Ecological studies often focus on average effects of environmental factors, but ecological dynamics may depend as much upon environmental extremes. Ecology would therefore benefit from the ability to predict the frequency and severity of extreme environmental events. Some extreme events (e.g., earthquakes) are simple events: either they happen or they don't, and they are generally difficult to predict. In contrast, extreme ecological events are often compound events, resulting from the chance coincidence of run-of-the-mill factors. Here we present an environmental bootstrap method for resampling short-term environmental data (rolling the environmental dice) to calculate an ensemble of hypothetical time series that embodies how the physical environment could potentially play out differently. We use this ensemble in conjunction with mechanistic models of physiological processes to analyze the biological consequences of environmental extremes. Our resampling method provides details of these consequences that would be difficult to obtain otherwise, and our methodology can be applied to a wide variety of ecological systems. Here, we apply this approach to calculate return times for extreme hydrodynamic and thermal events on intertidal rocky shores. Our results demonstrate that the co-occurrence of normal events can indeed lead to environmental extremes, and that these extremes can cause disturbance. For example, the limpet Lottia gigantea and the mussel Mytilus californianus are co-dominant competitors for space on wave-swept rocky shores, but their response to extreme environmental events differ. Limpet mortality can vary drastically through time. Average yearly maximum body temperature of L. gigantea on horizontal surfaces is low, sufficient to kill fewer than 5% of individuals, but on rare occasions environmental factors align by chance to induce temperatures sufficient to kill >99% of limpets. In contrast, mussels do not exhibit large temporal variation in the physical disturbance caused by breaking waves, and this difference in the pattern of disturbance may have ecological consequences for these competing species. The effect of environmental extremes is under added scrutiny as the frequency of extreme events increases in response to anthropogenically forced climate change. Our method can be used to discriminate between chance events and those caused by long-term shifts in climate.
Mark W. Denny, Luke J. H. Hunt, Luke P. Miller, and Christopher D. G. Harley. "On the Prediction of Extreme Ecological Events" Ecological Monographs (2009): 397-421.