Fishing for data : using krill fishery data to expand krill research

Antarctic krill (Euphausia superba) are the most abundant pelagic Antarctic crustacean. This large abundance has led to the expansion of the krill fishery to the point that today it is one of the largest fisheries in the world. The Commission for the Conservation of Antarctic Marine Living Resources (CCAMLR) regulates the Southern Ocean fisheries, including the krill fishery, by using the best available science to determine, and adjust, appropriate catch limits. Currently, data collected on board fishing vessels are primarily used to develop processes for optimising management strategies and determining catch limits. However, with expected climatic changes and increasing fishery pressure, expanding the capacity of the fishery data for use in biological and modelling analyses is crucial to better advise CCAMLR‚ÄövÑv¥s ecosystem-based management strategies. The studies within this thesis examined krill fishery data, and current modelling strategies used by CCAMLR, to identify the research potential of CCAMLR‚ÄövÑv¥s resources for the effective conservation of the krill population. Following the general introduction, this thesis presents four data chapters: In chapter 2, fishery data were used to reproduce results from published investigations that utilized research vessel data. It was determined that both data types are comparable for exploring variability in krill populations and can be analysed collectively for locations or seasons with sparse data availability. In chapter 3, fishing activities that occurred within narrow spatial and temporal windows of one another were examined to evaluate catch compositions and determine their external influencers. Overall, sample size is the biggest contributor to observed composition variability, and as a result, a minimum of 200 krill samples from a harvest should be collected for an accurate representation of the population. In chapter 4, the effects of moonlight on swarm formation and krill movement through the water column were investigated using fishery harvesting depths. The findings show a promising link between moonlight and harvesting efforts, possibly caused by changes in swarm formation as a response to a full moon. And in chapter 5, the proportional recruitment model used by CCAMLR for krill stock assessments was evaluated. Testing of the model highlighted deficiencies for scenarios with large variations in recruitment. Two alternative methods are presented for modelling krill recruitment. These models can support a wider range of scenarios, such as years of low recruitment. This thesis highlights that krill fishery data are essential to not only improve our current understanding of krill behaviour, but also to enhance management strategies in the face of expected future climate and harvesting challenges.