Suitable habitat prediction
We use a maximum entropy modeling algorithm implemented in the program Maxent v.3.4.1 (Phillips et al. 2006, Phillips and Dudik 2008) to predict suitable habitat. Maxent uses environmental variables from localities at which a species has been documented previously to build a predictive model of where else the clades may occur due to the presence of similar environmental conditions (Elith et al. 2011). To assess model performance, we calculated the average value of the area under the receiver operating characteristic curve (AUC) for training and test datasets (Swets 1988), AUC takes on values ranging from 0.5 (no better discrimination than random) to 1 (perfect discrimination).