Species distribution modeling of the South American cactus moth: The importance of treating SDMs as hypotheses in predicting invasions

Species distribution models (SDMs) have been developed for a wide range of invasive species, in part because they offer the possibility of predicting the species' ultimate geographic distribution using user-friendly software and widely-available data. This ease of use has led to an increasing misuse of SDMs, especially in their application to consumer invasions. Models constructed based solely on the bioclimatic envelope of an organism's native geographic range ignore the potential importance of biotic interactions constraining its distribution. Thus, when invasion leads to a change in the region of fundamental niche space occupied, we cannot distinguish evolutionary changes from ecological shifts due to simultaneous changes in biotic interactions. We show the potential utility of using various SDMs of the escaped biocontrol agent Cactoblastis cactorum (Berg) as competing hypotheses. We will examine niche shifts across the invaded ranges of C. cactorum to identify niche axes which are experiencing the greatest (and least) divergence from that of the native range. This information will allow us to determine which components of the species’ niche are most likely to allow creation of models that more accurately represent its current and future distributions. The results of this work will advance our insight into the dynamics of niche shifts, and allow for better predictions of invasive species, as well as pest response to climate change.