Drivers of variation in the predator-prey interaction between cod and capelin in the Barents Sea

Abstract

Cod (Gadus morhua) and capelin (Mallotus villosus) are commercially important species with key roles in the Barents Sea ecosystem. Cod is a major predator on capelin, and fluctuations in the capelin stock has influenced growth and reproduction of cod. Cod-capelin population dynamics are central to the fisheries management of capelin and play an important role in ecosystem models and assessments. The codcapelin interaction has mainly been studied at population level, but variation at smaller scales can have a large influence on population-level processes. In this thesis, I aimed to quantify and explain spatial and temporal variation in the cod-capelin interaction at different scales, with a focus on spatial overlap and consumption. The work combines theoretical modelling with statistical analyses of survey data collected over 12 years in late summer and winter. In Paper 1, we develop optimal foraging models to study individual-level mechanisms of cod prey selection and show that capelin is the most profitable of cod’s many prey species. Under assumptions of active prey search and homogeneous prey distributions, the models predict selective feeding on capelin at a rate that is limited by cod’s digestion rather than prey encounter. However, a comparison with field data revealed that cod’s feeding on capelin varies considerably between individuals. In Paper 2, we analyse cod-capelin spatial distributions with spatially explicit statistical models and identify the main overlap areas in late summer and winter. We find that cod has a weak aggregative response to capelin in both seasons, and that increasing population sizes and water temperatures have influenced a northward shift in the late summer overlap area. In Paper 3, we use statistical models to analyse cod stomach data from the overlap area in late summer. We find that a large proportion of the population-level diet is capelin, but individual consumption is highly variable. Variation in capelin density alone cannot explain variation in cod feeding since cod’s functional response to capelin quickly reaches saturation. In contrast, the vertical distribution of capelin strongly influences variation in cod feeding, especially at the Great and Central banks where the main feeding interaction takes place during daylight.