Leveraging A Spatiotemporally Large Dataset to Test Predictions of Life History Theory in Stream Fish Communities

ABSTRACT Identifying the mechanisms that underlie how organisms respond to current and expected hydrologic alterations is essential to understanding the effects of climate change on riverine systems. Life history theory is one such mechanism and forms clear predictions about how traits (e.g., body size) or specific groups of traits (strategies) will be selected for by hydrological conditions. However, few researchers have tested hypotheses informed by life history theory across significant spatial and temporal scales simultaneously, or considered how selection acts on all traits or specific strategies. Our goal was to use a stream fish community dataset spanning five river basins and over two decades in the National Forests of Mississippi to test the hypothesis that (1) life history traits would be selected for by environmental conditions, and (2) that life history strategies would be selected for by hydrological conditions. To test the first hypothesis, we used a permutation procedure to compare functional beta diversity between observed and randomised datasets. To test the second hypothesis across life history strategies, we first fit beta‐binomial models with life history strategy (identified through archetype analysis) as the dependent variable and predictors of streamflow as independent variables, and then identified interpretable models using model selection. Our results did not support the first hypothesis or the second hypothesis within opportunistic strategies. However, we did find evidence that equilibrium strategists were selected for by less variable hydrology. Our results fill an important data gap within the southeastern U.S. and suggest that the influence of life history traits on community structuring is not as influential within our study system as within previous studies. We hypothesise that this could be due to the lack of environmental extremes within the dataset. Future studies should expand on the current study by investigating how selection may act differentially on individual traits and life history strategies.