COS 79-8
Patterns of seed-dispersal mutualisms across temperate North America

Wednesday, August 13, 2014: 4:00 PM
Regency Blrm D, Hyatt Regency Hotel
Christopher M. Moore, Biology, Case Western Reserve University, Cleveland, OH
Stephen B. Vander Wall, Program in Ecology, Evolution, and Conservation Biology, Department of Biology, University of Nevada, Reno, Reno, NV
Sarah C. Barga, Natural Resources and Environmental Science, University of Nevada, Reno, Reno, NV
Amy Seaman, Montana Audubon, Helena, MT
Background/Question/Methods

Interactions between individuals belonging to species and guilds ultimately mediate ecological dynamics.  Mutualistic interactions are a subset of direct interactions where both participants incur an increase in fitness.  These types of interactions are found in virtually all communities and differently contribute to ecological structure and dynamics compared to other interaction types.  The goal of our study was to examine the spatial distribution of plants that putatively engage in mutualistic seed-dispersal interactions with animals, inferred from their fruit morphology, and correlate those patterns with various environmental gradients.  Our main a priori hypothesis was that higher-order patterns of mutualistic interactions covary with patterns of species richness and distributions, and along environmental gradients.  We tested this hypothesis by compiling previously published plant lists for 197 sites across the North American continent, then attributing dispersal and other life-history characteristics, such as growth form and higher-order taxonomic grouping, for each species.  Our main analytical approach employed multiple regression, model selection, and principal component analysis to detect relationships of mutualisms and mutualism types in space and across environmental gradients.

Results/Conclusions

Using incidence data from each site, plants with seed-dispersal mutualisms accounted for roughly one-sixth of all records and one-eighth of all species.  We found expected relationships of species richness at sites, including positive significant relationships with mean annual precipitation, elevational range, sampling area, and latitude.  Non-mutualist richness accounted for five-sixths of the records and largely mirrored patterns of species richness.  Plant species with mutualists showed stronger relationships with precipitation and latitude and no relationship with sampling area.  Dividing the mutualisms into dispersal modes, frugivory was the most abundant and drove most of the patterns of mutualisms.  The relative abundance of frugivory was highest in the eastern US, with seed-hoarding being relatively abundant in the eastern US and arid southwest.  Myrmecochory was of relative low abundance compared to other modes and showed the highest relative abundance in Appalachia and around the Great Lakes.  We further identified univariate diversity metrics and bivariate type-abundance distributions of mutualism types across sites.  Our main finding was that similar type-abundance distributions were observed for interactions as are normally found with species.  This is the first attempt to examine patterns of seed-dispersal mutualisms at this spatial scale and we hope this stimulates future research.