The impact of modelling choices in the predictive performance of richness maps derived from species-distribution models: guidelines to build better diversity models
Robert B. O'Hara
BIOTREE-NET, species composition, Species richness, stacked species-distribution models
1. The stacking of species-distribution models (S-SDMs) is receiving attention by conservation researchers because this approach is capable of simultaneously predicting species richness and composition. However, the steps required to build S-SDMs implies at least two choices that influence its predictive performancewhich have not been extensively assessed: the selection of themodelling algorithm and the application of a threshold to trans- form the species-distributionmodels into binarymaps to be added together to build thefinalS-SDM.Our goal was to provide guidelines concerning the best combinations ofmodelling algorithms and thresholds with which to build more accurate S-SDMs. 2. Wegenerated 380 S-SDMs of 1224 tree species in Mesoamerica by combining 19 distribution modellingmeth- odswith 20 different thresholds using presence-only data from the Global Biodiversity Information Facility.We compared the predicted richness and composition with inventory data obtained from theBIOTREE-NETforest plot database. We designed two indicators of predictive performance that were based on the diversity factors used to measure species turnover: a (shared species between the observed and predicted compositions), b and c (the exclusive species of the predicted and observed compositions respectively) and compared them with the Sorensen andBeta-Simpson turnovermeasures. 3. Our proposed indexes and the Sorensen index proved suitable as indicators of predictive performance for S- SDMs,whereas theBeta-Simpson turnovermeasure presented issues that would prevent its application to evalu- ate S-SDMs. 4. Some modellingmethods – especially machine learning and ensemble model forecasting methods performed significantly better than others in minimizing the error in predicted richness and composition. Our results also points out that restrictive thresholds (with high omission errors) lead to more accurate S-SDMs in terms of spe- cies richness and composition. Here, we demonstrate that particular combinations of modelling methods and thresholds provide results with higher predictive performance. 5. These results provide clear modelling guidelines that will helpS-SDMmodellers to select the appropriate com- bination of modellingmethods and thresholds to buildmore accurate S-SDMs, and thereforewill have a positive impact on the quality of the diversitymodels used to assist conservation planning.