COS 80-8
Disturbance and dispersal as drivers of beta-diversity: A synthesis of experiments

Wednesday, August 12, 2015: 4:00 PM
319, Baltimore Convention Center
Christopher P. Catano, Department of Biology, Washington University in St. Louis, St. Louis, MO
Timothy L. Dickson, Department of Biology, University of Nebraska Omaha, Omaha, NE
Jonathan A. Myers, Department of Biology, Washington University in St. Louis, St. Louis, MO
Background/Question/Methods

The effects of disturbance on site-to-site variation in species composition (beta-diversity) appear to be highly contingent. Disturbance is often observed to decrease beta-diversity through convergent selection of disturbance-tolerant species, resulting in homogenization of community composition. In contrast, disturbance can increase beta-diversity through two alternative processes: divergent selection of species across environmental gradients, or through stochastic effects on species richness in local communities. In addition, the effect of disturbance on beta-diversity may depend on the rate of dispersal among communities. We tested these hypotheses by synthesizing results from 23 factorial-field experiments that manipulated local disturbance and dispersal in plant communities distributed broadly across North America and Europe. Using a null-model approach, we tested the extent to which disturbance and dispersal interactively create non-random patterns of beta-diversity after controlling for observed changes in local species richness. We then conducted a meta-analysis using a random-effects model to assess general patterns across experimental studies.

Results/Conclusions

Disturbance and dispersal had opposing effects on beta-diversity. In some studies disturbance increased beta-diversity more than expected by chance, with overall trends suggesting divergent selection. Only one study showed evidence of convergent selection. In contrast, dispersal decreased observed beta-diversity, especially in disturbed communities. However, the effects of dispersal disappeared after controlling for observed changes in local richness. Our results suggest that disturbance can increase landscape level biodiversity through non-random processes, but acts synergistically with dispersal to homogenize species composition across a landscape. This study lends insights into how community assembly may be altered by changes in disturbance and dispersal – two processes heavily influenced by human-driven environmental change and relevant for habitat restoration.