Inclusion of biotic variables improves predictions of environmental niche models

Aim Species Distribution Models (SDMs) are correlative models that predict the occurrence or abundance of species in relation to predictor variables. SDMs have become an important part resource management and conservation biology yet they rarely incorporate species’ demography into their predictions. To explore possible influence biotic relationships explaining patterns distribution, explanatory power SDMs, we chose two intertidal shellfish with overlapping but different environmental preferences (Austrovenus stutchburyi Macomona liliana) modelled distributions without Location New Zealand. Methods The relationship between variables on our was investigated using Boosted Regression Trees (BRTs) increasing model complexity: (1) BRT were fitted each species; (2) co-occurring study taxa not being fitted; (3) variables, estimated patch fitted. Results A strong, non-linear effect Austrovenus observed only a weak Austrovenus. inclusion improved fit metrics for both species, as assessed by withheld evaluation data, markedly so Macomona. overall deviance explained increased, correlation predicted vs data increased variability these measures decreased. Main conclusions combination improvement performance changes is importance when predicting unsampled space (e.g. impacts climate change). Our approach improves classic integrating ecological theories how interactions can alter across gradients.