Assumption- versus data-based approaches to summarizing species’ ranges

This is the peer reviewed version of the following article: Assumption- versus data-based approaches to summarizing species’ ranges, which will be published in final form in Conservation Biology at http://dx.doi.org/10.1111/cobi.12801. This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Self-Archiving: http://olabout.wiley.com/WileyCDA/Section/id-828039.html#termsSpecies’ geographic distributions are mapped using various approaches for use in conservation decision-making. Some such mapping efforts have relied on modifications of coarse-resolution extent-of-occurrence maps to downscale them to fine resolutions for conservation planning. This contribution examines (1) the quality of the extent-of-occurrence maps as range summaries, and (2) the utility of refining those maps into fine-resolution distributional hypotheses. In both cases, we found significant problems: the extent-of-occurrence maps are overly simple, omit many known and well-documented populations, and likely frequently include many areas not holding populations. Refinement steps involve typological assumptions about habitat preferences and elevational ranges of species, which can introduce significant error in anticipating species’ true distributional areas; however, as no model evaluation steps are taken to assess predictive ability of models, “bad” models are not noticed. Whereas range summaries derived by these methods may be useful in coarse-grained, global-extent studies, their continued use in on-the-ground conservation challenges at fine resolutions is not advisable. On the other hand, data-driven techniques that integrate primary biodiversity occurrence data with remotely sensed data summarizing environmental dimensions, termed ecological niche modeling or species distribution modeling, with rigorous and quantitative testing of model predictions prior to any use. These data-driven approaches constitute a well-founded, widely accepted alternative with a minimum of assumptions