Modelling entanglement rates to estimate mortality of marine birds in British Columbia commercial salmon gillnet fisheries.

Incidental mortality of marine birds in fisheries is an international conservation concern, including in Canada where globally significant populations of vulnerable diving species overlap with coastal gillnet fisheries. In British Columbia (BC), commercial salmon gillnet fishing effort was historically very high (>200,000 days fished annually in the early 1950's), and although this fishery has declined, over 6,400 days were fished annually in the 2006-2016 decade. Observations of seabird bycatch within the commercial fishery, however, are limited in both scope (comprising <2% of cumulative effort 2001-2016) and in time (being available only from 1995 onwards and only for a small number of areas). Using onboard fishery observer data from commercial, test and experimental fisheries (1995-2016), we developed two models to estimate the number of marine birds captured per set in sockeye (Oncorhynchus nerka) and chum (O. keta) salmon gillnet fisheries employing a Generalized Linear Mixed Modeling (GLMM) approach in a hierarchical Bayesian framework, with observer data post-stratified by fisheries management area and year. Using estimates of total commercial fishing effort (estimated number of sets, 2001-2016) we applied the models to extrapolate annual take for the main bird species (or groups) of interest. Multinomial probability estimates of species composition were calculated based upon a sample of 852 birds identified to species that were associated with sockeye or chum fisheries, enabling estimates (with CIs) of potential numbers of the mostly commonly observed species (common murres (Uria aalge), rhinoceros auklets (Cerorhinca monocerata), and marbled murrelets (Brachyramphus marmoratus)) entangled annually in commercial sockeye and chum salmon gillnet fisheries throughout BC. Conservative estimates of annual losses to entanglement were greatest for common murres (2,846, 95% CI: 2,628-3,047), followed by rhinoceros auklets (641, CI: 549-770) and marbled murrelets (228 CI: 156-346). Populations of all three of these alcids species are currently in decline in BC and entanglement mortality is a conservation concern. Gillnet mortality has been identified as a longstanding threat to marbled murrelet populations, which are recognized as Threatened in the Canada and the United States of America (USA). In addition, 622 (CI: 458-827) birds from 12 other species were estimated to be entangled annually. We conclude that cumulative mortality from incidental take in salmon gillnet fisheries is one of the largest sources of human-induced mortality for marine birds in BC waters, a conservation concern impacting both breeders and visiting migrants.

Seasonal variability in vulnerability for Cassin's auklets (Ptychoramphus aleuticus) exposed to microplastic pollution in the Canadian Pacific region.

Marine plastic pollution is an emerging global conservation challenge, potentially impacting organisms at all trophic levels. However, currently it is unclear to what extent plastic pollution is impacting marine organisms at the population, species or multispecies level. In this study, we explore seasonal exposure (i.e., vulnerability) of Cassin's Auklet (Ptychoramphus aleuticus) to plastic pollution with exposure models during boreal summer and winter seasons. Based on these models, we infer exposure at the population level for this species, in the Canadian Pacific region where approximately 75% of the global population of this species breeds. The models quantify plastic exposure by determining seasonal core foraging areas and plastic concentrations found in those same areas. Core foraging areas were determined using a Generalized Additive Model based on at-sea observation data (collected year round: 1990-2010) and 50% Home Range Kernels based on aerial telemetry data (May and June 1999-2001). Plastic concentrations within these core areas were interpolated based on seawater microplastic concentrations from the summer of 2012. We found that during the boreal summer, Cassin's Auklets were exposed to relatively low concentrations of plastics. During the winter, auklet distribution shifted towards the coast where plastic concentrations are considerably higher. Model derived seasonal variability in exposure was consistent with necropsy results from bird carcasses recovered during the winter of 2014, and from a multiyear study on chick provisioning during the summer. Local oceanography likely plays a role in determining seasonal shifts in both marine bird as well microplastic concentrations, and hence exposure. As well, individual sensitivity (i.e., dose-dependent effect) may vary with annual cycles. Currently, research is focusing on determining how sensitive individual birds are to microplastic concentrations, and our models will help translate sensitivity found at the individual level to potential impacts at population or species level.