deakin research onlinedro.deakin.edu.au/eserv/du:30015951/arnould-divebehaviour-2001.pdf · brook,...

16
Deakin Research Online Deakin University’s institutional research repository DDeakin Research Online Research Online This is the authors final peer reviewed version of the item published as: Arnould, John and Hindell, Mark A. 2001 , Dive behaviour, foraging locations and maternal attendance-patterns in Australian fur seals (Arctocephalus pusillus doriferus) , Canadian journal of zoology, vol. 79, no. 1, pp. 35-48. Copyright : 2001, NRC Canada

Upload: others

Post on 05-Aug-2020

2 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Deakin Research Onlinedro.deakin.edu.au/eserv/DU:30015951/arnould-divebehaviour-2001.pdf · brook, Tex., U.S.A.). Upon capture, each animal was given an in-tramuscular injection (ca

Deakin Research Online Deakin University’s institutional research repository

DDeakin Research Online Research Online This is the authors final peer reviewed version of the item published as: Arnould, John and Hindell, Mark A. 2001, Dive behaviour, foraging locations and maternal attendance-patterns in Australian fur seals (Arctocephalus pusillus doriferus) , Canadian journal of zoology, vol. 79, no. 1, pp. 35-48. Copyright : 2001, NRC Canada

Page 2: Deakin Research Onlinedro.deakin.edu.au/eserv/DU:30015951/arnould-divebehaviour-2001.pdf · brook, Tex., U.S.A.). Upon capture, each animal was given an in-tramuscular injection (ca

Dive behaviour, foraging locations, and maternal-attendance patterns of Australian fur seals(Arctocephalus pusillus doriferus)

John P.Y. Arnould and Mark A. Hindell

Abstract: The dive behaviour, foraging locations, and colony-attendance patterns of female Australian fur seals(Arctocephalus pusillus doriferus) from Kanowna Island (39°10′S, 146°18′E) in Bass Strait, southeastern Australia, weredetermined throughout lactation during 1997–1999. Foraging-trip durations increased as lactation progressed, beingshortest in summer (3.71 ± 0.24 days; mean ± 1 SE) and longest in winter (6.77 ± 0.57 days,P < 0.05), but maternal-attendance periods did not differ in duration (1.70 ± 0.10 days,P > 0.5). Individual mean attendance periods and tripdurations were positively correlated (r2 = 0.21,P < 0.005). Diving commenced shortly after seals left the colony (2.6 ±0.4 h), was continuous for long periods (up to 36 h), occurred mostly during daylight hours, and lacked regular dielvariation in depth. The majority of dives (78%) were typically U-shaped and reached depths corresponding to the pre-vailing depths in Bass Strait (65–85 m), indicating that these animals forage mostly on the benthos of the shallow con-tinental shelf in this region. Such behaviour is unusual for fur seals but is reminiscent of that of some sea lion species.Mean dive durations varied between 2.0 and 3.7 min (maximum 8.9 min) and the theoretical aerobic dive limit (3.91–4.26 min) was exceeded on 17.3% of dives. Dive frequency (8.3 ± 0.6/h) and the proportion of time at sea spent div-ing (40.7 ± 2.1%) were weakly negatively related to the duration of the foraging trip (r2 = 0.07,P < 0.004, andr2 =0.13, P < 0.0001, respectively). Data from at-sea locations showed that lactating females forage almost exclusivelywithin Bass Strait during all seasons.

Résumé: Nous avons étudié le comportement de plongée, les zones d’alimentation et la présence dans la colonie chezdes femelles de l’Otarie d’Australie (Arctocephalus pusillus doriferus) dans l’île de Kanowna (39°10′S, 146°18′E),dans le détroit de Bass, dans le sud-est de l’Australie, pendant toute la période d’allaitement, de 1997 à 1999. La duréedes excursions de quête de nourriture augmente pendant toute la période d’alllaitement et c’est en été qu’elle est leplus courte (3,71 ± 0,24 jours; moyenne ± 1 erreur type) et en hiver qu’elle est le plus longue (6,77 ± 0,57 jours,P < 0,05), mais la durée de la présence maternelle ne varie pas selon la saison (1,70 ± 0,10 jour,P > 0,5). La duréede la présence individuelle et la durée des excursions sont en corrélation positive (r2 = 0,21,P < 0.005). Les otariescommencent à plonger peu après leur départ de la colonie (2,6 ± 0,4 h), plongent pendant de longues périodes conti-nues (jusqu’à 36 h), surtout durant les heures de clarté et la profondeur de leurs plongées ne suit pas de pattern parti-culier selon l’heure. La majorité des plongées (78 %) sont des plongées en U jusqu’aux profondeurs correspondant auxprofondeurs les plus communes du détroit de Bass (65–85 m), ce qui indique que les otaries cherchent leur nourrituresurtout dans le benthos de la plate-forme continentale de cette région. Un tel comportement est inusité chez une otarie,mais rappelle celui de certains lions de mer. La durée moyenne des plongées va de 2,0 à 3,7 min (maximum 8,9 min)et la limite théorique d’une plongée aérobie (3,91–4,26 min) s’est trouvée dépassée dans 17,3 % des plongées. Lafréquence des plongées (8,3 ± 0,6/h) et la proportion du temps en mer consacré aux plongées (40,7 ± 2,1 %) sont encorrélation négative faible avec la durée des excursions (respectivementr2 = 0,07,P < 0,004 etr2 = 0,13,P < 0,0001).Les données sur les sites en mer démontrent que les femelles qui allaitent se nourrissent presque exclusivement dans ledétroit de Bass, en toute saison.

[Traduit par la Rédaction] Arnould and Hindell 48

Introduction

Throughout most of the lactation period, otariid (fur sealsand sea lions) pups are completely dependent on their motherfor nutrition (Bonner 1984). For periods ranging from 4 to

more than 24 months (depending on species), lactating fe-males alternate between foraging trips to sea and regularnursing periods ashore at the natal colony (Gentry andKooyman 1986). As all the nutrients required for milk pro-duction are acquired by the females during these trips, their

Can. J. Zool. 79: 35–48 (2001) © 2001 NRC Canada

35

DOI: 10.1139/cjz-79-1-35

Received April 26, 2000. Accepted September 29, 2000.Published on the NRC Research Press website on December 21, 2000.

J.P.Y. Arnould.1 Marine Mammal Research Group, Graduate School of the Environment, Macquarie University, Sydney 2109,New South Wales, Australia.M.A. Hindell. Antarctic Wildlife Research Unit, School of Zoology, University of Tasmania, P.O. Box 252-05, Hobart 7001,Tasmania, Australia.

J:\cjz\cjz79\cjz-01\Z00-178.vpMonday, December 18, 2000 1:45:27 PM

Color profile: DisabledComposite Default screen

Page 3: Deakin Research Onlinedro.deakin.edu.au/eserv/DU:30015951/arnould-divebehaviour-2001.pdf · brook, Tex., U.S.A.). Upon capture, each animal was given an in-tramuscular injection (ca

foraging behaviour and how they apportion time spent at seaand nursing ashore directly influence their maternal-investmentpatterns (Trillmich 1996).

Since the development in the mid-1970s of electronic time–depth recorders (TDRs) for measuring diving activity andradiotelemetry methods for monitoring the movements ofanimals to and from the colony, there have been many stud-ies in which at-sea behaviour and foraging effort of lactatingotariids were investigated (e.g., Gentry and Kooyman 1986;Feldkamp et al. 1989; Boyd and Croxall 1992; Francis et al.1998; Mattlin et al. 1998; Thompson et al. 1998). Thesestudies have revealed two broadly divergent patterns for furseals and sea lions. Lactating sea lions generally undertakeshort trips (1–2 days) during which they exhibit a continuousdiving pattern with no diel variation, foraging mostly on thebenthos (e.g., Kooyman and Trillmich 1986b; Gales and Costa1997; Gales and Mattlin 1997; Thompson et al. 1998). Incontrast, lactating fur seals generally undertake longer trips(3–23 d) during which diving is mostly nocturnal and occurs inbouts to the deep scattering layer, with pronounced diel varia-tion in depth that reflects the vertical migration of their prey(e.g., Boyd et al. 1991; Harcourt et al. 1995; Georges et al.2000a). The exception is the northern fur seal (Callorhinusursinus), in which both foraging strategies have been ob-served (Gentry et al. 1986b; Goebel et al. 1991).

Detailed data are still lacking for numerous species, how-ever, making it impossible to determine whether these trendsare due to phylogenetic or environmental influences. Indeed,recent studies have shown marked geographical differencesin foraging-trip durations and diving behaviour within bothAntarctic fur seals (Arctocephalus gazella) (Boyd et al. 1994;Goldsworthy et al. 1997; Green 1997) and subantarctic furseals (Arctocephalus tropicalis) (Goldsworthy et al. 1997;Georges et al. 2000b) from widely dispersed colonies in theSouthern and Indian oceans, suggesting that local marineecosystem characteristics play an important role in determin-ing a species’ foraging behaviour (Gentry et al. 1986a andFrancis et al. 1998).

The Australian fur seal (Arctocephalus pusillus doriferus)is the largest of the fur seal species, with adult females andmales weighing, on average, 76 and 279 kg, respectively(Warneke and Shaughnessy 1985). It is a temperate-latitudeotariid with a lactation period lasting 10–11 months, butsome females may nurse a pup for a second or even thirdyear (Warneke 1982). The population is still recovering fromthe severe exploitation of the commercial sealing era (1798–1825), with annual pup production currently estimated atapproximately 14 000, well below the estimated presealinglevel of up to 50 000 (Warneke 1982; Pemberton and Kirk-wood 1994). The breeding distribution of the species is re-stricted to Bass Strait, a shallow body of water (averagedepth <85 m) between the southeastern tip of the Australianmainland and Tasmania, which is recognised as being of lowoceanic productivity (Murray and Parslow 1999). Natal colo-nies are currently located on just nine islands, but there ishistorical evidence that several other islands in Bass Straitalso once hosted breeding colonies (Warneke 1982; Pember-ton and Kirkwood 1994).

With its limited distribution, small population size, andslow rate of recovery, the Australian fur seal is in sharp con-

trast to the Cape fur seal (Arctocephalus pusillus pusillus),which has a population size of 1.7 million, has experiencedrapid population growth since conservation measures wereintroduced in 1893, and occupies a wide distribution alongthe coastline of Namibia and South Africa (Butterworth etal. 1995). The disparities between the subspecies may bedue to the substantial difference in marine productivity be-tween the Benguela Current, which runs along the southwestcoast of Africa, and the nutrient-poor waters of Bass Strait(Warneke and Shaughnessy 1985). However, relatively littleis known about the Australian fur seal, particularly its forag-ing behaviour (Hindell and Pemberton 1997; Hindell et al.1998), so it is not possible to ascertain the influence of itsnutrient-poor marine habitat on maternal-investment patternsor population demography.

The aims of this study, therefore, were to determine, inlactating females, (i) foraging-trip durations and maternal-attendance patterns; (ii ) dive behaviour; (iii ) foraging loca-tions; and (iv) how these vary according to season and stageof lactation.

Materials and methods

Study site, animals, and captureThe study was conducted between December 1997 and July

1999 on Kanowna Island (39°10′S, 146°18′E), the site of an Aus-tralian fur seal breeding colony with an annual production of ap-proximately 1600 pups (Arnould and Littnan 2000). The island issituated in central northern Bass Strait, where the distance to thenearest continental shelf edge (200 m depth contour) is 150 and280 km to the east and west, respectively (Fig. 1).

Fieldwork was conducted in 2- to 3-week periods every 2–3 months during which randomly selected nursing females werecaptured using a modified hoop net (Fuhrman Diversified, Sea-brook, Tex., U.S.A.). Upon capture, each animal was given an in-tramuscular injection (ca. 0.15 mg·kg–1) of the sedative Midazolam(Hypnovel®, Roche Products Pty Ltd., Dee Why, NSW, Australia)to facilitate handling and reduce capture stress. Once the sedativehad taken effect, the seal was transferred to a restraint board (Gen-try and Holt 1982). A subset of animals was weighed using aspring scale (200 ± 0.5 kg; Salter, Peterborough, U.K.), and mea-sured (straight-line length) to the nearest centimetre with a tapemeasure. Individual numbered plastic tags (Super Tags®, Dalton,Woolgoolga, Australia) were placed in the trailing edge of bothforeflippers and the seal was instrumented with electronic record-ing devices (see below) before it was released. The device was re-moved by cutting the fur beneath it upon recapture of the animal.All capture and handling of seals were conducted in accordancewith the principles and guidelines of the Canadian Council onAnimal Care.

Instrumentation, data collection, and statistical analysesTo monitor the attendance of individuals at the colony, a small

VHF transmitter (Sirtrack Ltd., Havelock North, New Zealand)was glued to the dorsal fur using quick-setting epoxy (RS Compo-nents, Corby, U.K.). The presence or absence of individuals wasrecorded by an automatic scanning receiver and data logger (RX-900E; Televilt Int. AB, Lindesberg, Sweden) located on a slopeabove the colony. The receiver was programmed to continuallyscan each frequency for 30 s every 10–12 min (depending on thenumber of frequencies monitored). As Australian fur seals oftenenter the water to thermoregulate without leaving the vicinity ofthe colony, only absences of greater than 6 h wereconsidered to beforaging trips. For logistic reasons, the deployment of transmitters

© 2001 NRC Canada

36 Can. J. Zool. Vol. 79, 2001

J:\cjz\cjz79\cjz-01\Z00-178.vpMonday, December 18, 2000 1:45:27 PM

Color profile: DisabledComposite Default screen

Page 4: Deakin Research Onlinedro.deakin.edu.au/eserv/DU:30015951/arnould-divebehaviour-2001.pdf · brook, Tex., U.S.A.). Upon capture, each animal was given an in-tramuscular injection (ca

was staggered into periods corresponding to late breeding season,postmoult, and midlactation.

A subset of individuals was also instrumented with geolocatingTDRs (MK7, Wildlife Computers, Redmond, Wash., U.S.A.) pro-grammed to sample depth (±1.0 m) and light every 5 and 60 s,respectively. The VHF transmitter and TDR together constituted<0.5% of the cross-sectional surface area and body mass of the ani-mals. Custom-written software (Dive©, Stewart Greenhill, MurdochUniversity, Wash., U.S.A.) was used to analyse dive behaviour,with all dives >2 m in depth considered potential foraging dives.While it is possible that many shallow dives (i.e., <10 m) are asso-ciated with travelling or nonforaging behaviour in an animal of thissize, Australian fur seals are often sighted in shallow areas farfrom natal colonies and their diet is known to include inshore-dwelling fish species (Gales and Pemberton 1994). Consequently,as no detailed information exists on the dive behaviour of femaleAustralian fur seals, it was considered prudent to set such a shal-low dive threshold. The software program classified the shape ofdives according to the proportion of time spent at depth and thevariation in depth during that time. Dives in which >5% of the totalduration was spent at the maximum depth and no divergences fromthis depth were made during that time were classified as U-shaped.Based on preliminary inspection of the dive traces, all other diveswere classified as V-shaped.

At-sea locations were determined by geolocation (Hill 1994).MultiTrace software (Jensen Software) was used to determine thetimes of sunrise and sunset, based on light levels logged every 60 sby the TDRs. Latitude and longitude were then calculated, usingthe same software, from local apparent midday and day length.Two locations were calculated daily, one based on the duration andtiming of daytime and one for nighttime. Locations were derivedfor each day that seals were at sea, with the exception of 2–3 weeks either side of the equinoxes, when the software could notdiscern latitude.

All at-sea locations were initially filtered using the iterativetechnique of McConnell et al. (1992) to remove any fixes that wereunrealistic in terms of the required rate of travel with maximumspeed set at 4 m·s–1.

Even under the best possible circumstances, geolocation algorithmshave a resolution of approximately 1° of latitude and longitude(Hill 1994). Given this, a conservative approach to quantifying at-sealocations was adopted. We calculated the number of times thatseals were located in 1° cells and presented these data as densitycontours using MapInfo® (Version 5.5) and Vertical Mapper® (Ver-sion 2) (MapInfo Corporation, New York, N.Y., U.S.A.). The po-tential errors in rates of travel between consecutive locations due tothe poor resolution rendered calculations of time spent in each cellproblematic, so the more simple index of hit per cell was used. The

© 2001 NRC Canada

Arnould and Hindell 37

Fig. 1. Australian fur seal (Arctocephalus pusillus doriferus) breeding colonies (r) in Bass Strait, southeastern Australia. The studywas conducted at Kanowna Island.

J:\cjz\cjz79\cjz-01\Z00-178.vpMonday, December 18, 2000 1:45:31 PM

Color profile: DisabledComposite Default screen

Page 5: Deakin Research Onlinedro.deakin.edu.au/eserv/DU:30015951/arnould-divebehaviour-2001.pdf · brook, Tex., U.S.A.). Upon capture, each animal was given an in-tramuscular injection (ca

resulting pictures provide a broad-scale indication of regions ofBass Strait and the surrounding ocean that were most visited by the seals.

Seasonal trends were analysed with the start of summer definedas 1 December, which is also the peak pupping date for the species(Warneke and Shaughnessy 1985; Pemberton and Kirkwood 1994).Statistical analyses were performed using the Systat® statistical soft-ware (Version 7.0.1, SPSS Inc., Chicago). The Kolmogorov–Smirnovtest was used to determine whether the data were normally distrib-uted and anF test to confirm homogeneity of variances. Unlessotherwise stated, data are presented as the mean ± 1 SE andresultsconsidered significant at theP < 0.05 level.

Results

Foraging-trip durations and maternal-attendancepatterns

Foraging-trip durations and maternal-attendance recordswere obtained from a total of 36 lactating female seals forperiods ranging from 37 to 286 days. A total of 664 foragingtrips, 687 maternal-attendance periods, and 599 completeforaging cycles (foraging trip plus subsequent onshore atten-dance) were recorded, spanning all seasons and stages of thelactation period. The fewest data were obtained in spring,when, during a period corresponding to the expected wean-ing phase (Shaughnessy and Warneke 1987), the majority(80%) of the 15 females instrumented with a VHF transmit-ter at the time departed the colony for exceptionally longtrips (see below). These absences were analysed separately,as it was assumed they did not represent episodes of forag-ing for the purpose of maternal provisioning of the currentpup.

There was no significant trend in the distribution of arrivaltimes at the colony (F[1,11] = 1.64,P > 0.8). Departures fromthe colony, however, were not evenly distributed in time(F[1,11] = 3.89,P < 0.0001), being more frequent in the earlymorning hours preceding sunrise and just before midday(Newman–Keuls test,P < 0.03; Fig. 2).

Foraging-trip durations varied greatly (range 0.31–21.79 days), the means for individual seals ranging from2.02 to 10.41 days. Maternal-attendance periods also variedconsiderably in duration (range 0.10–8.44 days), the meansfor individual seals ranging from 0.71 to 3.02 days. Therewas a weak but significant positive relationship (r2 = 0.21,n = 36, P < 0.005) between an individual’s mean foraging-trip duration and its mean attendance period. The variability inthis relationship may result, in part, from the influence of sea-son on foraging and attendance patterns (Fig. 3). Repeated-measures ANOVA for seals whose records spanned all seasons(n = 6) showed no significant difference in individual meanattendance periods between the seasons (F[3,15] = 0.749,P >0.5). The mean foraging-trip duration for individual seals,however, increased significantly between seasons (F[3,15] =6.679,P < 0.005). These results were supported by simpleANOVAs on three random subsamples of all individuals (inall casesP > 0.1 for attendance andP < 0.05 for foragingtrips). Using the data for all individuals, a Newman–Keulstest indicated that trip durations were significantly shorterin summer (3.71 ± 0.24 days) than in autumn (6.03 ±0.71 days), winter (6.77 ± 0.57 days), or spring (6.59 ±0.75 days). Trips in autumn, winter, and spring did not differsignificantly in duration. The attendance period for individ-ual seals, in all seasons combined, was 1.70 ± 0.10 days.

The durations of complete foraging cycles for individualseals also varied significantly between seasons (simpleANOVA of random subsamples and repeated-measuresANOVA, P < 0.04 in all cases) being shortest in summer(5.41 ± 0.44 days) and longest in winter (8.67 ± 0.68 days).However, as foraging-cycle duration varied with season butattendance periods did not, the proportion of time spent atsea was significantly less (P < 0.001 in all cases) in summer(63.9 ± 2.0%) than at other times of the year (range 74.4–76.2%).

In early to midspring (mean date 10 October ± 5.1 days),12 seals departed the colony on foraging trips that were con-siderably longer than their previous ones. The duration ofthese trips (34.9 ± 4.7 days; range 11.6–74.6 days) indicatesa shift in maternal provisioning behaviour and, if it is as-sumed that this represents weaning, pup age at weaning forthese individuals can be estimated at 313 ± 5.1 days. Inter-estingly, data from three of these individuals that were alsoinstrumented with TDRs indicate that although they werenot recorded at the colony during this postlactation phase,they regularly hauled out elsewhere. All 12 females returnedto the colony during the breeding period on the mean date of28 November ± 1.7 days. Data on their subsequentattendance periods (8.4 ± 1.0 days; range 4.6–11.1 days)

© 2001 NRC Canada

38 Can. J. Zool. Vol. 79, 2001

00:00 04:00 08:00 12:00 16:00 20:00 24:000

4

8

12

16 (a)

Local time

Mea

nfr

eque

ncy

(%)

00:00 04:00 08:00 12:00 16:00 20:00 24:000

4

8

12

16 (b)

Local time

Mea

nfr

equ

ency

(%)

Fig. 2. Mean frequency distributions (36 individuals) of times ofarrival (a) and departure (b) (Australian Eastern Standard Time)of lactating female Australian fur seals from the natal colony.Daylight hours are indicated by the horizontal bars.

J:\cjz\cjz79\cjz-01\Z00-178.vpMonday, December 18, 2000 1:45:32 PM

Color profile: DisabledComposite Default screen

Page 6: Deakin Research Onlinedro.deakin.edu.au/eserv/DU:30015951/arnould-divebehaviour-2001.pdf · brook, Tex., U.S.A.). Upon capture, each animal was given an in-tramuscular injection (ca

© 2001 NRC Canada

Arnould and Hindell 39

Fig. 3. Frequency distributions of mean maternal attendance periods (a) and foraging-trip durations (b) for individual seals during dif-ferent seasons (26 seals in summer, 20 in autumn, 19 in winter, and 17 in spring).

J:\cjz\cjz79\cjz-01\Z00-178.vpMonday, December 18, 2000 1:45:33 PM

Color profile: DisabledComposite Default screen

Page 7: Deakin Research Onlinedro.deakin.edu.au/eserv/DU:30015951/arnould-divebehaviour-2001.pdf · brook, Tex., U.S.A.). Upon capture, each animal was given an in-tramuscular injection (ca

were obtained from six individuals that thereafter resumedregular foraging trips.

The remaining three seals whose transmitters were work-ing during early spring continued to follow the same atten-dance patterns throughout the remainder of spring and earlysummer until the transmitters failed. They displayed no in-crease in attendance during the breeding season (i.e., therewas no evidence of a perinatal period).

Dive behaviour and foraging effortDive records were obtained from 13 individual females,

for a total of 121 complete foraging trips covering all sea-sons and stages of lactation and comprising 113 636 dives(Table1). An example of a typical foraging-trip dive recordis shown in Fig. 4a. All foraging trips were characterised byindividuals commencing diving soon after departing the col-ony (2.6 ± 0.4 h; range 0.8–6.0 h) and including long peri-ods of continuous diving (up to 36 h). While there wereperiods of up to several hours with no diving activity, therewas a lack of regular dive bouts. The majority of dives(78%) were typically U-shaped, with distinct descent, bot-tom, and ascent phases, and showed little variation in depthbetween sequential dives (Fig. 4b). The remainder of the re-corded dives were generally V-shaped in profile (i.e., had nobottom phase). Females continued to make foraging divesuntil shortly before hauling out (3.4 ± 0.3 h; range 0.8–12.0 h).

In 88 (73%) of the foraging trips, females made a seriesof dives (1–5) just prior to leaving the water after havingbeen at the surface since foraging ceased. The profile ofthese dives (see Fig. 4b) suggests that they were to thebenthos. In 70% of the foraging trips in which females madedives just prior to leaving the water, they hauled out at thecolony during daylight. In contrast, only 38% of foragingtrips without such dives occurred during daylight.

A summary of dive statistics for individual females is pre-sented in Table 2. Mean dive depths (all foraging trips com-

bined) ranged from 27 to 75 m (median 19–80 m), though>50% of the dives recorded were to depths of 65–85 m(Fig. 5). Maximum dive depths for all seals were between 76and 94 m, except that one individual (seal 9) dived to 164 m.This female regularly dived to >100 m on numerous tripsduring three seasons (spring, summer, and autumn). Therewere no apparent diel patterns in dive depth (ANOVA,P >0.07 in all seasons; Fig. 6). There were also no diel patternsin the frequency of diving in summer, autumn, or spring(ANOVA, P > 0.08 in all cases). In winter, however, divingwas significantly more frequent during daylight hours (F[1,11] =8.26, P < 0.0001, Newman–Keuls test,P < 0.02; Fig. 6).

Mean dive durations were between 2.0 and 3.7 min (me-dian 2.0–4.1 min), the longest dive recorded lasting 8.9 min.An estimated theoretical aerobic dive limit (ADL) was cal-culated for animals whose body mass was measured at firstcapture using the equation ADL (min) = 1.57·mass (kg)0.22

(Gentry et al. 1986b). The theoretical ADL, which rangedfrom 3.91 to 4.26 min, was exceeded, on average, in 17.3%(range 2.2–49.7%) of dives. The overall interdive intervalwas 4.1 ± 0.08 min (range 0.08–1216 min). However, 95 and99% of surface intervals were shorter than 10 and 60 min, re-spectively. When surface intervals more than 10 and 60 minin duration were excluded from the analyses, the interdiveinterval was 1.4 ± 0.01 and 2.0 ± 0.01, respectively.

Dive frequency (number of dives per hour), dive rate (inmetres per hour), and the proportion of time at sea spent div-ing have been shown to be good indicators of foraging effortin other otariid species (Boyd et al. 1991, 1994; Arnould etal. 1996), and were investigated in the present study (Ta-ble 2). The number of dives per hour for each individual was8.3 ± 0.6 (range 5.5–14.5), while the rate of vertical divingwas 996.6 ± 40.8 m·h–1 (range 676–1210 m⋅h–1). There wasa weak but significant negative relationship between the du-ration of a foraging trip and the number of dives per hour(r2 = 0.07,n = 121,P < 0.004) but not the rate at which ver-tical distance was covered (P > 0.5). Similarly, the propor-

© 2001 NRC Canada

40 Can. J. Zool. Vol. 79, 2001

SealNo.

Date TDRdeployed

Length ofrecord (days)

Body mass atdeployment (kg)

Length(cm)

No. offoraging trips

Mean triplength (days)

Total no.of dives

Mean no. ofdives per trip

1 1998-02-12 4 70.5 148 1 1.3 168 1682 1998-02-12 126 69.5 146 17 5.2 18 767 1 0433 1998-02-14 9 73.5 147 2 2.7 1 128 5644 1998-04-10 100 63.0 123 10 7.8 14 537 1 4545 1998-07-16 7 92.0 157 1 3.7 624 6246 1998-07-16 12 92.5 153 1 6.6 1 237 1 2377 1998-07-16 9 78.5 154 4 1.6 1 494 3748 1998-07-30 125 93.0 93 17 4.9 17 634 9809 1998-10-23 129 — — 15 6.0 18 535 1 23610 1998-10-23 42 — — 5 5.6 5 047 1 00911 1999-03-13 77 — — 14 3.4 6 539 46712 1999-03-13 76 — — 10 5.1 11 137 1 11413 1999-05-28 65 67.0 140 22 2.1 16 789 763

Mean 60.1 77.7 140.1 9.2 4.3 8 741.2 848.7SE 13.7 4.0 6.8 2.0 0.6 2 124.8 107.2

Table 1. Summary of TDR deployments for the 13 lactating female Australian fur seals (Arctocephalus pusillus doriferus) used in thisstudy.

J:\cjz\cjz79\cjz-01\Z00-178.vpMonday, December 18, 2000 1:45:34 PM

Color profile: DisabledComposite Default screen

Page 8: Deakin Research Onlinedro.deakin.edu.au/eserv/DU:30015951/arnould-divebehaviour-2001.pdf · brook, Tex., U.S.A.). Upon capture, each animal was given an in-tramuscular injection (ca

tion of time spent diving, 40.7 ± 2.1% (range 30–54%), wasalso weakly negatively correlated with foraging-trip duration(r2 = 0.13,n = 121, P < 0.0001).

To investigate the influence of season on dive behaviour,the means for all foraging trips from the same individualwere averaged within each season (4 individuals in summer,6 in autumn, 7 in winter, and 3 in spring). Mean and mediandive depths, mean dive duration, mean dive frequency, meandive rate, and mean percentage of time spent diving did not

differ significantly between seasons (ANOVAP > 0.1 in allcases).

At-sea locations and foraging rangeA total of 327 at-sea locations were available from 12 in-

dividuals after filtering, representing 169 days spent at sea.The number of locations for each of the four seasons rangedfrom 50 to 134, and came from between three and seven in-dividuals. The seals generally stayed within 500 km of the

© 2001 NRC Canada

Arnould and Hindell 41

Fig. 4. (a) Dive profile from a typical foraging trip by a lactating female Australian fur seal. (b) Examples of the two most prominentdive types identified (U-shaped benthic dives and V-shaped midwater pelagic dives) and the characteristic dives observed just prior tohauling out at the end of the majority of foraging trips. Times are given as Australian Eastern Standard Time.

J:\cjz\cjz79\cjz-01\Z00-178.vpMonday, December 18, 2000 1:45:36 PM

Color profile: DisabledComposite Default screen

Page 9: Deakin Research Onlinedro.deakin.edu.au/eserv/DU:30015951/arnould-divebehaviour-2001.pdf · brook, Tex., U.S.A.). Upon capture, each animal was given an in-tramuscular injection (ca

colony, with more than 90% of at-sea locations beingobtained within the shallow waters of Bass Strait, where thelocations were concentrated in areas within 100 km of thecolony (Fig. 7).

The seals dispersed most widely from the island duringspring and summer (Fig. 7). This is likely to be a result ofthe extended postlactation foraging trip. During these seasons,although locations were still concentrated in Bass Strait,some individuals spent time outside the strait on the westcoast of Tasmania and may have moved off the continentalshelf to forage in deep oceanic waters. Indeed, one of theseindividuals was seal 9, which dived to >160 m.

Discussion

Foraging and attendance cyclesFemale Australian fur seals arrived at the colony at all

times and departed to sea more often before sunrise or aroundmidday. This is in contrast to the pattern observed in mostother fur seal species, where females tend to return to thecolony mostly between sunrise and midday and start the ma-jority of their foraging trips between late afternoon and lateevening (e.g., Trillmich 1986; Trillmich et al. 1986; Boyd etal. 1990; Goldsworthy 1992).

© 2001 NRC Canada

42 Can. J. Zool. Vol. 79, 2001

Div

ed

ep

th(m

)D

ive

du

ratio

n(m

in)

Se

al

No

.T

ota

ln

o.

of

div

es

Me

an

SE

Me

dia

nM

ax.

Me

an

SE

Me

dia

nM

ax.

Div

efr

eq

ue

ncy

a

(no

./h

)D

ive

rate

a

(m·h

–1 )

Tim

esp

en

td

ivin

ga(%

)T

he

ore

tica

lA

DL

b(m

in)

%d

ive

s>

AD

L

11

68

68

.81

.77

77

93

.40

.01

3.7

4.8

5.5

77

13

14

.00

22

.12

18

76

76

2.4

0.2

74

86

2.6

0.0

12

.86

.18

.31

09

13

73

.99

4.3

31

12

85

7.7

0.6

62

76

2.7

0.0

32

.85

.28

.91

08

74

04

.04

4.0

41

45

37

71

.10

.17

58

33

.00

.01

3.0

6.0

7.8

11

54

39

3.9

11

4.1

56

24

74

.80

.78

08

23

.60

.04

3.7

6.9

7.1

11

07

43

4.2

52

3.2

61

23

75

7.4

0.4

61

88

3.7

0.0

33

.97

.17

.89

51

48

4.2

53

2.3

71

49

44

9.5

0.7

62

81

3.3

0.0

54

.16

.98

.91

03

35

44

.10

49

.78

17

63

47

2.4

0.1

78

84

3.1

0.0

13

.26

.38

.61

21

04

44

.26

3.9

91

85

35

50

.20

.35

11

64

2.2

0.0

12

.36

.88

.79

58

34

——

10

50

47

61

.80

.26

68

13

.40

.02

3.5

7.4

7.7

10

35

44

——

11

65

39

55

.10

.25

79

43

.20

.02

3.3

8.9

5.7

67

63

0—

—1

21

11

37

45

.70

.35

58

42

.20

.01

2.3

8.1

8.8

85

53

2—

—1

31

67

89

26

.60

.21

97

62

.00

.01

2.0

7.0

14

.51

02

85

33

.96

2.2

Me

an

87

41

.25

8.0

62

.88

9.1

2.9

3.1

6.7

8.3

99

6.6

40

.74

.08

17

.3S

E2

12

4.8

3.7

4.5

6.4

0.2

0.2

0.3

0.6

40

.82

.10

.05

5.4

aM

ean

for

all

fora

ging

trip

s.bT

heor

etic

alA

DL

calc

ulat

edfr

ombo

dym

ass

atca

ptur

e(s

eeth

ete

xtfo

rde

tails

).

Tabl

e2.

Su

mm

ary

of

sta

tistic

sfr

om

the

div

ere

cord

so

fth

efe

ma

leA

ust

ralia

nfu

rse

als

use

din

this

stu

dy

(da

tafr

om

all

fora

gin

gtr

ips

com

bin

ed

).

0

5

10

15

20

0 10 20 30 40 50 60 70 80 90 100 110 120 >120

(a)

Dive depth (m)

Fre

quen

cy(%

)

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 850

5

10

15

20(b)

Median dive depth (m)

Fre

quen

cy(%

)

Fig. 5. Frequency distribution of all dives recorded (113 636) from13 individual female Australian fur seals (a) and median divedepths during all the complete foraging trips recorded (121) (b).

J:\cjz\cjz79\cjz-01\Z00-178.vpMonday, December 18, 2000 1:45:37 PM

Color profile: DisabledComposite Default screen

Page 10: Deakin Research Onlinedro.deakin.edu.au/eserv/DU:30015951/arnould-divebehaviour-2001.pdf · brook, Tex., U.S.A.). Upon capture, each animal was given an in-tramuscular injection (ca

© 2001 NRC Canada

Arnould and Hindell 43

The departure of some females on very long foraging tripsin early October is consistent with the observation that thetotal number of Australian fur seals at colonies, particularlyadult females, decreases dramatically at this time (K. Cane1999, personal observations)2 and supports the notion thatthis corresponds to weaning. The estimated mean weaningage of 313 days (10 months) is less than the 11–12 monthspreviously estimated for Australian fur seals by Warneke andShaughnessy (1985). However, Rand (1955) also observed

very long foraging trips in some Cape fur seals duringSeptember–October, which precipitated weaning at approxi-mately 10 months of age.

Prepartum attendance at the colony has been found to bein the order of 1.5–2.0 days in several otariid species (Davidand Rand 1986; Higgins and Gass 1993; Lunn and Boyd1993). Therefore, the mean date of return from the extendedpostlactation foraging trips made by the instrumented fe-males in the present study, 28 November, is consistent with

30

40

50

60

70

80

30

40

50

60

70

80

30

40

50

60

70

80

30

40

50

60

70

80

00:00 04:00 08:00 12:00 16:00 20:00 24:000

5

10

15

20

Local time

Mea

nfr

eque

ncy

(%)

Mea

ndi

vede

pth

(m)

00:00 04:00 08:00 12:00 16:00 20:00 24:000

5

10

Local timeM

ean

fre

quen

cy(%

)

Mea

ndi

vede

pth

(m)

00:00 04:00 08:00 12:00 16:00 20:00 24:000

5

10

Local time

Mea

nfr

eque

ncy

(%) M

ean

dive

dept

h(m

)

00:00 04:00 08:00 12:00 16:00 20:00 24:000

5

10

Local time

Mea

nfr

equ

ency

(%)

SpringWinter

AutumnSummer

Mea

ndi

vede

pth

(m)

Fig. 6. Temporal distributions of mean dive frequencies and mean dive depths for female Australian fur seals during all seasons(4 seals in summer, 6 in autumn, 7 in winter, and 3 in spring). Daylight hours are indicated by the horizontal bars and times are givenas Australian Eastern Standard Time.

2K. Cane. 1999. Aspects of the behaviour of male Australian fur sealsArctocephalus pusillus doriferusduring the breeding season. B.Sc.(Hons.) thesis, University of Melbourne, Melbourne.

J:\cjz\cjz79\cjz-01\Z00-178.vpMonday, December 18, 2000 1:45:38 PM

Color profile: DisabledComposite Default screen

Page 11: Deakin Research Onlinedro.deakin.edu.au/eserv/DU:30015951/arnould-divebehaviour-2001.pdf · brook, Tex., U.S.A.). Upon capture, each animal was given an in-tramuscular injection (ca

the previously documented mean pupping date of 1 Decemberfor Australian fur seals (Warneke and Shaughnessy 1985).The subsequent attendance period (8.4 days) is similar to theperinatal period reported for several fur seal species (Boyd1991). Interestingly, the three instrumented females (20%)that did not undertake extended foraging trips during springalso showed no increase in attendance periods during thebreeding season and may have failed to pup that year. Thisis consistent with the observed pregnancy rate of 79% inCape fur seals (Guinet et al. 1998).

The only previous data on maternal-attendance patternsfor subspecies ofA. pusillusis during summer for Cape furseals and Australian fur seals in southern Bass Strait (Davidand Rand 1986; Hindell et al. 1998). The mean summerforaging-trip duration and attendance period of 3.7 and1.7 days, respectively, observed in the present study arewithin the ranges reported in these earlier studies, 1.3–4.0and 1.2–2.8 days, respectively.

Foraging-trip durations differed significantly betweenseasons,being shortest in summer and longest in winter. A

© 2001 NRC Canada

44 Can. J. Zool. Vol. 79, 2001

Fig. 7. Density plots of at-sea locations of female Australian fur seals, derived from geolocation, during summer (four seals, 48 loca-tions) (a), autumn (five seals, 134 locations) (b), winter (seven seals, 97 locations) (c), and spring (three seals, 50 locations) (d). Theplots are based on the number of locations per 1° grid expressed as a percentage of the total number of locations obtained during thatseason.

J:\cjz\cjz79\cjz-01\Z00-178.vpMonday, December 18, 2000 1:46:09 PM

Color profile: DisabledComposite Default screen

Page 12: Deakin Research Onlinedro.deakin.edu.au/eserv/DU:30015951/arnould-divebehaviour-2001.pdf · brook, Tex., U.S.A.). Upon capture, each animal was given an in-tramuscular injection (ca

similar trend has been observed in subantarctic fur seals atAmsterdam Island (Georges and Guinet 2000), where in-creases in foraging-trip duration between summer and winterare associated with increases in foraging range and divingeffort, indicating a reduction in local food availability (Geor-ges and Guinet 2000; Georges et al. 2000a, 2000b). In con-trast, the data on at-sea locations and diving behaviour forfemale Australian fur seals in the present study do not indi-cate an increase in foraging range or effort associated withthe observed increase in trip durations throughout lactation.

The duration of the maternal-attendance period in femaleAustralian fur seals did not differ significantly with stage oflactation or season. This is similar to the pattern observed innumerous fur seal species (David and Rand 1986; Gentryand Holt 1986; Goldsworthy 1992; Lunn 1993; Lea andHindell 1997; Georges and Guinet 2000). It has been sug-gested that the lack of change in duration of attendance peri-ods reflects the increasing sucking efficiency of the growingpups, which enables the females to increase the rate of trans-fer of the larger volumes of milk delivered as lactation pro-gresses (Gentry et al. 1986a; Georges and Guinet 2000). Aswith other species where trip duration increases but durationof attendance periods remains constant, the proportion oftime spent at sea by lactating Australian fur seals increasedthroughout lactation.

Despite the absence of significant variation in duration ofattendance periods throughout lactation, there was a positivecorrelation between an individual’s mean foraging-trip dura-tion and its mean attendance period. Similar observationshave been made in several other fur seal species (Boyd et al.1991; Lea and Hindell 1997; Francis et al. 1998; Georgesand Guinet 2000). These relationships are consistent withthe finding of Arnould and Boyd (1995) that females makinglonger trips deliver greater amounts of milk to their pup.

Dive behaviour and foraging effortThe most striking features of the dive behaviour of female

Australian fur seals in this study are the long periods of con-tinual diving that commenced soon after the seals left thecolony and the lack of regular diel variation in dive depths.In addition, diving occurred at all times (although it wasmore common during daylight hours in winter). Further-more, the majority of dives were U-shaped in profile. Thelack of variability in depth between sequential U-shapeddives strongly suggests that they were benthic in nature. Thisis supported by the observation that most dives were to 70–85 m, which corresponds to the prevailing depths in BassStrait (Murray and Parslow 1999). While benthic dives werethe most common, almost a quarter of all dives were V-shaped. This is consistent with the observations of Gales andPemberton (1994) that Australian fur seals feed on solitaryinshore species and pelagic schooling species known to in-habit both pelagic and benthic regions.

The characteristic short, shallow benthic dives made justprior to hauling out at the colony observed during many for-aging trips made by Australian fur seals may indicate predator-avoidance behaviour rather than reflect foraging activity.Great white sharks (Carcharodon carcharias) are the mainpredator of Australian fur seals (Warneke 1982) and areknown to use the silhouettes of seals at the surface as hunt-ing cues (Riedman 1990). The depth of the water around

Kanowna Island increases sharply within a short distanceand it is possible that the Australian fur seals in this studymade their final approach to the colony hugging the seafloorto avoid potential detection by sharks. A similar explanationhas been suggested for the behaviour observed in northernelephant seals (Mirounga angustirostris) leaving colonies inCalifornia (Le Boeuf et al. 1988).

The only previously available information on the dive be-haviour of Australian fur seals comes from a single adultmale observed in winter along the east coast of Tasmania(Hindell and Pemberton 1997) and six lactating females insummer (Hindell et al. 1998) in southern Bass Strait. In bothstudies there was no evidence of a diel pattern in diving ac-tivity or dive depth. Whereas the mean depth of dives madeby the adult male was only 20 m, 85% of dives had the U-shaped profile observed in the present study, which suggestsa benthic mode of foraging. Hindell et al. (1998) found thatnearly 50% of the dives made by lactating females in thesummer were to <10 m and suggested that such dives wereassociated with travelling rather than foraging behaviour.When these shallow dives are excluded, the majority of fe-males in their study dived to between 44 and 69 m, whichcorresponds to the prevailing bathymetry where the majorityof at-sea locations were obtained (Hindell et al. 1998) andsuggests that these females were also foraging mostly on thebenthos.

The predominantly benthic dive behaviour of female Aus-tralian fur seals contrasts with that of most other fur sealspecies, in which the majority of foraging dives made by fe-males are pelagic, are performed at night in well-definedbouts, and there is regular diel variation in dive depth(Kooyman and Trillmich 1986a; Trillmich et al. 1986; Goebelet al. 1991; Boyd and Croxall 1992; Harcourt et al. 1995;Francis et al. 1998; Georges et al. 2000b). The diel variationin dive depth observed in most fur seal species is thought toreflect the diel vertical migration of the species upon whichthey prey (Gentry et al. 1986a; Boyd et al. 1994). Only insome individual northern fur seals (“deep divers”) has a div-ing pattern with little diel variation in depth or frequencybeen observed, and this is also assumed to reflect a benthicforaging mode (Gentry et al. 1986b; Goebel et al. 1991).The diving activity reported in this study, therefore, resem-bles more the typical foraging behaviour of sea lions thanthat of fur seals (e.g., Gales and Costa 1997; Gales andMattlin 1997; Merrick and Loughlin 1997; Thompson et al.1998). A potential explanation for the differences in femaleforaging behaviour between the Australian fur seal and otherfur seal species is the lower quality of their foraging habitat.Indeed, California sea lions (Zalophus californianus) forag-ing in the productive continental shelf-edge region near SanMiguel Island, California, also display dive behaviour typi-cal of fur seals(Feldkamp et al. 1989). Alternatively, it ispossible that the shallow continental-shelf habitat of BassStrait supports an abundance of benthic prey easily accessi-ble to Australian fur seals.

The mean dive duration for female Australian fur seals(2.0–3.7 min) is greater than for most fur seal species (0.8–2.2 min) (Gentry et al. 1986b; Boyd et al. 1991; Harcourt etal. 1995; Francis et al. 1998; Georges et al. 2000b). Whilethe larger body size of Australian fur seals should allowgreater aerobic capacity and, therefore, potentially accounts

© 2001 NRC Canada

Arnould and Hindell 45

J:\cjz\cjz79\cjz-01\Z00-178.vpMonday, December 18, 2000 1:46:10 PM

Color profile: DisabledComposite Default screen

Page 13: Deakin Research Onlinedro.deakin.edu.au/eserv/DU:30015951/arnould-divebehaviour-2001.pdf · brook, Tex., U.S.A.). Upon capture, each animal was given an in-tramuscular injection (ca

for this difference (Gentry et al. 1986a), the theoretical ADLwas exceeded on an average of 17.3% of dives, a proportionthat is substantially greater than that reported for most otherfur seal species (<10%; Gentry et al. 1986b; Harcourt et al.1995; Boyd and Croxall 1992). Only in New Zealand furseals (Arctocephalus forsteri) from Open Bay Islands has agreater mean dive duration (4.2 min) been recorded in a furseal, and in these animals the theoretical ADL was exceededon 18.4% of dives (Mattlin et al. 1998).

The mean dive frequency recorded in the present study(8.3/h) is within the range previously observed in fur seals(1.5–18.1/h) (Gentry et al. 1986b; Kooyman and Trillmich1986a; Boyd and Croxall 1992; Goldsworthy et al. 1997;Georges et al. 2000b). However, the mean rate at which ver-tical distance was covered by female Australian fur sealsduring foraging trips (997 m·h–1) is the highest recorded forany fur seal species (46–460 m·h–1), including the Cape furseal (243 m·h–1), but similar to that observed in the Califor-nia sea lion (1014 m·h–1) and Steller sea lion (735 m·h–1)(Gentry et al. 1986b; Kooyman and Trillmich 1986a; Feldkampet al. 1989; Boyd and Croxall 1992; Goldsworthy et al.1997; Merrick and Loughlin 1997; Francis et al. 1998; Geor-ges et al. 2000b). Similarly, the proportion of time spent div-ing by females in the present study (41%) is substantiallygreater than in other fur seal species (14–26%) but withinthe range reported for sea lions (33–60%). These observa-tions may reflect the different foraging modes (i.e., long pe-riods of continual diving versus diving in bouts at night) anddifferences in the time between departure from the colonyand commencement of foraging (0.5–3.4 vs. 15–48 h) be-tween Australian fur seals (and sea lions) and other fur sealspecies (Gentry et al. 1986b; Kooyman and Trillmich 1986a;Feldkamp et al. 1989; Boyd and Croxall 1992; Goldsworthyet al. 1997; Merrick and Loughlin 1997; Francis et al. 1998;Georges et al. 2000a, 2000b). Weak but significant negativerelationships between foraging trip duration and measures offoraging effort were found in the present study. Similar rela-tionships have been observed in Antarctic fur seals, and ithas been suggested that females of this species making lon-ger trips may spend more time and energy searching forfood patches but are better at exploiting the patches, or thatthe patches are of higher quality, than females making shorttrips (Boyd et al. 1991; Arnould et al. 1996).

Foraging locationsThe relatively small number of seals in the study and the

poor spatial resolution of the geolocation techniques meanthat the data presented here can only be regarded as a pre-liminary examination of foraging locations in this species.Nonetheless, two important trends are apparent.

The first is that lactating females spend almost all of theirtime in the relatively shallow waters of Bass Strait and con-centrate their time in waters within 100 km of the colony.The second generalisation that can be made is that some fe-males do leave Bass Strait during a postlactation foragingtrip. In this study, only the western edge of Bass Strait andthe west coast of Tasmania were utilised at this time. This isan area of considerable oceanic upwelling (Lewis 1981) andmay offer better foraging to the seals. Other marine preda-tors, such as the Shy Albatross (Diomedea cauta), also pref-erentially use this area (Brothers et al. 1998). If this is the

case, however, it begs the question of why more females donot make use of this area, as it is certainly within the poten-tial foraging range of females from Kanowna travelling at3 m·s–1. Other fur seal species make much longer trips totheir foraging grounds (e.g., subantarctic fur seals), but inthese cases the waters that they traverse are relatively unpro-ductive. The patterns of at-sea locations in our study suggestthat the benthic and pelagic regions of Bass Strait providesufficient resources to lactating females, although certaincharacteristics of the environment require the seals to adoptforaging patterns different from those of other fur seal species.

ConclusionsFemale Australian fur seals from Kanowna Island in central

northern Bass Strait undertake foraging trips of increasingduration as lactation progresses, being shortest in summerand longest in winter. Whether this a response to increasingpup demand and (or) reduced food availability during thewinter months is not possible to determine at this stage. Theduration of maternal foraging trips is similar to that ob-served in other fur seal species. However, the foraging be-haviour of female Australian fur seals, with its characteristiclong periods of continual diving and lack of diel variation indive depth, is unlike that observed in other fur seal species.The majority of maximum depths recorded (<90 m) corre-spond to the maximum depth in Bass Strait, suggesting thatthese animals forage on the benthos of the shallow continen-tal shelf in this region. This is supported by the profile of thedives and the preliminary at-sea locations obtained by geo-location, which indicate that these animals forage almostexclusively within Bass Strait in all seasons. Whether thisunusual foraging behaviour, reminiscent of that of some sealion species, is representative of the species, or is restrictedto colonies in central Bass Strait is not known. Analysis ofthe dive behaviour and foraging locations of female Austra-lian fur seals from colonies close to the continental shelfedge is needed to clarify this. In addition, at-sea energy ex-penditure and foraging efficiency should be compared amongAustralian fur seal colonies and with other species of furseals (in particular the closely related Cape fur seal) and sealions to investigate whether this dive behaviour is an adapta-tion for successfully exploiting benthic prey or to a marginalforaging environment.

Acknowledgements

This research was supported with funds from the AustralianResearch Council and the Sea World Research and RescueFoundation Inc. Logistic support was generously providedby Parks Victoria and fieldwork was conducted under De-partment of Natural Resources and Environment (Victoria)research permits 978/003, RP-97-112, and 10000187. Theassistance of many fieldworkers, particularly C.L. Littnanand A.D. Morrissey, is gratefully acknowledged.

References

Arnould, J.P.Y., and Boyd, I.L. 1995. Temporal patterns of milkproduction in Antarctic fur seals (Arctocephalus gazella). J. Zool.(Lond.), 237: 1–12.

© 2001 NRC Canada

46 Can. J. Zool. Vol. 79, 2001

J:\cjz\cjz79\cjz-01\Z00-178.vpMonday, December 18, 2000 1:46:10 PM

Color profile: DisabledComposite Default screen

Page 14: Deakin Research Onlinedro.deakin.edu.au/eserv/DU:30015951/arnould-divebehaviour-2001.pdf · brook, Tex., U.S.A.). Upon capture, each animal was given an in-tramuscular injection (ca

Arnould, J.P.Y., and Littnan, C.L. 2000. Pup production and breedingareas of Australian fur seals (Arctocephalus pusillus doriferus)at Kanowna Island and The Skerries in northeastern Bass Strait.Aust. Mammal.22: 51–55.

Arnould, J.P.Y., Boyd, I.L., and Speakman, J.R. 1996. The relation-ship between foraging behaviour and energy expenditure in Ant-arctic fur seals. J. Zool. (Lond.),239: 769–782.

Bonner, W.N. 1984. Lactation strategies in pinnipeds: problems fora marine mammalian group. Symp. Zool. Soc. Lond. No. 51.pp. 253–272.

Boyd, I.L. 1991. Environmental and physiological factors controllingthe reproductive cycles of pinnipeds. Can. J. Zool.69: 1135–1148.

Boyd, I.L., and Croxall, J.P. 1992. Diving behavior of lactatingAntarctic fur seals. Can. J. Zool.70: 919–928.

Boyd, I.L., Lunn, N.J., Duck, C.D., and Barton, T. 1990. Response ofAntarctic fur seals to immobilization with ketamine, a ketamine–diazepam or ketamine–xylazine mixture, and Zoletil®. Mar. Mamm.Sci. 6: 135–145.

Boyd, I.L., Lunn, N.J., and Barton, T. 1991. Time budgets and for-aging characteristics of lactating Antarctic fur seals. J. Anim.Ecol. 60: 577–592.

Boyd, I.L., Arnould, J.P.Y., Barton, T., and Croxall, J.P. 1994. For-aging behaviour of Antarctic fur seals during periods of con-trasting prey abundance. J. Anim. Ecol.63: 703–713.

Brothers, N., Gales, R., Hedd, A., and Roberston, G. 1998. For-aging movements of the Shy albatrossDiomedea cautabreedingin Australia—implications for interactions with longline fisher-ies. Ibis,140: 446–457.

Butterworth, D.S., Punt, A.E., Oosthuizen, W.H., and Wickens,P.A. 1995. The effects of future consumption by the Cape furseal on catches and catch rates of the Cape hakes 3: modellingthe dynamics of the Cape fur sealArctocephalus pusillus pusillus.S. Afr. J. Mar. Sci.16: 161–183.

David, J.H.M., and Rand, R.W. 1986. Attendance behavior of SouthAfrican fur seals.In Fur seals: maternal strategies on land and atsea.Edited byR.L. Gentry and G.L. Kooyman. Princeton Uni-versity Press, Princeton, N.J. pp. 126–141.

Feldkamp, S.D., DeLong, R.L., and Antonelis, G.A. 1989. Divingpatterns of California sea lions,Zalophus californianus. Can. J.Zool. 67: 872–883.

Francis, J., Boness, D., and Ochoa-Acuna, H. 1998. A protractedforaging and attendance cycle in female Juan Fernandez furseals. Mar. Mamm. Sci.14: 552–574.

Gales, N.J., and Costa, D.P. 1997. The Australian sea lion: a re-view of an unusual life history.In Marine mammal research inthe southern hemisphere. Vol. 1. Status, ecology and medicine.Edited byM. Hindell and C. Kemper. Surrey Beatty and Sons,Chipping Norton, Australia. pp. 78–87.

Gales, N.J., and Mattlin, R.H. 1997. Summer diving behaviour oflactating New Zealand sea lions,Phocarctos hookeri. Can. J.Zool. 75: 1695–1706.

Gales, R., and Pemberton, D. 1994. Diet of the Australian fur sealin Tasmania. Aust. J. Mar. Freshwater Res.45: 653–664.

Gentry, R.L., and Holt, J.R. 1982. Equipment and techniques forhandling northern fur seals. Tech. Rep. NMFS SSRF 758, Na-tional Oceanic and Atmospheric Administration, Seattle, Wash.

Gentry, R.L., and Holt, J.R. 1986. Attendance behavior of northernfur seals.In Fur seals: maternal strategies on land and at sea.Edited byR.L. Gentry and G.L. Kooyman. Princeton UniversityPress, Princeton, N.J. pp. 41–60.

Gentry, R.L., and Kooyman, G.L. 1986. Fur seals: maternal strate-gies on land and at sea. Princeton University Press, Princeton,N.J.

Gentry, R.L., Costa, D.P., Croxall, J.P., David, J.H.M., Davis, R.W.,Kooyman, G.L., Majluf, P., McCann, T.S., and Trillmich, F.1986a. Synthesis and conclusions.In Fur seals: maternal strategieson land and at sea.Edited byR.L. Gentry and G.L. Kooyman.Princeton University Press, Princeton, N.J. pp. 220–264.

Gentry, R.L., Kooyman, G.L., and Goebel, M.E. 1986b. Feedingand diving behavior of northern fur seals.In Fur seals: maternalstrategies on land and at sea.Edited byR.L. Gentry and G.L.Kooyman. Princeton University Press, Princeton, N.J. pp. 61–78.

Georges, J.-Y., and Guinet, C. 2000. Maternal care in the subantarcticfur seals on Amsterdam Island. Ecology,81: 295–308.

Georges, J.-Y., Bonadonna, F., and Guinet, C. 2000a. Foraginghabitat and diving activity of lactating subantarctic fur seals inrelation to sea surface temperatures at Amsterdam Island. Mar.Ecol. Prog. Ser.196: 291–304.

Georges, J.-Y., Tremblay, Y., and Guinet, C. 2000b. Seasonal div-ing behaviour in lactating subantarctic fur seals on AmsterdamIsland. Polar Biol.23: 59–69.

Goebel, M.E., Bengtson, J.L., DeLong, R.L., Gentry, R.L., andLoughlin, T.R. 1991. Diving patterns and foraging locations offemale northern fur seals. Fish. Bull.89: 171–179.

Goldsworthy, S.D. 1992. Maternal care in three species of southernfur seals (Arctocephalusspp.). Ph.D. thesis, Monash University,Melbourne.

Goldsworthy, S.D., Hindell, M.A., and Crowley, H.M. 1997. Dietand diving behaviour of sympatric fur sealsArctocephalus gazellaand A. tropicalis at Macquarie Island.In Marine mammal re-search in the southern hemisphere. Vol. 1. Status, ecology andmedicine.Edited byM. Hindell and C. Kemper. Surrey Beattyand Sons, Chipping Norton, Australia. pp. 151–163.

Green, K. 1997. Diving behaviour of Antarctic fur sealsArcto-cephalus gazellaPeters around Heard Island.In Marine mam-mal research in the southern hemisphere. Vol. 1. Status, ecologyand medicine.Edited by M. Hindell and C. Kemper. SurreyBeatty and Sons, Chipping Norton, Australia. pp. 97–104.

Guinet, C., Roux, J.-P., Bonnet, M., and Mison, V. 1998. Effect ofbody size, body mass, and body condition on reproduction offemale South African fur seals (Arctocephalus pusillus) inNamibia. Can. J. Zool.76: 1418–1424.

Harcourt, R.G., Schulman, A.M., Davis, L.S., and Trillmich, F.1995. Summer foraging by lactating female New Zealand furseals (Arctocephalus forsteri) off Otago Peninsula, New Zea-land. Can. J. Zool.73: 678–690.

Higgins, L.V., and Gass, L. 1993. Birth to weaning: parturition, du-ration of lactation, and attendance cycles of Australian sea lions(Neophoca cinerea). Can. J. Zool.71: 2047–2055.

Hill, R.D. 1994. Theory of geolocation by light levels.In Elephantseals: population ecology, behaviour and physiology.Edited byB.J. Le Boeuf and R.M. Laws. University of California Press,Berkeley. pp. 227–236.

Hindell, M.A., and Pemberton, D. 1997. Successful use of a trans-location program to investigate diving behavior in a male Aus-tralian fur seal,Arctocephalus pusillus doriferus. Mar. Mamm.Sci. 13: 219–229.

Hindell, M.A., McCarrey, S., and Pemberton, D. 1998. Foragingecology of Australian fur seals in Bass Strait, and an assessmentof the potential for interactions with commercial fisheries. Re-port to Environment Australia from the Antarctic Wildlife Re-search Unit, School of Zoology, University of Tasmania, Hobart.

Kooyman, G.L., and Trillmich, F. 1986a. Diving behavior ofGalapagos furseals.In Fur seals: maternal strategies on landand at sea.Edited byR.L. Gentry and G.L. Kooyman. PrincetonUniversity Press, Princeton, N.J. pp. 186–195.

© 2001 NRC Canada

Arnould and Hindell 47

J:\cjz\cjz79\cjz-01\Z00-178.vpMonday, December 18, 2000 1:46:11 PM

Color profile: DisabledComposite Default screen

Page 15: Deakin Research Onlinedro.deakin.edu.au/eserv/DU:30015951/arnould-divebehaviour-2001.pdf · brook, Tex., U.S.A.). Upon capture, each animal was given an in-tramuscular injection (ca

© 2001 NRC Canada

48 Can. J. Zool. Vol. 79, 2001

Kooyman, G.L., and Trillmich, F. 1986b. Diving behavior of Galapagossea lions.In Fur seals: maternal strategies on land and at sea.Edited byR.L. Gentry and G.L. Kooyman. Princeton UniversityPress, Princeton, N.J. pp. 209–219.

Lea, M.-A., and Hindell, M.A. 1997. Pup growth and maternal carein New Zealand fur seals,Arcocephalus forsteri, at MaatsuykerIsland, Tasmania. Wildl. Res.24: 307–318.

Le Boeuf, B.J., Costa, D.P., Huntley, A.C., and Feldkamp, S.D.1988. Continuous, deep diving in female northern elephant seals,Mirounga angustirostris. Can. J. Zool.66: 446–458.

Lewis, R.K. 1981. Seasonal upwelling along the south-eastern coast-line of South Australia. Aust. J. Mar. Freshwater Res.32: 843–854.

Lunn, N.J. 1993. The reproductive ecology of female Antarctic furseals (Arctocephalus gazella) during lactation. Ph.D. thesis, OpenUniversity, Milton Keynes, England.

Lunn, N.J., and Boyd, I.L. 1993. Influence of maternal characteris-tics and environmental variation on reproduction in Antarctic furseals. Symp. Zool. Soc. Lond. No. 66. pp. 115–129.

Mattlin, R.H., Gales, N.J., and Costa, D.P. 1998. Seasonal dive behav-iour of lactating New Zealand fur seals (Arctocephalus forsteri).Can. J. Zool.76: 350–360.

McConnell, B.J., Chambers, C., and Fedak, M.A. 1992. Foragingecology of southern elephant seals in relation to the bathymetryand productivity of the Southern Ocean. Antarct. Sci.4: 393–398.

Merrick, R.L., and Loughlin, T.R. 1997. Foraging behavior of adultfemale and young-of-the-year Steller sea lions in Alaskan wa-ters. Can. J. Zool.75: 776–786.

Murray, A.G., and Parslow, J.S. 1999. Modelling of nutrient im-pacts in Port Phillip Bay—a semi-enclosed marine Australianecosystem. Mar. Freshwater Res.50: 597–611.

Pemberton, D., and Kirkwood, R.J. 1994. Pup production and distribu-tion of the Australian fur seal,Arctocephalus pusillus doriferusin Tasmania. Wildl. Res.21: 341–352.

Rand, R.W. 1955. Reproduction in the female cape fur seal,Arcto-

cephalus pusillus(Schreber). Proc. Zool. Soc. Lond. No. 124.pp. 717–740.

Riedman, M. 1990. The pinnipeds: seals, sea lions, and walruses.University of California Press, Berkeley.

Shaughnessy, P.D., and Warneke, R.M. 1987. Australian fur seal,Arctocephalus pusillus doriferus. In Status, Biology, and Ecol-ogy of Fur Seals: Proceedings of an International Symposiumand Workshop, Cambridge, England, 23–27 April 1984.Editedby J.P. Croxall and R.L. Gentry. NOAA Tech. Rep. NMFS 51,Seattle, Wash. pp. 73–77.

Thompson, D., Duck, C.D., McConnell, B.J., and Garrett, J. 1998.Foraging behaviour and diet of lactating female southern sea lions(Otaria flavescens) in the Falkland Islands. J. Zool. (Lond.),246: 135–146.

Trillmich, F. 1986. Attendance behavior of Galapagos fur seals.InFur seals: maternal strategies on land and at sea.Edited byR.L.Gentry and G.L. Kooyman. Princeton University Press, Prince-ton, N.J. pp. 168–185.

Trillmich, F. 1996. Parental investment in pinnipeds. Adv. StudyBehav.25: 533–577.

Trillmich, F., Kooyman, G.L., Majluf, P., and Sanchez-Griñan, M.1986. Attendance and diving behavior of South American furseals during El Niño in 1983.In Fur seals: maternal strategieson land and at sea.Edited byR.L. Gentry and G.L. Kooyman.Princeton University Press, Princeton, N.J. pp. 153–167.

Warneke, R.M. 1982. The distribution and abundance of seals inthe Australasian region, with summaries of biology and currentresearch. FAO Fish. Ser. No. 5, Vol. 4. pp. 431–475.

Warneke, R.M., and Shaughnessy, P.D. 1985.Arctocephalus pusillus,the South African and Australian fur seal: taxonomy, evolution,biogeography, and life history.In Studies of sea mammals insouth latitudes.Edited byJ.K. Ling and M.M. Bryden. SouthAustralian Museum, Adelaide. pp. 53–77.

J:\cjz\cjz79\cjz-01\Z00-178.vpMonday, December 18, 2000 1:46:12 PM

Color profile: DisabledComposite Default screen

Page 16: Deakin Research Onlinedro.deakin.edu.au/eserv/DU:30015951/arnould-divebehaviour-2001.pdf · brook, Tex., U.S.A.). Upon capture, each animal was given an in-tramuscular injection (ca