Hypoxia has emerged as a prominent research and management issue in coastal marine and large-lake ecosystems worldwide, primarily owing to human-driven eutrophication. Our understanding of hypoxia’s impacts on aquatic organisms, particularly pelagic fishes, is limited, however. Herein, we present findings from an investigation of hypoxia’s impact on two pelagic planktivores, the rainbow smelt Osmerus mordax and the emerald shiner Notropis atherinoides, in central Lake Erie, wherein hypoxia is a naturally occurring phenomenon but which has experienced a recent increase in spatial extent and duration. We hypothesized that the consequences of bottom hypoxia, both in the short- and long-term, would be less for emerald shiners (ES) than rainbow smelt (RS), owing to interspecific differences in evolved anti-predator behaviors (i.e., surface [ES] vs. bottom [RS] schooling during day) and thermal requirements for growth (i.e., warmwater [ES] vs. coldwater [RS]), as well as invasion history (and hence, exposure to hypoxia in Lake Erie; native [ES] vs. naturalized [RS]). To test this hypothesis, we combined a spatially-explicit bioenergetics-based habitat modeling approach with long-term (1990-2008) monitoring data and intensive field observations (during 2005) of fish spatial distributions, vertical migration patterns, feeding behavior, growth, and condition (health) before, during, and after hypoxia.
Our findings support our hypothesis, with our spatial modeling demonstrating that hypoxia diminishes sub-hypolimnetic areas of positive somatic growth (our proxy for habitat quality) for rainbow smelt, whereas habitat quality for emerald shiners was unaffected by oxygen availability. Intensive field observations supported our modeling results in that hypoxia caused rainbow smelt to reduce their use of cold bottom waters, forcing individuals into a thin layer in the metalimnion that was more oxygenated and warmer. In turn, reduced access to prey and optimal temperatures for growth caused rainbow smelt consumption, condition, and growth to be lower than expected. By contrast, emerald shiner vertical distribution, foraging, condition, and growth appeared unaltered by hypoxia. These short-term findings help to understand results from our long-term analyses, which demonstrated that fall condition of rainbow smelt was poorer in years with long versus short hypoxic episodes, whereas condition of emerald shiners was unrelated to hypoxia duration. We discuss the implications of our results for efforts aimed at forecasting the impacts of hypoxia on the world’s fisheries, highlighting the need to consider evolved physiological and behavioral adaptations and prior exposure to hypoxia.