Using Species Invasions to Test the Accuracy of Bioclimatic Models

Cary Institute Of Ecosystem Studies, Inc., Millbrook NY

Investigators

Abstract

Scientists often use models based on the climatic tolerances of species ("bioclimatic models") to predict the spread of newly introduced exotic species and to predict how the geographic ranges of species will change in response to climate change. Although such models are conceptually simple and widely used, many scientists object to bioclimatic models because they ignore interactions between species (like predation and competition), the genetic structure of species, dispersal, complex relationships among climatic variables, and ranges that are not in equilibrium with climate. Despite their importance and potential flaws, scientists have not evaluated the actual performance of bioclimatic models. I will use information on the spread of exotic species in Europe and North America to test the performance of bioclimatic models. I will select a wide range of species that have been introduced between the two continents and construct bioclimatic models based on the distribution of each species in its native range. These bioclimatic models will then be applied to predict the range of each species in the continent to which it was introduced. The predicted range will be compared to the actual range using several statistical measures, which will reveal the actual predictive abilities of bioclimatic models and diagnose likely causes of failures.