2.4 Habitat Modeling
We used an ensemble modeling approach to predict habitat suitability for both species. Ensemble modeling is a powerful approach that combines predictions from different models (Araújo & New, 2007). Moreover, the accuracy of the model is increased by fitting several suitability models, the uncertainty associated with using a single model is decreased, and finally, a range of predictions is explored across more than one set of uncertainty sources (Araújo & New, 2007).
Our ensemble models were created by averaging seven different models using the biomod 2 R package (Thuiller, Lafourcade, Engler, & Araújo, 2009). Biomod2 was chosen because it is a well-known and well-established software (Hao, Elith, Guillera‐Arroita, & Lahoz‐Monfort, 2019). These models included two regression-based models (Generalized Linear Model [GLM], and Multivariate Adaptive Regression Splines [MARS]) and three machine-learning models (Maximum Entropy [MaxEnt], Random Forest [RF], and Generalized Boosting Model [GBM].