Evaluating sensitivity of the stock assessment tool for the Antarctic krill fishery to seasonal trends in natural and fishing mortality

The central stock assessment tool for the Antarctic Krill fishery is a single-species model developed in 1996 named the Generalized Yield Model (GYM). This model mainly uses the proportion of recruits and their growth parameters to project fluctuations of the krill abundance under different levels of fishing pressure, maximizing harvest rates within the bounds precautionary harvest control rules. The GYM, and its implementation in R (Grym), provides a population model operating on a daily timesteps, allowing some input parameters with intra-seasonal mortalities, which permits modelling of within-year patterns in the population dynamics. We explored the effects of including within-year patterns in natural and fishing mortality on the harvest rates in krill, by incorporating the temporal presence of key predator species in main fishing areas, along with contemporary spatiotemporal trends of the fishing fleet. Our results show that the inclusion of intra-seasonal variations in these mortalities increased the precautionary yield. Fishing mortality had a greater effect on outputs than natural mortality, potentially increasing catch limits by 75,000 tons compared to the current baseline model parameters.  Additionally, results indicate that from the perspective of the krill stock, current harvest levels in Subarea 48.1 are more precautionary than in Subarea 48.2, related to the latter taking place in the peak summer months. We advise taking contemporary spatiotemporal fishing trends into account in future stock assessments, as well as considering models that allow implementation of additional ecosystem components to the Generalized Yield Model.