Using sea-ice to calibrate a dynamic trophic model for the Western Antarctic Peninsula

The pelagic ecosystems of the Western Antarctic Peninsula are dynamic and changing rapidly in the face of sustained warming. There is already evidence that warming may be impacting the food web. Antarctic krill, Euphausia superba, is an ice-associated species that is both an important prey item and the target of the only commercial fishery operating in the region. The goal of this study is to develop a dynamic trophic model for the region that includes the impact of the sea-ice regime on krill and krill predators. Such a model may be helpful to fisheries managers as they develop new management strategies in the face of continued sea-ice loss. A mass balanced food-web model (Ecopath) and time dynamic simulations (Ecosim) were created. The Ecopath model includes eight currently monitored species as single species to facilitate its future development into a model that could be used for marine protected area planning in the region. The Ecosim model is calibrated for the years 1996–2012. The successful calibration represents an improvement over existing Ecopath models for the region. Simulations indicate that the role of sea ice is both central and complex. The simulations are only able to recreate observed biomass trends for the monitored species when metrics describing the sea-ice regime are used to force key predator-prey interactions, and to drive the biomasses of Antarctic krill and the fish species Gobionotothen gibberifrons. This model is ready to be used for exploring results from sea-ice scenarios or to be developed into a spatial model that informs discussions regarding the design of marine protected areas in the region.


observations, regurgitated pellet analysis Cephalopods
Literature synthesis, gut contents, fatty acid analysis Low Rodhouse and Nigmatullin [24], Phillips et al. [25], Kozlov [26] Myctophids (Off shelf) Gut contents Low Pusch et al. [27], Pakhomov et al. [28] On-shelf fish Literature synthesis, Low La Mesa et al. [29], Barrera-Oro [30] N. rossii Gut contents Low Casaux and Barrera-Oro [31], Jones et al. [ [56] . The study type column provides a gross description of the field methods used to collect diet data. The confidence column relates to the confidence in the percentages of prey items reported in each study being precise for all predators of that model group in the region. Many of the diet studies referenced have small, spatially constrained sample sizes relative to their respective populations and therefore the percentage of mass in the diet was uncertain.
No detailed, regionally specific diet studies exist for the cetacean species included in the model. Killer whale (Orcinus orca) is a species that is composed of several ecotypes distinguished, inter alia, by diet [1,2]. At least two distinct ecotypes of killer whale occur in the WAP. Ecotype A, which feeds on minke whales (Balaenoptera bonaerensis and B. acutorostrata) and to a lesser extent elephant seals (Mirounga leonina), and the more abundant Ecotype B, which feeds on pack ice seals and penguins [1,2]. Fish-eating killer whales may also occur in the region [2].  [12][13][14].
Published diet studies do not adequately describe the percentages of prey items consumed by two pinniped species. A diet study of leopard seal (Hydrurga leptonyx) in the region [3] describes a high percentage of krill and a variety of fish in this predator's diet. Observations from elsewhere in the study region indicate that consumption of Antarctic fur seal (Arctocephalus gazzella) pups by leopard seals is a significant source of pup mortality [4,5,57].
The diet used in the Ecopath model includes consumption of Antarctic fur seals. Three diet studies have been conducted on southern elephant seals in the region, and all three agree that fishes and cephalopods are dietary staples [10,58,59]. However, the studies do not assign proportion of diet to cephalopods or fishes. In the current study, the diet of southern elephant seal is described as favoring cephalopods, with a significant portion of the diet coming from both myctophids and on-shelf fishes.
The seabird functional group represents a diverse group (see S1). There is wide variation in the reported diets of seabirds, though all species feed primarily in the near-surface marine environment [21][22][23]. The diet reflects a synthesis of the literature to create a representative diet for this group.
The multispecies groups for cephalopods and on-shelf fish represent diverse species and diets. Cephalopods are known to be opportunistic foragers [24] and important consumers of myctophids and mesopelagic fishes [24,26], though percentages of diet composition could not be found in the literature. The diet for cephalopods is split between euphausiid and fish groups.
Electrona antarctica serves as the example diet for myctophids as it is the best documented and most abundant myctophid in the region [27,39].
The diet of Antarctic krill (Euphausia superba) varies both by season [40,42] and life stage [43,44]. The diets of large and small krill presented here take that variability into account and attempt to reflect an average annual diet for each krill group. We assume that the benthic invertebrate group largely consumes detritus that has reached the seafloor. This is consistent with other Antarctic food web models [38,39,60] The initial diet matrix compiled from the literature is provided below. This diet matrix was incrementally adjusted to bring the model into balance.