Skip to main content
Fig. 6 | Annals of Forest Science

Fig. 6

From: Potential distributions of seven sympatric sclerophyllous oak species in Southwest China depend on climatic, non-climatic, and independent spatial drivers

Fig. 6

Results of variation partitioning for the predicted habitat suitability simulated by our ensemble models in terms of fractions of variation explained. Variation of the occurrence probability is explained by three groups of explanatory variables: CLIM (climatic predictors), NCLIM (non-climatic predictors) and SPAT (spatial predictors). a–c Unique effects of CLIM and NCLIM factors and SPAT variables, respectively. d–g Fractions indicating their joint effects. Negative values are shown as “< 0%” in the figure. The font-size increases with the intensity of the effect. Note that for large datasets like our study, negative explained variances are possible when explanatory variables explain no more variation than random normal variables would (Legendre 2008)

Back to article page