Importance of biotic predictors in estimation of potential invasive areas: the example of the tortoise beetle Eurypedus nigrosignatus, in Hispaniola

Climatic variables have been the main predictors employed in ecological niche modeling and species distribution modeling, although biotic interactions are known to affect species’ spatial distributions via mechanisms such as predation, competition, and mutualism. Biotic interactions can affect speci...

Full description

Bibliographic Details
Main Authors: Marianna V.P. Simões, A. Townsend Peterson
Format: Article
Language:English
Published: PeerJ Inc. 2018-12-01
Series:PeerJ
Subjects:
Online Access:https://peerj.com/articles/6052.pdf
Description
Summary:Climatic variables have been the main predictors employed in ecological niche modeling and species distribution modeling, although biotic interactions are known to affect species’ spatial distributions via mechanisms such as predation, competition, and mutualism. Biotic interactions can affect species’ responses to abiotic environmental changes differently along environmental gradients, and abiotic environmental changes can likewise influence the nature of biotic interactions. Understanding whether and how to integrate variables at different scales in ecological niche models is essential to better estimate spatial distributions of species on macroecological scales and their responses to change. We report the leaf beetle Eurypedus nigrosignatus as an alien species in the Dominican Republic and investigate whether biotic factors played a meaningful role in the distributional expansion of the species into the Caribbean. We evaluate ecological niche models built with an additive gradient of unlinked biotic predictors—host plants, using likelihood-based model evaluation criteria (Akaike information criterion and Bayesian information criterion) within a range of regularization multiplier parameter values. Our results support the argument that ecological niche models should be more inclusive, as selected biotic predictors can improve the performance of models, despite the increased model complexity, and show that biotic interactions matter at macroecological scales. Moreover, we provide an alternative approach to select optimal combination of relevant variables, to improve estimation of potential invasive areas using global minimum model likelihood scores.
ISSN:2167-8359