A large number of statistical approaches are being developed to forecast the potential future responses of species and communities to environmental change. But how good are those predictions?
Species distribution models or SDMs are the most commonly used tool to predict species’ likely future distribution shifts. Since we are using these models to predict events that are yet to happen, modeled predictions are subject to a high uncertainty. Arguably the best way to assess the ability of these models to predict species’ changes over time is to test whether they are able to predict changes that have already happened.
My research has focused on using historical data sources to test the predictive ability or transferability of SDMs over time and identify situations in which these methods can and can’t generate accurate forecasts.
Rapacciuolo G, Roy DB, Gillings S, Purvis A (2014).
Temporal validation plots: quantifying how well species distribution models predict species’ range
changes over time. Methods in Ecology and Evolution 5: 407-420.
Rapacciuolo G, Roy DB, Gillings S, Fox R, Walker K, Purvis A (2012)
Climatic associations of British species distributions show good transferability in time but low
predictive accuracy for range change. PLoS One 7: e40212.
Yates KL [and 49 others, including Rapacciuolo G]. Outstanding challenges in the transferability of ecological models. Trends in Ecology and Evolution. Online early.
Bouchet PJ [and 28 others, including Rapacciuolo G]. Better Model Transfers Require Knowledge of Mechanisms. Trends in Ecology and Evolution. Online Early..
Oliver TH, Gillings S, Girardello M, Rapacciuolo G, Brereton TM, Siriwardena GM, Roy DB, Pywell R, Fuller RJ (2012) Population density but not stability can be predicted from species distribution models. Journal of Applied Ecology 49: 581 - 590.