Population growth of Antarctic fur seals: limitation by a top predator, the leopard seal?

In other research associated with this study, (e.g., Bengtson et al. 1994) we monitored foraging and reproductive performance of Antarctic fur seals at Seal Island. During the 9 yr of research, we observed no evidence of extreme food limitation in fur seals such as occurred in 1977/1978, 1983/1984, 1990/1991, and 1993/1994 when krill were scarce around South Georgia (Croxall et al. 1988, Brierley and Watkins 1996). Preliminary analyses of measures such as growth rates of pups and durations of maternal foraging trips (J. L. Bengtson, unpublished data) indicate that food availability to fur seals at Seal Island is typically comparable to or greater than that at South Georgia (Lunn et al. 1993, Boyd et al. 1994). Thus, bottom-up forces may be less strong or exert themselves less frequently in the Elephant Island vicinity than at South Georgia.

Although these observations do not provide conclusive evidence that predation is a major force limiting fur seal population growth and recovery in the South Shetland Islands, we believe that top-down limitation should be given a greater focus in working hypotheses that guide further research on this population.

Relevance to other vertebrate predator-prey systems. - Compared with the well-studied terrestrial systems of vertebrate predators and prey in the northern hemisphere, such as wolves and moose (Messier 1994), lynx and hares (Sinclair et al. 1993), and mustelids and rodents (Hahski and Henttonen 1996), very little is known of the quantitative nature of the relationship between leopard seals and fur seals. However, it is possible to make qualitative inferences about the primary features of this relationship, and about whether leopard seals are capable of regulating fur seal populations, by considering the historical dynamics of the fur seals' recovery from exploitation.

As we discussed earlier, the fur seal population of the South Shetland Islands has recovered rather slowly from overharvesting. Nevertheless, this population did grow and the growth does not appear to be decelerating (Bengtson et al. 1990, Aguayo and Torres 1993), even in the Elephant Island area (Table 3). Long-term, steady growth rules out certain conceptual models for the predator-prey relationship between leopard seals and fur seals. For example, a scenario like that depicted in Fig. 5a is unlikely to reflect the conditions of the recovery in the South Shetlands. In that scenario, strongly density-dependent predation mortality would be capable of maintaining the fur seal population in a "predator pit" at low densities (e.g., Pech et al. 1995). Decelerating growth would be expected as the recovering population increased toward a low equilibrium regulated by density-dependent predation ([E.sub.P]).

Messier (1994) suggested that strong density-dependent predation of moose by wolves occurred because moose are spatially predictable, available year-round, and alternative prey species are occasionally or seasonally rare. Fur seals share some of these traits, but not others. They are spatially predictable for [approximately]4 mo during the breeding and pup-rearing season. However, because leopard seals prey mainly on young-of-the-year fur seals, which tend to disperse after weaning, fur seals may not be a reliable year-round food source. Leopard seals have many alternatives to fur seals as prey, including other seal species, seabirds, fish, cephalopods, and krill (Siniff and Stone 1985). Therefore, the rate of predation on fur seals by leopard seals may be less strongly dependent on prey density than the rate of predation on moose by wolves. Because relatively few individual leopard seals may be responsible for most of the predation of fur seal pups in the South Shetland Islands (Rakusa-Suzsczewski and Sierakowski 1993; L. M. Hiruki et al., unpublished manuscript) the predation rate may even be inversely dependent on density; the per capita rate of predation may decline with increasing fur seal density (Pech et al. 1995). Fig. 5b depicts an example of such a scenario; leopard seal predators consume a substantial fraction of the net production, allowing fur seals to slowly increase to a level at which they are essentially limited by some resource such as food or space (equilibrium density [E.sub.P]). The growth rate of the fur seals would be modest and constant or slightly increasing (accelerating) at low to moderate densities.