Stochastic life history modeling for managing regional‐scale freshwater fisheries: an experimental study of brook trout

Environmental heterogeneity can combine with evolutionary responses to create very dynamic and often locally independent populations across a landscape. Such complexity creates difficulties for managers trying to conserve populations across large areas. This study develops, applies, and tests the use of stochastic life‐history modeling and Monte Carlo simulation to assess management scenarios related to the realities of regional fisheries management and conservation. We apply this approach to the management of recreational brook trout (Salvelinus fontinalis) fishing; an activity that can severely impact species balance, abundance, and the size structure of fish communities. Specifically, the model incorporates population‐specific life‐history information (e.g., growth rate, reproductive effort, and survival) to allow forecasts of the impact of various management strategies and/or changes to environmental conditions on a population's ecological characteristics (e.g., size structure, abundance, and probability of persistence). Sampling was carried out in 16 water bodies spread across four sites in Atlantic Canada. Each water body was sampled in 2005 and reassessed in 2008. This sampling had two primary objectives: (1) define a significant proportion of life‐history variation of brook trout in Atlantic Canada, and (2) to test the precision and accuracy of model predictions of population responses to experimental exploitation and management changes. The model successfully predicted population responses to changes in adult survival in 12 of 13 populations having sufficient data for validation testing, while also proving to be a useful tool when engaging stakeholders regarding management options and their associated risk. We suggest that such models are cost‐effective and have great potential for informing proactive management of jurisdictions with numerous and diverse populations.