Extreme high-temperature events (hereafter ETEs), defined as rare and extreme anomalies in the historical distribution of temperature, can drive loss of local species, altering the biological diversity of landscapes. Research demonstrates that both the maximum temperature and rate of increase in temperature (rate) can play a role in determining whether a species survives a particular ETE. Less well understood is the effect that rate has on order of species loss and total possible variation in ecological outcomes resulting from ETEs. Are there combinations of surviving species possible under one rate that are not possible under another rate, regardless of the maximum temperature that is reached? If so, different rates of increase in temperature can drive rate-specific order of species loss and combinations of surviving species, increasing the total possible number of species combinations resulting from ETEs (across all combinations of rate and maximum temperature), and increasing the total possible degree of spatiotemporal variability (heterogeneity) in community composition observed in nature.
We exposed an experimental freshwater microcosm of four rotifers and six protists in long term coexistence (>4 years) to a 22°C increase in temperature at three rates differing by more than two orders of magnitude (+ 0.5°C hour-1, +0.5°C day-1, and +0.5°C week-1). We recorded the identify of surviving species every 0.5°C, allowing us to both observe the order of species loss and generate a table of surviving species combinations (from the temperature at which the first species was lost until the temperature at which the last species was lost).
Results/Conclusions
Order of loss of species from the community within each of the three rates of increase in temperature was highly consistent (nonrandom). Rate significantly affected order of loss, driving rate-specific combinations of survivors at different points in community disassembly. Differences in order of loss were driven primarily by the effect of rate on three of the ten species examined. Seven species largely retained their average relative (to each other) positions within the overall pattern of loss across the three rates. This provides evidence that the shape of rate response functions (i.e., the function describing the relationship between rate and mean temperature of loss) for many coexisting species can be remarkably similar across an extraordinary range of rate. Results from this study represent a novel contribution to our understanding of role of extreme temperature fluctuation (i.e. environmental stochasticity) in structuring community composition in natural systems across space and time.