Distribution of post-weaning Antarctic fur seal Arctocephalus gazella pups at South Georgia

ORIGINAL PAPER

N.L. Warren Æ P.N. Trathan Æ J. Forcada Æ A. Fleming

M.J. Jessopp

Distribution of post-weaning Antarctic fur seal Arctocephalus gazellapups at South Georgia

Received: 27 January 2005 / Revised: 8 June 2005 / Accepted: 13 June 2005 / Published online: 25 August 2005� Springer-Verlag 2005

Abstract Potentially some of the biggest gaps in ourknowledge about the ecology of Antarctic fur seals(Arctocephalus gazella) relate to juvenile animals. Weinvestigated the at-sea distribution of five male and fivefemale fur seal pups post-weaning. The study was car-ried out at Bird Island, South Georgia during two suc-cessive winters using satellite-linked platformtransmitter terminals (PTTs). Our results are analysed inrelation to pup sex and the physical environment andproductivity of those 2 years, as well as in the context ofour present knowledge of where post-breeding femalesand males forage. The available physical and biologicaldata during both of the winters of this study suggest thatboth years were not unusual. We report marked differ-ences between the sexes with male pups foraging sig-nificantly further away from land and their birth sitethan do females. The pups foraged in areas to the Eastof Bird Island seldom reported as foraging areas for theadult population. Also as winter progressed they showeda more oceanic distribution leaving the continental shelf,possibly to exploit a different prey source that was morereadily available in the upper water column.

Introduction

The Antarctic fur seal (Arctocephalus gazella) popula-tion at South Georgia has recovered from the brink ofextinction over the course of the twentieth century.

Initial indications of recovery started with the discoveryof a few small, isolated breeding colonies in the 1930s;these had increased to an estimated population of over1.5 million by the early 1990s (Boyd 1993). Today, thepopulation is thought to be in excess of 3 million (Bar-low and Croxall 2002).

A recent estimate of food consumption by Antarcticfur seals at South Georgia, assuming a diet mainly ofkrill (Reid and Arnould 1996), suggests that this popu-lation consumes approximately 3.84 million tonnes ofkrill annually (Boyd 2002a), based on the 1991 popula-tion estimates. This removal of krill from the SouthGeorgia region potentially brings fur seals into directcompetition with other species and with man. Compet-itive interactions with sympatric species are thought tobe increasingly important and certainly are not negligi-ble, especially for other krill specialists such as macaronipenguins (Barlow and Croxall 2002). Similarly, interac-tions between fishermen and seals, either through com-petition for resources or through incidental mortality,are increasingly recognised as an important issue inecosystem management; both entanglement in fishinggear (Arnould and Croxall 1995) and by-catch (Hooperet al. 2004) are now known to occur in South Georgiawaters.

Consequently, as the population of Antarctic furseals continues to expand and as interactions with othermarine predators and with man intensify, it is increas-ingly important to obtain information about the diet,feeding behaviour and foraging areas of these top pre-dators. In the Southern Ocean, such information isconsidered to be critical to the management process(Agnew 1997); this is because Southern Ocean fisheries,including that for Antarctic krill (a major constituent ofthe diet of Antarctic fur seal), are managed using eco-system-based methods (http://www.ccamlr.org/pu/e/pubs/am/toc.htm; CCAMLR 2002).

Many studies have investigated the at-sea distributionof Antarctic fur seals throughout their complete cir-cumpolar range (e.g. Staniland et al. 2004; Boyd et al.2002, 1998 in the Southern Atlantic; Bonadonna et al.

N.L. Warren Æ P.N. Trathan (&) Æ J. Forcada Æ A. FlemingM.J. JessoppBritish Antarctic Survey, Natural Environment Research Council,Madingley Road, Cambridge, CB3 0ET UKE-mail: [email protected].: +44-1223-221602Fax: +44-1223-221259

Present address: M.J. JessoppDepartment of Zoology, Ecology and Plant Science, UniversityCollege Cork, Lee Maltings, Cork, Ireland

Polar Biol (2006) 29: 179–188DOI 10.1007/s00300-005-0037-x

2001; Guinet et al. 2001 in the Indian Ocean; Robinsonet al. 2002 in the Pacific). However, most studies haveconcentrated on breeding females and very little workhas been carried out on males or on other demographiccategories.

Several studies on the foraging distribution of marinemammals have observed marked differences betweensexes (Boyd et al. 1998; Stewart 1997; Kovacs et al.1990, Hindell et al. 1991; LeBoeuf et al. 1993; Stewartand DeLong 1993) and age classes (Merrick and Loug-hin 1997; McConnell et al. 2002) of pinnipeds. Thesestudies highlight that conclusions about feeding behav-iour and foraging distribution could potentially bebiased if intra-specific differences are ignored. Theimportance of gaining a better understanding of fur sealdistribution across different age classes and sexes istherefore essential.

Potentially one of the biggest gaps in our knowledgeis that for juvenile animals. Very little is known aboutjuveniles after they leave their birth site until they returnto breed as 3-year olds (females) or 7-year olds (males).Consequently, in this study, we investigated the at-seadistribution of Antarctic fur seal pups in the periodimmediately after they were weaned. The study wascarried out at Bird Island, South Georgia during twosuccessive winters using satellite-linked platform trans-mitter terminals (PTTs). Our results are analysed inrelation to the sex of the pup and the physical environ-ment (bathymetry, sea surface temperature) and ameasure of primary productivity (SeaWiFS), as well asin the context of our present knowledge of where post-breeding females forage.

Materials and methods

Experimental design

Platform transmitter terminal are expensive both interms of the cost of the instruments and their operations(Boyd 2002b). Our experimental design was therefore acompromise between sample size and our intendedprotocol. Consequently to reduce variability associatedwith gender and inter-annual environmental differenceswe deployed tags on one gender only in each of ourstudy years, but were careful to monitor environmentalconditions in both years.

The study period extended from April until earlySeptember 2001 for males and from April to lateDecember 2002 for females. When comparing the maleand female distributions, we have only considered themonths in which data were available for both sexes, thusthe period of interest ranged from the start of April untilthe end of August in both years.

Deployments

The study was carried out from Freshwater Beach, BirdIsland, South Georgia (54�00¢S; 38�02¢W). Prior to

weaning and departure from the breeding beaches(Doidge et al. 1986), five male and five female pups werechosen randomly in April 2001 and 2002, respectively.The pups were caught using a well-established restrain-ing method (Boyd et al. 1998). Weight measurementswere collected and each individual was given a uniqueidentification marker with numbered tags (Dalton Sup-plies, Henley-on-Thames, UK). The 180 g satellite-linked platform transmitter terminals (PTTs; Model ST�18 25% duty cycled; Telonics, USA, packaged bySirtrack, New Zealand) were deployed to the fur of eachstudy animal in the mid-dorsal region between thescapulae using epoxy glue (Boyd et al. 1998). Total de-vice mass always represented less than 1.5% of the furseal body mass. The animals were subsequently released.No handling interval lasted more than 30 min

Data handling

The resulting Argos uplinks and locations were filteredto remove potentially unreliable records. Locationsdetermined by two or fewer uplinks were removed (allB and Z Classes). The remaining uplinks (Classes 3, 2, 1,0 and A) were deemed to be reliable (Vincent et al. 2002)and were further filtered by identifying fixes that wouldrequire an unrealistic rate of travel. The ‘maximumspeed parameter’ in the filter was set to 2.0 m s�1 (Boyd1996). The data were then plotted and analysed usingESRI ArcMap 8.3. The point distributions were trans-formed into density estimates using Kernel-basedmethods.

Using the coordinates of two points, the distancebetween these two were calculated using the great circlearc which is the shortest distance between two points ona spherical surface (Maling 1992).

Environment

The physical environment probably drives a large pro-portion of the observed biological variability at SouthGeorgia (Trathan et al. 2005); we thus compared thedifferences between the two study years in terms of SeaSurface Temperatures (SST). We also compared thebiological environment during the two years in terms ofsurface chlorophyll-a. These two environmental vari-ables were used to describe the conditions that the furseals were experiencing during both years. Data outputsfrom these analyses include monthly SST grids in unitsof degrees Celsius (�C) with an approximate spatialresolution of 4 km.

The SeaWiFS-derived estimates of surface chloro-phyll-a concentration were obtained from the GoddardDistributed Active Archive Centre using the standardOC4v4 chlorophyll algorithm, described by O’Reillyet al. (2001). The grids were in units of mg m�3 and hadan approximate spatial resolution of 9 km.

The area for the analysis of the SST and SeaWiFSaround South Georgia was set as 49–57� South and

180

20–44� West to incorporate the overall distribution ofthe fur seals during both winters.

Data used to trace the Antarctic polar front and theSouthern Antarctic circumpolar current, one of the fastmoving jets of the Antarctic Circumpolar Currents weretaken from Moore et al. (1999), Orsi et al. (1995) andTrathan et al. (1997).

To determine the extent of the differences betweenyears for these oceanographic variables, we analyseddifferences between the means of each month and yearby regression analysis. The same method was used forboth the SeaWiFS and SST data.

Five different models were fitted to the data usingSplus (Mathsoft Inc. Cambridge MA, USA):

– A simple linear model where SeaWiFS or SST are afunction of month, year and the interaction of monthon year.

– The same model without the interaction of month onyear.

– Two linear models with separate intercept.– Two linear models with a separate intercept and a

quadratic relationship– A simple linear model with a quadratic relationship.

These five models were compared using a model-order selection criterion based on parsimony, wheremore complicated models are penalised for the inclusionof additional parameters, named the Akaike Informa-tion Criteria (AIC); the model with the lowest AIC waschosen.

Results

Oceanography

There was no significant difference between years forthe SST values around South Georgia (Fig 1a,t3,6=1.71, P= 0.13; �1.7�C to +10�C). Both years,

Fig. 1 a Mean values and linearregression analysis (with 95%confidence levels) of SeaSurface Temperature aroundSouth Georgia from April untilAugust (the empty squaresrepresenting Winter 2002,whereas the full squares Winter2001). b Mean values and linearregression analysis (with 95%confidence levels) of SeaWiFSaround South Georgia fromApril until August (the emptysquares representing Winter2002, whereas the full squaresWinter 2001)

181

showed a linear decline in sea temperatures as winterprogressed.

A significant difference between years for the SeaW-iFS was recorded (Fig 1b, F3,6= 12.04, P= 0.006) andthis was to the 10�2 mg m�3 level; an amount we did notconsider to be biologically significant, knowing that thedifference between a productive and unproductive areain summer around South Georgia ranges from 15–20 mg m�3 to <0.3 mg m�3, respectively (Korb et al.2004). Winter productivity also declined over our studyarea from April to August.

We thus assumed that both winters did not differgreatly in terms of productivity and temperature andtherefore the weaned seals experienced similar environ-mental conditions.

Tracking duration and overview of movements

Male and female post-weaning pups in our study samplewere not significantly different in weight (t-test5= 2.57,P= 0.11; Table 1).

Between 4 April and 31 August, a total of 532 loca-tions were received (‘Males’= 414, ‘Females’= 118).Over this period the mean tracking duration was120 days for males (86–140 days) and 80 days for fe-males (24–141 days). Two of the PTTs deployed on fe-males stopped uplinking within 2 months ofdeployment, resulting in fewer locations.

Throughout the entire study period all animals re-mained south of the Antarctic Polar Front (Fig. 2a, b).The main focus of uplinks for both genders was centredaround Bird Island and the northwest continental shelf.They showed similar range in terms of latitude, from 51�to 55�S, but differed in longitudes, females going furtherwest (range from 42� to 32�W), whereas males foragedfurther east (range from 40� to 24�W). If we encompasstheir overall distribution within a rectangular box whichincludes the furthest locations to the north, east, southand west the male area covered 500,000 km2 and thefemale 360,000 km2, thus the male area was approxi-mately 39% larger than the region used by females.These areas overlapped over 220,000 km2 representing61% and 43% of the total female and male areas,respectively.

The furthest distance recorded from the natal site wasapproximately 900 km for male pups and 400 km forfemales or 750 km and 300 km from the nearest pointon South Georgia. In all months apart from April, whenmost individuals were still within the vicinity of BirdIsland (Fig. 3), there was a significant difference betweensexes in their range (Table 2, P<0.01) with males re-corded further away from their birth site than females(Fig. 3).

Over the course of the winter, the mean locations alsomoved eastwards; the majority of uplinks were west ofBird Island in April and shifted eastwards thereafter(Fig. 4 a, b, Table 2). Similarly, fewer locations werewithin the 200 m or 500 m isobaths of the SouthGeorgia Continental shelf as winter progressed (Fig. 4c) T

able

1Deploymentandtrackingcharacteristics

of10fursealpupsin

theirrespectiveyears.TheoverlappingperiodofAprilto

Augustwaschosenforthecomparativestudybetween

theopposite

sexes

Tags

Sex

Weight

(kg)

Satellite

tag

Date

Deployed

Date

of

firstuplink

Date

of

last

uplink

Number

ofdays

uplinking

Number

of

uplinks

Number

ofuplinks-

Aprilto

August

Maxim

um

distance

from

birth

site

Apr-

Aug(km)

Date

recovered

W6758

M17.9

1527

04/04/2001

04/04/2001

29/06/2001

86

53

53

758

Notrecovered

W6757

M16.1

1531

04/04/2001

06/04/2001

24/08/2001

140

63

63

614

Notrecovered

W6761

M17.7

1543

04/04/2001

04/04/2001

04/08/2001

122

79

79

231

Notrecovered

W6759

M17.3

30201

04/04/2001

07/04/2001

05/09/2001

151

110

106

903

Notrecovered

W6760

M17.2

30202

04/04/2001

04/04/2001

27/07/2001

114

113

113

658

Notrecovered

W6579

F15.8

23801

06/04/2002

10/04/2002

09/06/2002

60

24

24

301

Notrecovered

W6577

F18.4

23807

06/04/2002

07/04/2002

25/12/2002

262

80

20

265

27/12/2002

W6578

F15

23812

06/04/2002

16/04/2002

24/11/2002

222

90

53

407

Notrecovered

W6576

F13.5

23813

06/04/2002

10/04/2002

21/05/2002

41

11

11

252

Notrecovered

W6580

F15.2

23814

06/04/2002

06/04/2002

30/04/2002

24

10

10

258

Notrecovered

182

with the fur seals foraging away from the continentalshelf into deeper waters.

Discussion

Recent evidence suggests that large-scale physical pro-cesses rather than local factors govern much of the

observed variability in the northern Scotia Sea aroundSouth Georgia (Trathan and Murphy 2002; Trathanet al. 2003; 2005) and that such physical variability hasprofound consequences for some of the major biologicalcomponents in the marine ecosystem (Trathan et al.2003, 2005; Atkinson et al. 2004; Jessopp et al. 2004;Forcada et al. 2005). The forcing factors governingmuch of this physical variability originate outside theScotia Sea. Thus, anomalies in the physical environmentat South Georgia, particularly anomalies in SST corre-late well with El Nino processes in the western Pacific

Fig. 2 Distribution of female (a) and male (b) fur seal pups (A.gazella) during their first winter at sea. Hatched area representingthe extent of overlap

183

(Trathan and Murphy 2002) with the Pacific leadingSouth Georgia by 2–3 years.

Based on the correlation between the western Pacificand South Georgia, SST in both areas can be used todetermine whether the 2 years of this study wereanomalous. Based on this relationship, and the resultspresented above (Fig. 1a), it is evident that all SSTvalues between May and August in both years (2001,2002) were between the respective 20% and 80%quantiles of recorded values at South Georgia (1988–2004). Similarly, no major La Nina/El Nino events wererecorded in the Pacific in the 3-year period prior to thisstudy, again supporting the suggestion that both years of

the study were not unusually anomalous in terms oftheir temperature (Trathan and Murphy 2002).

Similarly, the levels of winter standing crop (chloro-phyll-a) were very low in both years, and though theydiffered from each other by approximately 0.1 mg m�3,the difference is unlikely to have been biologically sig-nificant, given the very large differences observed be-tween eutrophic and oligotrophic areas during thesummer months (Korb and Whitehouse 2004). Thesummer bloom evident in the region (Korb and White-house 2004) declines to a minimum over the wintermonths of April to August. For the period 1999 to 2004the range of average winter values varied between

Fig. 3 Monthly representationof the satellite tag uplinks inrelation to their distance fromSouth Georgia (including BirdIsland)

184

0.206 mg m�3 (during 2004) and 0.446 mg m�3 (during2000) with a mean of 0.268 mg m�3 (standard devia-tion=0.085); both study years fell within one standarddeviation of the average winter value. This again sup-ports the suggestion that both years of this study werenot unusual in terms of their productivity.

Annual indices of pup production and pup survivalat birth (Forcada et al. 2005) for the breeding seasonsfollowing the two winters of this study (summers of2001–2002 and 2002–2003), were between the respective20% and 80% quantiles of recorded values between1988 to 2004. A second species, the gentoo penguin(Pygoscelis papua), which is also an important predatorat South Georgia (Trathan et al. 2005), also showed asimilar response with indices of reproductive successfalling between the 20% and 80% quantiles of recordedvalues (1988–2004). This again indicates that both sea-sons were not unusually anomalous. Thus, the availablephysical and biological data during both of the wintersof this study, together with the breeding performance offur seals (and other predators) following each winterstudy period, suggests that both years were not unusual.

Some caution should be taken with the overall dis-tribution and differences in areas between male and fe-male pups as they could conceivably result from acombination of individual preferences unrelated togender, given the relatively small sample sizes. The dif-ferences in satellite tag performances were not thoughtto be gender related since male and female weaned furseals of this study were not significantly different bymass, the total device mass always represented less than1.5% of the fur seal body mass and the first and last tagsto cease uplinking were both deployed on females. Thevarying performances of the tags were thus attributed toother factors such as battery performance or tags lost atsea.

Our results show that throughout the whole of thewinter period during both years of this study, both maleand female post-weaning juvenile fur seals stayed southof the Antarctic Polar Front (Trathan et al. 1997) andwithin the waters of the Antarctic Circumpolar Current(ACC) (Orsi et al. 1995). The main area of activity forboth genders was centred at Bird Island and to thenorthwest of the South Georgia (Fig. 2a, b) in an areapreviously reported to be important for adult female furseals during both winter and summer (Staniland et al.2004; Boyd et al. 2002). This region was, however, mostused at the start of the winter with an easterly shift asthe season progressed. Within this western shelf region,most uplinks were within the 500 m depth contour.

Previous studies have shown that the diving behav-iour of adult female fur seals varies, depending upontheir foraging location (Staniland et al. 2004). Femalesforaging over the continental shelf dive both deeper andlonger than those diving over oceanic waters beyond theshelf break, with a considerably greater frequency ofdives <20 m in the oceanic waters (Median dive depthShelf= 40 m, Far Oceanic= 35 m n.s.) (Staniland et al.2004; Staniland and Boyd 2003). This is most likelyT

able

2Foragingtrip

characteristics

ofmale

andfemale

fursealpupsduringthewinters

2001and2002,respectively

April

May

June

July

August

Males

Fem

ales

Males

Fem

ales

Males

Fem

ales

Males

Fem

ales

Males

Fem

ales

Minim

um

range(km)

12

53

14

7113

11

207

58

Maxim

um

range(km)

494

258

903

301

860

350

658

407

698

349

Average

range(km)

121±

137

126±

89

329±

295***

152±

70***

359±

233***

208±

114***

460±

186**

306±

137**

604±

85***

257±

104***

Average

latitude(�)

�54.22±

0.77�53.40±

0.63�53.88±

0.49�53.47±

0.73�53.62±

0.77�53.35±

1.01�52.85±

0.81�52.35±

0.81�53.69±

0.44�53.21±

0.92

Average

Longitude(�)�37.22±

2.27�37.87±

1.84�33.51±

4.97�37.76±

2.06�32.76±

3.59

35.71±

1.97

�31.44±

2.60�34.66±

2.20�28.85±

1.34�34.79±

1.89

Number

of

uplinks

79

44

96

41

103

12

96

12

37

9

Number

ofPTTs

55

54

53

42

32

(Values

oftheaverages

are

mean±

s.d;averages

werecomparedusingt-teststheresultsofwhichwerenon-significantunless

indicatedotherwiseby*P<

0.05,**P<

0.01and***P<

0.001)

185

associated with the dietary items available in the watercolumn; indeed, Staniland et al. (2004) showed that fe-males foraging over oceanic waters consumed a greaterproportion of fish, than did females over continentalshelf waters. Staniland and Boyd (2003) suggest thatfemales foraging over oceanic waters have a higherefficiency in terms of the energy return per dive.

Post-weaning juveniles are approximately 50% of themass of adult females and only 14% of adult males;consequently, such a difference in size and mass shouldimpact significantly upon their overall diving and for-

aging capacity (Wartzok 1991). McCafferty et al. (1998)in their study on the diving behaviour of pups found nosignificant differences in maximum dive depth betweengenders. They further showed that prior to weaning, themaximum-recorded depth range of pups (26 m) wasclose to the range of depths to which adult females mostcommonly dive whilst foraging (22 m). However,McCafferty et al. (1998) revealed that such extensivedives were uncommon and therefore, on body massalone, we would anticipate that post-weaning juvenileswould generally dive to depths that were less deep than

Fig. 4 Proportion of uplinkseast or west of Bird Island andthe longitudinal shift duringwinter (a, b, c) Proportion ofuplinks within the 200 m and500 m isobaths over thecontinental shelf of SouthGeorgia

186

those of adult females. Thus, post-weaning juvenilesforaging over the shelf in an area intensively used byother fur seals and higher predators would need to in-crease their diving capacity rapidly and/or target dif-ferent prey items.

After April, the majority of uplinks were to the eastof Bird Island and over deeper waters (Fig. 4). This ra-pid move away from continental shelf waters is consis-tent with post-weaning juveniles seals possibly targetinga different prey source that was more readily available inthe upper water column. The reasons for this easterlyshift are unclear but could be partly associated with thedistribution of Antarctic krill, a major component in thediet of adult fur seals and possibly in that of juveniles.Although the distribution and biomass of krill has beenlittle studied during the winter, summer studies suggestthat there is a greater biomass of krill at the eastern endof South Georgia, compared with the western end(Brierley et al. 1998). Evidence from diet studies (fromBird Island) (Reid 1995) together with information fromthe fishing industry (Trathan et al. 1998; Reid et al.2004), suggest that krill is present throughout the winter.However, there is usually a decrease in the occurrence ofkrill in the diet of fur seals during winter (Reid 1995). Itis not yet clear whether the shift in scat compositionbetween summer and winter recorded on Bird Island(Reid and Arnould 1996; Reid 1995) reflects seasonvariability or the different sex ratio hauling out andfeeding around Bird Island during those times of year.

The fact that winter diet analysis (Reid 1995) mainlyrepresented adult and sub adult male Antarctic fur sealsand reported a greater consumption of shelf living fishspecies such as Champsocephalus gunnari and Lepido-notothen larseni as part of the diet, suggest yet again aprobable difference for the weaned pups foraging inoceanic waters. Similarly, fishing vessels target krillaggregations at greater depth over the course of thewinter (Reid et al. 2004), suggesting that krill may be-come less accessible during some periods of the winter.

Only during the early stages of the breeding season,do adult males and females Antarctic fur seals overlapspatially; otherwise they show a geographic sexual seg-regation (Boyd et al. 1998, Staniland 2005). Fur seals donot recruit until age 3 (females) or 7 (males) and untilthis age are superficially morphologically similar (al-though some differences in body composition were no-ted in Arnould et al. 1996, but not in Rutishauser et al.2004), the juveniles of our study had a similar body mass(P= 0.11, t-test 5= 2.57). A priori, and at least untilthey recruit, both male and female post-weaning juve-niles may be expected to show similar diving and for-aging behaviour. Consistent with other markeddifferences between the genders observed days afterbirth, when male pups play-fight, a behaviour seldomobserved in females (Warren and Jessopp, personalobservation); the geographic separation observed be-tween sexes and reported here, could suggest that gen-der-specific behavioural differences in fact develop earlyin life.

The performances of our satellite tags and the lack ofinformation about the distribution and biomass of krill(or any other fur seal prey) during winter, does not allowus to make firm conclusions about the reasons for ourobserved results. However, we have shown that both thephysical and biological environments were similar be-tween years as was the observed level of breeding successeach subsequent summer season. Consequently, basedon these multiple indicators, together with the knownsexual segregation of adults, we suggest that the ob-served differences in our satellite uplinks could be theresult of early sexual segregation of fur seal juveniles.

Acknowledgements We wish to thank all staff of the British Ant-arctic Survey research station on Bird Island for all their supportand help. Janet Silk for helping with ArcGIS. Keith Reid and IainStaniland for useful suggestions and two anonymous referees. Wethank the SeaWiFS Project (Project Code 970.2) and the GoddardEarth Sciences Data and Information Services Centre / DistributedActive Archive Centre (Project Code 902) for the production anddistribution of satellite data products and the Physical Oceanog-raphy Distributed Active Archive Center (PO.DAAC) at theNASA Jet Propulsion Laboratory, Pasadena, CA http://pod-aac.jpl.nasa.gov for the distribution and production of the SSTdata.

References

Agnew DJ (1997) Review: The CCAMLR ecosystem monitoringprogramme. Antarct Sci 9:235–242

Arnould JPY, Croxall JP (1995) Trends in entanglement of Ant-arctic fur seals (Arctocephalus gazella) in man-mande debris atSouth Georgia. Mar Pollut Bull 30:707–712

Arnould JPY, Boyd IL, Socha DG (1996) Milk consumption andgrowth efficiency in Antarctic fur seal (Arctocephalus gazella)pups. Can J Zool 74(2):254–266

Atkinson A, Siegel V, Pakhomov E, Rothery P (2004) Long-termdecline in krill stock and increase in salps within the SouthernOcean. Nature 432: 100–103

Barlow KE, Croxall JP (2002) Seasonal and interannual variationin foraging range and habitat of macaroni penguins Eudypteschrysolophus at South Georgia. Mar Ecol Prog Ser 232: 291–304

Bonadonna F, Lea MA, Dehorter O, Guinet C (2001) Foragingground fidelity and route-choice tactics of a marine predator:the Antarctic fur seal Arctocephalus gazella. Mar Ecol Prog Ser223:287–297

Boyd IL (1993) Pup production and distribution of breedingAntarctic fur seals (Arctocephalus gazella) at South Georgia.Antarct Sci 5:17–24

Boyd IL (1996) Temporal scales of foraging in a marine predator.Ecology 77:426–434

Boyd IL (2002b) The measurement of dispersal by seabirds andseals: implications for understanding their ecology. In: BullockJM, Kenward RE, Heils RS (eds) Dispersal ecology, BritishEcological Society Symposium. Blackwell Science, Oxford, pp72–88

Boyd IL (2002a) Estimating food consumption of marine preda-tors: Antarctic fur seals and macaroni penguins. J Appl Ecol39:103–119

Boyd IL, McCafferty DJ, Reid K, Taylor R, Walker TR (1998)Dispersal of male and female Antarctic fur seals (Arctocephalusgazella). Canad J Fish Aquat Sci 55:845–852

Boyd IL, Staniland IJ, Martin AR (2002) Distribution of foragingby female Antarctic fur seals. Mar Ecol Prog Ser 242:285–294

Brierley AS, Watkins JL, Goss C, Wilkinson MT, Everson I (1998)Acoustic estimates of krill density at South Georgia during theeleven austral summers between 1981 and 1998. CCAMLR,WG-EMM 98/51, Hobart

187

Doidge DW, McCann TS, Croxall JP (1986) Attendance behaviorof Antarctic fur seals Arctocephalus gazella. In: Gentry RL,Kooyman GL (eds) Fur Seals: maternal strategies on land andat sea. Princeton University Press, pp 102–114

Forcada J, Trathan PN, Reid K, Murphy EJ (2005) Climatechange, environmental forcing and nonlinear demographic re-sponses in Antarctic fur seals. Ecology (in press)

Guinet C, Dubroca L, Lea MA, Goldsworthy S, Cherel Y, Duh-amel G, Bonadonna F, Donnay J-P (2001) Spatial distributionof foraging Antarctic fur seals Arctocephalus gazella in relationto oceanographic variables: a scale-dependent approach usinggeographic information systems. Mar Ecol Prog Ser 219:251–264

Hindell MA, Burton HR, Slip DJ (1991) Foraging areas ofsouthern elephant seals, Mirounga leonina, as inferred fromwater temperature data. Aust J Mar Freshwater Res 42:115–128

Hooper J, Reid K, Agnew D (2004) Incidental seal entanglementson trawl vessels fishing for krill in CCAMLR subarea 48.3.CCAMLR, WG-EMM 04/31, Hobart

Jessopp MJ, Forcada J, Reid K, Trathan PN, Murphy EJ (2004)Winter dispersal of leopard seals (Hydrurga leptonyx) environ-mental factors influencing demographics and seasonal abun-dance. J Zool 263:251–258

Korb RE, Whitehouse M (2004) Contrasting primary productionregimes around South Georgia, Southern Ocean: large bloomsversus high nutrient, low chlorophyll waters. Deep-Sea Res PartI 51:721–738

Korb RE, Whitehouse MJ, Ward P (2004) SeaWiFS in thesouthern ocean: spatial and temporal variability in phyto-plankton biomass around South Georgia. Deep-Sea Res Part II51:99–116

Kovacs KM, Jonas KM, Welke SE (1990) Sex and age segregationby Phoca vitulinaconcolor at haul-out sites during the breedingseason in the Passamaquoddy bay region, New Brunswick. MarMammal Sci 6:204–214

LeBoeuf BJ, Crocker DE, Blackwell SB, Morris PA, Thorson PH(1993) Sex differences in diving and foraging behaviour ofnorthern elephant seals. Symp Zool Soc Lond 66:149–178

Maling DH (1992) Coordinate systems and map projections.Pergamon, Oxford

McCafferty DJ, Boyd IL, Taylor RI (1998) Diving behaviour ofAntarctic fur seal (Arctocephalus gazella) pups. Can J Zool76:513–520

McConnell B, Fedak M, Burton HR, Engelhard GH, ReijndersPJH (2002) Movements and foraging areas of naive, recentlyweaned southern elephant seal pups. J Anim Ecol 71:65–78

Merrick RL, Loughlin TR (1997) Foraging behavior of adult fe-male and young-of-the-year Stellar sea lions in Alaskan waters.Can J Zool 75:776–786

Moore JK, Abbott MR, Richman JG (1999) Location anddynamics of the Antarctic Polar Front from satellite sea surfacetemperature data. J Geophys Res 104:3059–3073

O’Reilly JE, Maritorena S, Mitchell BG, Siegel DA, Carder KL,Garver SA, Kahru M, McClain C (2001) Ocean color chloro-phyll algorythms for SeaWiFS. J Geophys Res-Oceans103:24937–24953

Orsi AH, Whitworth TI, Nowlin WDJ (1995) On the meridionalextent and fronts of the Antarctic Circumpolar current. Deep-Sea Res 42: 641–673

Reid K (1995) The diet of Antarctic fur seals (Arctocephalus gaz-ella) during winter at South Georgia. Antarctic Science 7:241–249

Reid K, Arnould JPY (1996) The diet of Antarctic fur seals Arc-tocephalus gazella during the breeding season at South Georgia.Polar Biol 16:105–114

Reid K, Agnew DG, Warren NL, Owen E (2004) An initial analysisof the characteristics of Antarctic krill taken by the commercialfishery and Antarctic fur seals during the winters of 2002 and2003 at South Georgia. CCAMLR, WG-EMM-04/62, Hobart

Reynolds RW, Smith TM (1994) Improved global sea-surfacetemperature analyses using optimum interpolation. J Clim Res7: 929–948

Robinson SA, Goldsworthy SG, van den Hoff J, Hindell MA(2002) The foraging ecology of two sympatric fur seal species,Arctocephalus gazella and Arctocephalus tropicalis, at Mac-quarie Island during the austral summer. Mar Freshwater Res53:1071–1082

Rutishauser MR, Costa DP, Goebel ME, Williams TM (2004)Ecological implications of body composition and thermalcapabilities in young Antarctic fur seals (Arctocephalus gazella).Physiol Biochem Zool 77 (4):669–681

Staniland IJ (2005) Sexual segregation in seals. In: Ruckstuhl KE,Neuhaus P (eds) Sexual segregation in vertebrates. CambridgeUniversity Press, Cambridge (in press)

Staniland IJ, Boyd IL (2003) Variation in the foraging location ofAntarctic fur seals (Arctocephalus gazella) and the effects ondiving behavior. Mar Mammal Sci 19:331–343

Staniland IJ, Reid K, Boyd IL (2004) Comparing individual andspatial influences on foraging behaviour in Antarctic fur sealsArctocephalus gazella. Mar Ecol Prog Ser 275:263–274

Trathan PN, Croxall JP (2004) Marine predators at South Georgia:an overview of recent bio-logging studies. Memoirs of NationalInstitute of Polar Research 58:118–132

Trathan PN, Murphy EJ (2002) Sea surface temperature anomaliesnear South Georgia: Relationships with the Pacific El Ninoregions. Journal of Geophysical Research 107:10.1029/2000JC000299

Trathan PN, Brandon MA, Murphy EJ (1997) Characterization ofthe Antarctic Polar Frontal Zone to the north of South Georgiain summer 1994. Journal of Geophysical Research 102:10483–10497

Trathan PN, Everson I, Murphy EJ, Parkes G (1998) Analysis ofhaul data from the South Georgia krill fishery. CCAMLRScience 5:9–30

Trathan PN, Brierley AS, Brandon MA, Bone DG, Goss C, GrantSA, Murphy EJ, Watkins JL (2003) Oceanographic variabilityand changes in Antarctic krill (Euphausia superba) abundance atSouth Georgia. Fisheries oceanography 12:569–583

Trathan PN, Murphy EJ, Forcada J, Croxall JP, Reid K, ThorpeSE (2005) Physical forcing in the southwest Atlantic: ecosystemcontrol. In: Bold IL, Wanless S (eds) Management of marineecosystems. Zoological Society London (in press)

Vincent C, McConnell BJ, Ridoux V, Fedak MA (2002) Assess-ment of Argos location accuracy from satellite tags deployed oncaptive gray seals. Mar Mammal Sci 18:156–166

Wartozk D (1991) Physiology of behaviour in pinnipeds. In: Re-nouf D (ed) Behaviour of Pinnipeds. Chapman and Hall,London, pp 236–299

188