A foodweb model to explore uncertainties in the South Georgia shelf pelagic ecosystem

Foodweb models provide a useful framework for compiling data on biomass, production, consumption and feeding relationships. They are particularly useful for identifying gaps and inconsistencies in the data, and for exploring plausible scenarios of change. We compiled data on the pelagic foodweb of the South Georgia shelf, which is one of the most intensively studied areas in the Southern Ocean. The data suggest that current average annual copepod production is three times that of Antarctic krill and that flying seabirds and fish are, respectively, responsible for 25% and 21% of local krill consumption. The most striking inconsistency was that estimated consumption of fish was 5 times their estimated production. We developed a static mass balance model of the foodweb representing one of many possible solutions to the inconsistencies in the data. The model included sufficient fish biomass to balance the original consumption estimate, and consequently fish became the main krill consumers. Nonetheless, only 74% of local krill production was consumed by predators, suggesting that there are additional mortality sources that we did not explicitly model. We developed further models to explore scenarios incorporating plausible climate-driven reductions in krill biomass. In scenarios with unchanged predator diets, an 80% reduction in krill biomass resulted in a 73% reduction in vertebrate biomass. However, when predators with diverse diets were able to switch to feeding on alternative zooplankton prey, total vertebrate biomass was maintained at current levels. Scenarios in which 80% of krill biomass was replaced with copepod biomass required 28% more primary production because the estimated consumption rate of copepods is higher than that of krill. The additional copepod biomass did not alter the consequences for vertebrates. These scenarios illustrate the wide range of potential consequences of a shift from a krill to a copepod dominated system in a warming climate. They suggest that both maintenance and dramatic reduction of vertebrate production are plausible outcomes, although the former requires major changes in predator diets.