| Maintaining sustainable fisheries through future changes in climate poses a serious challenge for humans. Despite uncertainties in climate projections and associated ecosystem responses, patterns of past responses to climate variability can provide important clues about the features of populations, landscapes, and ecosystems which are likely to matter for sustaining fisheries into the future. I analyzed historical datasets and paleo-records of sockeye salmon (Oncorhynchus nerka) and Atlantic cod (Gadus morhua), to explore how spatially-structured fish populations have responded to past changes in climate across a range of spatial and temporal scales.;At fine spatial scales, sockeye salmon populations in Bristol Bay, Alaska, have not responded coherently to interannual climate variability. Spatially-proximate populations varied asynchronously in their dynamics through time and showed differential sensitivities to life-stage specific climate drivers, despite a general tendency for populations to respond positively to warming trends.;To investigate patterns of variability in salmon abundance over longer time scales, I used paleolimnological methods to infer past sockeye densities in 21 Alaskan lakes. This revealed that salmon abundances have varied extensively over the past 500 years, even prior to the onset of commercial fishing. Interdecadal modes of variability, which dominated 20th century catch records, were dwarfed by lower frequency trends or oscillations which were not synchronized across lakes. This provides further evidence that populations are unlikely to show coherent responses to future climate change.;Finally, I investigated how a spatially-structured group of Atlantic cod on the Skagerrak coast of Norway has responded to changes in climate over the past century. The size of juvenile cod in autumn was greatest after warm springs and cool summers, indicating opposite effects of seasonal temperatures on body size. Mean length and long-term length trends varied by fjord, suggesting important spatial variation at sub-regional scales.;Taken together, these results show that we should not expect fish populations to respond coherently to climate change. Rather, responses will vary depending on features of habitat, life histories, genetics and local ecology. By maintaining population diversity and heterogeneous landscapes, we can increase the likelihood that some components of fisheries will be productive into the future. |
