Predicting responses to temperature variation: Ecology and Evolution in trophic systems

Background/Question/Methods

Temperature variation is ubiquitous in natural environments and this variation has important consequences for vital processes within organisms and at the level of populations and communities.   Recently, much effort has been devoted to measuring and predicting how global climate change will impact various aspects of species’ ecology, including fitness or other performance measures.  However, much of this work has considered: i) changes in the mean of temperature (and/or other climate variables), with no associated vision of how higher moments of the distribution will impact populations and communities; and ii) single populations in the absence of a larger community of interacting competitors, prey and predators.    

Using a set of well-described relationships between temperature and performance  for ectothermic insects we examine how changes in moments of the temporal distribution of temperature (e.g. mean, variance, skewness, etc.) independently impact populations and we assess how the concerted changes in these moments will ultimately affect performance.  We then measure the extent to which the temperature/performance relationships of these populations deviate from those which would optimize their performance in current and future environments and we ask how the incorporation of species interactions (e.g. competition, predation) alters the optimal temperature/performance relationships. 

Results/Conclusions

Consistent with previous work, we find that increases in mean temperature typically lead to an increase in performance of species inhabiting temperate regions and a decrease in performance of species inhabiting tropical regions.  Increases in temperature variance tend to echo this relationship, whereas increases in temperature skewness tend to negatively impact all species.  Interactions that occur when all three moments are increased together (as is predicted for many regions) yield strong reductions in performance due to a phenomenon known as Jensen’s inequality and the impact of extreme events.   When measuring the extent to which populations deviate from their evolutionary optimum, we find no clear differences between temperate and tropical regions and we demonstrate, using theoretical models, that competitive and tropic interactions can generate a contrasting selection pressure that limits a population’s ability to conform to the characteristics of it’s environment.  This work demonstrates that understanding responses to climate change requires better resolution of both the finer details of environmental change and the type of interactions important to any particular species’ performance.