Evaluating the Evolutionary Implications of Natural and Anthropogenic Selection with Predictive Individual-Based Modeling

Anthropogenic selection has recently been implicated as a causal mechanism for observed changes in phenotypic trait values. In commercially exploited aquatic species, concern mounts that size selective fishing gear may be causing significant decreases in size-at-age or alterations in life history, and possibly jeopardizing the sustainability of these species. Sockeye salmon (Oncorhynchus nerka) are a culturally and economically important species that are subject to a suite of natural and anthropogenic selection pressures while completing their migration to natal spawning streams. Current harvest practices are hypothesized to cause changes in average length, body-depth and timing of arrival on spawning grounds, across generations. However, concurrent natural selection processes may help to mitigate the evolutionary impacts of selective harvest and buffer against significant phenotypic change. A holistic understanding of selective pressure from both of these sources is imperative, as sustainable resource management is contingent upon knowledge of the relationship between harvest policies and population response on both proximate and evolutionary time scales.

The objective of this study is to synthesize in-stream survey and fishery data into an individual-based simulation model, which will provide a mechanistic picture of harvest and natural selection and predictions for resultant evolutionary trends. This predictive framework will have two components, modeling marine and freshwater salmonid life history stages. These components will simulate the fishery, predatory, stranding and sexual selection processes that affect overall fitness. Statistical approximations for the probability of survival and successful mating will employ length, body-depth and timing of arrival as predictor variables. This study will focus on a model population of sockeye salmon spawning in Hansen Creek, a tributary of the Wood River system (Bristol Bay, Alaska) where extensive fishery and in-stream survey data are available and permit evaluation of evolutionary trends.

This simulation model will provide a basis for inference about changes in average phenotypic values over multiple generations under current selection regimes, within the Hansen Creek population. However, the model is designed to be generally applicable to any population for which data on these same selection processes are available. The predictive framework will also be used to evaluate the efficacy of alternative fishery management strategies from an evolutionary standpoint, with respect to the temporal regulation of effort and gear size restriction. The proposed project represents a novel approach to assessing the evolutionary impact of anthropogenic selection, because it will model the effects of natural selection simultaneously and incorporate empirical data collected in situ.