Towards Process-based Range Modeling of Many Species. (November 2016)

• Record Type: Journal Article
• Authors: Evans, Margaret E.K.; Merow, Cory; Record, Sydne; McMahon, Sean M.; Enquist, Brian J.
• Is Part Of: Trends in ecology & evolution. Volume 31:Number 11(2016)
• Page Start: 860
• Page End: 871
• Publication Date: 2016-11
• Subjects: data fusion -- ecological forecasting -- hierarchical model -- inverse modeling -- species distribution models; Ecology -- Periodicals; Evolution (Biology) -- Periodicals; 576.8
• DOI: 10.1016/j.tree.2016.08.005
• Languages: English
• ISSNs: 0169-5347

Abstract

Abstract : Understanding and forecasting species' geographic distributions in the face of global change is a central priority in biodiversity science. The existing view is that one must choose between correlative models for many species versus process-based models for few species. We suggest that opportunities exist to produce process-based range models for many species, by using hierarchical and inverse modeling to borrow strength across species, fill data gaps, fuse diverse data sets, and model across biological and spatial scales. We review the statistical ecology and population and range modeling literature, illustrating these modeling strategies in action. A variety of large, coordinated ecological datasets that can feed into these modeling solutions already exist, and we highlight organisms that seem ripe for the challenge. Trends: Process-based modeling can improve understanding and prediction of species' ranges. Modeling trade-offs impede process-based range forecasting of many species. Hierarchical and inverse modeling strategies can help overcome these trade-offs.