Does hypoxia have population-level effects on coastal fish? Musings from the virtual world

Hypoxia is often associated with increasing nutrient loadings and has clear mortality effects on sessile organisms, but its population effects on mobile organisms in coastal environments are uncertain. The evidence for hypoxia having population level effects is laboratory experiments, many examples of localized effects in nature, a few population-level examples, fish kills, and intuition. Despite the perception by many people, none of these provide conclusive evidence of widespread population responses to hypoxia. We synthesize the results from seven ecological simulation models that examined how low dissolved oxygen (DO) affected fish at the individual, population, and community levels. These models represent a variety of species, simulate the dynamics at a range of temporal scales and spatial scales, and impose a variety of subsets of possible DO effects. Several patterns emerged from the accumulated results. First, predicted responses were large in simpler models, and small to large in more complex models. Second, while the main effects of increased hypoxia were generally small to moderate, there were instances of relatively large indirect effects and interaction effects. Indirect effects involved growth and mortality responses due to altered spatial distribution (rather than due directly to DO) and food web interactions. Interaction effects were larger responses to hypoxia when other factors were at certain levels (e.g., responses at low versus high fish densities). Interactions also occurred when the predicted responses were larger than would be expected by the sum of the separate effects. Third, accurate information on exposure and degree of avoidance of low DO were critical unknowns. Our interpretations should be viewed as suggestive rather than definitive. The patterns described were based on a collection of modeling results that were not designed to be compared to each other. A quick look at other models seems to confirm our patterns, or at minimum, does not contradict our patterns. Quantifying the effects of hypoxia on fish populations, whether large or small, is critical for effective management of coastal ecosystems and for cost-effective and efficient design of remediation actions. The potential for interaction and indirect effects complicates field study and management. Improving our predictions of the effects of hypoxia on fish populations and communities has moved from a computational issue to a biological issue. We seem to be making progress on monitoring and modeling movement behavior, but progress is slower in food web theory and empirical research and in quantifying interspecific interactions and habitat quality in terms of process rates that relate to population dynamics.