VARIATION IN THE FORAGING LOCATION OF ANTARCTIC FUR SEALS (ARCTOCEPHALUS GAZELLA) AND THE EFFECTS ON DIVING BEHAVIOR

This study investigated how female Antarctic fur seals adapt their foraging behavior, over time scales of days, to spatial unpredictability in the distribution of their food. Lactating Antarctic fur seals are central-place foragers that feed on highly patchy but spatially and temporally dynamic food. We measured the foraging distribution of 28 fur seals to test whether variation in foraging trip durations was reflected in variation in the location of foraging and the diving behavior of seals at sea. Based on the maximum distance travelled from the breeding beach, three categories of foraging trips were denned: those to the continental shelf area ( n = 12, median = 71 km), to oceanic water ( n = 11, median =164 km), and to farther offshore oceanic waters ( n = 5, median = 260 km). Trip duration and mean surface speed were positively correlated with the maximum distance travelled from the breeding beach. Seals on longer trips spent proportionally less of their time submerged, but there was no significant difference in the total number of dives or the total time spent foraging by seals in relation to trip duration. Evidence from this study and previous work investigating energy gain suggests that an animal on a longer foraging trip could potentially have a higher mean energy return per dive than a similar animal on a shorter foraging trip. Evidence presented suggests that the type of foraging trip (near or far) is not predetermined by the animal but may be a simple response to the stochastic distribution of the resources available.     

POPULATION SIZE AND BREEDING SEASON OF THE ANTARCTIC FUR SEAL ARCTOCEPHALUS GAZELLA AT HEARD ISLAND—1987/88

Breeding colonies of the antarctic fur seal Arctocephalus gazella on Heard Island (53°10'S, 73°30'E) are situated on the sheltered northern and eastern coasts on flat vegetated terrain near streams and pools. Pupping in the 1987/88 summer began on 21 November, with 90% of births in 26 d. The median birth date was 11 December. Pup counts at Heard Island made in seven breeding seasons from 1962/63 to 1987/88 show an exponential rate of increase of 21%, which may be inflated due to undercounting in early years. The total of 248 births in 1987/88 represents an exponential increase of 37% since the previous year, but pups may have been undercounted then. Based on the number of pups born, the breeding population is estimated at 870–1,120. During the breeding season, the largest number of animals ashore was 835. Many non-breeding fur seals began hauling out from early January and 15,000 animals were estimated to be ashore by late February, a fat larger number than expected from the size of the breeding population. Both the breeding and non-breeding components of the population may be augmented by immigration. The source of immigrants may be undiscovered breeding colonies of this species in the northwestern sector of the Kerguelen Archipelago or the concentration at South Georgia. Further censuses are required at Heard Island to monitor the population growth.     

AGING LIVE ANTARCTIC FUR SEALS AND SOUTHERN ELEPHANT SEALS

This study describes a method for extracting post-canine or incisor teeth from live antarctic fur seals ( Arctocephalus gazella ) and southern elephant seals ( Mirounga leonina ) respectively and their use to determine age in a field situation. Dental elevators were used to loosen the teeth from the alveolus and periodental ligament. Most teeth were removed within l-2 min and a total of 214 and 81 teeth were collected by this method from antarctic fur seals and southern elephant seals respectively. No seal recaptured at intervals up to a year after a tooth was extracted showed signs of infection or distress related to removal of the tooth. Teeth were thin-sectioned for the purpose of aging. In both species cementum growth layer groups were a more satisfactory indicator of age than dentinal growth layer groups. Estimates of age from cementum growth layers were confirmed for Antarctic fur seals using seals which had been tagged as pups up to 16 yr before sampling.     

GROWTH RATES OF ANTARCTIC FUR SEALS AS INDICES OF ENVIRONMENTAL CONDITIONS

The growth rates of Antarctic fur seal ( Arctocephalus gazella ) pups estimated from weighing cross-sections of the population were compared with measured/ inferred changes in the availability of their main prey species, Antarctic krill ( Euphausia superba ) from 1989 to 2000. There was no relationship between growth rate and mass at weaning and there were counter-intuitive indications of higher growth rates in years of low krill availability. Biases reflecting changes in the component of the population available for sampling appear to invalidate the widely held assumption that interannual differences in growth rate can reliably be derived from differences in the slope of a linear relationship based on cross-sectional population samples. A new index was developed, based on the deviation of pup mass at age in each year compared to the multiyear mean, that was not dependent on assumptions of linearity. The indices of growth deviates produced a more logical relationship with other indices of pup development and related more appropriately to variations in prey availability. The potential impact of methodological biases on the interpretation of growth rate suggests that comparisons of growth rates should not rely on assumptions regarding the underlying growth pattern.     

MONITORING THE PREY-FIELD OF MARINE PREDATORS: COMBINING DIGITAL IMAGING WITH DATALOGGING TAGS

There is increasing interest in the diving behavior of marine mammals. However, identifying foraging among recorded dives often requires several assumptions. The simultaneous acquisition of images of the prey encountered, together with records of diving behavior will allow researchers to more fully investigate the nature of subsurface behavior. We tested a novel digital camera linked to a time-depth recorder on Antarctic fur seals ( Arctocephalus gazella ). During the austral summer 2000–2001, this system was deployed on six lactating female fur seals at Bird Island, South Georgia, each for a single foraging trip. The camera was triggered at depths greater than 10 m. Five deployments recorded still images (640 × 480 pixels) at 3-sec intervals (total 8,288 images), the other recorded movie images at 0.2-sec intervals (total 7,598 frames). Memory limitation (64 MB) restricted sampling to approximately 1.5 d of 5–7 d foraging trips. An average of 8.5% of still pictures (2.4%-11.6%) showed krill ( Euphausia sulperba ) distinctly, while at least half the images in each deployment were empty, the remainder containing blurred or indistinct prey. In one deployment krill images were recorded within 2.5 h (16 km, assuming 1.8 m/sec travel speed) of leaving the beach. Five of the six deployments also showed other fur seals foraging in conjunction with the study animal. This system is likely to generate exciting new avenues for interpretation of diving behavior.     

DIVE DATA FROM SATELLITE TAGS AND TIME-DEPTH RECORDERS: A COMPARISON IN WEDDELL SEAL PUPS

The diving behavior of juvenile Weddell seals, Leptonychotes weddellii , was monitored simultaneously with time-depth recorders (TDRs) and satellitelinked time-depth recorders (SLTDRs). Recovered TDRs provided a complete record of the depth and duration of all dives, while data received from SLTDR tags via the ARGOS satellite system were compressed into the number of dives in each of six depth or duration bins. The dive information from the two types of tags was compared to determine if data compression, processing, and transmission influenced the data received.
While only half of the dive data collected by TDRs was also received from the SLTDR tags, the chance of receiving SLTDR data was independent of when diving occurred, when data was transmitted, and the subsequent dive activity. In addition, the number of dives in each depth and duration bin was an accurate representation of the actual dive behavior. Therefore, SLTDR tags were judged to provide data qualitatively similar to that provided by TDRs. The accuracy of seal locations provided by Service ARGOS was estimated by comparison to Global Positioning System (GPS) locations, and the average position error found to be significantly greater than predicted by Service ARGOS or reported in other studies (LCO locations ± 11.4 km, LC1 ± 5.0 km).     

INDICES OF BODY CONDITION AND BODY COMPOSITION IN FEMALE ANTARCTIC FUR SEALS (ARCTOCEPHALUS GAZELLA)

An attempt was made to develop simple, inexpensive, rapid means of determining body composition in Antarctic fur seals ( Arctocephalus gazella ). Measurements of total body water ( TBW ) and total body lipid ( TBL ), obtained by hydrogen isotope dilution, were compared to the results of bioelectrical impedance analysis ( BIA ) and morphometric indices of body condition in 52 adult females. TBW was weakly correlated with BIA measurements of resistance ( v = -0.30, P < 0.03). Conductor volume (length2/resistance) was more highly correlated with TBW ( r = 0.75, P < 0.0001) and the inclusion of mass into the predictive equation improved the correlation further ( r = 0.95, P < 0.0001). A body condition index (mass/length) previously used in pinniped studies was positively correlated to TBL ( r = 0.77, P < 0.0001) validating its use as a relative index of condition. However, body mass alone was highly correlated to TBW ( r = 0.94, P < 0.0001) and appears to provide a simple, rapid means of estimating body composition in adult females. This technique may also be applicable to juvenile male Antarctic fur seals.     

SUCCESSFUL USE OF A TRANSLOCATION PROGRAM TO INVESTIGATE DIVING BEHAVIOR IN A MALE AUSTRALIAN FUR SEAL, ARCTOCEPHALUS PUSILLUS DORIFERUS

This study reports some of the first foraging behavior data collected for male fur seals. A nonbreeding male Australian fur seal, Arctocephalus pusillus doriferus , captured at a commercial salmon farm in southern Tasmania, Australia, was relocated 450 km from the site of capture. The animal was equipped with a geolocating time-depth recorder that recorded diving behavior and approximate location for the 14.4 d that it took the seal to travel down the east coast of Tasmania and be recaptured at the salmon farm. During its time at sea, the seal spent most of its time over the relatively shallow shelf waters. It spent 30% of its time ashore on a number of different haul-out sites. The deepest dive was 102 m and the maximum duration was 6.8 min. "Foraging" type dives made up 31.2% of the time at sea and had a median duration of 2.5 min and a median depth of 14 m. The seal performed these dives more commonly during the latter part of its time at sea, while it was on the east coast. Unlike other fur seal species studied to date, there was no evidence of a diurnal foraging pattern; it made dives at all times of the day and night.     

A PROTRACTED FORAGING AND ATTENDANCE CYCLE IN FEMALE JUAN FERNÁNDEZ FUR SEALS

Previous studies of fur seals suggest that the attendance patterns and consequent temporal patterning of energy transfer from mother to pup follows a latitudinal cline. While data from subpolar, tropical, and some temperate latitude species support the postulated cline, data for the temperate latitude Juan Fernández fur seal do not. Maternal foraging trips and associated visits ashore were the longest of all otariids studied to date. The first foraging trip postpartum averaged 10.2 d ( n = 51 females, range 1–22.5), foraging trips combined averaged 12.3 d ( n = 100, range 1.0-25.0), and visits ashore averaged 5.3 d ( n = 91, range 0.3-15.8) over the three seasons of study. Only duration of lactation was intermediate between subpolar and tropical strategies as predicted. Dive records suggest that these females feed almost exclusively at night at depths of less than 10 m and at distances of more than 500 km offshore. The prey species of this fur seal, primarily myctophids and squid, migrate to the surface at night and are patchily distributed. Foraging trip length varied between years in conjunction with shifts in seasurface temperature and type of prey consumed. We suggest that distribution of prey, irrespective of latitude, dictates foraging patterns of fur seals and leads to the exceptionally long foraging trips and visits ashore observed in this species.     

INDIVIDUALITY IN THE VOICE OF FUR SEAL FEMALES: AN ANALYSIS STUDY OF THE PUP ATTRACTION CALL IN ARCTOCEPHALUS TROPICALIS

Like most otariids species, the Subantarctic fur seal breeds on land in large, dense colonies. Pups are confronted by the long and repetitive absences of their mother throughout lactation. At each mother's return, pups have to find her among several hundreds of congeners. This recognition process mainly relies on acoustic signals. We performed an acoustic analysis on 125 calls from 20 females recorded during the 1999–2000 breeding season on Amsterdam Island (Indian Ocean). Ten variables were measured in both temporal and frequency domains. To find the acoustic parameters supporting individual signature, we assessed the differences between individuals using Kruskall-Wallis univariate analysis of variance. For each variable, we also calculated the potential of individuality coding (PIC) as the ratio between the between-individual coefficient of variation and the mean value of the within-individual coefficients of variation. We found that the frequency spectrum, the characteristics of the frequency modulation of the initial and middle part of the call and the call duration exhibit an important individual stereotypy (PIC values ranging between 1.5 and 3), whereas features relative to amplitude and the frequency modulation of the final part of the call are weakly individualized (PIC values between 1 and 1.2).     

BACULAR AND TESTICULAR GROWTH AND ALLOMETRY IN THE CAPE FUR SEAL ARCTOCEPHALUS P. PUSILLUS (OTARIIDAE)

The baculum in Arctocephalus p. pusillus reaches up to 14.1 cm in length, 13.5 g in mass, and 1.3 g/cm in density (= mass/length). A pubertal growth spurt occurs between 2 and 3 yr of age, when bacular length increases by 28%, mass by 124%, and density by 77%; concurrently, body length increases by 14%. A second, weaker spurt occurs at social maturity (9-10 yr of age). Testes grow most rapidly between 1 and 2 yr of age, when testicular length increases by 29%. After 3 yr of age, growth in bacular and testicular length slows, and bacular mass continues to increase approximately linearly. Bacular and testicular lengths average 6.8% and 3.4% (respectively) of body length in adults, compared with 9.9% and 5.7% in the promiscuous harp seal ( Pagophilus groenlandicus ). Bacular length, mass, and density, and testicular length, are positively allometric to body length over growth; bacular length is isometric to testicular length. Among animals of the same age, bacular length and mass are positively allometric to body length in young animals, with negative allometry or isometry thereafter; testicular length is isometric to body length in young animals and negatively allometric thereafter. Patterns of early growth and allometry of the baculum and testes are interpreted as adaptations for mating opportunities, years before territoriality is possible. The baculum and testes of adult Cape fur seals and other otariids are small compared with those of most phocids, because sperm competition among male otariids is weak.     

INVESTIGATING THE BIASES IN THE USE OF HARD PREY REMAINS TO IDENTIFY DIET COMPOSITION USING ANTARCTIC FUR SEALS (ARCTOCEPHALUS GAZELLA) IN CAPTIVE FEEDING TRIALS

The analysis of pinniped scats has been used to quantify their diet, using prey remains to identify species and to estimate the numbers and sizes of prey consumed. There are, however, potential biases involved with scat analysis and, for this method to be used successfully, these biases need to be quantified. Thirty-six Antarctic fur seals ( Arctocephalus gazella ) were fed meals of exclusively either fish, squid, or krill and their scats were collected and analyzed. The major sources of error in the analysis of prey remains from scats were the differential erosion and passage rate of items in relation to their size. However, using simple correction functions, such as those which model otolith erosion, it is possible to reduce these errors. Using plastic beads as dietary markers showed recovery rates were negatively related to their size. Larger squid beaks had lower recovery rates than smaller beaks, but there was no size-related bias in the recovery rates of krill carapaces. Only 33% of the squid beaks and 27% of the otoliths originally fed were recovered from the scats. Recovery rates were greater for squid (77%) and fish (50%) eye lenses and these structures gave a better estimate of numbers eaten. Differences found between experimentally derived and published regression equations (used to calculate prey sizes eaten from prey remains) highlights the need for regression equations based on local prey characteristics, if these are to be used with any success to describe the prey eaten.     

DEVELOPMENT OF HEMOGLOBIN, HEMATOCRIT, AND ERYTHROCYTE VALUES IN GALÁPAGOS FUR SEALS

We studied the ontogeny of hemoglobin concentration, hematocrit and erythrocyte counts in the Galapagos fur seal ( Arctocephalus galapagoensis , Heller 1904). Two hundred and fifty-three animals were sampled between the ages of 22 d and > 8 yr, of which 46 were adult females. Body mass increased steadily with age from 6.1 ± 1.2 kg in 1-mo-old pups ( n = 27) to 28.5 ± 3.3 kg in adult females. Even adult females increased in mass with age. Hemoglobin (Hb), hematocrit (Hct), and red blood cell (RBC) values all increased in a logarithmic fashion with age up to 2 yr. Blood values for pups were Hct: 35.5 ± 4.1%; Hb: 12.9 ± 1.3 g/dl; RBC: 4.1 ± 0.3·10⁶/μl. Half-year-old fur seals (Hct: 42.1 ± 3.2%; Hb: 15.7 ± 1.3 g/dl; RBC: 4.9 ± 0.5·10⁶/μl; n = 50) were the oldest age group to show significantly lower blood values than adult females ( P < 0.001 for all three parameters). Yearling blood values (Hct: 47.2 ± 3.6%; Hb: 17.3 ± 1.6 g/dl; RBC: 5.6 ± 0.4·10⁶/μl; n = 56) did not differ significantly from those of adult females ( P < 0.32; P < 0.26; P < 0.23, respectively). Blood values of adult females were lower than those of 2-yr-olds (Hct: 49.6 ± 2.4%; Hb: 18.5 ± 1.2 g/dl; RBC: 5.7 ± 0.3·10⁶/μl; n = 31). These differences were significant only for RBCs ( P < 0.003). Up to the age of 1 yr, age was the best predictor for blood values, thereafter mass tended to be a better predictor. Female juveniles between the ages of 150 and 600 d had higher blood values than same-age males. The relationship of blood value development to diving activity is briefly described and the results are compared to values of other marine mammals. Ontogeny is discussed in relation to the development of these blood values in terrestrial mammals.