You are here: Home Results Task 5.5.1. Protocol for evaluating models

Task 5.5.1. Protocol for evaluating models

Task lead: Antoine Guisan, Université de Lausanne, CH

Our results showed that there are tight links between change in habitat suitability and change in species functional traits but only for some of the species under study. Tests for the role of prevalence, latitudinal range and spatial autocorrelation on the quality of predictions for 8 modelling techniques reported mixed results depending on the geographical distribution of species. Finally, a state-of-the-art protocal for validation of habitat distribution models (including a review best practices, strategies, metrics…) has been produced.

Publications

Marmion, M., Luoto, M., Heikkinen, R.K., Thuiller, W. (2009) The performance of state-of-the-art modelling techniques depends on geographical distribution of species. Ecological Modelling, 220, 3512-3520

Thuiller, W., Albert, CH., Dubuis, A., Randin, C. & Guisan, A. 2010 Variation in habitat suitability models do not always relate to variation in species' plant functional traits. Biology Letters, 6, 120:123

Maiorano, et al. (In prep) Evaluating Species Distribution Models: Why, What and How.

Document Actions