Testing mechanistic explanations of observed correlations between environmental factors and marine fisheries

Based on observed correlations, marine fisheries are often hypothesized to depend on environmental factors. Since correlations are unreliable as a predictive tool, it is desirable to seek mechanistic explanations for observed correlations. This paper considers methods available for testing such mechanistic explanations. As a specific example, we consider the optimal stability window, proposed as a mechanistic explanation of observed correlations between the survival of North Pacific salmon stocks and the state of the atmosphere over the North Pacific in winter, as applied to the coastal waters and fisheries of southern British Columbia, Canada.