity differences between the surface and bottom waters on the Northern Queensland shelf probably rarelyexceed 0.1 ' , Local variations may occur during this time after a rainfall, but these would be exceptions and of short duration, This is most likely true of the southern shelf also as the SE Trades remain dominant over the year and are capable of keeping the water column mixed.’ The speculative nature of these remarks is surprising if Brandon had data of his own on which he could have based more firm statements. Brandon also reviewed some of Orr’s observations and speculated about lagoon salinities north of 14S during the wet season. DISSOLVED OXYGEN Orr (1933a) made regular measurements of dissolved oxygen at all depths at Low Is, and found relatively small variations with time or depth, the water generally being undersaturated. His results may be summarised as: Mean 0, conc. 90", of values Mean sat. 90", of values ¢ml/1) between: fh) between: Surface 4.6 4.3 to 4.9 95 92 to 99 28m 4.55 4.110 4.8 93 88 to 97 Slightly lower values were observed at all depths. from January to April than at other times, but the seasonal variation was small, Orr considered that there was a significant oxygen demand by the particulate material and as this was continually mixed through the column, the demand showed little or no variation with depth most of the time. Only in calm periods was there a possible greater reduction of oxygen content in the deeper water than in the upper waters, Orr remarked on the large diurnal changes in oxygen content which occur in the shallow waters of the reef flats but considered that the volume of water involved was too small for the lagoon waters to be significantly affected away from the immediate vicinity of the reef, Kinsey & Kinsey (1967) measured oxygen content over and near Heron Is, reef and showed that while large changes, particularly between day and night, occurred over the reef (7.5 to 1,5 ml)1) the changes at 2m depth only 100 m outside the reef showed a range of only 4.66 to 4,92 ml1 (95 to 102", saturation), The value 15 km SE of Heron Is. and 5 km from the nearest reef was 4.76 ml] (97"., saturation), very similar to the Low Is. values. Brandon (1973) reported no measurements of dissolved oxygen. OTHER CHEMICAL PROPERTIES Orr (1933a) reported that dissolved phosphate showed ‘no seasonal character’ although his data table showed variations from 0 to 8 ugmil. Most of the values were between 3 and 5 jgm/] with an average of 4 jigm/1 with no significant difference through the water column, and no systematic seasonal variation, 36 A few measurements were made of silicate, giving values between 40 and 110 zgm 1. A very few measurements were made of nitrate but these were regarded as uncertain because of long storage. Orr’s (1933a) measurements of hydrogen ion concentration gave pH valucs of 8.26+0.01. Brandon (1973) stated that he made pH measurements during September-November 1967 but as no significant variations or correlations with other data were found the measurements were discontinued, TRANSPARENCY Orr (1933a) made a secchi disc measurement at each Low Is. station and at the other stations in the lagoon and outside the reef. ‘The Low Is. values ranged from 3.5 to 25 m, averaging 12 m, with no seasonal variation perceptible but apparently a correlation in which low secchi disc depths were associated with high wind speeds and lasted for a few days after such a wind. From Orr’s data it can be shown that the secchi disc depth (S) in metres is related to the wind speed (W) during the previous 24 hours approximately by: (a) W in knots : for W=0 to 8 kts, S=27-2W, for W=8 to 15 kts, S=19-W, (b) Winm/s : for W=0to 4.5 m/s, S=27-4W, for W =5 to 9 m/s, S=19—2W. The standard deviation of the original points about these mean lines was about + 2.5m. Hedley (1925b) recorded secchi disc depths of 2 and 2.5 m respectively at the jetties at Townsville and Cairns, while in the lagoon he recorded values of 5 to 33 m. The reduction in transparency with increase of wind speed was attributed by Orr to stirring up of the fine grey sediment characteristic of inner passages of the lagoon (clay size fraction about 35",,). Fig. 20 Seasonal variation of water density (as ot) at 28 m depth, and difference (A a1) between 28 m depth and surface at Low Is., 16.4 S (data from Orr, 1933a). 255 “t TIME (AT 28m) Ag (BOTTOM — SURFACE), TIME 37 Approaching the outer reefs, in Trinity Opening, secchi disc depths of 11 to 30 m were recorded, being associated by Orr with the coarser bottom material there. To seaward of the outer reef values from 14 to 40m were measured. (I have observed the same correlation between poor visibility in the presence of fine bottom sediments in a lagoon contrasted with good visibility and coarse bottom sediment near or outside the reef in many areas, e.g. Glovers Reef, Tikehau, Majuro and Nomo1. This is the case even when the lagoon is much calmer than the reef area.) DENSITY Orr (1933a) presented a density (as ,) versus time plot for the Low Is. data. As the surface value curve is very similar to the salinity, time plot (Fig. 15B), it has not been reproduced here; however, Fig. 20 presents the 28 m depth a, values-versus time as well as the difference (Ac,) between 28 m and surface values. This graph shows the marked decrease of o, in summer, while the Ac, curve gives an indication of the gravitational stability of the water (Ac, Az). Actually, in the summer most of the change of density Ac, took place in the top 10 m (5 m for the cases where Ao; was greater than 1) and so the stability of the upper layer is generally greater than indicated by this curve. Fig. 21 T,S, time diagram for surface and 28 m depth at Low Is. station, 16.4 S (data trom Orr, 1933a). SALINITY “oo 32 34 36 30 1 @,=20 Nn Ww TEMPERATURE °C 25 20 38 The similarity of the salinity time and the density time curves led Orr to state that ‘It is salinity rather than temperature which is the determining factor in the stability of the sea in the Barrier Reef lagoon.’ This statement is misleading in that it is only true for one third of the year, the wet season from December to March; at other umes, temperature plays a major role in determining density. This is evident from Fig. 21, in which the monthly average values of temperature and salinity at Low Is. are plotted on a T,S diagram and connected in temporal sequence. Curves for the surface and for 28 m are presented. At the surface, for December to March the change is mainly of salinity, and the ¢, change can be attributed to this factor, but for July to December the change is almost isohaline, and the a, change must be due to change oftempcrature, For the period March to July both temperature and salinity play apart. At 28 m, salinity plays an even smaller role than at the surface. Fig. 21 also shows clearly the uniformity of the water column during the winter dry season because the T,S time curves for all depths le between the surface and 28 m ones which are almost coincident. PROPERTIES IN OTHER AREAS New Caledonia barrier reef lagoon /22 S, 166.4 E) H. Rotschi and Y. Magnier (personal communication) of O.R.S.T.0.M., Centre de Noum¢ea, made observations of water properties for a year in 1961-62 along a section from the Tontouta River in the Baie St Vincent (22 S, 166 E), west coast of New Caledonia, to the barrier reef. The water depths ranged from 6 m at the inner station to over 200 m near the reef. ‘Vhe rainfall in this region is typical of the west coast of New Caledonia, with a January maximum then falling steadily to November (5:1 ratio) before rising again, the annual average being 1100 mm (compared to twice this on the east coast). The property distributions showed a sequence from salinity stratified water during the wet season, January to April, to a relatively homogeneous water column in the drier season, August to November. At stations near the middle of the Baie St Vincent (Stn 3, depth 21 m) and near the reef (Stn 6, depth 100 m) the mean differences in the water column from the surface to 20m depth were: Stn 3 (shore) Stn 6 (reef) Sfe. to 20m Season: Wet Dry Wet Dry AT(C ) 0.8 0.2 0.2 0.3 AS(..) 0.6 0.1 0.3 0.2 AO, (m! 1) 0.1 0.1 0.3 0.1 At the deeper station (Stn 6) the property changes deeper than 20 m were small compared to those for the upper 20 m. These results are similar to those found at the Low Is. and neighbouring stations (Orr, 1933a) of the Great Barrier Reef, with a well-mixed water column except during the wet season. P. Bourret (personal communication) of O.R.S.T.O.M., Centre de Nouméa, made available surface temperature and salinity measurements for 10 months in 1966 at Pam in the north of New Caledonia (20.3 S) about 6 km from the mouth of the Diahot River in a region with an annual rainfall of 1500 mm with a January maximum and October minimum (12:1 ratio). A T,S, time plot of these data was very similar to that for the centre zone of the Great Barrier Reef (Fig. 23) but with a less pronounced February salinity minimum. The temperature and salinity ranges for the period February to November were 7C and 4.5). respectively, with the temperature minimum in July and salinity 39 maximum in October November. The larger salinity range at Pam than in the Great Barrier Reef data is attributed to the fact that Pam is closer to the river and in a more restricted waterway than the inshore channel of the Barrier Reef lagoon for which Fig. 23 data were obtained. The correspondence between these two sets of results for the New Caledonia lagoon and those for the Great Barrier Reef lagoon suggests that neither is a unique region and that there may be many features in common between the various barrier reef lagoons, Tropical estuaries Relative to the speculation that estuarine circulations may develop in the lagoon after heavy rainfall, Yves Magnier, O.R.S.T.O,M., Centre de Nouméa, drew attention to two studies made by himself and Piton in Madagascar (Baie d’Ambaro, two years’ observations, Piton & Magnier, 1971; and Baie d’Ampasindava, one year’s observations, Magnier & Piton, 1972) which suggest that the estuarine circulation hypothesis is reasonable. These bays are subject to a precipitation and river runoff regime very similar to that of the North Queensland coast, although they do not have any barrier reef outside. Ambaro is an open bay while Ampasindava is narrower. The runoff into the bays had a maximum of about 500 m*,s in February—March falling in exponential fashion to less than 10 m? s in October-November and then rising rapidly. The vertical profiles of water properties were similar to those at Low Is. (Orr, 1933a) with vertical differences as: Ditference Season (Baie d’Ambaro) Sfc. to 15 m Wet (March) Dry (Sept.) ATIC) 1.0 0.50 AS¢ ) 10.0 0.02 Aa 7.5 0.15 (Most of wet season changes were in the upper 5 m). Calculations for an estuarine circulation suggested outflow speeds in the upper layer of about 4 km day. The observations in the Baie d’Ampasindava gave very clear indications of estuarine circulation. In the dry season (e.g. October) the water was isohaline to within 0,1 from top to bottom, and had a uniform dissolved oxygen content of 4.5 ml.1 and nitrate-N of less than 0.2 jig at 1, By mid-December, when the runoff had started, there was a salinity difference of 5 between the surface and 20 m depth, and the oxygen content had decreased to 4 ml] in the bottom 10 m (40 m water depth) near the head of the bay. By February (runoff maximum) the bottom water oxygen content had fallen to less than 2.2 ml 1 and nitrate-N increased to over 6 yg at/l. These marks of stagnation at the bay head were clear indications of the development of an estuarine circulation. The surface outflow driven by the river runoff entrained and carried out salt water which was replaced by a subsurface inflow, trapping a pocket of water below the surface at the estuary head (e.g. Pickard, 1961). As the river runoff decreased in April-May and the surface outflow consequently diminished, the subsurface inflow decreased and the stagnant pocket of water at the bay head dissipated, the water column becoming homogencous by July. Therefore, in the Great Barrier Reef where the runoff characteristics are similar to those in Madagascar, with a sharp rise in January, the associated development of an estuarine circulation with outflow of the surface layer and inflow developing below may be expected. The formation of a pocket of stagnant water is typical of the conditions at the head of a bay and probably would not occur in the open lagoon of the Barrier Reef. The real extent of the estuarine circulation is something which will have to be determined by observation. 40 Piton & Magnier (1971) also described very clearly, from results at a 50-hour station, the effect of wind in causing mixing. For the first 24 hours, the wind averaged 2 m’s and the water (12 m depth) remained distinctly stratified with a salinity difference top to bottom of 1.6)... The tidal amplitude was 3 m at this time. ‘When, in a period of about 2 hours, the wind rose to 7 m/s and within 9 hours the water had become completely isohaline from surface to bottom. The salinity stratification was redeveloping within 6 hours of the wind starting to decrease from 7 m/s to 4 mis. 41 VI T,S Characteristics of Surface Waters TS DIAGRAMS Brandon (1973) presented ‘VS scatter diagrams for the surface waters, which fairly well represent the water column much of the time. In such diagrams, the 15S combination for each station fora cruise is indicated by a dot, and the dot distributions are then studied to see if they show any systematic groupings which might assist in describing the distribution in space of combinations of P\S properties. A comparison of such diagrams fora series ol cruises may then give some ideas about relations between regions and about changes with time. However, one must remember that at the surface, temperature and salinity are not conservative properties and ifa cruise extends over a significant period of time the seasonal changes of the properties may distort the distribution of points on the diagrams. Pive T,S scatter diagrams, for spring (Sept. to Nov., 1967), autumn (March to April, 1908) and winter (June to July, 1968) are given by Brandon (1973). Unfortunately, wthough the individual points are shown on the PS diagrams, the locations of most of the stauions are not given in the paper. On the diagrams, the geographical locations are indicated only in very general terms. Rather than reproducing Brandon’s five scatter diagrams (which use different scales), mean lines have been drawn through his groups of points and presented together on one diagram (Mig. 22) to facilitate comparison between locations and seasons. (it should be noted that the actual points scatter over a zone of width | O.) to 10.2 (for isothermal water) or over a zone of width +0.5C (for isohaline water) about the mean lines.) Because the locations for the points on Brandon’s diagrams are located for broad areas only, there was some difficulty in drawing the mean curve for the 10.5) to E8.5 S part of the Inner Shelf( March April), curve A.2. Note also that the June July cruise extended only to 17 S, whereas the others extended to about 20S. Some of the features shown by Mig. 22 are: Gi) the seasonal temperature and salinity cycles for lagoon waters in the sequence of curves AL, A.2, ALS (see also the next section), (bya significant scasonal PYS evele for Coral Sea waters just outside the northern reef’ (curves ©.2, 0.3), (¢) the large salinity gradient across the shelf for the autumn (A.2 at about 16.5 S, B.2, (32), (d) the evele in the Vorres Strait (1, 1.2, D.3) including the low salinity (1.2) due to eastward transport of low salinity water from the Arafura Sea and northern Gulf of Carpentaria during the NW? monsoon. One notices the gap between curves A.3 and C.3. As the latter represents a transect from outside the reef to the shore, one would expect the shoreward end of C.3 (low temperature) to approach the 17 S end of A.3 which is for the inner (shoreward) part of the lagoon. Brandon stated that “Phe P\S values on the shelf south of Cape Grenville (12.0 8) to Princess Charlotte Bay (14.3 S) are not shown but they reflect the trend of gradually increasing salinities from the low values registered during the monsoon season’. 4D Possibly the A.3 curve represented values during June while C.3 represents later values in July when the salinity on the shelf had increased seasonally. Another feature of Brandon’s T,S scatter plots is an apparent bunching of points geographically. For instance, on his Fig 8 (Sept.—Oct.) there is a gap with no points in the temperature dimension between 18 and 19 S, and in the vicinity of 13.5 S (latitudes taken from his text description). In his Fig. 11, there is a gap of 0.6C. and 0.4 at 22.8, of 0.7 between 19 and 18.5 S,and ofabout 1C and 1 at 14S. If these gaps in the occurrence of water properties are real, representing step increases in properties over short distances, rather than resulting from spatial gaps in the station distribution, they present an interesting feature to be explained. Fig. 22. Mean T,S, curves for surface waters of areas of the Great Barrier Reef lagoon, the Coral Sea just outside the reef, and the ‘Vorres Strait, for 1967-68 (data trom Brandon, 1973): Inner shelf: A.1—Sept., Oct., A.2—Mar., Apr., A.3 June, July, Centre shelf: B.1—Nov. (10.5 13.5 8), B.2—Mar., Apr. (14.5 -17 S), Coral Sea: C.1--omitted (no data), C.2—Mar., Apr. (14.5 -16.5 8), C.3 July 16.7 8S), Torres Strait: =D.1—Oct., D.2—Mar., Apr., D.3—July SALINITY "Joo 34 32 33 30+ 1 —L 1 Gan ° 4 OD. 10:3, 37 D2 ss _ — a, a 14°S 4 B.2 aw, nee 185° oP oo ™ WW a —_ = 4 25 a = Ww ra SCATTER POINTS ABOUT MEAN CURVES 4 F 4 20+ 43 T,S, TIME DIAGRAMS The T,S, time diagram for the Low Is. data (Fig. 21) has already been discussed. From Brandon’s (1973) data via the present Figs. 17 and 19, T,S, time diagrams of monthly mean values are presented in Fig. 23. (Note that these data are not homogeneous in that the monthly temperature data (Fig. 17) are for latitudinally wider zones (4 to 5 degrees) than are the salinity data (Fig. 19, 1 degree wide zones). However, the only effect of this will be to emphasize the reduction in salinity in the wet season for the centre zone.) An indication of the scatter about the monthly points is shown in the diagram (+0.8C , and +0.2 for north and south zones and +0.4 for centre zones). Fig. 23. T,S, time diagrams for surface waters of the Great Barrier Reef lagoon, shore side, for zones as: ‘Temperature Salinity North zone: 10.5 14.5 S, 11 —12 S, Centre zone: 14.6 —19.5 S, 16 —17 S, South zone: 20.5 —24.5 S, 22 —23 S, (data from Brandon, 1973) SALINITY “%/oo 34 32 33 35 36 30-4 i 1 i 1 7 1 1 1 a Fee, NORTH JAN + fee CENTRE dAN SN MAR pec Je ae AR t. Se t=20 PAPR 4 21 / MAY, APR 22 JUNS oO ° 4 Ww MAY a = ‘4 i= at a wos = WW = 23 JUL 4 Jun& x MEAN SCATTER: AT = £0-8° ALL AS = £0-2%» N@S —4 pose hide * 0-6%e. CENTRE ——---l 20 44 Data from some stations taken from F. V. Degei in August 1965 (CSIRO, 1968a) between 22 and 24 S fit these curves within the scatter given, being about 0.5C warmer on the average and 0.2').. more saline. Fig. 23 reveals very well the cycle of temperature and salinity changes in the lagoon, and the different emphasis in the three zones. The most obvious feature is the greater extent of salinity variations in the north and centre zones than in the south, related to the rainfall cycle. The second feature shown is the greater difference in minimum temperature between the zones than in maximum temperature. The third feature is that, as remarked for Low Is., the annual cycle of density is determined both by temperature and by salinity changes; in fact, in the south zone the temperature plays the major role because the salinity change is quite small compared to that in the other zones. 45 VII Water Properties Outside the Reef INDIVIDUAL PROPERTIES OUTSIDE THE REEF Orr (19334) made a few measurements of water properties just outside the outer reef line at‘Prinity Opening (16.4 S) cast of Low Is., at Papuan Pass (15.8 S) about 75 km north of Low Is., and at Cook’s Passage (14.5 S$) about 200 km north. At ‘Trinity Opening in October and November (dry season) the surface temperature and salinity values were essentially the same as at Low Is. at the same ume. At Papuan Pass in March (end of the wet season) the temperature was the same and the salinity greater than at Low Is. by 1.2. jandat Cook’s Passage in February (wet scason) the temperature was 1.6C higher and the salinity 1.7 higher. Che oxygen content was essentially the same at Low Is. and outside the reef in all three cases. Below the surface all properties were much the same as at the surface to 50 to 100m depth (wind mixed layer). Below this a thermocline extended to the maximum depth sampled (to LO Cat 600m), ‘There was a salinity maximum of about 35.7 at 100-200 m and an associated oxygen minimum of about 3.5 ml 1. Orr regarded the salinity maximum as unexpected but it was, of course, characteristic of the Pacific Subtropical Lower water of wide distribution in the south-west Pacific (see Part 2.—Western Coral Sea). As the depth of the wind-mixed layer outside the reef (up to 100 m) is greater than that of most of the lagoon (Mig. 3B), itis the properties of only this layer which are significant in exchanges between the sea and lagoon. Also, because the layer is well-mixed, the surface values for the sea outside provide most of the information needed when considering the effects of the outer sea on the lagoon waters. Brandon (1973) discussed conditions outside the Reef but presented no data of his own, relying on other accounts and particularly on three sections into the Coral Sea made from the Unutaka Maru in December 1967 (no reference given) starting at 13 ,16.5 and 18.5 S. He was particularly concerned about Orr’s (1933a) and Maxwell’s (1968) suggestions on the possibility and significance of upwelling outside the Reef. He concluded that there was some evidence in the Umitaka Maru data for limited upwelling, possibly from between 50 and 100 m depth. I do not think that the data show this; and Brandon himself, carlicr in his paper, had accepted Wyrtki’s (1960) suggestion of downwelling against the Reef. (It should be noted that Brandon’s technique (1973, section V.B) for calculating geostrophic currents in the upper layers is not acceptable dynamically, nor is his subsequent treatment of these currents as absolute currents.) A feature of the Unutaka Maru data as presented by Brandon, but upon which he did not comment, was the almost complete absence of an upper mixed layer. The temperature decreased almost linearly from the surface to 250 m at all three lines of stations (from about 28 to 17 —20 C), Salinity data were not shown but density decreased steadily from the surface to 250 m witha slightly smaller rate of decrease in the upper 30 m. This lack of a marked mixed layer is unexpected in this region and when the SE trades are still blowing (December). Lo T,S, TIME CHARACTERISTICS OUTSIDE THE REEF From the 1966-74 Monthly Oceanographic Charts, Tasman and Coral Seas (CSTRO, 1974), Coral Sea surface temperature and salinity values just outside the reef have been estimated for the three zones: north (10 to 14.5 S), centre (15 to 19.5 S) and south (20 to 25 S). The data in the charts are averaged by 1 squares for each month. Because of the limited number of observations in most months in the north-western Coral Sea, interpolated (and sometimes extrapolated) isotherms and isohalines had to be used in most cases. The procedure was to estimate the range of valucs for each month (1966-74) for each zone, reject the highest and lowest months, and take the mean of the remainder and also estimate a range (+) which included all or most of the remaining values, rounding off to the nearest 0.5C or 0.1. The mean values for the three zones are presented in T,S, time plots in Fig. 24 (the centre zone plot is displaced to the left to reduce crowding). It should be noted that data are missing or inadequate to make estimates of temperature or salinity values outside the Reef for 32",, of months for the north zone, 26",, for the centre and 6"), for the south. In many cases, particularly in the north, the number of actual values for the water properties was small and the interpolated values cannot be considered very reliable. A conspicuous feature is that the north zone shows the lowest salinities, rather than the centre zone as is the case for the lagoon waters. This is probably due to the seasonal eastward flow through Torres Strait of low salinity Arafura Sea water under the influence of the NW monsoon (Rochford, 1959), together with a possible reduction of salinity due to river runoff from Papua New Guinea (Scully-Power, 1973a). Fig. 24 TS, time diagrams for Coral Sea just outside Great Barrier Reet, mean of 1966-74 with indication of scatter about mean (data from CSIRO Atlas, 197-4). SALINITY "loo 36 30 it L a \ NorTH SCATTER (10~14-5°S) 7 a FEB, / JAN wi \vec on HK 0 —_—__ NORTH | AMAR APR FEBgJAN vA r DEC NOV Vi wi ] oO 222 APR \iov \\4 P 474 4 rs ‘pat \A W 9 \ a \ g DEC ra CENTRE \ oct > s (15-1955) aprl |\ CENTRE uw 4 AMAY . a / | \ = JUN SEP! \nNov tw AOct 2 ] oP 1 23 AUG may, 25+———_ °. 0 e—! T JUL | | ASEP I | I | jun! | 4/23 JUN “A AUG . DET -—se— \ SOUTH, SOUTH \ [ize 27° Ss) Z| Are | SEP ee + uh 2% M 24 "auG 22 _| Orr's (193 $a) data for stations just outside the Reef compare quite well with these TSS, Hine diagrams except for bis 18 March values outside Papuan Pass (north zone) where he measured 34.6 against the value of 33.0 from the CSIRO data. Brandon’s ( 1973) values (Hig. 22) agree with Mig. 24 if one associates his 14.5 to 16.5 S Coral Sea values with my centre zone (15) to 19.5 8) values. Values obtained from FV. Degei in August 1965 (CSIRO, L968a) just outside the reef between 17 and 18 S and at 24 S fit the T,S, time curves of Lip. 24 within the scatter indicated. ‘PEMPERATURE AND SALINITY VARIATIONS ACROSS ‘VHE LAGOON FURTHER REMARKS ‘Phe difference in temperature and salinity between inshore lagoon waters and Coral Sea waters has also been estimated, by subrracting the Coral Sea values in Big, 24 (CSIRO dati) from the inshore values of Fig. 23 (Brandon’s data), Vhe results are shown in Fig. 25. Some of the expected features of the shore to Coral Sea salinity changes are evident, e.g. in the dry season the salinity decreases seaward in the north (and possibly in the south) but not inthe centre zone, and in the wet season the salinity increases seaward in the centre zone. Av unexpected feature in the north is a decrease in salinity to seaward in the late wet season (March and April), ‘Uhis feature is a consequence of the marked drop in salinity in March and Aprilof the Coral Sea water (Hig. 24) which was not taken inte account in the discussion above. ‘The CSERO (197E) data on which this March and April salinity drop ts based are very scattered for this region and, in the absence of direct evidence, the features should be regarded us uncertain (i... both the low salinity in the Coral Sea just off the Reef in March and April and the consequent decrease in salinity from shore to sea in the north yone at the sume time). ‘Vhe few available data on shore to sea changes (plotted in Fig. 25) show rough agreement, although the large values for salinity increase to scaward in the centre zone, wet season (Brandon’s data), occur later than the curve suggests. “The smaller salinity differences in the south zone are to be expected from the smaller and less concentrated rainfall there than in the north and centre. Kor temperature dilferences, the few values available from Orr and Brandon are also plotted on Fig. 25, Again Brandon’s value in July in the centre zone is much higher than the curve values, but Orr’s centre zone values are smaller aad close to the curve except for November. ‘lhe remaining conspicuous feature is the marked temperature rise to seaward in the winter in the south zone, Having derived those curves and discussed their features and Orr’s and Brandon’s data, a note of caution is in order. Looking at the two figures (23 and 2-44) from which Fig. 25 was derived, one notes the information on scatter about the points (about +0.8C and t 0.4 ). These seatter magnitudes are indicated on Vig, 25 and suggest that possibly the only features which are statistically significant are the salinity increase to seaward in the centre zone in summer and the temperature increase to seaward in the south zone in winter, If property gradients from shore to sea turn out to be important, it will be necessary to verify or redetermine the Mig. 25 information by direct measurement along transects. “Che best way to carry out such measurements would probably be to use continuously recording instruments mounted below the surface in order to avoid the errors introduced by short term variations (see Fig. 15 for examples of variations associated with weekly sampling or the graph of daily temperature readings presented by Moorhouse, 1933). In his Ph.D. thesis, Brandon (1970) included the oceanographic data obtained during his three cruises in Cape Moreton and copies of Unitaka Maru data as follows: Brandon (1) Cruise GBR-DBI 67: 17 Sep.-21 Nov. 1967 trom 27.0 S to 10.5 S and return to 15.0 S, surface samples and vertical profiles, 48 (2) Cruise GBR-DB2/68: 17 Mar.-17 Apr. 1968 from 10.6 S to 27.0 S, surface samples only. (3) Cruise GBR-DB3/68: 15 June-29 July 1968 from 16.8 S to 10.5 S and return to 17.2 S, surface samples and vertical profiles. Umitaka Maru (4) Stations UM6715-6730, 6-12 Dec. 1967 in the Western Coral Sea from 12.0 S to 18.5 S, surface to 1000 m or deeper. Brandon did not present tables of his data obtained on transverse sections across the lagoon from the yacht C-Gem (Brandon, 1970, p. 11). It would probably be profitable to review Brandon’s data in detail. Fig. 25 Seasonal variations of salinity and temperature differences across Great Barrier Reet lagoon— inshore value minus Coral Sea value. Full line—AS from Figs. 23, 24; Dashed line—AT from same Figs., Points—triangles AS, dots AT (from data of Orr, 1933a, and Brandon, 1973). MER AUTUMN WINTER SPRING SUM- JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC CES ADE Sie FA 1 ri ae 1 a 1. CENTRE 14-6°19-5° 1 ° 0 2 a s—_$ 4 e 0 Om erg nt. atop risa: beeie yet weeny ae Sa Px. 7 AT AS 4 son SN Hg ON af ce lea | h a ‘ ee ‘é a oo No AT _y ) -2 vy rear’ s @ TEMPERATURE 3 ry AcALINITY 1, SOUTH 20-5°-25° 1 A ak s4S ——A 4—__,_ AS, bette eet, Ree aE eee a me Sesion rok ai 0 “Joo eet Te eg - Pea \ _-O~ ae i ° y Pe a _ AT \ of ce? oy 4 \ -2 Odp say -3 ESTIMATED SCATTER: Ss be -—-—H4 49 VIII Currents GREAT BARRIER REEF LAGOON The main sources of information on currents are the Austra/ia Pilot (1962, 1973) and Sea areas around Australia (Roy. Neth. Met. Inst., 1949), both of which basc their statements on ship’s estimates from their navigation records. In addition, Woodhead (1970) made some drift float measurements in the Capricorn Channel area. According to the Pilot, the total current is a combination of that due to the wind and that due to the tide. Except in narrow passages, the wind driven current 1s stated to be the major component. Over most of the lagoon this is due to the SE trade winds and is therefore to the north or north-west, setting fairly along the main channels witha speed of 0.25 to 0.6m s but somewhat less south of 20 8, The reversing tidal currents are superimposed on the wind-driven current. In the open lagoon, this means that there will be a general north or north-west set varying in speed semi-diurnally with the tide. In narrow passages the tidal currents may be stronger and the total current may change direction four times in cach 25 hours. In the north zone, the current due to the NW monsoon wind is less strong than that due to the tides, generally resulting in irregular currents in this zone in December to March with, perhaps, a southerly tendency less than 0.4 mis. There are a number of remarks in the Pi/ot about currents in various places along the lagoon, and it is suggested that anyone interested in a specific locality should refer to that publication for any data which may be available. Such information however, comes from an unspecified number of observations irregularly spaced in time and can, at best, be regarded as only an indication of possible water movements. In Torres Strait, the tidal currents are strong and variable. There is considered to bea net set (mean flow) to the west from March to November but an eastward set during the NW monsoon in January and possibly also in December and February, according to the Pilot. Hamon (CSIRO, 1958) states that there is a net flow to the east between December and March and to the west between May and October, the annual mean transport being zero. The maximum net transport is 0.9 x 10°m?* s in February, (The U.S, Navy Pilot Charts (U.S.N.O.0., 1955) show a westward flow through Torres Strait even in December to February but this must be an error.) Sea areas around Australia (Roy, Neth. Met. Inst., 1949) presents information on currents in the Reef area in two forms: (a) monthly charts of mean current vectors in one degree squares and (b) current roses for two locations in the lagoon. The data were obtained from ships’ logs from several countries and for different periods between 1880 and 1939, As Wyrtki (1960) has pointed out, vector averages give too small values for the mean speed and no information at all on variability. In addition, the number of observations in the Barrier Reef lagoon area 1 squares varies from 1 to 23 per month, South of 19 S there are usually 5 or more observations for each square but north of that there are often none or only 1 or 2 observations. All months in the first set of charts above, with five or more current observations in the lagoon, were reviewed, With that criterion there is no information from 10 to 15 8, From 15 to 20 S there isa net How to the north for most of 50 the year, with southward flow in October to December. South of 20 S there is a nct flow south, which is contradictory to the information in the Pilot, although the mean speed is only about 10 km per day (0.1 m/s). Probably only limited weight should he given to this source of information for scientific use. Its main value is presumably to alert masters of vessels to regions where significant currents have been observed in the past (and therefore might be in the future). The second presentation is limited to two locations for the lagoon, at 19 S, NE of Townsville, and at 23.5 S, NE of Rockhampton. For these positions, current roses are given with frequency of occurrence and speed for 16 directions for each month. Most directions are represented for most months but the predominant ones are: Direction sector of flow Av. obs. ——= = Position per month Calm W to N EtoS Av. speed 19S M755 49 16", 33", 32", 04ms 2258S 150.1 E 52 10", DF 39", 0.4ms Atl9 S, the W to N direction of flow is favoured for the first half of the calendar year and the Sto E direction for the second half; at 22.5 § the E to S direction is favoured all year. It is hard to reconcile this information with the P/or’s preference for a predominant N or NW set, but equally hard to ignore it. The E to S tendency at the 22.5 S location is consistent with Woodhead’s results, to be described next. Woodhead (1970) made some measurements of water movements in the Swain Reefs to Great Sandy Is. region (22 to 24 S) by means of surface drifters consisting of a22 cm diameter polypropylene plate with a 1 m long rod perpendicular to it and ballasted to float with most of the rod submerged. The drifters were dropped in bundles of a dozen from an aircraft. To determine ‘the extent to which the surface drifters were affected by wind driven movements of the surface layers’ Woodhead compared the deduced drift directions for drifters released near Heron Is. with the wind at that station. During the period of the experiment, the directions rowards which the wind was blowing were in the quadrant from NW to SW while the drifters were found in a quadrant from W to 8. Woodhead concluded from this that ‘the prevailing winds had not had predominant effects on the movements of the drifters.’ In this conclusion, Woodhead ignored the fact that a wind-driven current would tend to be directed somewhat to the left of the wind, i.e. more toward the direction in which the drifters moved, than in the actual wind direction. He also used only the 0900 hr wind at Heron Is. whereas at that time of the year the winds tend to back in the afternoon, their direction then tending towards the W to S quadrant, near to the observed drifter motion, In other wards, it is probable that the drifter movements were not really independent of the local wind, nor is it probable that they should be (e.g. see von Arx’s (1948) comment quoted below, and results under “he barrier reef lagoon off Noumea, New Caledonia’.) Drifters were released at 5 to 7 stations each on four lines oriented at about 060 true starting close to the mainland as shown in Fig.26, Lines 2, 3 and 4 were laid on 18 September 1966 and line 1 on 3 October. Abour 25", of the drifters were recovered, mostly on the central Queensland coast between 23.5 and 26'S, but some from as far as southern N.S.W, As usual, only the release and recovery points were known, while the time from release to recovery was an upper limit to the drift time, so that any calculated drift speed was a lower limit. Woodhead gives only two specds—to Jervis Bay (35 S) at 0.39 mis (32 kmiday) and to ‘450 miles of NSW coastline’ with a mean speed of 0.3 m's (25 km/day). Most of these travels would be in the East Australian Current. Ui 7 mS Qn Le 1527, 153°E iy “N wee a ae } Ba ne a ” eee r XN \ Ne? 5 \ \ f SAUMAREZ \ na J REEF wo 6) ~ x Qs SWAIN, } it \N N \™s rf < 2 P | x NNSA 3%) REEFS N F \ ae » SA. x 4 roe SN NON ~ om | a9%s phy =. x j al a™ Ey \ we i v : feet =~ 7 Zi : wi ; f. aN OY NY\ Ufo / Phi? scsi ~ 7 QUEENSLAND | SEA | | 7 a Af } 25 “—d | oe 25 GT. SANDY IS. | £7 Ss a | : yer \ BRISBANE g : a TASMAN | einen RO v as 7 SEA NEW SOUTH WALES 145° Fig. 28 The Coral Sea, and divisions used in this review. I Introduction Rotschi & Lemasson (1967) reviewed in considerable detail the knowledge available on the oceanography of the whole of the Coral and Tasman Seas up to about 1965. The present work is a review up to 1975 of the main features of the physical oceanography of that part of the Coral Sea proximate to the Great Barrier Reef and thus likely to attect the properties of the Barrier Reef lagoon waters. The formal boundaries of the Coral Sea are set out in Limits of oceans and seas (Int, Hydrog. Bureau, 1937) and are shown in Fig. 28. For the present review the limits adopted are the formal boundaries on the west and north (the Australia and Papua New Guinea coasts), the 155 E meridian, and the 25 S parallel as shown by the dashed line in Fig. 28. This part is referred to as the ‘Western Coral Sea’, abbreviated to W’CS. In addition, ‘Central’ and ‘Eastern’ Coral Sea divisions have been established for convenience in reference. ‘he eastern and southern boundaries for the WCS have been chosen arbitrarily and areas outside of them will be considered when necessary to understand the Coral Sea itself. No water masses are formed in the Coral Sea and, in fact, the source areas of most of the Coral Sea water masses are at considerable distances, including areas north of the equator, the eastern Pacific and the Antarctic. Most of the flow into the WCS is across 155 E, north of 20 S, with a seasonal inflow to the North-west Coral Sea through Torres Strait during the NW monsoon; there is outflow in the north to the Solomon Sea and in the south to the East Australian Current. This review will be concerned with the upper waters, from the surface to about 1000 m depth, as this is adequate to cover exchanges with the Barrier Reef lagoon, However, the data available are quite limited both in space and time. It is believed that the main features of the water masses are known but, as is the case in most of the oceans, their variation with time is incompletely understood and there is much to learn yet about the circulation. The chief papers on the properties and circulation of the waters deeper than the Antarctic Intermediate (about 1200 m) are those by Rochford (1960c) and W’yrtki (1961a, 1962a). These and other papers were well summarised in the review by Rotschi & Lemasson (1967) to which reference may be made. There have been no significant deep water studies in the Western Coral Sea since that review. The sequence will be to describe the topography of the region briefly (the deep bottom topography is of little concern here) and the climate, then to summarise the main descriptions available of the water mass characteristics and inferred flow patterns, and the determinations of the circulation by observation and dynamic methods, and finally to summarise the present knowledge of the region. 63 II ‘Topography In Fig. 29 are shown the 200 and 2000 m isobaths in the Western Coral Sea, with the 1000 m isobath added where the other two are significantly separated (data from Mammarickx er al/., 1974). The main features are the Queensland Plateau off the Great Barrier Reef with depths less than 2000 m (much of it less than 1000 m), the Papuan Plateau in the northwest, and the Coral Sea Basin with depths of more than 4500 m. The shallow area of the Chesterfield Is. is just outside the WCS but with depths of less than 200 m in the vicinity of the main inflow it must have a significant influence on the sub-surface circulation to the WCS. The connection northward to the Solomon Sea is limited to 200 m depth between Papua New Guinea and 154 E but has depths to 3800 m between 156 E and the Solomon Is. To the east between the Solomon Is. and the Chesterfield Is., the sill depth is greater than 3000 m. To the south, the sill depth between the Chesterfield Is. and the Barrier Reef is about 3300 m but this is in a very narrow gap and the main trough is less than 1500 m deep. The connection through Torres Strait to the west is less than 20 m deep. Typical dimensions for the WCS are given in the Appendix. 64 of: 145 PAPUA | NEW GUINEA M SOLOMON wa BASIN < 2008 °'N 2 > owe 10° 29 ee. ?: f PAPUAN ) 0 ttre PLATEAU oO | 15° ANS "QUEENSLAND — ‘4 ( 2. EWILLis 5. , . : Ne CAIRNS” “PLATEAU? ae TOWNSVILLE™ aeNe CHESTERFIELD 20° Ti: u / QUEENSLAND Roy 4 25°S DEPTH CONTOURS METRES 145° 150°E Se) fi 155° Fig. 29 Bathymetry of the Western Coral Sea, based upon J. Mammerickx er a/., South Pacific, Sheet 11, revised 1974. 65 III Climate WINDS ‘he main wind systems affecting the Coral Sea are the SE trade winds for most of the year and the NW monsoon winds in summer. Weather on the Australia Station (RAAF, 1942) presents quarterly charts for the WCS with wind frequency but no speeds. Sea ureas around Alustratia (Roy. Neth. Met. Inst., 1949) gives wind information in two forms, monthly charts of the ‘General Air Circulation’ giving the vector mean wind by 1 squares, and monthly charts with wind roses by 3 squares with frequency and wind speed for 16 directions, To provide a simpler picture of the wind character for this review, the wind rose charts have been used to prepare four quarterly mean wind rose charts for December—February, March-May, etc. ‘lo simplify the diagrams the 16 directions have been reduced to 8 (NNE combined with NE and attributed to NE, ENE combined with E and attributed to E, etc.). The results are shown in Fig. 30. In this figure, the wind roses represent the information available for the sea areas within the nominal 5 squares. At the bottom left of the square is shown the percentage frequency of occurrence of calms, and at the top right is the mean wind speed in m/s for the prevailing direction in that square. These diagrams display the main features of the wind systems. During the period from March to November (Hig. 30B,C,D) the SE trade winds predominate over the entire area; in December to February (Fig. 30A) the effect of the NW monsoon is evident to 15S. It will also be noted (Fig. 30C) that in winter, as the SE trades move north the westerlies intrude into the area and give rise to increased westerly components south of 20 S. Vhe General Air Circulation charts by 1 squares (Roy. Neth. Met. Inst., 1949) emphasise these features where the winds are predominantly from one direction, but are not very helpful where there is much variety in direction because the procedure of vector averaging conceals varicty in direction (as well as tending to give too low a mean wind speed). However, these charts include the number of observations available by | squares and this is sometimes illuminating. For instance, the region to the east of the Great Barrier Reef between 10 and 20 S is notably deficient in observations, many squares having none at all. Wyrtki & Meyers (1975) prepared monthly mean surface wind stress charts for the area from 30 N to 30 S and from 125 E to 75 W at 10 meridian intervals. The 145 and 155 E sets show the stress effect of the wind very well for the WCS. At 155 E, the northern limit of the NW directed stress (of the SE trades) moves from about 14 S in February to north of the equator in September. Vhe maximum stress values occur in July at 15 S. At this time the stress has veered until at 25 S itis directed almost north (cf. Fig. 30C).The effect of the NW monsoon is scen by the wind stress at 145 E having a predominant easterly directed component of stress as far south as 14 S in January and February, but being small or north-westerly directed for the rest of the year at these latitudes. 66 CYCLONES On the average, two cyclones per year affect the WCS. They originate in the Intertropical Convergence Zone between about 8 and 18 S inthe northern Central and Eastern Coral Sea and are most common in January to March, with less frequent occurrences in December and April. They initially move to the west; those north of 12 S tend to continue west while those south of that latitude tend to curve southeast. Fig. 30 Wind roses for the Coral Sea for four quarters: (A) Dec.- Feb., (B) Mar.- May, (C) June-Aug., (D Sep.-Novy. Bars show percentage frequency of wind direction: wind blows toward junction of frequency bars. Bottom left figure is frequency of calms; top right figure is mean speed of prevailing winds (data from Roy. Neth. Met. Inst., 1949), aw 4 AS ° TOWNSVILLE MEAN PREVAILING WIND SPEED m/sec. "oe ¥ 5 G WESTERN PERCENTAGE Mounanee. PERCENTAGE FREQUENCY OF CALMS 0 50 100 AUSTRALIA 67 Presumably the main oceanographic effects of cyclones at sea are to cause some reduction in salinity from the associated rainfall, together with an increase in the mixed layer depth due to the strong winds. Rougeric & Donguy (1975) observed a significant heat loss attributable to a cyclone during the ‘Gorgone I’ cruise. Cyclone Diane remained stationary from 10 to 15 December 1972 at 15 S, 163 E. Crossing this area on 16 December the thermograph on N.O, Coriolis registered a drop in surface temperature from 27 to 25 C and evidence of a decrease in temperature relative to neighbouring stations which indicated a loss of heat of 9000 cal. cm?, There was no upwelling, simply the decrease of temperature. ATR TEMPERATURE Sea areas around Australia (Roy. Neth. Met. Inst., 1949) gives mean monthly air temperatures from ship observations, and a few isotherms are included in Fig. 34. The isotherms are roughly zonal in alignment. Temperature values in the WCS range from over 28 Cinthe north to 24 Cin the south in the summer, and about 26 C in the north to 18 C in the south in the winter. RAINFALL Using ground station data, supplemented by deductions from satellite cloud-cover photographs, Taylor (1973) prepared an Arlas of Pacific Island Rainfall with monthly and annual charts. Fig. 31 shows the estimated maximum (February), minimum (September) and annual total values for the WCS from Taylor’s Atlas. Most of the monthly charts have the same pattern of rainfall as the annual one, with increasing rainfall to the north-east to a maximum at about 10 S$, 168 Enorth of the New Hebrides. In the south-west Pacific, the mean monthly rainfall ar this maximum off the New Hebrides is almost uniform throughout the year at about 400 mm per month, while the minimum rainfall area lies off the Queensland coast, migrating from about 25 S, 160 E in February March to 10 S, 145 E in August—October, According to Taylor’s Ar/as, the mean rainfall over the WCS ranges froma monthly maximum of about 250 mm in February toa minimum of 50 mm in September. It should be noted that the only ground station in the Coral Sea from which Taylor had data was Willis Island. Kilonsky & Ramage (1975) described a technique for estimating open ocean rainfall from the distribution of highly reflective clouds (HRC) from visual satellite picture mosaics of the tropical Pacific Ocean, having determined the correlation between this parameter and measured rainfall at coral island stations all less than 30 m high, Although there was considerable scatter about their (linear) regression relationship between HRC and measured rainfall, the correlation was reasonably significant and offers a method for obtaining open ocean rainfall over large areas almost simultaneously. Kilonsky & Ramage estimated rainfall with this technique for the period May 1971 to April 1973, and compared it with earlier estimates for the 20 N to 20 S band of the Pacific. All estimates agreed in showing a peak at about 5 N and little variation with latitude from 0 to20 S. However, there were differences between the rainfall amounts, Taylor’s (1973) values being among the highest. Kilonsky & Ramage considered Taylor’s estimates too high, possibly because his ground truth stations included a number of ‘high’ islands where orographic effects are known to increase the rainfall, From HRC charts for three years, May 1971 to April 1974, made available to O.R.S.T.O.M., Nouméa, by C. S. Ramage, rainfall has been estimated for 5 to 20 S, 150 E (the western limit of the charts for these latitudes) to 160 E. The mean annual rainfall for the three years is shown in Fig. 31D for comparison with Taylor’s values (Fig. 31C). Taylor’s values are higher than the 68 WILLIS IS. r B07 SS LTOWNSVILLE™ oe ——~++_ 20 FEB. oe SEP. (MAX.) (MIN.) RAINFALL (mm) 450 RAINFALL aaa key ak F T 142° 150°E ANNUAL TOTAL RAINFALL ANNUAL TOTAL RAINFALL i= FEB./ MAR. MEAN CLOUD Fig. 31 (OKTAS) (OKTAS) et OCT./ NOV. MEAN CLOUD Estimated rainfall (mm), Coral Sea: (A) Maximum monthly, (B) Minimum monthly, (C) Annual (A,B,C from Taylor, 1973), (D) Annual (from Kilonsky & Ramage, 1975). Cloud amount (oktas): (EF) Maximum, (F) Minimum (from Atkinson & Sadler, 1970). 69 holonsky Ramage techoique values by about 250 minyrin the south-west of the common wea, by over L000 mime yr in the middle and by 500 mimyyr in the north-east (Solomon S ay The monthly values from Kilonsky & Ramage show a seasonal ¢ ycle north of 18 S with theumaimuniin Pebruary Marchand the minimum in September October. Uhe ratio of Mastin tom imtinas harpest 3.5) atabout LOS, 155 and decreases to the south (to about Ear 20S, 150 1) Por the WES, probably the best estimate at present for open sea rainfall would be to Wise 8O" of Baylor's | 1973) elas values. Powever, thimust be realised that, where satellite photographs are the source of miformation, both the Vaylor and the Kilonsky-Ramage Values aie based on only afew years of data for the open sea. CLOUD Seaareas aremid Australia (Roy. Neth. Met. Tnst., 1949) gives the cloud amount by the nes squgtres as used for wind roses, while Ramage (1970) gives the mean cloudiness for January and July 1967 determined from weather satellite observations. Both sources indicate a small deere: scam cloud cover from about 5 oktas (cighths of sky covered) at LOS to boktas at 2o 8, with sliphtly lower values near the coast and higher at 155 KE. Amore recent analysis by Atkinson & Sadler (1970) showed a tendency for cloud amount tobe miaxmnalin February March and minimal in October: November over the Western Coral Sea, with lower values prevailing south of 15 $ than north of this latitude. Mean values (i oktas) for approximate 5 areas are given in lig, 31h, WTLLIS ISLAND Some datelor this location (position in Fig. $14) were included in Part | but as this station isclose to the contre af the WCS and is alow ishind, so that data should be representative oF open sea conditions, some of the statistics will be summarised here (for the 1922-41 period): JO Vear Wean Notes Aw temperature: Max. 28.3 C January Long-term extreme values Min, 23.9 °C July August were +5.5C relative to these (ef. + 12C at Cairns) Raantall Max. 283 mim Vebruary Rainy days = 10/month Min. 17 mm October Rainy days = 1/month Annual 1180 mm Wind Annual 9°), from NE Mean speed 6.7 m/s, all 30" from i directions (cf. coastal 17°, from SE stations | to 4 m/s). 10"), from S 8°. various Weather on the Australia Station RAALK, 1942) states that ‘rainless periods of 20 to 40 days are usual during the dry season and spells of 50 to 80 days have been recorded particularly trom June te November, while in the wet season dry spells of 10 to 15 days are Hotuncommon’, “Chis implies that rainfall is concentrated in time (3 days per month on the averape from: May to December) which would contribute to variability in time of surbace salinity. a) It should be noted that the meteorology and climatology of Willis Island is currently under study by the Bureau of Meteorology using all available data (personal communication, Dr J. W. Zillman, Superintendent, Physical Research). Some aspects of the SE trades of the Coral Sea have been discussed by A. B. Neal (1975) using regular and special observations at Willis Island. Neal states that ‘the station is fully equipped for surface observations which are taken routinely every three hours (except midnight) and include sea temperatures at 0600 and 1800 hours measured at buoys moored in deep water beyond the western reef’. 71 IV Water Masses, Properties and Deduced Flow Paths INTRODUCTION—SALIENT CHARACTERISTICS For orientation purposes, a brief sammary of the salient characteristics of the WCS water mass properties will be given before reviewing the papers upon which our present knowledge is based. Fig. 43 shows examples of the temperature and salinity distributions at the surface during the early winter, the gross features being a decrease in temperature and an increase in salinity from north to south, and a tongue-like distribution of the isopleths near the Australian coast south of about 25 S (the start of the East Australian Current system). The basic change to summer conditions is an increase in temperature of 2C in the north and 5 or 6C_ in the south, with the disappearance of the low salinity in the Gulf of Papua by early summer and its reappearance later (January). This low salinity is attributed to Arafura Sea water driven through Torres Strait during the NW monsoon, but river runoff from Papua New Guinea may also contribute significantly. It should be noted that the data on which Fig. 43 is based are from the only near- simultaneous set for the WCS. They were obtained during three Royal Australian Navy Research Laboratory cruises in May—July 1968 (Scully-Power & France, 1969a, b,c; area shown in Fig. 45A), and described by Scully-Power (1973a, b), With the exception of the description of the Central Coral Sea by Rougerie & Donguy (1975); O.R.S.T.O.M. ‘Gorgone 1’ Cruise, November—December 1972, see Fig. 45A) all other descriptions or analyses of the western half of the Coral Sea have been based on more limited data distributed over several years, although there are data for other cruises (‘Tule’ 1965, Shoyo Maru 1973, and ‘Gorgone 2” 1975) for parts of the WCS. The ‘Gorgone 2’ data are currently being analysed. In the vertical, Fig. 32A shows an almost ‘textbook’ example of the structure in the upper 1200 m of the WCS, using an R.A.N.R.L. station from the 1968 cruises referred to above. The location of the station (CS 4/2 Stn 14) is shown in Fig. 45A. The major features are: Temperature (a) an upper mixed layer of as much as 150 m depth from the surface; (b) amonotonic decrease in temperature from the bottom of the mixed layer to below 1000 m, forming a permanent thermocline with its maximum gradient between the mixed layer and about 500 m depth. Salinity (a) an upper mixed layer as for temperature; (b) an upper salinity maximum, usually sub-surface from 50-250 m depth but reaching the surface in the south of the region (Subtropical Lower water); (c) a salinity minimum at 650-1100 m depth (Antarctic Intermediate water). Density Below the mixed layer, this increases with depth, opposite to the temperature change. Density changes are determined more by changes of temperature than of salinity. 72 = 24 26 28 0 - 2 4 6 mic! Ss loo To. 30 °c oF 4 ka. { MIXED LAYER Moe <— _ SAL.MAX.(SUBTROP. LOWER) 20044 _ <— 0, MIN. <— SECONDARY _.4004 0, MIN. E = 600-4 A o, <— 0, MAX. a <— SAL. MIN. (ANT. INTER) aol Ds 1000 4 Ki, 1200+ | 1 *loo 30°44 | 4 1 L 1 4 % Ast (B) 0-75m. 4 100-24 392 S) SAL.MAX. => Vaal (SUBTROP. LOWER) = J. --25 297 204 ES 200 { 2 726 202 xz a ul 4 > weet 107 ra 10°4 4 728 12 in SAL.MIN. UAE oct if Me Oe 1 / PA [INTER “1200 y: o- + T T Tr T 34 35 36 *foo 5-4 1 4 | 1 _t = 0, MAX 0-75m. es KG) ' z 600 100 oO coral 400 [ roy 800 150 wo 200 2) 1000— az 1200— f upper 0; MIN. SECONDARY 02 MIN. 34+ Fig. 32 (A) Typical vertical protiles of temperature (T), salinity (S), density (a, ) and dissolved oxygen (OQ) for the Western Coral Sea (R.A.N.R.L. Cruise CS4/2, Stn. 14 (15.0 S, 149.5 E), June 1968) (data from Scully-Power & France, 1969b), (B) T,S diagram for the same station, (C) S,O, diagram for the same station. 73 Dissolved Oxygen (a) an upper oxygen minimum at 150-500 m depth, typically a little deeper than the upper salinity maximum (at 170 m in Fig. 32A); (b) an oxygen maximum at 400-800 m, typically shallower than the salinity minimum. A minor feature which has been mentioned only by Scully-Power (1973a) is the occurrence of a double oxygen minimum (in Fig. 32A the second one is at 300 m). Scully- Power stated that this was observed at about 25", of the 94 stations in May-July 1968, apparently occurring with no clear geographic distribution (but see the note 1n the next paragraph). Scully-Power did not discuss it further but this feature also appears in the Central Coral Sea ‘Gorgone 1’ data (Donguy e7 a/., 1972b). In fact, multiple maxima and minima frequently occur below the upper salinity maximum and, on a single vertical profile, it may be difficult to single out any one as the significant minimum. This feature should be studied; possibly itis related to successive inflows of Subtropical Lower water. Fig. 33 shows the distribution of salinity and dissolved oxygen for a north-south vertical section through the WCS from near Port Moresby to east of Brisbane, the station positions being shown in Fig. 45A. The salinity and oxygen features described above are shown, except the mixed layer which is too thin to show clearly on the depth scale used. The boundaries of the Subtropical Lower and Antarctic Intermediate waters have been taken as 35.5 and 34.5! respectively following Scully-Power (1973a). The subsidiary oxygen minimum mentioned above appears in the section of Fig. 33 at seven successive stations and seems to be a continuous and systematic feature. The oxygen maximum in the upper part of the Antarctic Intermediate water is very conspicuous in the south, and the Fig. 33 Vertical sections of salinity and dissolved oxygen for a north-south section of the Western Coral Sea, May-July 1968 (data from Scully-Power & France, 1969a, b, c} station positions shown in Fig. 45A). 146° SALINITY °/co 5 155° € "5 10°S 20° 25 278 0 - : | LS See 5 1 eye ee §5—-——-—-- ~~ gece PR aka fe keaayart, Ta CORE ~y yea ee SuetRorica, ower! * * 8 4 a a & ye 400 -|_——$-_-____ 3 0%¢ E Se = 600 B4b=- oT an a ~hL NE SPR Dy TORRE MOG NOTNOS Si WIRINS CON OC MO Sn Pr 800 o Tal 4 a, y Bi PB arial . ¢ bane ANTARCTIC INTERMEDIATE ee eee 1000 eee sou A 1 Bett a merere CORE See + NS Pat Ling ww en eed 1200 fo ee nf a + OFF PORT MORESBY EAST OF TownsvitLe OFF ariseane—t J 2vISSOLVED OXYGEN (ml./ 1) 10°S 1 20° 25° oti i fl jl p88 —— — — — — o 9 : ° == 0S Se —_— 200-4 - ao ao 0 0 2 © 2 8 8 8 ORS eg ° ° a UPPER OXYGEN MINIMUM CORE SUBSJDARY 0, MINM. est Cs | 74004 > nat ° o g 1 , c |e te Ir... [i4--- trey Bre a a er’ 2, a ae BA Bb yy F 600-4 I aL ae ee * my 4 re td / uy OXYGEN MAXIMUM CORE ~~ ~~~ 2g, 800 OO =1 I~ aan 19004 aN 1200-L 5 low oxygen northern component of the Subtropical Lower water in the north, The Subtropical Lower water mass is seen to be at the surface in the south. Vertical profiles, such as those in Fig. 32A, are generally used only in the first stages of checking data; most of the analysis of water masses has been carried out with characteristic diagrams, generally T,S and S,O, diagrams, although T,O, and S, phosphate diagrams have been used in the SW Pacific. To show how such diagrams relate to the profiles, Figs. 32B,C present T,S and $,O, diagrams corresponding to the vertical profiles of Fig. 32A. The linear portion of the T,S curve between 250 and 600 m in Fig. 32B corresponds to Sverdrup’s (1942) Western South Pacific Central water. his is not recognised as a specific water mass in the Coral Sea, while Sverdrup did not recognise the Subtropical Lower water as a specific water mass, although he did describe the subsurface {upper: salinity maximum. It should also be noted that the curves of characteristic diagrams are notoriously non-linear in depth scale (see Fig. 32B,C) and depth interpolation should not be carried out on such a diagram alone. It must be realised that the area on a T,S diagram representing a water mass does not indicate its volume. For instance, the Antarctic Intermediate water is represented by a very compact area in Fig. 61 compared to the Subtropical Lower water, but the former has about twice the volume of the latter in the Western Coral Sea (see Fig. 33). Because the depth of a particular water mass may vary from place to place, it is often difficult to determine the distributions of such masses on surfaces of constant geometrical depth (apart from the water-air boundary surface) and most studies have been carried out using either the ‘core method’ (Wust, 1935) or isentropic analysis (Montgomery, 1938). In the former, a ‘core’ is defined as the level where a property reaches an extreme value in the vertical direction, and a plot of the depth of this core for a region defines a surface called the ‘core layer’. The location where a property has an extreme value on the core layer is called the ‘origin’ and flow is assumed to take place along the core layer in the direction of increase or decrease of the property (due to mixing with water having less extreme values). In isentropic analysis, flow and mixing are assumed to take place along surfaces of constant density (¢;), which approximate to isentropic surfaces, and distributions of conservative properties (e.g. T, representing heat, or salinity) and to some extent quasi-conservative ones (e.g. oxygen) indicate flow patterns. These procedures help to identify both sources of water and flow patterns to and in a region. The flow patterns may also be studied by direct measurements of currents or indirectly by the geostrophic method. SURFACE WATER CHARACTERISTICS Surface temperature—mean distribution Sea areas around Australia (Roy. Neth. Met. Inst., 1949) gives monthly charts of sea and air temperature by 1 squares, with interpolated isotherms at 1C intervals. Fig, 34 is based on this source. The smaller scale charts in Sverdrup ef a/. (1942) agree with the Dutch data. The numbers of observations in each 1 square in the Dutch atlas vary from none at all in many squares east of the Great Barrier Reef, to less than ten per month in most of the WCS, and to a hundred or more near major ports or in major shipping lanes. The main ‘holiday’ areas are listed in the legend: There is considerable scatter between neighbouring squares and the preparation of the isotherms must have required a considerable amount of personal judgment. According to this atlas the water temperature in the WCS ranges from 26 to 29 Cin summer (30 C appears in a few individual 1 squares) and from 21 to 26 Cin winter, and averages about 0.4C_ higher than the air temperature above it. Away from the land, the isotherms are oriented zonally, Approaching the Barrier Reef they trend equatorward from April to September (winter, Fig. 34B), i.e, the water near the 75 Reef is cooler than that further offshore at the same latitude, South of 20 S, a southward tongue of warmer water associated with flow into the East Australian Current is evident both in winter and summer. This feature is much more conspicuous in the current itself south of the WCS area covered in Fig. 34. The Monthly Oceanographic Charts, Tasman and Coral Seas, 1966-74 (CSIRO, 1974), referred to hereafter as the ‘CSIRO Atlas’, present monthly accumulated temperature and salinity observations for the surface waters. The individual charts present the station positions and interpolated isotherms at 1C intervals. The station density was quite variable and for some months large areas were blank. In these cases, some of the interpolations to prepare the isotherms became essentially extrapolations, which are always uncertain, especially in a region as complicated as the Cora] Sea. In addition, there were some areas, particularly east of the Barrier Reef, where there were few or no station positions for any year (see below and Fig. 37 for more details). Mean monthly surface temperature charts were prepared from the Atlas but, as will be explained shortly, their significance is limited because of the large variations with time. The procedure used was to select a month and to trace the positions of a particular isotherm on one diagram for all years, then drawing a mean isotherm for that month for the whole period 1966-74. To reduce the uncertainties associated with the interpolations on the charts, only those portions of isotherms lying between two stations not more than 100 km (about 1 of latitude) apart were used. Values inside the Barrier Reef were excluded. The procedure was repeated for several isotherms for the months of January, April, July and October. This process indicated that the annual range of surface temperature values was about 20 to 30 C. The positions of individual isotherms extended over very large areas when the nine years of data were combined. Fig. 35 shows the areas of occurrence of the isotherms 24 , 26 ,28 and 30 C, representative of the bulk of the values occurring in the WCS. It is clear that the WCS is a region of considerable variability of surface water temperature (and of salinity, as will appear shortly). It is evident that attempting to Fig. 34. Mean surface water (Tw) and air (T,) temperatures, Coral Sea: (A) Summer (Jan. ), (B) winter (Jul.) (data from Roy. Neth. Met. Inst., 1949). Note: There are few data for most months in areas (10 -13 S, 144 -146 E), (15 -20 S, 148 —152 E), (10 -20 S, 155 -160 E). 141° 150° 160°E 141° be ¥ : ae 150 160°E ° es TOWNSVILLE reer, SUMMER | JANUARY WINTER JULY AUSTRALIA determine a mean picture of surface conditions by combining data from a number of cruises is likely to be of dubious value. Mean isotherms (Fig. 36) were drawn to show the trend of seasonal variation over the area, rather than to specify mean positions of isotherms for each month. In the main body of the WCS and within the limitations of the scatter shown in Fig. 35 there is evident: (a) location; and Fig. 35 TEMPERATURE r (°C) —_ \—__**" ___a 20° - 130% the usual trend to warmer temperatures in summer and cooler in winter at each Surface temperatures, Western Coral Sca, areas of occurrence of 24 > 26,28 and 30 C water for January, April, July and October 1966-1974 (data from CSIRO, 1974). Note: For all months there are gaps in the data off the Great Barrier Reef, see Fig. 37, and some shaded areas should probably extend further west. yi? 160°E lene, °. TOWNSVILLE AUSTRALIA 77 (b) the shape of the isotherms from a roughly zonal orientation in the north to a tongue shape in the south associated with the start of the East Australian Current. A criticism is appropriate here. If one had available only the material (isotherm spot positions) from which Fig. 35 was prepared it is doubtful if the tongue form in Fig. 36 would have been drawn. It was prompted by the individual monthly charts, most of which show this feature clearly. However, the tongue is quite narrow and it shifts slightly east or west (i.e. transversely) from year to year so that the accumulated isotherm positions on the tracing spread over an area, such as the 24 C area in Fig. 35, October, and obscure the tongue. Fig. 36 Surface water temperatures, Western Coral Sea, mean isotherm positions for January, April, July and October, 1966-1974 (data from CSIRO, 1974). Note: The termination of some isotherms in mid-sea is probably due to gaps in data coverage. T T —T 7s T + + tae SS ————— 30°C ° %o % b Se a aA: 4 ° 4 _ _a ~ XV 2° «fl 28° | APR. -20°4 MEAN 26% L ISOTHERMS \ | (°C) \26° / \ | 24 1 yt 4 P.N.G. LY fF 10° 4 \ \ | V oct a BS) | 2 i 24° | 20°4 TOWNSVILLE™ ast 4 7 To AUSTRALIA \22° WA / / \ / 20 30% 1 a L_/Z 78 The termination of isotherms in mid-sea is usually due to gaps in data coverage rather than to the occurrence of convergences or divergences. The isotherms in the Dutch atlas (Fig. 34) and those from the CSIRO Atlas data (Fig. 36) differ in two respects. The tongue form in the south is much more evident in the curves derived from the CSTRO Atlas than in the Dutch ones, and the CSIRO values are generally 1to2C higher than the Dutch ones for corresponding areas. With regard to the shape, it is likely that the construction of the curves of Fig. 36 in the present study was influenced by the shapes of the individual monthly isotherms in the CSIRO Atlas where the author of the Dutch atlas isotherms may have been working with the totality of his data (by 1 squares) and not had access to ready-drawn individual charts for inspiration, The reason for the difference in mean temperature values is not known. Of the two figures, Fig. 36 indicates the character of the isotherm shapes in the area, although in a very smoothed form as the isotherms for individual months show numerous small features or irregularities. Fig. 35 is the more realistic for describing the environment on a long-term basis and probably for relating to biological observations of fixed features such as coral reefs. For comparison with biological records of a specific (past) expedition one might be able to find data in the relevant CSIRO Atlas chart for the period 1966-74 (and later years in due course), but for future projects it would be wise to plan to include physical (and chemical) measurements in parallel with any biological ones. Finally, when using Figs. 35 or 36 one should bear in mind that the distribution of observations over the area available for preparing the CSIRO Atlas was far from uniform. Fig. 37 shows qualitatively the density of observations for a typical month. It will be seen that there are some areas for which there are no observations at all, notably and unfortunately off the Barrier Reef. As mentioned previously, this is why several isotherms terminate in this region. An alternative treatment of the CSIRO data is offered in the next section and may be sufficient for many purposes. Fig, 37 Density of observations of surface temperature and salinity fora typical month, Western Coral Sea, 1966-1974, in CSTRO, 1974. (Refer to Figs. 35, 36, 39, 40.) yar? 150° 160°E 204 TOWNSVILLE “Ss OBSERVATION SoNK fo de 774 DENSITY BV ies “| 5 OR MORE YEARS anh 2704 YEARS ———=— SINGLE TRACK 30°S : 1 79 Rochford (1973) examined surface temperature and salinity values in the T'asman and Coral Seas for 1966-70 for annual and longer term variations. He determined the annual range of mean monthly temperature as shown in Fig. 38A. The range increased from 2C in the northeast (Solomon Is.) toabout 5C._ at the Barrier Reef (and 8C_ in the south zone of the Lagoon). Ina 1 square off the Swain Reefs (A in Fig. 38A) there was little or no indication of any long-term change in mean temperature, The surface water property distributions for 1968 obtained by Scully-Power (1973a) and shown in Fig. 43 will be described later to maintain the continuity of that 1968 data set information and after the possible source areas for waters in the Coral Sea have been discussed. The daily temperature measurements at Willis Island reported by Hogan (1925) were taken over the reef structure at various stages of the tide and were reported to be affected by solar heating. Unfortunately, therefore, they are not likely to be of much value in describing short-term changes and are not summarised here. Surface salinity—mean distribution As no chart of mean surface salinity values with any detail in the WCS was available (the chart in Sverdrup er al., 1942 has only the southern winter and that has only one isohaline for the whole of the WCS) one was prepared from the CSIRO Atlas in the same manner as for the temperature charts. This indicated that the salinity values were mostly between 34.0 and 35.7 in the main body of the WCS with seasonal low values to less than 32 in the North-west Coral Sea. The values below 34 were usually recorded off Port Moresby in December to April. They may have been typical of the whole Gulf of Papua but there were insufficient observations in the Gulf to determine this. After April the minimum salinity in the WCS increased each month to about 34.5 in August then started to decrease again. The maximum value was usually 35.4 to 35.5 in the south-east corner of the WCS. Fig. 38 Meanannual ranges of temperature (AT) and salinity (AS) in upper 5 m, Western Coral Sea (based on Rochford, 1973). 150° 160°E 141° 150° 160°E n i n =i 2 ow 4 é) L 10° 3 2 TOWNSVILLE- AT(C°) AS (°loo) ANNUAL ANNUAL o) 130°s Cy As for temperature, the positions of individual isohalines extended over very large areas during the nine-year period. Fig. 39 illustrates this, showing the areas of occurrence of three isohalines, 34.6, 35.0 and 35.4 |.., representative of the bulk of the values observed. The mean positions of the isohalines are shown in Fig. 40, again to show the trend of seasonal variation rather than to specify geographic positions for mean monthly values. In the main body of the WCS and within the limitations of the scatter of the values, Fig. 39 shows that the salinity distribution for 35 |. and higher in the south-east half of the WCS is fairly steady, while most of the changes take place in the north-west half. One of the most obvious features is the occurrence of very low salinity water in the North-west Fig. 39 Surface salinity, Western Coral Sea, areas of occurrence of 34.6, 35.0 and 35.4 water for January, April, July and October, 1966-1974 (data from CSIRO, 1974). See note to Fig. 35. SALINITY ("/oa) AUSTRALIA 81 Coral Sea in the monsoon season, generally attributed to inflow from the west through Torres Strait coupled with runoff from the river system of Papua New Guinea. (The relative influence of the low salinity Arafura Sea water west of Torres Strait and of the fresh water from the rivers is not known yet, but further comments are offered in the section on Surface waters—Scully-Power.) In the north-east sector, the lowest salinitics occurred at the end of the monsoon season, ¢.g. April, Fig. 40. his has been confirmed in recent studies by Donguy & Henin (1975a) who showed that, for the four-year period 1969-73, a salinity minimum ‘ig. 40° Surface salinity, Western Coral Sea, mean isohaline positions for January, April, July and October, 1966 1974 (data from CSIRO, 1974). See note to Fig. 36. JAN. MEAN ISOHALINES (Soo) JUL. 82 progressed south-east across the Solomon Sea (around 10. S, 155 E in Fig. 40), occurring in March at 7 S to May at 12+ 8. Donguy & Henin demonstrated that this salinity minimum was the result of local rainfall associated with the development of the westerly (monsoon) winds in this region rather than of advection of low salinity water from the north of Papua New Guinea as had been thought previously. The figures for mean salinity distributions derived from the CSIRO Atlas are believed to be the only ones available for the WCS. A similar comment to that for temperature 1s appropriate, i.e. for a description of the long-term environment Fig. 39 is most realistic, while Fig. 40 shows the character of the temporal variations rather more clearly. In addition, the earlier comments referring to Fig. 37 on the uneven area distribution of observations should be noted. Less confidence should be placed in the significance of the isohalines of Fig. 40 than in the isotherms of Fig. 36, particularly for the January picture (Fig. 40) for which the north- eastward turn at the east end of the 35.0 and 35.5 isohalines is uncertain. Rochford (1973), in his study of annual and longer term variations, showed that the annual salinity range (Fig. 38B) varied from more than4d — in the Gulf of Papua in the North-west Coral Sea to less than 0.3. in the south-east. The long-term salinity values showed a tendency to decrease by about ().2 in six years off the Swain Reefs (area A in Fig. 38A) and a smaller decrease with time off Great Sandy Is. (area B, Fig. 38A). A conspicuous feature was a cycle with marked minima occurring at 2-year intervals at these two locations off the Queensland coast. The reasons for these long-term changes were not known, Mean seasonal variations of surface temperature and salinity by areas In an endeavour to better describe the local variations of surface propertics with season, the Western Coral Sea was divided into six approximately ‘5 areas’ as shown in Fig. 41. The mean surface temperature and salinity for each area (for the sea outside the Reet) was estimated by eye from the CSIRO Atlas for each month and the long-term mean for 1966-74 then calculated. The temperature time and salinity time curves are presented in Fig. 41A—D and the corresponding T,S, time curves are plotted in Fig. 41, F from the smoothed curves above. (There were insufficient data for area A, Gulf of Papua, to plot any curves.) It is evident that clear and similar variations with time occur in all areas, with differences mainly in the magnitudes. For temperature, the mean values and annual ranges are slightly higher in the eastern areas than in the western ones, while for salinity the eastern areas have slightly smaller ranges but essentially the same mean values as the western ones. Noticeable is the small range of salinity variation in area F. It must be stressed that these are mean value curves. The average standard deviations about the monthly means were about +0.5C and +0.3). ,; while cach individual monthly value was a mean of a range of values within the ‘5 area’. These monthly ranges were from +0.5 to +2C andfrom +0,1,. to +2 (occasionally in areca B). Itshould also be noted that the T,S,t diagrams of Fig, 41 are not identical with those of Fig. 24 in Part 1, for which values were selected close to the Reef, not for the sea areas as a whole. Sources of surface waters in the Coral Sea—Rochford No water types are formed in the Coral Sea; thus the waters therein all come from external sources. In one of the first studies of the water masses in this area, Rochford (1959) analysed the surface characteristics in the Coral and Tasman Seas using T,S,f plots (T,S diagrams with the frequency of occurrence (f) of T,S combinations included to make the display semi-quantitative,; Montgomery, 1955). Mean curves were drawn through the areas of most frequent occurrence of T,S combinations. (Rochford used chlorinity (Cl) in his paper, but for convenience this has been converted to salinity (S) by the current relationship S = 1.80655 Cl.) 83 (AREA“a” INSUFF. a ae DATA) ae Vet re 22° Fig. 41 Seasonal variations of surface water temperature and sa degree areas: (A,B) Temperature time, CSIRO, 1974). inity, Western Coral Sea, by approximate 5 (C,D) Salinity time, (E,F) T,S, time diagrams (data from 3 g 84 Rochford made the basic assumption that the effects of local evaporation and precipitation were negligible compared with those due to advection; i.e. salinity was regarded as conservative even in the surface layer. The validity of this assumption is dubious in view of Taylor’s (1973) figures for rainfall in the Coral Sea showing that precipitation is significant (1500 mm yr) and that the monthly amounts vary by a factor ot five over the year. There is no information available on evaporation in the Coral Sea. Measurements or estimates would be very desirable. Presumably Rochford also regarded temperature, i.e. heat content, as conservative but he made no mention of this, Most of the data available were from surface sampling from merchant ships, although these only travel along restricted lanes across the seas. For instance, in Rochford’s area 9, which corresponds closely to the defined W’CS, there was only one shipping lane, from Gt Sandy Is. to the south-east tip of Papua New Guinea, and only 24 samples per month on the average. T,S,f plots and mean T,S curves were prepared by Rochford for four quarters (November-January, etc.). The general character of the mean curves is indicated in Fig. 42A which is an envelope of all the T,S curves with the Coral and Tasman Seas not differentiated. Rochford divided this envelope into three main zones; (1) above 28 C where there was little annual temperature change but considerable salinity change (presumably the north Coral Sea area), (2) from 28 to 19 C where salinity increased as temperature decreased, with greater annual changes of temperature than above 28 C, and (3) below 19 C where salinity decreased with decrease of temperature and where there were considerable annual ranges of variation (mostly Tasman Sea samples). Rochford compared these surface T,S curves with vertical T,S curves for stations in the area between Australia and 155 W for winter (envelope in dashed lines in Fig. 42B) and summer (envelope in dashed lines in Fig. 42C). The mean surface T,S curves for the Coral Tasman Seas are shown in full lines in these figures and the latitude scales at the left give some indication of which parts of the surface T,S curves are pertinent to the present review of the WCS. From these T,S curves Rochford determined three major and three minor water types, mixtures of which could produce the vertical or horizontal T,S curves of Fig. 42A, B, C, These types were identified as in Table 5. The major types and the Arafura Sea water are indicated on Fig. 42B, C by number. The two remaining water.types (5 and 6) have not figured significantly as such in subsequent discussions of the Coral Sea waters. Rochford’s description of the surface T,S characteristics does indicate the probability of the four water types (1,2,3,4) referred to above being the main contributors to the Coral Sea surface waters. (It should be noted that the word water ‘type’ here does not imply a large volume of completely homogeneous water as represented by a point on a TS diagram (Sverdrup’s definition) but rather the mean temperature and salinity of a water body, it being understood that the range of values about the mean is small compared to the Table 5. 1T,S characteristics of Coral Tasman Sea water types (from Roch- ford, 1959) Temperature Salinity Water Type as C) ; i oa Major: 1, South Equatorial (from N of 10 8S) 28,2-28.8 34.7 2. West Central South Pacific (from 15 -20 S, 150 W) 26.0 36.5 3. Sub-Antarctic Surface (from S of 45 S) 9.0-11.8 34.7 Minor: 4, Arafura Sea 28 -29 34.1 or less 5, Tropical high salinity 29 —30 35.6 6. Temperate high salinity 20 -21 36.5 or more 85 30° + T 28°4 Sy [Sa S) : \ So 1 Oc DD Son 1. 100% ul20- | Z ‘ SOUTH a 7} el // 50% fe EQUATORIAL. F s] iG on | = NEW HEBRIDES At) 71 N (en F 104 A 4 4 we \ te r T T * 34 35 36 NEW ° SALINITY (°/oo) 205 Tee tote en = “w. Sentra NLy SUMMER ? s. PASI 3s oD SURFACE FLOW LE gk , 307— — —+ -10 = Je “4 "% A ] CORAL SEA } 20 “ | MEAN ] {0 a0 / 140° 150° 160° E i a ° WINTER 7 730 20% ¥ r 7 1 SOURCE L40°S AREAS Z 7 TS 10° “ ENVELOPE B “f 3’ + e tT 1 34 35 36 “loo Pi 1 100% 307—= === — “, z 5 meee I % WS aH5 ae v N NX 30° ) / L. .| SUMMER if 20 7 7 yf L40°S Po WINTER SURFACE FLOW We td 37 AUSTRALIA 1 7 t A ra —— ‘a = hae 140° 150° 160°E 170) Vig. 42. (A) Envelopes of surface T,S values, Coral Sea, (B) Full lines—mean surface T,S curves, winter, Coral Sea; dashed lines—envelope of vertical TS curves for source areas; numbers—four major water types contributing; (1) South Equatorial, (2) West Central South Pacific, (3) Subantarctic Surface, (4) Arafura Sea, (C) As for (B) but summer, (D) Summer, and (E) Winter flow patterns, for water types of (B). (All adapted from Rochford, 1959.) 86 differences between those of the water types contributing to the region.) Presumably for the WCS the two higher temperature types (1 and 2) would be the most important. (‘This is evident in Scully-Power’s data (1973a) shown in Fig. 44.) On this basis, Rochford then used the method of the mixing triangle (ref. Appendix) on the T,S diagram to determine the percentage of cach of the three major external water types contributing to the make-up of each sample in the Coral and Tasman Seas and to deduce the surface flow patterns for the four water types as shown in Fig. 42D, E. In summer (Fig. 42D) the flow was primarily meridional except for some westerly flow of West Central South Pacific water at about 25 S. For late winter (Fig. 42E) the flow was chiefly zonal with a strong westward flow of South Equatorial water across 160 E into the Coral Sea and inflow from the south-east of West Central South Pacific water (and some outflow to the cast at about 30 S of a mixture of these two waters). The mixture of South Equatorial and West Central South Pacific waters formed at about 27 S was called ‘Coral Sea’ water by Rochford. He did not specify the water characteristics in his paper but from his diagrams they would appear to be about 27 Cand 35.6. The Arafura Sea water was recognised in the North-west Coral Sea at this time and might be expected also to occur later (February to April). Rochford also examined the total phosphorus content of the surface waters and considered that it could also be used as an identifying water property. In particular, he noted that the phosphorus content of the West Central South Pacific water was low (0.1 to 0.16 wg at 1) compared to that of the South Equatorial and Sub-Antarctic waters (0.65 to 1.0 ug at) 1). Surface waters—Scully-Power The most systematic set of observations for the WCS was made by Scully-Power (1973a, b) for the winter period, between 4 May and 20 July 1968, when three cruises comprising 94 stations were completed. The station positions are shown in Fig. 43A. Because of the coherence of this data set and its coverage of the W’CS, it will be discussed in some detail. The surface temperature and salinity distributions are presented in Fig. 43A, B to show that a reasonably simple pattern may be present when the data are comprehensive and quasi-simultancous (although the circulation did not appear to be simple: see later). ‘These distributions resemble the very much smoothed mean distributions, Fig. 36 and 40, constructed from the CSIRO Atlas data. The isotherm and isohaline patterns were similar except for the feature due to a low salinity at one station in the Gulf of Papua, The surface density pattern was very similar to the surface temperature pattern, 7, values increasing from 22 in the north-west to 25 at the south. The T,S and S,O, diagrams for the surface waters showed small scatter (Fig. 44, full lines), the surface T,S values being almost all within the envelope of vertical T,S curves for the WCS (dashed line). Scully-Power discussed the low-salinity waters in the North-west Coral Sea (area ab in Fig. 44) and suggested that continuing river runoff into the north-west Gulf of Papua was probably the major factor but that there might also be pockets (‘clouds’ would be a better word) of Arafura Sea water still present from the NW’ monsoon inflow through Torres Strait. No figures are at hand for the runoff of the extensive river system of Papua New Guinea which empties into the north-west Gulf of Papua but it is noted (Brookfield & Hart, 1966) that there are some curious features of the rainfall around the Gulf. At Daru (Fig. 28) the maximum rainfall period is from January to April (57",, of the annual total) whereas from about 143.5 to 146 E, as typified by Kikori, the maximum is later (May to September), when 56", of the annual total falls. To the east of 146 E the pattern reverts toa January to March maximum. The annual total at Kikori is 5760 mm compared to 2075 mm at Daru, and 995 mm at Port Moresby. The high rainfall area extends inland 87 °. TOWN SVILLE 204 Y SURFACE TEMPERATURE (°C) STATION POSITION 30°5 L 1 =i, [A] 150° 160°E Been eae ; Cy 4 SOLOMON 1S. 10 $4 & 2 Q 4 — se 2 Se CALEDONIA (°/oo) p a AUSTRALIA : f OF a 35-7 Lie 1 = 74 | B | Fig. 43 (A) Surface temperature and (B) surface salinity, Western Coral Sea, May-July, 1968 (Scully- Power, 1973a). 88 TEMPERATURE ‘C DISS. OXYGEN (ml./1.) TEMPERATURE °C Fig. 44 ° c T T T T | Lee to’s Se ee ENVELOPE OF SURFACE VALUES bis <5 - att o,2 “oN t= \ 254 Rell 23° ENVELOPE OF “eX VERTICAL T,5S \ L20'S CURVES 25° 247 | 5 Poe SURFACE LL T,S | 204 L30°s Es clA T T r a 4 S 34 SALINITY 35 36 %loo 1 L 41 1 5 30 25° 20° 15° a ENVELOPE OF SURFACE VALUES- SURFACE S,0, B 4 T T =I =‘) 4 34 SALINITY 35 36 loo 30° 1 -1_____~- m _| 3 oT oy “Z oN 4H 4 vA : N\ 5 rm pees ( \ 2 | wl ae SURFACE T,S 20°S a i ‘ ___I[c| (A) T,S and (B) S,O, diagrams for surface waters, Western Coral Sea, May-July, 1968 (Scully- Power, 1973a), (C) Comparison of surface T,S characteristics for several cruises in the Western Coral Sea: (1) Shoyo Maru, Oct.-Nov. 1973, (2) ‘Gorgone 1’, Nov.—Dec. 1972, (3) ‘Tule’, Jan. 1965, (4) ‘Gorgone 2’, May-June 1975, (5) R.A.N.R.L., May-June 1968. 89 along a strip lying at about 300 true from Kikori. These differences in rainfall pattern along the north side of the Gulf of Papua are remarkable in the relatively short distances and may be responsible for some of the peculiarities of salinity patterns which have been noted in this arca (e.g. CSTRO Atlas, April 1969, May 1972, etc.). Water of type 6 (Fig. 44A, B) was identified as South Equatorial (surface) water (Rochford, 1959), while type ¢ was identified as basically the southern component of Subtropical Lower water. The distribution of properties at the surface was consistent with a westward flow into the Coral Sea north of about 19 S, with most of the flow turning north on approaching the Barrier Reef, the division being at about 19 S also. This flow pattern agreed with the geostrophic circulation discussed later. (Data from a May 1972 R.A.N.R.L. cruise in the North-west Coral Sea are currently being studied—personal communication, P, D, Scully-Power.) Other surface data in seasonal sequence Envelopes of surface T,S characteristics for five cruises which covered substantial areas of the WCS (i.e. more than a single line of stations) have been assembled in Fig. 44C for comparison. The cruises used are listed below and the areas covered are shown in Fig. 45A, B: Spring : Curve 1 — Shoyo Maru, Oct.-Nov. 1973 (Far Seas Fish. Res. Lab., 1973), Summer : Curve 2 — ‘Gorgone 1’, Nov.—Dec. 1972 (Donguy er al., 1972b, NODC 350077; only stations west of 159 E and south of 11 S used), Curve 3 — ‘Iule’, Jan. 1965 (O.R.S.T.O.M., 1965, unpublished data, NODC 350081), Winter : Curve 4 — ‘Gorgone 2’, May—June 1975 (O.R.S.T.O.M., 1975, un- published data; only stations west of 159 E and south of 11 S used), Curve 5 — R.A.N.R.L., May-July 1968 (Scully-Power, 1973a; only stations north of 20 S used). For temperature there is some indication of a seasonal change, the January cruise (curve 3) having the highest temperatures. However, it is by no means clear-cut as the ‘Gorgone 2’ and R,A,N.R.L. cruises (curve 4 and 5) were at about the same time of year (winter) but the former show higher temperatures by about 1.5 C . For salinity, the shaded, low-salinity part of the R.A.N.R.L. data envelope (curve 5) was for the waters of the typically low salinity Gulf of Papua, further north-west than any of the other cruises. The difference in salinity between the two ‘Gorgone’ cruises may be attributed to the different seasons, ‘Gorgone 1’ (curve 2) being at the end of winter (high salinity period) while ‘Gorgone 2’ (curve 4) was after the wet season when low salinities are now known to occur systematically in the Solomon and north Coral Seas due to local rainfall (Donguy & Henin, 1975a). In addition there may also be some effect due to possible year-to-year variations. The larger range of salinity for the R.A.N.R.L. cruise was probably because this penetrated further into the low salinity Northwest Coral Sea, but the lower temperature for the R.A.N.R.L. cruise is probably indicative of year-to-year changes because the two envelopes are entirely separate in the temperature direction despite the fact that the two cruise areas overlap between 12 and 20 S. This emphasises (as has been shown already) that year-to-year variations in the surface layer characteristics can be significant. Speed of movement of water masses—Rochford Fig. 41 shows that the lower salinity (South Equatorial) water moves south in the summer. Rochford (1973) gave two examples of estimates of the speed of southward 90 ° 204 (@oscouee \ | 1972 \ NOV-DEC. \ | b | | news %\ \ HEBRIDES \ : ©, : “t, 0\ TOWNSVILLE ===> on Oo _| . g 5 Q 7 CRUISE CALEDONIA oF AREAS ¢ / oo SS RA.N.RLL. wa R.A.N.R.L. STATIONS CS4 i USED FOR FIG. 34. 1968 Pe ‘ 4 a 5: eae “ CORAL SEA 1968 4 BOUNDARY e JULY a 4 AUSTRALIA y a f J T T T T . 150° 160° pa | = | LB IN, R.S.7.0.M | “sorGoNe 2” | MAY-JUNE 1975 4 0.R.5.1.0.M “TULE” JAN.1965 | % 7 | BE Sty ie Bie ° t “SHOYO MARU* a S OCT-NOV.1973 oa 7 ° 20-1 1 aN ! a 1 1 a) Fig. 45 Cruise areas, Coral Sea: (A) R.A.N.R.L. Cruises CS4/1,/2,:3, May-July 1968 and locations of stations for Figs. 32 and 33 (Scully-Power, 1973a), O.R.S.T.O.M. ‘Gorgone 1’ cruise, Nov.—Dece. 1972 (Rougerie & Donguy, 1975), (B) O.R.S.T.O.M. ‘Tule’ cruise, Jan. 1965; Shoyo Maru cruise, Oct.-Nov. 1973; O.R.S.T.O.M. ‘Gorgone 2’ cruise, May-June 1975. 91 penetration of this water from the southward advance of lower salinities in successive months on the CSIRO Atlas plots. Fig. 46 shows these estimates for January to April 1966 and 1970. They yielded speeds of the order of 22-44 km/day or 6 to 12 degrees of latitude mo along the main flow directions and about 4 km/day perpendicular to the main flows. Mixed layer depth Because there is often a wind-mixed layer in the Coral Sea of depth comparable to those of the passes through the Great Barrier Reef (less than 100 m), the surface water characteristics may often be sufficient to characterise the WCS waters when considering exchange between the Sea and the Lagoon. Therefore the mixed layer depth was Fig. 46 Southward progress of low salinity (South Equatorial) water in the Western Coral Sea in January-April, 1966 and 1970 (Rochford, 1973). LAX : FEB \S_---- MAR \ \FEB / MAR \ i Sof SURFACE Tit FLOW oo 170° TOWNSVILLE AUSTRALIA 92 estimated from the R.A.N.R.L. 1968 cruise records (Scully-Power & France, 1969a, b, d) and plotted on a chart; an attempt was made to contour this feature. No simple pattern was found, if the rules of contouring were followed strictly, because of the sharp changes between some pairs of stations. A broad look at the field of values resulted in Fig. 47, which indicates that the mixed layer was deepest near the Reef and shallowest along a line south-east from the Gulf of Papua. Wind speeds during the three cruises varied from 2 to 15 m/s and averaged 6 m/s. No significant correlation between layer depth and wind speed was found. An earlier R.A.N.R.L. cruise in November—December 1966 (Scully-Power, 1969a) between Cairns and Papua New Guinea indicated mixed layer depths of 20 to 70 m, which were smaller than the 1968 values on the direct line between Cairns and Port Moresby but greater to the east of this. In April-May 1967 (Scully-Power, 1969b) in the same area, values north of 12 S were between 0 and 25 m while at 13 and 14 S they were 20-35 m. Winds were from 2.5 to 9 m/s during the 1966 cruise, and 4 to 6 m's at those stations where it was recorded in 1967, For the O.R.S.T.O.M. ‘Gorgone 1’ cruise in the Central Coral Sea in November—December 1972 (Donguy er al., 1972b), the shallowest sample below the surface was taken at 50 m nominal depth and only at one or two stations was there any indication of a mixed layer reaching to this depth. This is consistent with Scully-Power’s data in the region of overlap (see Figs. 47 and 4). Fig. 47 Surface mixed-layer depth, Western Coral Sea, May-July 1968 (data derived from Scully-Power & France, 1969a, b, c). T \ TT T SS T ~] 150 .) 160°E v YS KS, PORT__ " %, \ MORESBY o a \ —Y 7 , Nos NO ‘75-100 DATA : TOWNSVILLE” nr r20s iN MIXED LAYER DEPTH (m.) AUSTRALIA ap el L - : —_ 93 The Shovo Mari (Far Seas Fish. Res. Lab., 1973) made 75 bathythermograph casts in an area 13 —17 S, 144 -151 E (north-east of Cairns) in October-November 1973, the temperature values being reported at 25 m intervals from the surface. From these it appeared that the mixed layer depth varied from 0 to 50 m, the values along the Reef being 0-25 m. The wind speed was mostly from 4 to 12_m s and there was some indication of greater mixed layer depths being associated with greater wind speeds. One concludes that, as for other characteristics, the mixed layer depth is very variable in the WCS. SUBSURFACE WATER MASSES AND FLOW PATTERNS Introduction The main contributors to the study of the subsurface water masses in the Western and Central Coral Sea have been Rochford (using isentropic analysis), Wyrtki and Scully- Power (core analysis), and Rougerie & Donguy (isentropic analysis), Papers by these authors wiil be discussed substantially in chronological order. Isentropic analysis—Rochford The 27.2 o, surface. Rochford (1960a) discussed aspects of intermediate depth waters on the 27.2 o; surface, which is at 700 to 900 m depth in the WCS and near the core of the Antarctic Intermediate water. When considering this paper it should be noted that the region north of 25 S and west of 162 E (i.e. an area greater than the entire WCS) was represented by only one line of 10 stations (Umuraka Maru, 1959, unpublished, and one Planet station) at about 156 E. For these stations, on this surface, the temperature was close to 6 C, salinity was about 34.45 | , dissolved oxygen increased from about 3.7 ml] in the north of the line of stations (mid-WCS) to 4.05 ml] in the south, and phosphate-P increased from 1.8 to 2.05 yg at/1 from north to south. For the analysis of waters on the 27.2 oc, surface, Rochford placed considerable emphasis on S,PO, relations and identified three water types as: Salimity Phosphate-P ( / (yg at 1) (7) Pacific Equatorial Intermediate 34.60 3.10 (8) South-west Pacific Intermediate 34.70 0.50 (9) Antarctic Intermediate 34.02 1.88 He contoured the percentages of each on the 27.2 o, surface using the mixing triangle method. The result showed water type 7 entering the WCS from the north through the Solomon Sea, type 8 from the north-east between the Sclomon Islands and New Caledonia, and type 9 chiefly from the east by the same route (but entering the Tasman Sea from the south). The major outflow from the Coral Sea on this surface was to the south along the Australian coast (East Australian Current), The 25.0 o, surface. Rochford (1969) averaged salinity and oxygen data by 5 squares for the south-west Pacific (0'-50°S, 140°E-160°W) to study the origin and circulation of water types on the 25.0 a, surface which is close to the core depth of the Subtropical Lower water. He identified four types (essentially subdivisions of this water) as: 94 Type Namie Salinity Oxveeu Origin ( | (mill 36.00 3.50 Central South Pacific 5 A’ : Tropical low salinity 35.25 3.10 North Equatorial Pacific B_: Subtropical high salinity 35.75 5.15 Central Tasman Sea B’ : Subtropical low salinity 35.46 5.15 West of North N.Z. North of 15 S, types A and A’ drifted west near the equator but east around 20 S; south of 35 S, types B and B’ drifted cast and north. Types A and A’ were essentially the northern component of the Subtropical Lower water, while types B and B’ were southern component waters. A comparison of winter and summer values was possible in some areas and indicated that: (a) tropical waters (A,A’) drifted south along the Australian coast in the summer to winter period; (b) subtropical waters (B,B’) drifted north into the Central Coral Sea in the winter to summer period. Core layer analysis—Wyrtki Wyrtki (1962a) carried out a core layer analysis using fairly homogeneous data from CSIRO Gascoyne cruises in 1960 and 1961 in the area between the Australian coast, New Zealand, Fiji and the Solomon Is. supplemented by the Unuraka Maru data for 1959 tor the east edge of the WCS as used by Rochford (1960a, b) and also some Orsom I/] data south-east of New Caledonia (Rotschi, 1960c), Current nomenclature for the subsurface waters in the Coral Sea, as used in the introductory paragraph and subsequently, is based on this paper by Wyrtki. Fig. 48 (A) T,S and (B) T,Oz diagrams for Western South Pacific waters (adapted from Wyrtki, 1962a). (Tt T T 2001] oe T ; 1 25°C 4 SUB-TROP. /. SUB-TROP = LOWER \/ LOWER eee” RNQRTHERN \/( SOUTHERN) 1300 QQUL ‘ ve Si . A | b L 20° 4 VY, Ast= * G 200 UPPER ff Uf OXYGEN | 02 Ly “ SATURATION = L 15° 4 MINIMUR AT1ATM. 4 & re = e INTL TTT Ey iE = - SUB-ANT-- ra ~=""|INTER- 7 ~—OXYGEN____ I | Fe | == MAXIMUM———— WESTERN = oRVGEN—_ SOUTH PACIFIC MAXIMUM“ 27 60 bt So L ANTARCTIC INTERMEDIATE | 5° | | 7 WATER < T,S T, 0, 27°5 [A| . [B| T T T a = T T T SALINITY (°/a00) 3 OXYGEN (ml/I) 95 Wyrtki used 'T,S and TO, diagrams for his analysis. Fig. 48A is a simplified version of his ‘I'S diagram for the salinity core-layers, replacing the original station points with general shading to indicate concentrations of points, and omitting the deep and bottom water points. Fig. 48B simplifies his T,O, diagram in a similar manner. From these and other characteristic diagrams Wyrtki distinguished three main water masses in the upper 1100 m. Using the T,S diagram he identified; (10) Surface water: T > 24 C, S from 34.0 to 35.6) , o from 24.0 to 25.5 (discussed by Rochford, 1957-1959, and not discussed in Wyrtki’s paper or included in his T,S diagrams), (11) ‘Subtropical Lower’ water: Salinity maximum, T from 18 to 25 C, S from 35.5 to 36.0, core depth 50 to 150 m in the WCS, (12) ‘Antarctic Intermediate’ water : Salinity minimum, T from 4.2 to 6.0 C,S from 34.37 to 34,53, core depth 700 to 1000 m in the WCS. The T,O, diagram revealed further features of the water masses: (a) The Subtropical Lower water clearly subdivided into a lower oxygen component (3 to 4ml/J) at stations in the northern part of the region where the salinity maximum was below the surface (‘northern component’ of the Subtropical Lower water), and a higher oxygen component (above 4.4 ml.]) from subsurface and surface levels and found at stations further south (‘southern component’ of the Subtropical Lower water), (b) between 24 and 12 C, below the Subtropical Lower water, a layer of minimum oxygen in the vertical was found (the ‘Upper Oxygen Minimum’), (c) between 10 and5.5 C wasa layer with maximum oxygen in the vertical (Oxygen Maximum’), Wyrtki argued that the oxygen minimum was formed mm situ by biochemical reduction, thus qualifying it as a separate water mass, but that the oxygen maximum resulted simply from vertical mixing of the lower part of the Antarctic Intermediate water with low oxygen water below it, and therefore did mot qualify as a separate water mass. Examining the distribution of properties on the core layer of the upper salinity maximum (Subtropical Lower water), Wyrtki showed that two salinity maximum (>35.9 , ) water masses were apparent, one entering the Coral Sea from the east in the vicinity of the New Hebrides and the other via the Tasman Sea between New Caledonia and New Zealand (Fig. 49A). The oxygen content in this core showed low values (~ 3.2 ml) 1) associated with the northern salinity maximum and high values (5 ml 1) with the southern salinity maximum (Fig. 49B). The salinity and oxygen distributions suggested that the boundary between these two maxima lay approximately along 20 S to the north of New Caledonia and thence west-north-west toward the Gulf of Papua. The core layer of the northern component sloped upward from about 150-200 m in the northern Coral Sea (Fig. 49A). The southern component was at the surface at about 30 § in the Tasman Sea and sloped down to about 100 m at 20 S, Wyrtki then identified the source of the Coral-Tasman Seas Subtropical Lower water as the Subtropical Surface Water of the South Pacific formed between 15 and 25 8, 100 to 150 W whence the southern component flowed west between 23 and 33 at the surface, retaining its high oxygen content, while the northern component sank and spread to the west north of 23 S, oxygen being consumed en route. The above remarks refer to the Subtropical Lower core layer, Wyrtki suggested using 35.0 /- as limits to enclose, above and below, the Subtropical Lower water mass. With this definition, the southern component would extend from the surface to 400 m depth and the northern from 50 to 300 m. Such relatively shallow masses would be expected to follow the surface circulation and Wyrtki considered that the general movements of the Subtropical Lower water inferred above 96 SUB-TROP. LOWER SAL, MAX: ----DEPTH OF CORE(m) FLOW ° 442° ye 170 er f _L 1 1 Bi ’° E| NORTHERN LOW 02 IN COMPT. X oS \ QS 4 SUB-TROP. LOWER OQ DIST. 48 \ SOUTHERN ON SAL. MAX. HIGH 02 NX CORE ml/l comrt sa] — 30°S . 4 ° ° 150 160°E 170° 142° — SALINITY --- DEPTH OF CORE (m.) distributions and flows: (A) and (D) Subtropical Lower salinity maximum core distribution in salinity maximum core, (C) and (F) Antarctic Intermediate on positions (Wyrtki, 1962a), Fig. 49 Coral Sea property layer, (B) and (E) oxygen salinity minimum. Dots represent stati 97 (Fig. 4919) were in accord with his description of the surface circulation (Wyrtki, 1960, and in the surface circulation section following). Wyrtki showed that the flow of the oxygen minimum water mass between 150 and 500 m( <3.0 ml lin the north and 4.4 ml lin the south), shown in Fig. 49E, was similar to that of the salinity maximum above it. he next core, the oxygen maximum above the salinity minimum, was not regarded as a separate water mass, as explained above, and no effort was expended in tracing its movement, Wyrtki next examined the salinity minimum layer, the Antarctic Intermediate water. An interesting feature of this water mass was that the greater part entered the Coral Sea not from the south but from the east, between the New Hebrides and New Zealand (Fig. 49F). This was evident both from the salinity distribution (Fig. 49C) and from the oxygen distribution at the level of the salinity minimum (Fig. 49F), with values of 4.4 ml/l on entry north and south of New Caledonia decreasing to 3.8 ml] in the Solomon Sea. The depth of the core decreased from 1000 m near 30 S to 700 m in the northern Coral Sea (Fig. 49C). (Rochford, 1960a, had shown a similar distribution in his discussion of properties on the 27.2 o, surface which is close to the salinity minimum core.) In summary, Wyrtki listed the properties of the subsurface water masses, The values relevant to the WCS are presented in Table 6. Table 6. Properties of subsurface water masses in the Coral Sea Water Mass Feature Ti ©} S/ ] Ostml li Depth (me Subtropical Lower S max, Northern Compt. (lower O,) 21-24 35.7 —35.95 3,2-4.2 125-150 Southern Compt. (higher O,) 18-25 35.75-35.85 4,5-5,2 50-125 Upper Oxygen Minimum O, min. 12-21 34.9 -35.9 2,7-4.6 150-300 Antarctic Intermediate O, max. 5.4-9.0 34.4 —34.8 4.0-4.7 500-900 S min. 4.2-6.0 34,42-34.47 3.9-4.4 700-1000 Continuity of water masses along the western boundary of the Tasman and Coral Seas—Rochford Bearing in mind the limited number of stations on which Wyrtki’s (1962a) study above was based, it is interesting to review Rochford’s (1968a) description of the water masses along the Australian coast, using a few more stations. The station positions (pre- dominantly for the summer season) are given in Fig. 50A which shows the isohalines in the core layer of the Subtropical Lower water in the WCS and the oxygen distribution by shading. The arrows show Rochford’s interpretation of the flow, on the assumption that the oxygen gradients were too high to be the result of local consumption but must be maintained by flow. The ‘oxygen poor’ water may be identified as Wyrtki’s northern component of the Subtropical Lower water and the ‘oxygen rich’ water as the southern component. The depth of the salinity maximum layer increased from about 140 m in the North- west Coral Sea to 200 m at 20 —25 S before rising to the surface at about 30 S. Wyrtki (1962a) showed the layer rising steadily from the North-west Coral Sea to the surface at about 30 S and did not show any deep patch comparable to Rochford’s 200 m one. Fig. 50B shows the salinity values in the core of the Antarctic Intermediate salinity minimum. This layer had a depth of over 800 m round the margin of the North-west 98 200m.-7 CONTOUR aor] SUB-TROP. | LOWER WATER SALINITY "oo OXYGEN: 3-5ml/l 4-0 45 * STATION \ POSITION T 1S0°E FLOW: ——> (FROM SALINITY & ~~... TOPOGRAPHY S > OF CORE LAYER) 34-46 %/oo 20°S | ANTARCTIC INTERMEDIATE WATER SALINITY °/oo DEPTH LESS THAN 1000m. 1 it Fig. 50 Water masses and deduced flow patterns in the Western Coral Sea: (A) Subtropical Lower water, salinity and oxygen distributions, (B) Antarctic Intermediate water, salinity distribution (Rochford, 1968a). Coral Sea, doming to less than 700 m in the centre of this area and then increasing to over 900 m in the south-east corner of the WCS. In suggesting flow directions (Fig. 50B), Rochford was perhaps influenced more by the topography of the core layer than by the salinity distribution. A comparison of Wyrtki’s (1962a) and Rachford’s (1968a) results is appropriate at this stage. Rochford’s salinity values for the Subtropical Lower water were about 0,05 to 0.1 lower than Wyrtki’s, his oxygen values about 0.3 ml/1 lower, and his core layer depth greater by about 50 m in the north-west and 100 m in the south-east. For the Antarctic Intermediate Water the salinities and depths were similar in the two analyses. The general character of the flow for the Subtropical Lower water was essentially the same in Rochford’s as in Wyrtki’s interpretations, but for the Antarctic Intermediate water Rochford showed a clockwise circulation in the North-west Coral Sea while Wyrtki did not commit himself to any direction of flow in this region. Rochford suggested that the southward flow across 20 S branched off from the North-west Coral Sea circulation but it could also be a southward branch of the original westward flow into the Coral Sea. Rochford pointed out that while the Subtropical Lower water had a wide range of oxygen values in the WCS (from 3.5-6.5 ml/l) the Antarctic Intermediate Water had a much more restricted range (3.9 to 4.3 ml/l), more closely related to the southern component of the Subtropical Lower water than to the northern component (although only marginally so). He concluded that the major flow of Antarctic Intermediate water into the Coral Sea occurred south of Fiji and the New Hebrides (i.e. south of about 20 S), which agreed with Wyrtki’s (1962a) deduction, and that little flow occurred at 800-1000 m from the Solomon Sea south into the Coral Sea in contrast to the inflow at shallower depths. Core analysis—Scully-Power The May-July 1968 data set taken and analysed by Scully-Power (1973a, b) was introduced earlier and the surface water characteristics described there. The subsurface water mass descriptions are now presented. Subtropical Lower Water (upper salinity maximum). An example of the variation of properties with depth (Fig. 32) has already been described, using one of Scully-Power’s stations in the WCS. This pattern was typical for the area, the differences being chiefly in the depth of the mixed layer and the southward increase of salinity in this layer, so that at some of the southern stations the upper salinity maximum was at the surface. Accompanying the increase in mixed-layer salinity there was a decrease in temperature so that the start of the thermocline was less sharp and the density discontinuity layer less intense, Fig. 51A shows the envelope of all vertical T,S curves for Scully—Power’s WCS stations, showing the well-defined character of the water column over the whole WCS area. The usual main features were the upper salinity maximum and the salinity minimum below this. The depths of these features vary over significant ranges (see Fig. 33) despite the tightness of the envelope. (This is one of the reasons for using the T,S and other diagrams as analytical tools.) The salinity maximum occurred at o, = 24.75 in the north, increasing to o, = 25.20 in the south, while the salinity minimum occurred close to o, = 27.20 everywhere in the WCS. The upper salinity maximum (Wyrtki’s 1962a Subtropical Lower water) was below the surface north of 27 S and at or close to the surface with the upper 200 m being nearly isohaline south of 27 S. The core salinity value was 35.59 to 35.78 ; except for two lower values close to the Barrier Reef. Fig. 52 shows the distribution of (A) salinity and (B) dissolved oxyger in the salinity maximum core layer. These distributions were consistent with an inflow of a low oxygen, relatively high salinity, water from the northeast and a higher oxygen water inflow from 100 SALINITY (°/o0) , 35 DISS. paien (ml/1) 07 = SUB-TROP. LOWER am WATER SUB-TROP ae LOWER CORE UPPER 500-4 | -SALINITY MAXIMUM 20° (60-240m) 200 L mas lu 2 4 aa 21000 4 & 154 4 pe a a Ww a a lw = a WwW = 104 Al 1500-4 a Z SALINITY ° / MINIMUM 55 27 7 (650-1150m.) 7] / 2000 4 /\ Ly 275 i J IL L | (A) T,S (B) 0,,Z SALINITY (°/o0) 355 35:8 1 L 1 L 1 | 54 J Ss SE COMPT. COMPT. 4 4 E = | Saiteet CENTRAL ’ CORAL SEA | Ot A comer. COMPT. 4 x C 5° wn WwW o| O ar 79) Z10 e — w 5 NEW GUINEA --> See ~ oN . COASTAL<20 & OID ~~HLs oP | 74 o| @ COMPTON eos = Doe ie > 20 Pett SN 8 ONE = iJ Paso eg eee ee COMET t 1. ~~"NW 30° COMPT (C) S,0, Fig. 51 Envelopes of (A) T,S and (B) O,,z curves, Western Coral Sea, May-July 1968, (C) S,O, diagram for Subtropical Lower water components, Western Coral Sea, May-July 1968 (Scully-Power, 1973a). 101 the south (or south-east—the distribution of stations here was such as to leave this uncertain, although the salinity distribution suggested a northerly flow close to the Australian coast). Over most of the area the salinity maximum core layer depth was between 110 and 200 m depth and the thickness of the layer (for salinities over 35.50) average 180 m. In the core, the temperature ranged from 19.3 to 23.3 C, and dissolved oxygen from 3.2 to 5.0 ml 1. In discussing the water masses Scully-Power then reviewed in some detail the characteristics of the Subtropical Lower water in the WCS; as this, together with the surface water, is likely to be significant in considering exchanges with the Barrier Reef lagoon, his conclusions will be summarised. Fig. 52 (A) Salinity and (B) oxygen values in the core layer of the Subtropical Lower water, Western Coral »-July 1968, (C) Salinity and (D) oxygen values in the core of the Antarctic Intermediate estern Coral Sea, May July 1968 (Scully-Power, 197 3a). Sea, ! water 35° | \ |. TOWNSVILLE pa y 4:25 4 45 ~ S (oo) O, (ml / 1) Qars - SUB-TROP. + SUB-TROP. d a LOWER LOWER 8 VALUES OVER 35-0°/oo S (loo) 0, (ml / 1) + ANT.INTER. VALUES OVER 34:0 “Joo = er F [D] 102 ANT.INTER. T 4 The salinity distribution in the core layer of the Subtropical Lower water (Fig. 532A) showed three regions of high salinity values (to above 35.73) referred to fram thvir locations as the north-east, southern and north-western components (Fig. 31CG), The other main feature was the presence of a uniform salinity mass (35.61 to 35.65 in the centre of the area, named by Scully-Power the ‘Central Coral Sea component’, The north-east component evidently represented an inflow into the Coral Sva of northern component Subtropical Lower water (Wyrtki, 1962a). The southern component had its highest salinity of 35.78 close tothe shore at27 S. Scully-Power suggested that this high salinity water might originate in the Barrier Reet lagoon and flow south, quoting CSIRO (1968a) for evidence of water over 36.00 in the Lagoon and Woodhead (1970) for southward flow. However, Scully-Power did not explain why there was lower salinity water between the exit from the Reef lagoon (presumably by Capricorn Channel at 23 S) and the location of the southern high salinity water at 27 -28 S. The T,S diagrams for Reef lagoon waters (Fig. 23, Part 1) show that in the southern zone, surface water of T=25.6 C,S=35.75) ando,=25.25 occurs in July. These properties are almost identical with those in the W'CS 100-125 m depth salinity maximum at Scully-Power’s July stations 9 and 11 at 27 S, 154.5 and 156 E, ice. T =20.67 C, S=35.78'|. , ¢,=25.20. The agreement is almost too good because such water offshore would require some time to get there and would probably have its properties modified by mixing en roure. The lagoon waters earlier in the year are usually warmer and less dense, e.g. May, T = 24.5 C, q=23.7 and in June, T = 22.0 C, ¢,= 24.7. The Fig. 23 values are means and undoubtedly both more and less saline waters occur from time to time in the southern lagoon. However, a more telling point against lagoon origin of waters studied on the 1968 July cruise was that, according to the CSIRO Atlas, the lagoon salinities in April to July 1968 were all below 35.5 and densities were less than 24.5, significantly lower than the WCS Subtropical Lower valucs outside the Reef. It 1s also doubtful if the lagoon could produce sufficient quantities of high salinity water to supply the southern WCS. Discussing the north-west high salinity component, Scully-Power (1973b) advanced arguments against (a) inflow through Torres Strait at any time, (b) formation 77 sirm, and (c) flow from the Barrier Reef lagoon, He concluded that it must have come from the north-east at an earlier time, possibly in October-November at the end of the SE trades season. Data from earlier cruises in the area (Lockerman & Scully-Power, 1969) supported this suggestion. The Barrier Reef lagoon was dismissed as unlikely to provide sufficient high salinity (35.70+ |) water. The CSIRO Atlas indicates that water of sufficient salinity, over 35.8, was present in the lagoon in January 1968 and October-November 1967 but it is agreed that this is an unlikely source for large quantities of this water even though outflow from the lagoon does occur in the north zone (see Part 1). The characteristic which chiefly tells against the lagoon as a source of the high salinity Coral Sea water is the latter’s low oxygen content (<3,.5 ml 1), although Scully- Power did not mention this point. Scully-Power identified some minor components (Fig. 51C), a high oxygen south-east component, a New Guinea coastal component and a low salinity component occurring close to the coast at 23-25 S, possibly resulting from mixing with lower salinity surface water. The south-east component was essentially Wyrtki’s Subtropical Lower water, southern (high oxygen) component. This entered the WCS at 21-24 S, 157 E, and temporarily divided at 23 S to the north-west and south-west (Figs. 52, 58B) but rejoined to flow south along the Australian coast. Scully-Power regarded the Central Coral Sea component as being derived from the low oxygen, north-east component (Subtropical Lower water, northern (low oxygen) component) after vertical mixing with the waters above and below it. Antarctic Intermediate Water (intermediate salimty monmum). The salinity minimum was apparent on the S,z profiles (e.g. Fig. 32) but more conspicuous on the T,S diagrams (e.g. 103 Hig. 32A, 51A) with a very limited range of values of only 0,05 at the core layer. The salinity distribution at the core layer (ig. 32C) showed low salinity water entering the Coral Sea from the north-cast and turning south along 155 E. Scully-Power considered that this diagram also showed a secondary tongue of low salinity water entering from the south-castat about 24 S, and described the low salinity region in the north-west as a ‘cell’, presumably because there could be no inflow of intermediate water there. The inflows were identified as Antarctic Intermediate water. The core layer of this water lay between 700 and 900 m over most of the WCS, increasing to L100 m in the south and for the small cell in the north-west. Using 34.50 as the maximum value for this water mass, it had a mean thickness of 375 m. In the core, the salinity varied from 34.42 to 34.47) , temperature from 4.95 to 6.20 C and dissolved oxygen from 3.8 to 4.3 ml 1. Upper oxygen nininum, Some 73",, of the stations showed a single upper oxygen minimum (Fig, 51B), the remainder showing a double minimum below the salinity maximum. The distribution of oxygen on the oxygen minimum surface was very similar to that in the core of the salinity maximum (Vig. 52B). Scully-Power considered that this indicated that the two water masses moved with the same flow pattern. Oxygen waximum, All stations showed an oxygen maximum about 200 m above the salinity minimum of the Antarctic intermediate water (Fig. 32). Scully-Power agreed with Rochford (1960b) that on the basis of available information, this oxygen maximum water could not be identified as a separate water mass. Isentropic analysis—Rougerie & Donguy ‘Gorgone 1’ 1972, late winter. The most recently published account of waters in the Coral Sea is that by Rougerie & Donguy (1975) for the Central Coral and Solomon Seas between 2 and 20 S, 153 to 163 E, with 72 stations completed between 14 November and 20 December 1972 (O.R.S.T.0O,M, ‘Gorgone 1’ 1972 cruise), i.e. at the end of the SE trades season (see Fig. 31), This cruise overlapped the area of the R.A.N.R.L. May-July 1968 cruises by Scully-Power in the region 12. to 20 S$, 153 to 163 E (Fig. 45A). During this cruise, the general characteristics of the water structure for the part of the WCS covered were similar to those described in previous accounts. The Subtropical Lower water had a core layer salinity of over 36.0 north of the Solomon Is., 35.9 to 35.8 /. inthe Solomon Sea and decreased to 35.65 |. inthe centre and south of the Coral Sea, T,S and 8,0. diagrams are shown in Fig, 53 for typical stations representative of these three regions, the station positions being shown in Fig. 45A. ‘The northern salinity maximum occurred at a thermosteric anomaly of about 340 cl/t (corresponding to a, =24.5—see Appendix), while the southern maximum was at about 300 cl/t (a, =25.0). ‘These two waters correspond to Wyrtki’s northern and southern components of the Subtropical Lower water and were also clearly distinguished by their oxygen content. Rougerie & Donguy located the boundary between these two components at about 14 § with a region of overlap at about 15 to 18 S$, 153 to 156 E. At thermosteric anomalies of 230 to 90 el/t (¢,=25.7 to 27.2) the linear part of the T,S diagram showed a mixture of Subtropical Lower and Antarctic Intermediate waters (Fig. 53A and cf. Fig. 51A) common to all the ‘Gorgone 1° stations and similar to the R.A.N.R.L. ones. The S,O, structure was, however, much more complex as is evident in Fig. 53B. The strong oxygen minimum around Stn. 65 north of the Solomons can be traced back to the low oxygen layer in the eastern Pacific near Peru (‘Vsuchiya, 1968). The upper part of this oxygen minimum layer passed through the gaps in the Solomon Is. chain and thence to the north part of the Coral Sea. The oxygen maximum of the Antarctic Intermediate water was also evident in the S,O, diagram (Fig. 53B), occurring at 500 to 700 m with values as high as 4.5 ml) in the Coral Sea. Rougerie & Donguy (1975) referred to this as ‘Coral Sea Water’ (34.6 to 34.8" ).., 4.0 to 4.4 ml/1D, not to be confused with Scully-Power’s ‘Coral Sea Component’ of the Subtropical Lower water (about 35.6 | , 4.0 m1). 104 oO 30°C r : = 4 A STN.25 STN.64 —~-~ NORTH OF Ast “~~. SOLOMONIS ~ - | 400 | SOLON SE * “a” SEA 7 GORGONE j Su8- NOV.—DEC. 1972 / TROP. -| 300 Z—~ LOWER SAL.MAX. 20° | _| 200 (S) ea ° pie Ww eet jeq =) = Ps w ra 100 a ere = ee WwW - FE _| 50 pe es 104 4 ANT. hs a INTER. 5°} SAL. MIN. al T,S ra 4 7 °o 2 T T T T 1 34 35 SALINITY %/oco 36 L 4 n | 1 __ B| SOLOMON SEA ake eee | STN.25 3 COR 4 a ry = “CORAL SEA” _NORTH OF = SOLOMONS_. E WATER ~ = pore STN.64 z ANT. ; 5 INTER. i i ao aS SUB- TROP. >* correct) (; He Bae sole LM NOS \ LOWER 3 \ (SOUTH) 7) te ; 7) aT Q\ ._) SuB-TROP. a INTER. \\ ep yrae: (EQUAT.) te Z (NORTH) 34 ™~ a 4 \ of ae a Ss, 0, SO. EQUAT. O2 MIN. 2 | | 1 1 | Fig. 53 (A) T,S and (B) S,O, curves for selected stations, Solomon and Coral Seas, ‘Gorgone 1” cruise, Nov.—Dec. 1972 (Rougerie & Donguy, 1975). Station positions in Fig. 45. 25:0 26-0 27.1 27-6 105 ‘The salinity minimum of the Antaretic Intermediate water was found to havea salinity between 34.15 and 34,55 and oxygen content of about 4,2 ml lin the Coral Sea whereas that inthe Solomon Sea was about 4.0 ml land north of the Solomon Is, about 3.5 ml I. It was concluded that as Reid (1965) and Johnson (1973) had suggested, the Antarctic Intermediate water could enter the Coral Sea both from the south-east and also from the north-east passing, south of the Solomon [s, after making a wide circuit through the equatorial regions. ‘Grrgone 21975, late summer, Although the analysis of the ‘Gorgone 2’ data (7 April to 7 May 1975) is not complete, some comparisons with the “Gorgone | late winter cruise can be made, Along the 156 Ef meridian during ‘Gorgone 2’ the upper layer of salinity less than 35 extended to 16 8 in the Coral Sea and was twice as deep (60 m average) as during ‘Gorgone 1°. Uhis southward spreading of the low salinity South Equatorial water may be attributed to the NW monsoon wind stress, in consequence of which the New Guinea Coastal Carrent attains its maximum strength at this time (Wyrtki, 1961) and brings into the Solomon Sea water from the western equatorial Pacific whose salinity is further decreased by heavy raintall in the Solomon Sca. Below the surface, the 35.5 isohaline in the transition zone with the Subtropical Lower water (southern component) was at 160 m depth in the Solomon Sea but rose sharply to the south until in the Coral Sea it was 50 to 60m shallower than during ‘Gorgone 1’, Atthe core of the Subtropical Lower water, the salinity was not over 35.8 and the vertical salinity gradient was much less than during ‘Gorgone L’, In these tropical waters where the salinity has a significant effect on the density field, a result was that during ‘Gorgone 2’, in April-May, the vertical stability was relatively small so that vertical advective movements were able to produce ‘doming’ at 12 to 13 S and 17 to 18 S. In November—December 1972 (“Gorgone 1”) the doming was slight and contributed little to increasing the typically low nutrient values of the upper waters. The dissolved oxygen distributions were generally similar for the two cruises. However, the oxygen maximum of the subequatorial water which was present at 300 m depth to the north of the Solomon Is. and between 10 and 12 S during ‘Gorgone 1" had almost disappeared at this southerly position during ‘Gorgone 2°. In the summer, therefore, it appears that the extension to the south-east of the upper waters has an effect as deep as 300 m and partially prevents the movements into the Solomon Sea of low oxygen water from the eastern Pacific. ‘This depth of 300 m is that at which the T,S curves for the two cruises come together, perhaps indicating the limit to the depth of penetration of climatic effects. Consistency of subsurface water characteristics In order to estimate the consistency with time of subsurface water characteristics in the WCS, envelopes of T,S data from 50 to 1200 m depth were prepared for each of nine cruises within the area 10 to 19 S, 144 to 159 E (although not all cruises covered the whole area). The cruises used were from the five sets listed in ‘Other surface data in seasonal sequence’ from 1965 to 1975 together with two other R,A.N.R.L. cruises for the North-west Coral Sea, i.e. cruise CS1 for Nov.—Dec. 1966 and CS2 for Apr.-May 1967 (Scully-Power 1969a, b). The 1968 R.A.N.R.L. data set was divided into its three constituent cruises to make nine envelopes in all. The T,S envelopes were very similar in position and width on the diagram, the main differences occurring above 20 C, i.e, above the salinity minimum or above about 200 m, Seven of the envelopes are shown in Fig. 54, the other two duplicating the November and May periods (envelopes 2 and 5). Table 7 below shows the mean position of the centre lines of the envelopes and the scatter in position about the mean. The variation in mean T,S characteristics was about +1C or +0,1 above 20 C, ic. above the salinity maximum, and only +£0.2C or +0.02 below 20 C, 106 SALINITY °/oo ‘i 35 35 35 35 35 35 5 36.0 TEMPERATURE °C Toh et br PPP ore or asONDJ FMAM J MONTH SURFACE TEMPERATURE 1 2 3 4 Ba 58 of AREA d (FIG.42) RANRL SHOYO MARU GORGONE 1 ULE GORGONE2 CS4/1 C5.4/2 cS4/3 1973 1972 1965 1975 1968 1968 1968 41STNS 30 STNS 25 STNS 22STNS 24 STNS 12STNS 12STNS Fig. 54 T,Senyelopes for data from 30 to 1200 m depths (700 monly for curve Lim the northern part of the Western Coral Sea {11 to 19 $). Time of vear shown on inset with surface temperature curve for area d (sve Fig. 41). ‘The widths of the envelopes were much the same for all the data sets, the average width perpendicular to their centre-lines being equivalent, above the salinity maximum (1.¢. above about 22 C or 150m), to 1.5C or 0.15 on the T’,S diagram and below 20 C or 200m. to less than 1C or 0,1 (i.e. less than 0.2.¢,). The envelope width in Fig. 61 is wider than these values because it covers data from all three constituent R.A.N,R.L. cruises together and from a more extensive area, both north, west and south, than do the data from the diagrams in Fig. 54. Bearing in mind that the cruises took place in different seasons and were distributed over a ten-year period, the above analyses indicate that below the surface layer, the TS characteristics of the major part of the WCS show only small changes with time, T,O,and S,O, envelopes were prepared but these were generally not as well defined as the T,S envelopes and it was not possible to compare the cruises meaningfully for these combinations. On the T,O, envelopes the upper minimum and lower maximum of oxygen were present but not clearly defined as there was much more scatter than for the T,S plots. (The envelopes had a width in the oxygen dimension of 50 to 75", of the total range of oxygen values below the surface (1 to 1.5 ml)] width compared to a total range of about 2 ml/l)). The envelopes became narrow (0.2 to 0.4 ml/l) only below the Antarctic Intermediate water. For the 8,0, diagrams, when all the data from a cruise were plotted no characteristic envelope was evident in most cases. This is to be expected in view of the convoluted nature of even individual S,O, curves (¢.g. Fig. 32C, 53B). The exception was for the Shove Maru cruise for which both the T,O, and the $,O, envelopes were narrow in the oxygen dimension (only 0.2 to 0.4 ml/1). For this cruise the Subtropical Lower water was entirely northern (low oxygen) component, whereas some southern (high oxygen) component was present for the other cruises. The probable reason was that the Shovo Maru cruise covered a smaller area than the other cruises and was restricted to the northwestern part of the WCS (west of 151 E) whereas the others extended as far as 7 further east and 2 further south. 107 Wyrtki (1962a) noted that the oxygen content of many of the southern component Subtropical Lower water samples in the Coral and Tasman Seas was above the (one atmosphere) saturation value for their temperature (Fig. 48). In the northern part of the WCS this was rarely the case in the Subtropical Lower water which was chiefly the northern component whose oxygen content was well below saturation. However, many of the samples in the upper waters above the salinity maximum in the nine cruises discussed above did have an oxygen content above saturation. These results indicate that most variation occurs above the salinity maximum whose depth in the WCS is generally 200 m or less. Because the variations in the upper layer are presumably due chiefly to climatic changes, this suggests that the limit to downward penetration due to such effects may be less than the 300 m estimated from the analysis of the ‘Gorgone 1 & 2’ results described earlier. The reason for this depth limitation is probably because the layer immediately above the salinity maximum generally has a maximum of gravitational stability and this acts as a discontinuity layer restricting downward turbulent transfer of water properties. Table 7. Mean of centre lines of T,S data envelopes below 50 m depth for nine cruises from 1965-75 for Western Coral Sea, and range of positions about the mean Co-ordinates Co-ordinates of mean line Rees of mean line Range A about ee _ about Temp. ( C) — Sal. ( ) mean ( ) Sal. ( ) Temp. ( Cj) = mean (C ) 25 35.40 +0.10 34.5 28.5 +0.5 20 35.62 +0.03 35.0 27.0 +1.2 15 35.24 +0.02 35.5 24.2 +0.6 10 34.75 +0.02 35.5 17.8 +0.3 5 34.46 +0.02 35.0 12.5 +021 34.5 6.8 +0.2 34.5 4.2 +0.2 108 V Circulation INTRODUCTION In Chapter IV, the distributions of water properties were described together with inferences on the flow paths of the various water masses but with few estimates of the speed of travel being available from this source. ‘his chapter will review the available information on the velocity of water motion, i.e. the path followed together with some estimate of the speed—either directly in distance per unit time or as volume transport per unit time across a vertical section (units discussed in the Appendix). The sources of this information have been: (a) ship’s navigation logs in which the difference between the course and speed through the water and the track and speed made good over the ground was attributed to water motion (for the surface layer only), or (b) dynamic considerations related to the distribution of density of the water (geostrophic method) for the surface and sub-surface waters. As far as is known, no direct measurements of currents are available for the WCS (although there are some for the East Australian Current further south). SURFACE CIRCULATION Wyrtki, 1960, 1962b—Current atlas and geostrophie circulation The most ambitious surface current description for the Coral and Tasman Seas was by Wyrtki (1960), based on the Adlas of surface currents, south-western Pacific Ocean (U.S.N.H.O., 1944) with reference also to Sea areas around Australia (Roy. Neth. Met. Inst., 1949) and other atlases. W’yrtki presented the surface currents for an areafrom5 to 48 S,142 to180 E intwosets of twelve monthly charts, one showing current arrows with speed indicated and the other in the form of streamlines. The latter show the direction of the current at each point along their length but do not indicate the speed as directly as the coded current arrows. (It should be noted that streamlines show an instantaneous pattern of flow and are not the same as trajectories, the path followed by particles over a period of time, except in the steady state when the circulation does not change with time. If the thickness of the moving surface layer remains constant, then crowding of streamlines implies increase of speed and vice versa, but if the layer thickness varies, this indication of speed will be obscured.) Although Wyrtki’s current charts look very complete (and are frequently used as references), later studies have revealed differences in many aspects. This is not surprising, because the material on which the charts were based was inhomogeneous (many sources) and had numerous gaps in coverage of the area, so that considerable interpolation must have been necessary to produce the apparent complete cover of Wyrtki’s charts. Therefore, only parts of three charts will be reproduced here, selected as samples to show the character of the current patterns deduced by Wyrtki. These are in Fig. 55A,B,C, for the months of January, April and August. A selection of specd values, in km/day has been 109 added from the current arrow chart. The major features of the WCS surface flow patterns according to Wyrtki’s charts were: (a) westward or south-westward flow (into the WCS) across 160 E from 10 to 25 $ all year (Wyrtki’s “Trade Drift’), (b) at 155 E,from15 to 25 S,a south-west or south flow all year (the start of the East Australian Current); at 155 E, from 10 to 15 S, flow to the west or south-west from May to December, to the north in January, to the south in February, and west or north-west in March and April, uw wi Fig. Surface flow patterns for (A) January, (B) April, (C) August, and (D) regions of convergence and divergence, Coral Sea (Wyrtki, 1960), Numbers represent speeds in km day. SPEEDS IN km./day " 4 150°E 110 (c) outside the Barrier Reef (an area notably lacking in information in the original Atlases) the current was shown as: (i) at9 to 11 S (Torres Strait), west from April to November, and east from December to March (NW monsoon}, Gi) from 11 to 19 §, north or north-east from March to November, and south or south-east from December to February, (ui) between 20 and 25 S, south-east all year, except for a weak northerly flow at 20 S in June. In Fig. 55 the January chart is reasonably representative of W'yrtki’s December to March charts, the April chart for April to June and the August chart for July to November. Typically, the original January and August charts and their respective periods showed relatively simple circulations, while the April to June ones were more complex, Wyrtki deduced regions of convergence and divergence from the starting and termination of streamlines (although this seems to be a somewhat uncertain procedure in view of the subjective manner in which the streamline charts were prepared). In general, as Fig. 35D shows, there appeared to be areas of divergence in the north-east and areas of convergence in the west against the Barrier Reef and in the south at the start of the East Australian Current. This might imply sinking at the Reef but probably not much because the water column is very stable below the mixed layer. Most of the convergence here resulted in a division of flow north-west and south-east parallel to the Reef as shown for August (Fig. 55C). The position of this division was at about 15 Sin February, 18 to 20 Sin March to September, and then moved north again to about 14 8 in December. There was no division in January when Wyrtki had a strong south-east streamline pattern emanating from Torres Strait and continuing down the coast. (Note that this did not imply that Torres Strait water reached as far south as the East Australian Current because the speeds were not sufficient—only about 17 to 22 km day or 5 to 6 degrees of latitude per month. ) The eastwood flow through Torres Strait was attributed to NW monsoon wind stress, and Wyrtki also showed a strong flow into the north of the Coral Sea from the Solomon Sea in February and March, For most months, Wyrtki showed a significant flow over or through the Barrier Reef between 19 and 23 S into the lagoon and then out again north of Gt Sandy Is. (24 S) to join the East Australian Current. He did not comment on this, nor has anyone else. It implies a considerable flushing of lagoon waters by WCS waters and could be a significant feature in the exchange of these waters if it is true. This feature warrants investigation in connection with Barrier Reef lagoon oceanography. It may be noted in passing that the Australia Pilot, Vol. IV (1962) used four of Wyrtki’s charts but had the July and October charts reversed, Ina later paper on the geopotential topography (see Appendix) and circulation in the Western South Pacific, Wyrtki (1962b) combined data from January to April from three years and showed the topography of various pressure surfaces relative to the 1750 db surface (db=decibar, see Appendix). Unfortunately, during this period of the year considerable changes take place between months in the surface circulation according to Wyrtki’s (1960) charts (cf. Fig. 55A,B). The surface circulation from his 1962 calculation showed the convergence to the East Australian Current as in all months, a southeast flow in the Solomon Sea as for the February chart (and possibly March), and an eastward flow to the south of Papua New Guinea as for January and March to April in his monthly charts. Essentially, he had a clock-wise circulation occupying the WCS, and an anti- clockwise one in the Central Coral Sea(13 to 30 8,155 to 165 E) with little net westward flow across 160 E into the Coral Seca, which was surprising. Wyrtki commented that the westward surface flow across 160 E in his monthly charts (“Trade Drift’) was due to the wind stress and would not appear on the distribution of geostrophic currents, citing a dil similar situation in the Indian Ocean. He maintained that the wind stress flow should be added to the geostrophic flow (see Appendix) giving a gencral westward flow into the Coral Sea across 160 E between about 10 and 20 S, although this might be less regular than the subsurface flow. (This argument is of doubtful validity.) In summary, the January to April (3 years) compilation agreed with the monthly charts in the southeast corner of the WCS (south of 15 S, east of 155 E) all the time and with the March chart over the whole area, but disagreed with the January and February charts in the WCS (and with the January and April charts in the Solomon Sea). This was probably as good agreement as could be expected for a variable period of the year for the surface circulation. Ina recent calculation of the dynamic topography for the whole Pacific, Wyrtki (1974) presented mean topographies for the sea surface relative to 1000 db for two-month periods, The contours in the diagrams were at 10 dyn cm intervals which is rather coarse for this small region but there was a gencral tendency for inflow to the Coral Sea across 160 E from 10 to 15 S, and outHow from 15 to 20 S at the surface. Rotschi, 1958-61—Earlier O.R.S,T.O.M. cruises Five cruises by the Institut Francais d’Océanie (now O.R.S.T.O.M., Centre de Noumea, Section d’Occanographic) (Rotschi, 1958a,b, 1959c, 1960d, 1961b) were referred to by Wyrtki (1962b) and these have been reviewed together with a sixth cruise (Rotschi, 1961c) for the area between the Solomon Is., the New Hebrides and New Caledonia, a little to the east of the present area under review. All but one of these cruises (May 1960) showed a westward flow at the surface approaching 160 E between the Solomons (11 S) and about 14 S. Between 14 8 and New Caledonia (20 S) the surface flow was sometimes to the west, sometimes to the east. The net transport to the west above 1000 m varied from 16 to 37 sy. (See Appendix for definition of | sv.) These earlier cruises, together with later ones, for the area between 10 to 19 S,155 to 165 E have been re-examined by Donguy & Henin (1975b) and their deductions will be reviewed later. Takahashi, 1959, 1960—Eddy in Coral Sea Takahashi (1959, 1960) showed a cyclonic (clockwise) eddy in the eastern Coral Sea which does not seem compatible with Wyrtki’s or O.R.S.T.O.M. results. The southward flow (Takahashi) at 162 C might well be associated with Wyrtki’s flow into the Coral Sea but his northward flow at 15 S, 150 E was opposite to the most persistent feature of all twelve of Wyrtki’s (1960) charts and his 1962 results. Actually, it is doubtful if Takahashi had sufficient stations to delineate his eddy fully, (Note that this eddy is incorrectly referred to as a ‘counterclockwise’ eddy by Rotschi & Lemasson, 1967, possibly a misinterpretation of Takahashi’s use of the phrase contra solem, ret. Sverdrup e al., 1942, p. 437.) Donguy, Oudot & Rougerie, 1970—Review of 1956-68 northern Coral Sea data Donguy et al. (1970) reviewed the surface and subsurface circulations of the Solomon and Coral Seas, using data from 32 cruises between 1956 and 1968, and assembled current schemes for the summer and winter seasons. For the summer (Fig. 56A) they identified from north to south (modified in the light of Donguy & Henin’s 1975b findings), the following currents and related water masses: (a) Equatorial Current to the west—Subtropical water, northern component (Sverdrup et al., 1942, called this the ‘South’ Equatorial Current), (b) New Guinea Current to the east—Equatorial water, continuing as 112 (c) South Equatorial Counter Current to the east, (d) South Equatorial Current to the west Subtropical water, southern component, (e) South Tropical Counter Current to the east (from the Coral Sea, ref. Donguy & Henin, 1975b), (f) East Australian Current to the south-east. Fig. 56 Currents, Coral and Solomon seas, (A) surtace, summer, (B) surface, winter, (C) 150 m, summer, (D) 200 m, winter (Donguy et a/., 1970). Abbreviations for current names; E.C. = Equatorial Current, N.G.C.=New Guinea Current, S.E.C,C.=South Equatorial Counter Current, S.E,.C. = South Equatorial Current, $.1.C.C. = South Tropical Counter Current, E.A.C. = East Australia Current. a) == Ec: T>27°¢ — $=345-35-0-F Ss Xe [ ma i SURFACE PLE PRAY 2°57] SURFACE SUMMER ; 4 WINTER T T [EAC oi T t BAL Ht 4 140° 150° 160°E 140° 150° 160°E n tl l i o° 4 1 1 i L 4 C t= +° 10 vr Eco i— ] (ii SECO | | | iain is) oe s $= 358 | L “$= 35-7 SEC: | § <35-7 Tq S.T.C.C: q Logs r 150m. C Ry 77°] 200m. SUMMER cH WINTER i. 1 i a Pye ye ye i nl l { — SUB-TROPICAL EH SUB-TROPICAL T=C° $=%se —— (norTH) | ESUATORIAL LH (SOUTH) (Tse", ) 113 ‘he surface water of the Coral Sea in the summer was identified as mainly the low salinity equatorial surface water from the north, together with a component of Arafura Sea water coming through ‘Vorres Strait. In the winter (lig. 56B) the main feature was a considerable extension to the west of the South Liquatorial Current due to the SH trades. Part of this South Equatorial Current {lowed northwest through the Solomon Sea to join the Equatorial Current and part contributed to the Kast Australian Current. ‘Uhe Subtropical Counter Current was slightly further south and more saline (34.8 to 35.0) ) than insummer. The contribution of the Equatorial Current to the South Equatorial Current was also more saline (34.5 to 34.8) ) than in summer. The speeds in these currents were generally small, | to 3 cm/s. Most of the surface water in the Coral Sea in the winter was the Subtropical Surface water of the South Pacific diluted by some Equatorial water. Tt must be borne in mind that the above description was of broad features and that significant differences in detail between years may be expected, for example as mentioned in the next section, Rougerie & Donguy, 1975 —Gorgone J" 1972 cruise, northern Coral Sea In November—December 1972 during the ‘Gorgone I’ cruise (Rougeric & Donguy, 1975), regarded as a winter circulation period because it was still in the SE trades regime, the South Equatorial Counter Current was evident only north of the Solomon Is. (not south as well) and the South Tropical Counter Current which was conspicuous in 1956 between 15 and 18 S was, in 1972, only evident just north of New Caledonia (see also the next section). The strong westward flow across 160 E was clear in 1972 from 10 to 18 S. (A curious feature which often occurs in the surface dynamic topography about 16 to 17 8S, 156 Eis a cyclonic gyre of diameter about 300 km. ) Donguy & Henin, 1975b—Review of flow at 158 and 163 E Donguy & Henin (1975b) assembled the available data to 1972 (13 cruises) for the arca between New Caledonia and the Solomon Is., forming two lines of stations close to 158 E and 163 E respectively. They determined mean dynamic heights relative to 1000 db for the two sections, omitting two cruises which were notably different from the others. Fig.57 shows the currents identified across the two sections from the mean dynamic heights. The presently unnamed current (U in Fig.57) north of the Chesterfield Is. was possibly identified with the northerly flow to the west of New Caledonia described by Scully-Power (1973b) and the next section, At the right in Fig. 57 are given the maximum mean speeds at the surface relative to 1000 db, the maximum mean speeds in the upper 100 m, the maximum observed speeds and estimates of the volume transports for two of the currents. (As the northern limit of the South Equatorial Counter Current was not determined, its volume transport was not calculated.) One of the cruises omitted from the mean dynamic height calculation was the 1972 ‘Gorgone 1’ cruise which showed westward flow across the whole section at 158 E.and westward over most of the section at 163 E. Scully-Power er a/., 1969, 1973a, b—Western Coral Sez Lockerman & Scully-Power (1969) calculated geostrophic currents for a limited area in the North-west Coral Sea for the end of November 1966 and the end of April 1967. At the surface in November there was a clockwise circulation with casterly flow at 25 to 40 cm/s between 10 and 12.5 S and a westerly flow of 20 to 30 cm/s between 14 and 15S. In April a similar circulation was observed with speeds of 20 to 60 cm/s. The November circulation was opposite to Wyrtki’s (1960) surface current chart which showed anti-clockwise circulation in November with a flow west through Torres Strait, and a weak clockwise circulation in December. Lockerman & Scully-Power’s April circulation was basically in agreement with Wyrtki’s April and May charts, and with Wyrtki’s (1962b) geostrophic estimate from January to April data which showed a clearly 114 developed clockwise circulation at the surface centered at 13 8, 148 Eto the SSE of Port Moresby. The most complete coverage of the Western and Central Coral Sea, albeit only for the winter season, was that described by Scully-Power (197 3a, b, c) based on the three cruises in May, June and July 1968 (Fig.45A) and on data for May, June 1971 (R.A.N.R.L. Cruise CS7, unpublished data). In Fig.58A, which also indicates the station positions, the surface dynamic topography relative to 1500 db is presented to show the circulation. Scully-Power noted that the surface circulation was rather complex but drew attention to two features: (1) although there was considerable westerly flow into the Coral Sea across 160 E, most of this water flowed out to the north into the Solomon Sea rather than turning south across 20 S to feed the East Australian Current. This was not in keeping with usual ideas of flow in this region, e.g. Wyrtki (1960, 1962b), (2) an anticlockwise eddy was present between 20 and 26 S, off the Capricorn Channel area of the Reef. In addition, attention is drawn to: (3) the very strong flow south at about 24 S close to Gt Sandy Is. It is noted that, although not mentioned by Scully-Power, the U.S.N.O.O. Pilot Charts for the South Pacific (1955) show a similar feature to item (1) for winter. For September to May (summer), the main surface flow shown is west or south-west across 160 E with much of this flow proceeding south across 20'S to feed the East Australian Current system, but for June-August (winter) the westward flow across 160 E divides at about 25 S, part going south and part north across 20 S. The pattern resembles Scully- Power’s surface and 150 m winter flow patterns with the division at 25 S rather than at Vig. 57 Currents and transports in the Central Coral Sea (Donguy & Henin, 1975). Abbreviations for current names: $.H.C.C.=South Equatorial Counter Current, S.E.C.— South Equatorial Current, S.T.C.C. = South ‘Tropical Counter Current, U Unnamed current. P wT T T —T S SOLOMON Qs. v s _ SPEED (cm/s) VOLUME \ MAX. MEAN MAX He aibcag: Wit — 4 SFC, UPPER OBS. (sv) QS o 1000db.100m MAX. MEAN a | - S.E.C.C. ° S.EC.C. 4 20 45 < 4 SEC, sec<| HEARILES SEC. 20 20 45 67 35 ° L15° & 4 - S.T.C.C. S.T.C.C. N S.1.C.C, 15 15 25 5 2 u<| _ CHESTERFIELD 7 — LIS. Q CALEDONIA 155° 160° 165°E = 4 1 115 1 Sas he found ib. Phese Pilot Charts show a southward component of flow in summer fronpthe Solomon Seainto the North Coral Seaas do Wyrtki( 1960, 1962b) and Donguy ef wh (1970. However, a cumous feature in the Pilot Charts is that in summer (the December Mebruary chart) a significant eeesfeoerd current is shown through ‘Vorres Stran, contrary to all other accounts, ‘Vhe Pilot Charts show speeds of 20. 35 km. day Oo) OS kts) in the body of the Western Coral Sea, inereasing to 25 55 km day at the south toward the Bast Australian Current, Surprisingly, the ociydn of the strong flow to the south (item 3 above) was not explained although) in discussing the water musses Scully-Power referred to the possibiliy of bigh salinity water flowing from the Barrier Reef lagoon. [fe referred (1973b) to this water as flowing southward Gnside the reef) and escaping at the southern extremity, (Ehis possibility was discussed earlier with the conclusion that the lagoon was nota likely source of large quantities of water tothe WCS.) [tis possible that this feature oupht bave been a transicat one. Pn any event it depends on the observations ata single station (St £3, Cruise CS 3, Scully-Power & France, 1969¢), as the four stations north of this were in much shallower water ess than 420 mi) than the reference level of 1500 db used for the geostrophie calculations, Scully Power also pointed out that the topographic high at about 25 S, 155 FE and associated anticlockwise circulation also appeared on Wyrtki’s (1962b) pattern for late summer (hanuary to April) and, although less intense in winter, was probably a permanent feature of the WES circulation pattern. Pnally for the surface circulation, Scully-Power described some results for the North-west Coral Sea for the summer of 1966 and winters of 1907 70, allof which showed u northerly surface flow over the deep water off Cairns 17S) which turned cast when approaching the Papua New Guinea coast. ‘Phe surface speeds varicd from 5 to -bS ems, SUBSURFACE CIRCULATION Hamon, 1958—Intlow to the northern Coral Sea from the cast Hammon (CSIRO, 1958) estimated the flow into the Coral Sea between the Solomon Is. and New Caledonia from the 1 square current vectors in Seu areas around liscralia (Roy, Neth. Met. Inst., 1949), obtaining a mean speed of 7 em,s and a volume transport of 9 sv assuming a layer depth of 200m, Hamon emphasised that the choice of 200 m for the upper liyer flow was only an estimate and that the volume transport should only be reparded as an order of magnitude value. Ttwas about one-half of the mean of the values below calculated by Wyrtki. Wyrtki, 1962b—Geostrophic circulation in the Coral Sea Wyrtki (1962b) also determined the geopotential topographies of the 100, 200, 400, 700, and 1100 db surfaces relative to 1750 db, (‘Vhese can be taken as approximating the same depths in metres, ) At 100 and 200 m, the westward flow into the Coral Sea was better developed than at the surface, being stronger at 200m than at 100m. The circulation at these depths compared well with the spread of the Subtropical Lower Water (Wyrtki, 1962a) as described previously. ‘The westerly flow across 160 E was still evident at 400 m, and at 700 m where it consisted of Antarctic Intermediate water, At 1100 m there was little relief in the dynamic topography indicating no distinctive circulation at that level, Rotschi, 1958-61—O.R.S.1T.0O.M. earlier cruises, volume transports Between 100 and 500 m all six cruises (see above) showed a Westward flow into the Coral Sea, For these cruises, Wyrtki (1962b) showed the 300m flows which were characteristic of the 100 to 500 m layer. He regarded the 100 m and deeper dynamic llo topographics and associated geostrophic circulations as indicative of the total {low below 100 m but for the surface the wind stress flow had to be added. in his opinion. From the O.R.S.T.0.M. data Wyrtki (1962b) calculated the volume transport ‘at about 163 E) into the Coral Sea between the Solomon Is (11 S)and about 18 $8 where the reef system extending northwest from New Caledonia ends, and obtained, from the surface to 1000 m; Noy. 1956 23 sv Fob. 1960 2 25 54 June 1958 26 sv May 1960 12 sv Nov. 1958 6 sv Aug. 1960 15 sv Mean 18 sv The transport maximum occurred at 100-300 m depth and on two occasions there was asmall net transport to the east between the surface and 100 m. The speed of this flow was small, only 3 to 5 cm)s at its maximum. Donguy, Oudot & Rougerie, 1970—Review of 1956-68 northern Coral Sea data In their review of the Coral Sea circulation, Donguy e7 a/. (1970) found that below the surface, at 150-200 m (Fig.56C, D), the same currents were present as at the surface but the South Equatorial Current penetrated well into or across the Coral Sea in both seasons, and the salinities were higher than at the surface. The speeds in the vicinity of 160 FE averaged about 3 cms. Rougerie & Donguy, 1975—'Gorgone 1° 1972, late winter During the ‘Gorgone I’ cruise (Rougerie & Donguy, 1975), the flow below the surface was generally similar to the flow at the surface except that the (eastward) counter currents were weaker and shallower than the westward currents. At about 150 m (salinity maximum) and at 300 to 500 m (oxygen minimum) the westward flow predominated. The South Equatorial Counter Current was present north of 10 S but the South Tropical Counter Current had practically disappeared. At 500 to 700 m (oxygen maximum) the flow was zonal in the eastern and southern parts of the Coral Sea but the part north of 13 & turned to the north into the Solomon Sca at about 155 E. The westward transport of the South Equatorial Current was 37 sv across 153 E. The South Tropical Counter Current carried 3 sv at 153 E. Rougerie & Donguy (unpubl.)—'Gorgone 2° 1975, late summer Although the analysis is not complete, some preliminary comparisons between this *‘Gorgone 2’ late summer and the ‘Gorgone 1” late winter results can be offered. The end of the summer is characterised by the marked development of eastward Hows. Between 10 and 12 S the South Equatorial Counter Current reaches 50 cm:s at the surface and still has a speed of 20 cm/s at 200 m. Its depth decreases eastward and on leaving the Solomon Sea the 20 cm/s isotach is at only 100 m depth. Between 15 and17 S$ at the end of the summer the South Tropical Counter Current is slower and shallower than the South Equatorial Counter Current but shows the same decrease in speed and depth as it progresses eastward. Its southern boundary coincides with a thermohaline front where the salinity exceeds 35.0 | and the temperature falls below 28 1. Dynamic height calculations relative to 1000 db indicated the presence in this zone during ‘Gorgone 2’ of a trough identical to that observed during ‘Gorgone I’ and at the same position, 16.5 S, 156 E. This continuing dynamic structure is associated with the presence here of an isohaline, nutrient-poor layer which is about 100 m deep and was of considerable horizontal extent during ‘Gorgonc 1’. This trough is at the centre of a cyclonic gyre with which the eastward flow appeared to be closely related. The continuance of the gyré may be associated with the flow but it is possible that it 1s only when the SE trades weaken (in the summer) that the waters start moving to the east and develop the South Tropical Counter Current, This was inconspicuous during ‘Gorgone 1’ (at the end of the SE trades season) but reached a speed of 15 to 35 cm:s at the surface 117 Fig. 118 oS SOLOMON a ° 20 SURFACE 4 FLOW STATION POSITIONS « 1968 4 1971 58 Geopotential topography and flow patterns relative to 1500 db, Western Coral Sea, May-July 1968, (A) surface, (B) 150 db level (Scully-Power, 1973a). during ‘Gorgone 2’. During this cruise, the South Equatorial Current, squeezed between the two counter currents and opposed by the NW’ winds, had a surface speed of less than 20 cm’s. In the north-west of the Solomon Sea the westward flow had a similar speed but below the surface had a core of about 100 m thickness with a speed of 30 cm s. Scully-Power e7a/.,1969, 1973 a,b,c—Geostrophic circulation May-July 1968 and volume transport budgets, Western Coral Sea Lockerman & Scully-Power (1969) found clockwise circulation in the North-west Coral Sea at the 900 db level (about 900 m) in November 1966 with speeds of 5 to 8 cm s, and in April 1967 they found what appeared to be the northern half of a clockwise circulation with speeds of 15 to 20 cm/s. This meant that the whole column from the surface to 900 m had a clockwise circulation, Wyrtki’s (1962b) calculations for this region showed little movement below 400 db. Fig.58B shows the circulation at about 150 m, the level of the core of the Subtropical Lower (salinity maximum) water in the WCS, from Scully-Power’s (1973a,b) study of the winter regime, The circulation at this level was simpler than at the surface (Fig. 58A) but had the same main features. The division of the main westward flow at 18.5 S when it reached the Barrier Reef was clearer than at the surface where it occurred at about 19 S, close to the average for May to July in Wyrtki’s (1960) charts. Scully-Power pointed out that the dynamic topography slopes at 150 m were not very different from those at the surface, and he considered that the flow pattern at this level was probably representative of the mean flow of the whole upper layer, the irregularities in the surface pattern probably being due to local wind stress variations. The circulation at 900 db (Antarctic Intermediate water) relative to 1500 db was basically the same as at 150m although much reduced in intensity (speeds about one-fifth of those at 150 m), and the strong flow to the south off Gt Sandy Is. was absent. Scully-Power (1973c) went on to calculate volume transport budgets for the Coral Sea by selecting quasi-synoptic (i.e. near simultaneous) sections and calculating the flow across each of these from the surface to 250 m and surface to 1500 m depth (Fig. 59A). This showed a westward flow of 37 sv across 157 E (CD) into the WCS. Of this 28 sv Fig. 59° Volume transports in sverdrups (10" m*s), (A) Western Coral Sea, May-July 1968, (B) North- western Coral Sea (Scully-Power, 1973c). MaAY- JULY , 1968 VOLUME TRANSPORTS , O5 if <— 0-250m 0: May—JuNeE | — =0-1500m. : ie 4 IN SVERDRUPS (108 m)3/s) Howe owt te the perth ote the Solomon Sea across BO dae 59A), 9 sy inte the North wet tanh oa Ca andonty sv south CADG, appiarently gito the Barner Reef Lagoon. Australian Current wits Up phe Ten the cost achoss Pad Posouthead 2005 CD bo Phe patterns was the same for the PhO tie setts How ob Oy cress JOS bCr iite tie | Hyped bayer Mew above 290 bea Ph Adviet an September LO70, dati trom a ROALN OR OD. cruise (Wood, 1971b, Howobwestacid How ners LOS TOO in bap S9A, from the Solomon ds. todo Sanda mid New Cadedouim Phe westward flow across Hd tron clock wise eddy Tetween bo Had tia peed ab aboot $00 CSubtropical Power water, while the northware How dcross se Hon pit the Solomon Sea had iis sient ababout the same depth be Hippo diya How depos Ths section beri weil aid tothe southy. Phese subsurkice flows conpesponidted ta the substihice Components of the South bquatonal Current described Dy Donpay cha C1970) sore ponetratnig mite the WS aid sete north-west through the Sobor se Sealy Power (Od bofescabed tinisport caleulitions for the North west Coral Sea for (he ct Gb POG God winters of L967, L9G aiid 1970, all of which showed a il Hertherby sanbiee How of Cadtaes Cotte cast meu thie Papua New Crumen coast “Phe Hhsaihtee fhow had the sande poatbenin, always a northerly Component flow alonp the worth Ouest Cost Phe Chins ports calcibited between tielividuidl stations are shownin Hig MOB TT ii beabaee here was the aribility Walk tine, a chimietersic of the WES an penetabonebol the orth western Gworal Sea qa particuha Scully Power remarked on ‘the ont Common featite beri the westerly flow ai the south part ofthe (North west Coral Seb regen adel tlhe easter by: (low: qoctheot has! Toston Sealy Power) Pod sop comsrdered thatthe rest strikiny feature of the SOHO ree Ene WC OS ae The north westerly outflow: tothe Solomon Seaanal the lek ebooutherty Haw aeross (OOS aaecantrast Lo Previous Comee pis, Gosthy from simmer cata, eb the westerty How between the Solomon ds aad New Caledonia ite the Coral Sennorth COFCO OS Con CO ie occa pby a Combintions bast Ads trib Carrent to the south along the Ato thin coast ie Bead, P9ob, Wayrtha, L900, Melb. Scully Powe ported out that at Comey cre TO ster db COS ERO, Pod were added to the cataased by Woy ree CP9OCD Tb Conte Ol Phe South west flower the (oval Seu durimny summer would be DrohOu up ato da sciies ab wate lock wise ecleties ate PhecusstHiphon ofa econtindous How to Hh sor Thom Po Soni Rave to be tected Phe sipeested iistemcbaserres of southward HOOT ip hock Wine coldiies Oripinatinp ab about bo Sa stammer and about 20 San Wonton Phe abso cored (hat the winter pattern apreedl with that proposed by Donguy eral O98 Or althodeh there wits no evidenee this dato the South) Propical Counter Current, Sealy Power Wert on to discuss postible aspects ob the dynes of the bast Ate tratan (tb bent as aowestern botociry carrent bat this ts beyond the scope of this beviewW Pome ge P8Ad, POR, P70, P87, PO ane Ciodbeeyy & Bg. 1971, 197 ba, by have descobed andl discussed) Vartotis aspeets of his Carent and referenee mn be made to (her prapers 4 Pinally, Scully Power commented that the North western Coral Sea was the most compres abe ob ihe WS Phe stipestedt thittas the thaw (hrouph Corres Strait Gaax inn ee Tite as O 9 oa CSTRO: Pos. was an order ot mmivnitide less than the other tows edhodhited for the bea chap 99D) the Northwestern Coral Sea could be considered a Closed Dati amel the Hows ds beri in peastrophie baltce durin the winter. In the sHHe tr, However cea Novontber, December boo, ba S88), a southward flow lon the slope oth Cartas worl be tecded to bakice the tow beget. Phe results of the study of He May PR OACNOR OE cruise the North west Coral Sea are awaited with interest. PE UISTUNG CEIAR AG PERIS PLCS Ob THE WES PERN CORAL SEA Hos terest tacdise Senily Power's valuime thainsports fo estimate the flushing Clune beristis ab the WS, be the coliions between the volume flows ane the volumes of io water in the basins. This has been done for three areas of the WCS, the North-west Coral Sea, the main body between about 12 S and 20 S, and the southern part between about 20 S and 29S. These areas were chosen to enable Scully-Power’s (1973c) volume transports to be used. It has been assumed that the transport in the upper 1000 m would be 90", of Scully-Power’s values calculated to 1500 m, and the Reef areas have been assumed to be 100 m deep. The figures obtained are shown in Table 8, For areas 2 and 3 the annual inflow is greater than the basin volume of the upper layers, which implies, to a first approximation, that the water will be replaced in less than one year (replacement time’). In estaurine oceanographic jargon these basins are ‘well flushed’ for such large areas. The value of 5 years replacement time for the 0-250 m layer in area 1 follows from the net transport of only 0.6 sv into the area (across section AB, Fig, 59A) which appears unreasonably low by comparison with the larger values (by a factor of ten) across this section for other cruises (Fig. 59B). A more likely replacement time would be about 0.5 to 1 year, similar to the values for the other areas. It will be appreciated that these estimates, which extrapolate values from one three- month period to annual values, must be regarded as only approximate; but as there is no indication that the circulation changes drastically during the year, they are believed to be realistic, Naturally, year-to-year variations are to be expected. UPWELLING OFF THE EAST AUSTRALIAN COAST Godfrey (1973) made some calculations of transport relative to 1300 db off the east Australian coast between positions approximately as follows: (A) 18 S, 147 E;(B) 19 S, 157 E; (C) 28 S, 158 EF; (D) 39 S, 155 E; (E) 38 S, 150 E, using such data as were available near to those points, Across ABC there was a very small net outflow above 100 m and an inflow from 100-1300 m of about 18 sv, with its maximum at 300 m. There was an indication that most of this inflow was across the north-south section BC. Across CD there was a strong outflow (to the east) with its maximum at the surface and most of the flow above 300 m; across DE there was a less strong outflow with its maximum near the surface. The sub-surface inflow and strong surface outflow implied upwelling between the coast and the line ABCDE, probably within 50 km of the coast. The vertical flow was calculated to have its maximum at 250 m depth with a value of 7.5 sv, corresponding to a vertical speed of about 10 m/day which is a high value for upwelling. It should be noted that the upwelled water was not high in nutrient content. Typical values for P-PO, were: Depth: Sfc 100 m 200 m 400 m P-PO,: O.1 0.15 0.3 1.0 fg at/l. Table 8. Basin volumes and volume flows into the Western Coral Sea Sfe. to 250m 250 to 1000m = ~— Replacement time = (incl. reefs (excl. reefs) basin volume|inflow Area ass, to be per yr 100 m deep.) (xlO in) (x10'*m } Sfe. to 250- 250m 1000 m (years | (years ) NW Coral Sea Vol : 1.0 2.3 5 1 (west of 148 E) Inflow/yr : 0.2 2.4 Main WCS Vol : 2.0 5.9 (148 E to 157 E. 0.4 0.9 12 to 19.5 S) Inflow/yr_: 5.3 6.9 South WCS Vol : 1.2 2.8 (west of 157 E, 0.5 0.4 19.5 to 29 S) Inflow/yr : 2.2 6.9 12] VI Summary WATER MASSIS Te Phe precious parageaphie have reviewed the majorty of the papers relating to the Corl Goa and that oniatenal on the Solomon ane Paani Seas necessary for physical continnity and toaeiate the Coral Seacamfbormation to possible source areas of ita water Hasnes Paps 60,6} and Oo cumimatine, fromthe above review, tre main ¢ haracterinties of Heowater al the WC S ateelh, most weight haa been placed on cata from the RA NUR, 1008 ciiines abeacribedh by Sealy Power and an those parte of the ORS E.O,M Gorgon croimes which overlapped the WOES and were described by Rougerne and Pevtageeny Phe usual way to cummiiine water tiias Chardetermtien in numerically i tuble forma, Hhoalbonnatie ds todoctgiaphically aeing charieterintie diagrams, While the Gible form rs Comsonmont tor ceferenee did for quotition it fain some dinndvantigers, Phe fist in that the Fangencol properties ta a Water didas and the relations between mitnnes are lens enny to appreciahe nimerically than ina diagram Phe second in that a ible cain be minleading Por instance, the conventional Gable fon deaciphon of the Linear part of the PyS Climaetoristien am big TEA would be Eo wo 2, 8 = $d, F908 which NpPecilien a Pectanghoon the DS taggin whereas the actual PS combinations observed in the sea are limited toa diagonal stop occupying only 10") of the rectangulie area, Phe preemion posibb with the aumenical statement in rarely needed. Accordingly Chin summary of Water propertion in presented in graphical torn Suibace wateta Chanter diaeran Pag G0 shown feat the envelope of aetual surtace water PS Chatartetiaties Gor the WS dowinter Clath fie) from Scully Power (197 4a, 6), Po his Onvelope tas boon addecba seate of hitttinte becaune the surface TS values are distributed alone the covelope fromthe Gailbot Papua valies at the high temperature, low salinity end tothe valies semth ot Brisbane atthe low temperature, high salinity ead Toadidifien, the followin Dave been added on Pig. a0 fa) atentiniate af simmer sittace EUS valiies (dashed line) bused on GCSTRO Athas (PO SA) data, (hr TS hataeterintics af external waiter andes which probably contribute to the tke upoat the WS nities waters In bie G3, the envelope at sietace water TS valuew for the two westerly lines of Srationa tithe Coral Sea lari the ORS OTLOOM, "Gorgone 1 cruive is shown, This Severs a amaller area ta the diagram than that for the ROASN URAL. cruises, probably Docatine the stations dacd covered a smaller area of (he Coral Sea and particularly because Hettinger eaten dite the North wert Coral Sea where were found the low salinity valiien Which extended the ROA NOR OD envelope OOo laerdiis Cor the siilace waters are aot presented because they generally do not wiv tomer information than the TS diagram Phe reason is that the surtace waters te are usually very close to beta saturated with disselved oxy yen andy as the satiratron watlie is determined chictly by fomperatire, an S.O). drayran for surbace waters ts Dastealhy an SVP diagram with increasing, oxygen rephiciag decreasiag fempeniiare Cbeetise the saturation value for oxygen deercases as LOMpPeralybe Heres 3) Geographic distributions and seasonal eariadions. Pie presen titions ob saehice Compe rainy and salinity in Figs. 35, $6, 38 and lO are the best that ean be piven at the present They should be used subject to the Hmitations discussed im the text ‘The semi-geographic presentations of Pag. EP aiy be adequate for same purposes, and also show the s asonal cycles well, bothoas Taine and Same plots andoas combined PyS,time diagrams. Hip. 6000 ES diagrams, Western Coral Sea, sanhice waters, Wiiter Are SUE an Ee SPOT ITE SOuEeG WaIEO ns Key: Wester Coral Seaoy Coral Sea surtace Rochford, P29), WEN winter envelope Sent Power, 19740), SUAL summer envelope cestimated) TPosternab water mitsses Mratira Sea oa (Rochford, 1959), River rune type be Scully Power, Foe kr South Pqatoriat oo Rec htord, 1959), d cSeully Power, 1Oddai, 6 (Roupene & Donpuy, bee, West Central South Pacitie Rochford, 1959), 2 CRatschi & Temassen, P90 Subantirette sariiee he Rechbard bry (Rotschi& Lemiasson, 1907) SALINITY ("/on) 35 IG low Re) Ww « 3 ‘g fod WwW a = iW Ke 25 4 | h WESTERN i CORAL SEA ° SURFACE WATERS TS. Subsurface waters Characteristie diasrans. Pips. OL and 62 present the subsurface property envelopes for TS and S,O., ORS ZE.O.M lo “Gaorpone L 1200m depth during the R.ALN.R.E. 1972 (summer) cruises. ‘Phe greater extent of the oxygen 1968 (winter) and the Hina and maxim arests inthe RLALJN.RE.. data is probably due to the larger area of the Coral Sea covered. In the SO. diagrams, the major difference is the lower oxygen Pap Ob GA TS amd By) SoO. dhapisins, subsurface water, winter, Western Coral Sea (data from Scully- Power, 197 $4, Scully Power & Prance, b969n, b,c) 40 1 i] t i T Oo A 34 35 36 ¢ SU ot i RF ac E SL ~ Ss \ ‘el N\ \ SUB-TROP. non ein WESTERN Sere . < SUB-TROP. . CORAL SEA POWER 250m 20 WINTER (SOUTH) 4 a) til i ~ OXYGEN MINIMUM «t (150-450m.) ~ 25 / | til & ba ep ) 1h 7 4 5 if ir 5 eer io OXYGEN MAXIMUM 7 a 160°E (500-850m ) vt / Fr / VV ANTARCTIC INTERMED (650-1150) 9 \\ F KEte AREA NES COVERED ' . BY THESE | 27 TS; 5,02 DIAGRAMS ; 1 1 1 1 1 aN oe a 34 SALINITY “Joo 35 36 5 l i L 4 - n 4 B OXYGEN MAXIMUM __ ~SUB-TROP. LOWER ; (SOUTH) £ Pomies = ay. tu ree <— CENTRAL CORAL SEA 2 F ANTARCTIC + SUB-TROP. LOWER «x 4 INTERMED Ny ‘| oO NN \ “oe s> ” \ \ ++ « —-SUB-TROP. LOWER \\ SOUTH (NORTH) ao \ OXYGEN MINIMUM | NORTH & NORTH WEST 3 t L L 1 eR eee et], content of the Subtropical Lower water, southern component, in the “Gorgone I” data. This is probably because the cruise area did not extend as far south as the R.A.N.R.L. one, and therefore did not pick up as much of the higher oxygen southern component. (/ Auch of the Subtropical Lower water in the R.A.N.R.L. cruise area south of the ‘Gorgone I’ area had oxygen values of 4.5 to 5.0 ml/1.) Fig. 62 (A) T,S and (B) S,O, diagrams, subsurface water, summer, West Central Coral Sea (data from Rougerie & Donguy, 1975, and ‘Gorgone I’ data record, Donguy ef al., 1972b). ° 30 T T T T T A 34 35 36 Cy. SURFACE - \ ON 7 en eee VN a tN \ eae Sod sit NA | — \ 4 a \ _— \ \ Sst= “= — es SS 490 23 Xi 1 -SUB-TROP. WEST CENTRAL SY LOWER CORAL SEA : CNORTHD - ° ce SUB-TROP. | 225 20° SUMMER aul eae mJ ee i? (SOUTH) oO ° eR r —~ OXYGEN WwW Vie Z MINIMUM Ss ie gL (160-300m.) re st eT LO ff 54 eA x + 300 25 of Fs te. STNS.8-20,46-520NLY 4 WW a vars = 4 7 WwW 4 7 e 4 7 7 a va T T T T T ° “JF feat NG © 165°E ° 10°4 yo he AHN my 104 OXYGEN MAXIMUM arg Sa = (550-770m.) ia \ Oe ; ry wk 4 L ANTARCTIC et 5 2 8 INTERMED. ><, | Uf CORAL SEA \ | (750-1000m.) +r 20°s4 “4 TN AREA 2 4 a S parallel, but itis difficult to discount evidence of southward tlow across 20 Satsome times. At the surface, some inflow (eastward) through ‘Torres Strait occurs dug the NW monsoon toe (ve or three months, with westward outflow for the rest of the year, However, the volume transport cither way is probably only about 1 sv, which is much smuaitler than the other flows in the WCS, Many of the details of these circulations are uncertain, In particular, in the North-west Coral Sen the results of Lockerman & Scully-Power (1969) show a clear clockwise circulation at Che surtace and at 200 db whereas Scully-Power (197 3a) shows anticlock- Wise circulation atthe surface and 900 db and less clearly at 150 db. In addition, it must be expeeted that variations tn the flows will take place from time to time, and that the surface circukidion will be subject to local short-term variations due to wind stress variations, aa) From the flow budgets calculated by Scully-Power an analysis suggests that the waters of the upper 1000 m of the Western Coral Sea are replaced in one-half to one year, i.e. the region is well flushed. Fig. 63 Mean circulations, Western Coral Sea, surface: (A) summer, (B) winter; subsurface: (C Subtropical Lower water, core depth 50-250 m, layer thickness 200 m (> 35.5 ,(D) Antarctic Intermediate water, core depth 650-1150 m, layer thickness 400 m ( < 34.5 ). Abbreviations for current names as for Fig. 56. SURFACE WINTER SURFACE SUMMER CIRCULATION ANTARCTIC INTERMEDIATE WATER SUB-TROPICAL LOWER WATER Appendix Units, conversion factors and glossary 1. Length: Ponautical mute | onmil 6080 ft 1.85 km I derree hititude 50.8 nmi 110.7 km (mean for LO 25 8), At latitude 10 Ih 20 25 30S I dep long 59 58 57 55) 52. nimlbapprox. HO 108 105° 101 96 km approx. ‘Vypical dimensions of the Western Coral Sea (to 155 FE): Tancar: N-S: 15 dep lat 900 nml 1700 km approx. PW: lO dep tong 600 nmi 1100 km approx. Reef area Deep ocean ‘Votal Area sx 10° km? 11.5 x 10° km? 15x 10° km? Sto ea! 11.5 x 10! ms 15 x 10'' m? Volume O.33x 10! im! 11.5 x 10'' m! 12x 10' m? cass. LOO m deep) (to LOOO m depth) 2. Speed: Mariners normally use one knot Lam hr. Oceanographers sometimes use knots but usually use |} cos for slower currents and |imes for faster ones. While these units are satisfactory for comparison of currents, they are not immediately related to the physical sizes ol seas or everyday units of time. Phe following approximate conversion factors may be convenient for reference: Wcms Okm day 60 kin week 260 km/ month 0.5 deg lat week 2.4 deg lat month Pkt S2ems hbkm day 310 kim week 1350 km month 3 deg lat week 12 deg lat/ month 3. Volume transport or flow: The usuabunitis LOS m Ss. Kor compactness this is called | sverdrup = 1 sv =1 x 10° mis. (Ref. Pickard, 1975), 4. Density and 0.3; specific volume and A: Phe density (2) ofa sample of seawater at atmospheric pressure is usually expressed in unitsofy em but occanographers normally use o,(sigma-t) defined as @, = (v9 — 1)x 10°. 1 pising em then o,isinmy cm*, However, it is usual to omit the units foro, and treat it as a pure number. (Strickly speaking this is correct because operationally the quantity measured by oceanographers and called p is really the specific gravity, Le. the ratio of the density of a seawater sample to that of pure water, which is a pure number. This fact is often forgotten, ) 128 An alternative quantity, specific volume z=1 p, is used for dynamic oceanography calculations and related to this is the thermosteric anomaly, A,, which is often used in place of o;. The two are directly related as: 23 24 25 26 27 27.2. 28 488 392 297 202 107 88 12clt. Thermosteric anomaly is usually stated with units of centilitres tonne (cl t) although it also is really non-dimensional. 5. Glossary: Cyclonic Isopleth Decibar (db) Dynamic height Geostrophic flow Mixing triangle method clockwise (in the southern hemisphere). a general term for a line joining points of equal value on a two- dimensional plot, e.g. a contour line. Used when there is no specific name (e.g. isotherm, isohaline) in common usc. unit of pressure, | bar =10 db=1000 mb=750 mm Hg. (In the sea, a pressure of 1 db is close to that exerted by a column of lm of sea water, so that the 100 db level is close to the 100 m depth etc.) really a unit of work, despite its name. If data from a number of oceanographic stations are available, the dynamic height of a given pressure surface may be plotted on a chart and contoured. Then (if friction is ignored) flow on that surface will be parallel to the contours of dynamic height and, in the southern hemisphere, the direction will be such that the greater dynamic height will be on the left when facing in the direction of flow. The statement ‘dynamic height (or topography) of the 150 db etc. surface relative to 1500 db’ (for example) implies that it is assumed that the water at the 1500 db pressure level is stationary so that the speed may be calculated at the 150 db etc. levels. For more detail, refer to a text on dynamic oceanography, c.g. Sverdrup et al. (1942), Neumann & Pierson (1966). literally ‘earth turned’; steady flow compatible with a balance between pressure forces due to the mass distribution and the (virtual) Coriolis force due to the rotation of the earth (and ignoring friction forces). — See, for example, Rochford (1959, 1960a). 129 References Mote These references relate primarily te the Phyo oceamopraphy of the cea, amd refer Disa cally alvhorgh nor ewe hisavely, te the area west ot 15 Podbnertheb oS ta the coustoat Austrabiaanad of Papuan Mow Cranned INSON, Food) & SADEPR, J 6 (b970) ad sradient deved comm cdnarts Ni Weather Service, US. Au TH charts NOI Aiea tatines een the Lapis Poree, St Panis, Mao ho pp ATIROSSE ALT, M fhe paavaty ool LPO as) SOULE TL Pompenture and spect Japan Ovos fpures. between acl Vestiatie Pransecriphion of Prov Soe NOS WE AP, tO od? AUIROSSTOVEL, M Vatieiun (1998). Surface tempernture of the older records trom fixed Seis, statrons Prey Soe NOS AR Oo POR tba AUIS TRAL EA DEE Ob TD EETONGE, NAVY OPT ET (hogy Fastralnan National Tide Tables, (970 Aust Plydrop Pubh vt AUIS TRAT EA Dread On NATIONAL. DEVEL OPMEN E1909), Recercatclavrala'y resomees (vtheame flare and tondersround resomrees F968 Canberra, COnpinal nat seen.) AUS TRALTTN per On NATIONAL DEVELOPMENT (L907) oiilay of clastratian Second Series, Surface vetiees Camnberna AUIS TRAE EN perl or NATIONAL, DEVI OPMENT, WAV EER RESOURCES COMING TE CLOT Streamer Ofori ony De 2909 Aust Crowt Publ Seeviee, Canberra, BOLAND POM & TIAMON, BV. (L970), Phe Past \usteati Current, P9605 L908, Deep Sea a ee | whiter Kesotireansy redter ve BRANDON, Py bo Chogoy Gaeat Barner Reet and the Queenshind con Herta shelly Atsteatin PhD) thesis, Univ. of Michigan, Ann Arbor 204 pp, BRANDON, Db (hardy. Waters of the Great Barrier Reet Provinces, Che Fopp. Ike 24d No Jones & ORO Tideany Bravo aad geodon of coral preety, Mab dy Creodawv 0, GNeademic Press, NY) RREETSTE AD MER AE PY (100) Vol Ti tydropraphie Dept, Lond. (See alse GRIEVE BRREPATN, MIN. OF DEERENCHE, HY DROGRAPTEIO DEPT, Loe yy BRT ETSED AD MER AT EY (19802), Staytradia Pilon, Vol TV Plydropeaphie Depry Tand. (See alsa: GREAT RREEATN, DETR OE DEEENGCH, TY DROGRAPTITO DEPP Log) Occunography of the stastralia Malor, 1a BROOK TTD, WH. ©, & HART, 1D. (1960). Raimfall im the tropical south-west Pacific, Dept. Geogr, Publ, G3 (1966), Res. Sch. Pac, Stud., Aust, Nat, Univ., Canberra, 25 pp., Mb fip., 2 tables HhOUD, PLE. (1952). Preliminary report) on yeology and marine environments of Outoa Atoll, Gilbert Istinds. “lof Res, Bull, No, 12,735 pp SIRO, AUSTRALIA (1951), Hydrographic and plinktological observations by BOR, V, Warreen in southecastern Australian waters, 1938 49, OQeeunon, Sta, Last, ISTRO, AUSTRALIA (1957), Surtice sampling in the ‘Pasman and Goral Seas and) the south eustern Tndiin Ocean, 1956, Oceanogr, Sta. Last sd. SSTRO, AUSTRALIA (1958). Australian ments prepared forthe Unesco conference on the oceanography of the Pasman and Coral Seas held at Cronulla, 9 14 August 1958, Dao. Mist Ocea now, Rep, 18, SSTRO, AUSTRALIA (19590), Surface sampling in the ‘Tasman and Goral Seas, 19597, Qeeanogr Sta, Lave 30, SSIRO, AUSTRALIA (19596). Oceanic in vestigations in castern Australian waters. TILM.A. Ships Queenborough, Quickmatch and Warregvo, L958, Qeeanonn, Sta, Lase 43, ISTRO, AUSTTRALTA (1959¢), Scientific reports of wecruise on FLMLALS, Queenborough and Quick- matelt, L958. Oceanogr, Rep, 24. ISERO, AUSTRALIA (1959), Scientific reports oferuises on FLM.A.S, Warreyo, 1957 and 1958. Oceanogr Rep. 25, SSERO, AUSTRALIA (1900), Surtace sampling in the Coraland ‘Tasman Seas, 1958, Oceanogr. Sta, List 39, SSERO, AUSTRALIA (1961a), Oceanic in vestigations in-castern Australian waters, RLV. Derwent Hrnter, 1959, Oceanogr, Sta, Lasts, SIRO, AUSTTRAL TA (L90Tb), Coastal hydrologi- cal sampling at Rottnest Tslind, W.A. and Port Moresby, Papua during the EGY, (1958 59), and surface sampling inthe Tasman and Coral Seas, 1959, Qeeanoun, Sta, Last dO, SSTRO, AUSTRALIA (l9Q2Q0), Surface sampling nthe ‘Tasman and Coral Seas, 1860, Oceanogr, Nea. List 50 ISTRO, AUSTRALIA (19026), Oceanographical observations am the Pacific Ocean in 1800, TE MAAS. Gascovne, Cruises GE 60 and G2 60, Oveanoyr, Cruise Rep, 9. docu CSIRO, AUSTRALIA (1963). Oceanographieal observations in the Pacitic Ovean tm 196, H.M.A,LS, Guseavwe, Graise G1) Ol, Oceanogr Craise Rep. 3. CSIRO, AUSTRALIA (19674). Oceanographical observations in the Pacitie Oceun in L9od. H.M.A.S, Gascoyne, Cruise G3 61. OQveanoyr. Crise Rep. 12. CSIRO, AUSTRALIA (1967b). Oceanographical observations in the Pacilic Ocean, PLM.ALS Gascoyne, Cruise G1 02, Qveanogr. Crise Rep. 13. CSIRO, AUSTRALIA (1968a). Investigations by FB. oY. Degei in Queensland. waters in 1965, Oceanogr. Sta. List 73. CSIRO, AUSTRALIA (1968b), Occanographical observations in the Pacific Ocean in 1963. H.M.A.S. Gascoyne, Cruise G3 63, Oceanogr. Cruise Rep. 26. CSIRO, AUSTRALIA (1968e). Oceunographical observations in the Pacific Ocean in 1905. H.M.A,S, Gascoyne, Cruise GO 65, Oceana. Cruise Rep. 48. CSIRO, AUSTRALIA (1974). Monthly oceano- graphic charts, Tasman and Coral Seas (from 1966 on). Div, Fish. Oceanogr,, Cronulla, (Una- publ.) DONGUY, J. R. & HENIN, ©. (19750). Surface waters in the north of the Coral Sea. «lust, 7, Mar, Mreshw. Res. 26, 293 296. DONGUY, J. R. & HENIN, C. (1975b). Evidence of the South ‘Tropical Counter Current in the Coral Sea, Austr. J. Mar. Freshw. Res. 26, 405-409, DONGUY, J. R., HENIN, C. & ROUGERIL, I (1974a), Surface water in the northern Coral Sea, (Abst.) p. 7 in BL R. Stanton (Hd.), Proc, Reg. Workshop on ‘Circulation Studies in the South- West Pacific.” Wellington, 197-4. DONGUY, J. R., HENIN, C. & ROUGERTE, V, (1974b). Surface water exchanges between the Coral Sea and the Pacitic Ocean, (Abst.) pp. 8,9 in B. R. Stanton (Ed.), Proc. Reg. Workshop on ‘Circulation Studies in the South-West Pacific.” Wellington, 1974. DONGUY, J. R., HENIN, C., ROUGERIE, b & LE CORVAISIER, A, (19724). Surface con- ditions in the South ‘Tropical Pacific from Australia to Polynesia. p. 7l oa Re Vraser (Comp.), Oceanography of the South Pacifte 1972,N.Z. Nat. Comm. for Unesco, Wellington, 524 pp. DONGUY, J. R., HENIN, C., ROUGERIE, 1, LABOUTE, P. & DE GEOFTROY, B, (1972b). Résultats des observations physico-chimique de la croisi¢re ‘Gorgone |’ du N.O. Coriolis. Centre O.R.S.T.O.M. de Nouméa, Sér. occanoge. (Un- published data report.) (U.S.N.O.D.C. 350077). DONGUY, J. R., OUDOT, CG. & ROUGERIE, (1970). Circulation superficiclle cl subsuper- ficielle en Mer du Corail et a 1700, Gat, O.R.S.T.O.M, Sev, ockanagr. 8&4 20, EASTON, A. K. (1970). The rides of the continent of Australia, Res. Pap. 37, Horace Lamb Centre oceanogr, Res., p- 320. ENDEAN, RB, KUNNY. ROSS PEPETEN SON, W (L950a). Vhe ecology ant distetbution ot ia tertidal organisums on the toeky shores of the Queenshind mainland, elo J. Ve La esin Rey. 7,88 1 bo ENDIEAN, R. STEPHENSON, WS KRENNYSR (1956b). ‘Phe ecology and distribution of it tertidal organisms on cerbun osbinds off the Queensland coast, dase Fo Mar, bred, Nes, 7 WT M2. VAR SEAS FISTIERTES RESEARCH LABORATORY, JAPAN (1973). Suurion data. Shoyo Maru ertise, 2 Over L973 7 Jan. Loe Bish. Agency of [:ipan. GALLAGHER, B.S, SHIMADA, IK. MM. GONZALEZ, Foo, JR & STROUP, ED (L071. ‘Vides and currents in Banning Atoll Lagoon. Pact, Ser 25, 191 205. GARDNER, J. V. (1970), Subrmaring geology af the western Coral Sea. Geol. Sue. clue. Bull, ST, 2599 261-4, GODEREY, J. S.(1973a). On the dynamics of the Western boundary in Bryan and Cox's 1 1008 numerical Deep-Sea Key. 20, 1043 1058- GODEREY, J. S.(1973b). Comparison of the Hast Australian Current with the western boundary flow in Bry:in & Cox's (L968) numerical model ocean, Deep-Sea Res. 20, 1059 1070, GODEREY, J. 8. & ROBINSON, A. R.( 1971). Phe East Australian Current as a free inertial jet. 7. model ocean. GREAT BRITAIN. MIN. OF DEFENCE, HYDROGRAPIHG DEPT (L973). clerralia Pilot, Vol, TUL. (Replaces elsrralia Pilot, Vol. LIL, 1960 and Vol. IV, 1962), HAMON, B. V- (1965). Phe Hast Australian Gur rent, 1960. 64, Deep-Sea Res. 12, 899 92. HAMON, B, V. (1908), “Pemperature structure in the upper 250 metres in the t Australian Current area. elise. J Mar. Preshee. Res, 19, 9] 99. HAMON, B, V. (1970), Western boundary currents in the South Pacitic, pp. 50 59 a7 WS. Wooster (Ed.) Scientific exploration of the Sonth Pacific, Nat. Acad. Sci., Wash., D.C. HAMON, B. V, (1973). The East Australian Cur- rent. pp. 35 3007 R, Braser (Comp.) Oeeanon- raphy of the South Pacific, 1972, NZ. Nat. Comm. for Unesco, Wellington, 524 pp. HAMON, B. V. (1974). Recent work in physical oceanography in the Coraland ‘Vasman Seas. pp. 1b 17 iv BL OR. Stanton (Hd), Proc. Reg Workshop on ‘Circulation studies in the South West Pacific’, Wellington, L974. HAMON,B. V.& GRIEG, M., A. (1972), Mean sea- levelin relation wo geodeue hind levelling around Australia. 7. geoph. Res. 77, 7157 7162, HAMON, B. V. & STACKY, B.D. (1900). Sea levels around Australia during the International Geophysical Year. sr. 7, Mar, Freshaw. Res, (1, 269-281. HEDLEY, GC. (19254). "Phe natiteal destruction of a coral reef. Rep. Gr. Bare. Reef Cree, 1, 3540, 131 HBDILRY Cl) 1925b. An opacity meter. Rep. Gr Bure. Reef Crrer 1, G7 68. HEDLEY, ©. 1925e) “The surfice temperature of Maorctoir Bry Barr, Reef Cttee. 1, Ly 150 HIGLEY, FO 1907). Qevaiite cinentanian parterns ol te east codstof clastrafia, CSIRO Aust, Div, hish, Oceanogr, Peeh, Pap. 24 Rep. Gir, HOGAN, J, (1925). Record of termperatures obser- vedal Willis Ushiod daring the evclone season of PO22 23. Rep. Gir Barr. Reef Crtee, 1b) 46, INTERNATIONAL HYDROGRAPHIC BUREAU (1957). Lauuty of veces anid seas. Tt. Lydrogr, Bur, Spee. Publ. 23, 2nd Bdn., 25 pp. JAPAN PISTIERELS AGENCY (1957). EOQUAPAC. ocenhoytaphic and meteorological data. SHiove Alar cruise, 20 Aug. 3 Nov. 19560, (Onganal net seen, | JAPAN HYDROGRAPHIC OPTIC (1958), HOUAPAC oceanographic and meteorological data (preliminary report). Sarid, 25 July 29 Aug. 1956, (Orginal aor seen, ) [ARRIGH, EL, RADORK, R., KRAUSE, GG. & RUAL., P. (1975). Courants dans le lagon de Noumen (Nouvelle-Caledonie). (Unpubl. MSS.) OLR.SSTLO.M, Noumea, Juillet 1975, JOLINSON, RoE. (1973). Antarctic Intermediate water i the South Paciic Ocean. pp. 56 6977 R. Vraser (Comp,) Oceanography: of the South Pacific 1972. NZ. Nat. Comm. tor Unesco, Wellington, 524 pp. KENNY, R. (1974). Inshore sea surface tempera- tures at Townsville, last, J Mar. resize. Res. 28,1 5. KILONSRKY, Bo J. & RAMAGE, ©. S, (1975), A technique for estimating tropical open-ocean ramtall from satellite observation. Dept. of Met., Univ. Hawai, Contr. 75-b4(Unpubl. MSS.) KIMBALL, LL 11.(1928), Amount of solar radiation that reaches the surface of the earth on land and on the sea, and methods by which itis measured. Mon. Weath, Rew, 56, 393-449, KINSEY, DW. & KINSEY, B. E. (1967), Diurnal changes in oxygen content of the water over the coral reef platform at Heron Is. dust, 7, Mar. Mreshiw, Res. 18, 23-44, LEGAND, M. & RO'TSCHI, H. (1962). Bilan des recherches oceanographiques frangaises en Mer du Corail entre 1956 et 1961, Cah, ORS TOM, Ser, oceanogr, 14, 703-718, LOCKERMAN,R. CG, & SCULLY-POWER, P. D. (1969). Temperature-salinity velarionships and associated geostraphie etreulation and sotnd velocity calculations for tro seasons in the Western Coral Sea, Roy, Aust. nav, res. Lab. ‘Vech. Note 1 68. MAGNIER, Y. & PITON, B. (1972), La circulation en Baie d’Ampasindava (Madagascar) et ses implications biochimiques, Cah, O.R.S.T.O.M. Ser. oceanogr. X, 75-97, MAMMARICRX, J,, CHASE, ‘Tl, E., SMITH, S. M. & VAYLOR, I. L.. (1974). Bathymetry of the South Pacific, Sheer 11, revised to 1974, Inst. mar. Resour, tech, Rep, 44A, Scripps Inst, Oceanogr., Calif. 132 MAXWELL, Wo G. EL (1908), Atlas of the Great Barrier Ree}. (Lilsevier, N.Y), 258 pp. METEOROLOGICAL OFBICE (1949). Monthly sea strface temperatures af clustralian and New Zealand waters, Marine Branch, Met, Of, Air Min., Pond. MICHIEL, A., COLIN, C., DESROSIERS, R. & OUDOT, ©. (1971), Observations sur Phydrologie et le plancton des abords et de la zone des passes de Patoll de Rangiroa (Arehipel des ‘Puamoto, Océan Pacifique Central), Cut. ORS. OAL, Ser. octanogr., EX, 375 402. MONTGOMERY, R. B. (1938), Circulation in the upper layers of the southern North Atlantic deduced with the use of isentropic analysis. Pup. ples. Occanogr, 6, 155, MONTGOMERY, R. B. 11955), Characteristics of surface waters at Weather Ship J, Deep-Sea Res. Suppl. Vol. 3, 331 434. MOORHOUSE, EF. W. (1933). Phe temperature of the waters in the anchorage, Low Isles, Ser. Rep. Gp. Barrier Reef Hxped. 1928 29, Br. Mus. | Nat. Hist.), (1 0-4),98 101, NEAL, A. B. (1975). Some aspects of the south-east trades of the Coral Sea (MSS of presentation to 1th Amer. Met. Soc. Vech. Conf, on Hurricanes and ‘Tropical Meteorology, Florida, May, 1975). Res. & Dev. Branch, Bur, of Met., Melbourne. NEUMANN, G. & PIERSON, W. J. (1966), Principles of physical oceanography. \Prentice- Hall, NoJ.'. NEWELL, Bo 8, (1971). The livdrological environment of Moreton Bav, Queensland, 1967 68. CSIRO Aust., Div. Fish. Oceanogr. ‘Tech, Pap. 30. NEWELL, B.S. (1973). Aydrology of the Gulf of Carpentaria, 1970. 71, CSIRO Aust., Diy. Fish. Oceanogr. ‘Tech, Pap. 35. ORR, A. P. (1933a). Physical and chemical conditions in the sea in the neighbourhood of the Great Barrier Reet. Ser. Rep, Gr. Barricry Reef Exped. 1928-29. Br. Mus. | Nat. Hist.) If (3), 37-80. ORR, A. P. (1933b). Variations in some physical and chemical conditions on and near Low Isles Reef. Set. Rep. Gr, Barvier Reef lixped, 1928-29. Br, Mus, ( Nat. Hist.), 11 (4), 87-98, PICKARD, G. L. (1961). Oceanographic features of inlets in the British Columbia mainland coast. 7, Fish. Res. Bd. Can, 18, 907-999. PICKARD, G. L, (1975), Deseriptive plivgicul oceanography, Second Edn. (Pergamon, Oxford.) 214 pp. PITON, B. & MAGNIER, Y, (1971). Les regimes hydrologiques de la Baie d’Ambaro (nord-ouest de Madagascar), Cah. O.R.S.T.O.M,, Ser, oceanogr. IX, 149-166. RAMAGE, C. S. (1970). Meteorology of the South Pacific tropical and middle latitudes, pp. 16-29 in W. S. Wooster (Ed.) Scientific exploration of the South Pacific, Nat. Acad. Sci., Wash,, D.C. REID, J. L. (1961), On the geostrophic flow at the surface of the Pacific Ocean with respect to the 1000 decibar surface. Te//us 13, 489-502. REID, J. L. (1965), Lnrermediare waters of the Pacific Ocean. Johns Hopk, oceanogr. Stud, No. 2, 85 pp. ROCHFORD, D, J. (1957). The identification and nomenclature of the surface water masses in the ‘Tasman Sea (data to the end of 1954). lyst. 7. Mar. Freshi. Res. 8, 369-413. ROCHFORD, D. J. (1958a). Total phosphorus as a means of identifying cast Australian water masses. Deep-Seu Res. 5, 89-110. ROCHFORD, D. J.(1958b), Thre seasonal clrenlarion of the surface water masses of the Tasmait arnt Coral Seas. CSIRO Aust., Div. Fish. Oceanogr. Rep. 16. ROCHFORD, D. J. (1959), The primary exrernul carer masses of the Tasman and Coral Seas CSIRO Aust., Div. Fish, Oceanogr, ‘Pech. Pap, de ROCHFORD, D. J. (1960a). The intermediate depth waters of the Tasman and Coral Seas. 1. The 27.20 a surface. Aner. J. Mar. Fresfim. Res. 11, 127-147. ROCHFORD, D. J. (1960b). The intermediate depth waters of the Tasman and Coral Seas. IL. The 20.80 a, surface. Aust, J. Mar. Freshz. Res. 11, 148-165. ROCHFORD, D. J. (1960c). Some aspeets of the deep circulation of the Tasman and Coral Seas. Aust... Mar. Freshw, Res. 11, 166-181. ROCHFORD, D. J. (1962). The intermediate depth waters of the Tasman and Coral Seas. IIT. Succession of water types east of Port Hacking in 1957-59. Aust. J. Mar, Freshiw. Res, 13, 61-88, ROCHFORD, D, J. (1966). Some hydrological features of the eastern Arafura Sea and the Gulf of Carpentaria in August 1964. sTis7. 9. Mar. Freshw. Res. 17, 31-60. ROCHFORD, D. J. (1968a). The continuity of water masses along the western boundary of the ‘Tasman and Coral Seas. Aust, 7. Mar. Preshiw. Res. 19, 77-90. ROCHFORD, D. J. (1968b). Origin and circulation of water types on the 26,00 sigma-t surface of the southwest Pacific. Aust, 7, Mar. Freshw. Res. 19, 107-127. ROCHFORD, D. J. (1969). Origin and circulation of water types on the 25.00 sigma-t surface of the southwest Pacific. Aust. #.. Mar. Freshw. Res. 20, 105-114, ROCHFORD, D., J. (1973), Patterns of change in surface salinity and temperature of the Tasman and Coral Seas, 1966-71, pp. 105-112 m7 R. Fraser (Comp.) Oceanography vf the South Pacific 1972, N.Z. Nat. Comm. for Unesco, Wellington, 524 pp. ROCHFORD, D. J. (1974). Some preliminary results of the Tasman and Coral Seas ships of opportunity sampling program of the CSIRO Division of Fisheries and Oceanography, pp. 12-13 in B, R. Stanton (Ed,) Proc. Reg. Work- shop on ‘Circulation studies in the South-West Pacific’. Wellington, 1974. ROTSCHI, H, (1958a), Orsom III. Rapport tech- nique de la croisiére 56-5. Oceanographie physi- que, O.R.S.T.O.M., L.,0. Rapp. Set. 9. RO'VSCHI, H.1958b), Orson TT, Resultats de la croisiere ‘Astrolabe’, lere pattie, Qecunographie physique. O.R.SJE.O.M,. LO, Rupp Serb ROVSCHI, 1f.)1959a). Hydrologic et dyniqiie duoonmordest de la Mer du Coral Gaul, ORS LOWE, Ser oceanoun, VL, 727 TF 90). ROUSCHI, fl. 61959b). Orson) 77. Resulmaits de la croisiere ‘Boussole’, Jere partic. Occunugraphie physique. O,R.S-T.O.M., J2.0. Rupp. Sit 12. ROTSCHI, H. (1960%. Recents progres des recherches occunographiques entreprises dans le Pacifique Sud-Ouest. Cul, OLR SLT OLAL, Ser ovcunour, NIL, 249 207. ROTSCHI, TL. (1960b). Sur Ta virculation ct les masses Weau dans le nord-est de la Mer du Corail. GR. elcud. Sei., Paris, 251, 903-907. ROTSCHI, UH. (19600), Orson ITE. Resultats de la croisiere “Choiscul’, lere partic, Oceanographic physique. O.R.S/T.OLM., 2J6O. Rapp. Set. 15. ROTSCHIL, EL. 1960d), Qryon TET. Resultits de ta croisi¢re ‘Dillon’, lére partic, Oceanographic phypique. O.R.S.T.O.M., £160 Rupp. Set. 18. ROUSCHI, H, (196la), Contribution franguis en 1060) a la connaissunce de la Mer du Corail. Oceanographic physique. Cal. ORSON, Sér, accuivgr, NTT. 434-453- ROTSCHL, H. (1901b}, Résultats des observations scientifique de “Viare’ croisiere Lirrecasteais, O.R.S.T.O.M,, 2.1.0. Rapp. Sct. 21. ROTSCHI, H. (1961e). Orsou TH. Resultats de ke croisiere “Epi, lere partic. Oceanographic physique. Q.R.S.T.O.M., L200. Rupp. Set. 22, ROTSCHI, H. (1961d). Influence de la divergence des Salomon sur la répurtition de certaines proprictés des eaux, CLR, lead. Sei. Parts, 253, 2539 2561, ROUSCHI, H. (19624). Sur la divergence des Sulomon, CLR. lead. Sci. Paris, 255, 2799 2797 ROTSCHI, H. (1962b). Oxygene, phosphate et gaz carbonique total en Mer du Cotail. Deep-Sea Res, & 181-185, RO'SCHI, H, | 1963a). Sur les flux eau entre la Nouvelle-Calédonie et Varchipel des Salomon. CLR. Acad, Set., Paris, 256, 2009-2012. ROTSCHI, H,. & LEMASSON, L, (1967), Oceanography of the Coral and ‘Tasman Seas. Ann. Rev. Oceanogr. mar, Biol, 3, 49-97. ROUGERIE, F, & DONGUY, J. R, (1975). La mer du Corail en regime Walizeé de sud-est, Call. O.R.S.T.O.M. Sér- océanogr. XII, 49 67. ROYAL AUSTRALIAN AIR FORGE (1942), Wearher on the Australia — stutfon, local information. Publ. 252, Vol. T1, Melbourne. ROYAL NETHERLANDS METEORO- LOGICAL INSTITUTE (1949). Seca areas around “lustralia, Oceanogr, met. Duta 124, SCHOTT, G. (1935). Geograpltic des Indischen nud Srillen Oseans, (C. Boysen, Hamburg, } SCULLY-POWER, P. 1). (1969a). Oceanographic crue: Coral Sea, Noveniber-Deceniber 1966, Roy, Aust. nay, res. Lab, Tech. Memo, 14 69, SCULLY-POWER, P, D. (1969b). Oceanographic cruises Coral Sea, ipril-May 1967, Roy. Aust. nay, res. Lab, Tech. Memo, 15 69: 133 SCULEY-POWER, P.DL 197 3a), Oceanography of ta Coral Seas Pheiwinter rginne, Roy. Aust. nay. tes. Lab. Pech, Note 13 73. SCULEY POWER, POD. (1973b). Ovcinography ofthe Coral Sea; the winter regame, pp. 129 133 WOR. Praser (Comp.) Qecanography af the South Pacific 1972. NZ. Nat. Comm. tor Uneseo, Wellington, 26 pp. SCULLY POWHR, PLD. (L973e), Coral Sea flow budgets in winter, “ler J. Mar, Prestie. Res, 24, 204 21% SCULLY POWER, PL oD, & FRANCE, J. ©. (1969a), Oceanographic cruises Goral Sea, May 1968. Roy. Aust. nave res. Lab. “Pech, Memo, (Ext) 2°09. SCULLY-POWER, PL oD. & PRANCH, J.C. (19096). Occanovraplic crises Coral Sea, June 1968. Roy, Aust. nav. res. Lab. ‘Vech, Memo, (Ext.) 3.09 SCULELY-POWER, P, Do & PRANCE, J. C. (19096), Qceunovraplie cruises Coral Sea, July 1968, Roy. Aust. nay, res. Lab. ‘Tech, Memo (hixt.) 4.69, SVANTON, BR. (1975). Bibliography of Uwe pliysical oceanography of the Tasman and Goral Seas ta 1974. Mise. Publ. 06, N,Z. oceanogr. Inst, Wellington, 36 pp. SVODDART, 1. Re Ed.) (1906), Reef stares ar eld toll, Maldive Islands. Atoll Res, Bull, No. 116, 122 pp. STODDART, DR. id.) (1975), Phe Great Barrier Reefs a preliminary bibliography, Dept. Geog. Cambr, Univ, (in press). SVERDRUP, Th, JOHNSON, M. W, & BLEMING, R. 1. (1942), The oceans, their phusics, chemistry and general biology, (Prentice- Hall, N.J.) L060 pp. TAKAHASHI, ‘V. (1959).) Hydrographical re- seurches in the western equatorial Pacific, Mom. Fae. Mish, Kagoshima, 7, 141 147. TAKAHASHI, ‘V. (1960). Existence of a contra salem vortical motion in the Coral Sea, Rec. peeanogr, Whs, Jap. 5,92 54. TAYLOR, R. ©. (1973). An atlas of Pacific islands rainfall, Hawaii Inst. Geophys. HIG-73-9, (un- publ. MSS) TRAN'PER, D. J. (1958). Reproduction in Aus- tralim pearl oysters (Lamellibranchia). Aust. 7. Mar, Mresh. Res, 9, 509-574. "TSUCHIYA, M. (1968). Upper waters of the in- revivopieal Pacific Ocean, Johns Hopk. oceanogr. Stud. No. ++, 50 pp. U.S. NAVY HYDROGRAPHIC OFFICE (1955), lias of surface currents South-western Pacific Qevan. TO, Publ, 508, Wash. D.C, (reprinted L968), U.S. NAVAL, OCKANOGRAPHIC OFFICE (1955). ettlas of Pilot Charts, South Pacific and Indian Oceans. U.S.N.O,O, Publ, 107, Wash., D.C, ULS.S.R. (1950), Morskot Atay, 1, Navigatsionni- Creograficheskii,, Voennomorskoe Ministerstvo, RY sheets. 134 VAUX, D. (1970). Snrface temperature and salinity Jor Australian waters, 1961-65. CSIRO Aust., Diy. Fish, Oceanogr. Alas No, 1. VON ARX, W. S, (1948). The circulation systems of Bikini and Rongelap Lagoons. J'rans. Amer. nzeophys, Un, 29, 861-870. WEBB, D. J. (1973a). Tidal resonance in the Coral Sea. Narure 243, 511. WEBB, D. J. (1973b). On the age of the semi-diurnal tide. Deep-Sea Res. 20, 847-852. WOOD, J. E. (1969). Oceanographic cruise Coral Sea, Arafura Sea and Java Trench, April-May, 1969, Roy. Aust, nav, res, Lab. Tech. Memo. (Ext.) 12/69. WOOD, J. E. (197 1a), Oceanographic survey: Coral Sea, April-Mav 1970. Roy. Aust, nav, res. Lab. ‘Tech, Memo, 21/71, WOOD, J. E. (1971b), Oceanographic survey; Coral Sea, August-September 1970. Roy. Aust. nav. res. Lab. ‘Tech. Memo, 22/71. WOODHEAD, P. M. J. (1968). Sea surface drift between central Queensland and New Zealand. Aust. J. Sci, 31, 195-196, WOODHEAD, P. M. J. (1970). Sea-surface circu- lation in the southern region of the Great Barrier Reef, Spring 1966. Aust. J. Mar. Freshw. Res. 21, 89-102. WUST, G, (1935), Die strarosphare, Wiss, Ergebn. dtsch. atlant. Exped. Meteor., Bd. 6, 1 Tiel, 2 Lief, 288 pp. WYRTKI, K, (1960). Surface cireulation in the Coral and Tasman Seas. CSIRO Aust. Div. Fish. Oceanogr. ‘Tech. Pap, 8. WYRTKI, K. (1961a). The flow of water into the deep sca basins of the western South Pacific Ocean, Aust. J. Mar, Freshw. Res, 12, 1-16. WYRTKI, K. (19616), Physical oceanography of south-east’ Asian waters. Scientific results of marine investigations of the south China Sea and the Gulf of Thailand 1959-61. NAGA Rep. 2 (Univ. Calif., Berkeley) 195 pp. WYRTRKI, K, (1962a). The subsurtace water masses in the western South Pacific Ocean. Aust. F. Mar. Freshw. Res, 13, 18-47. WYRTKI, K, (1962b), Geopotential topographies and associated circulation in the western South Pacific Ocean, Aust. J. Mar. Freshw. Res, 13, 80-105, WYRTKI, K. (1962c). The oxygen minima in relation to ocean circulation. Deep-Sea Res. 9, 11-23. WYRTKL, K. (1967). Circulation and water masses in the eastern equatorial Pacific Ocean. Int. J. Qceanol. Limnol, 1, 117-147. WYRTREI, K. (1974). The dynamic topography of the Pacific Ocean and its fluctuations. Hawaii Inst, Geophys. HIG-74—5 (unpubl. MSS). WYRTKI, K, & MEYERS, G. (1975), The trade wind field over the Pacific Ocean. Part 1. The mean field and the mean annual variation. Hawaii Inst. Geophys. HIG-75-1, (anpubl. MSS). Some sources of aerial photographs of the Great Barrier Reef region (Dr J. Veron, personal com- munication): 1. Central Mapping Authority, Brisbane & Sydney 2. Commonwealth Department of Services & Pro- perty, Brisbane 3. Commonwealth Aerial Photography, Division of National Mapping, Department of National Resources, Canberra 4. RAAF Base, Laverton, Victoria R76/597 Cat. no. 76 3960 1