COS 96-4
Geographic variation in the factors controlling plant species ranges across the New World

Thursday, August 14, 2014: 9:00 AM
Regency Blrm C, Hyatt Regency Hotel
Naia Morueta-Holme, Bioscience, Aarhus University, Aarhus C, Denmark
Brody Sandel, Department of Biological Sciences, Aarhus University, Aarhus C, Denmark
Brian J. McGill, School of Biology and Ecology / Mitchell Center for Sustainability Solutions/Mitchell Center for Sustainability Solutions, University of Maine, Orono, ME
Peter M. Jørgensen, Missouri Botanical Garden, St. Louis, MO
Nathan J. B. Kraft, Department of Biology, University of Maryland, College Park, MD
Jeffrey E. Ott, Department of Biology, University of North Carolina, Chapel Hill, NC
Robert K. Peet, University of North Carolina
Irena Simova, Center for Theoretical Study, Charles University in Prague and Academy of Sciences of the Czech Republic, Praha, Czech Republic
Cyrille Violle, Centre d'Ecologie Fonctionnelle et Evolutive, CNRS, Montpellier, France
John C. Donoghue II, Ecology and Evolutionary Biology, University of Arizona, Tucson, AZ
Brian J. Enquist, Ecology and Evolutionary Biology, University of Arizona, Tucson, AZ
Jens-Christian Svenning, Department of Bioscience, Aarhus University, Aarhus C, Denmark
Background/Question/Methods

Species distributions can be limited by a range of factors, whose relative importance has been hypothesized to vary spatially. For instance, one hypothesis states that climate poses a gradually stronger control away from the equator. Similarly, water related factors are thought to be more important in controlling species ranges in the tropics, while the importance of temperature increases towards temperate and arctic regions. Here, we use the largest botanical dataset assembled to date to (1) quantify and test for geographic variation in the controls of broad-scale distributions of plant species across the New World, and (2) assess to which degree climate gradients and disjunctions influence the observed geographic variation in range controls. We use species distribution modeling to assess the importance of climatic and non-climatic factors (represented by spatial filters) in determining the broad-scale distribution of >17,000 plant species. Aggregating results across all species, we then map geographic variations in the relative importance of range controls, and use spatially explicit models to assess the role of mean annual temperature, annual precipitation and climate disjunctions in driving these spatial patterns.

Results/Conclusions

Across the New World, we find strong spatial patterns in the relative importance of climatic and non-climatic factors controlling species distributions. The importance of climate – and temperature in particular – increases away from the tropics as expected, but is lowest at mid-latitudes rather than in the tropics. Non-climatic factors appear to be most important in regions characterized by disjunct climates (i.e. that re-occur at distant locations within the study region). This finding emphasizes the role of dispersal limitation for the distribution of species at broad scales. While strong climatic gradients are the most important control of plant ranges at high northern latitudes, dispersal limitation between disjoint areas of similar climates plays a strong role in other regions, which consequently might be more sensitive to species invasions. Spatial variation in range controls calls for caution in non-geographic generalizations about the relative importance of determinants of species ranges.