PS 60-80
Deer herbivory affects phenology, seed production, and plant size in three prairie legumes

Thursday, August 8, 2013
Exhibit Hall B, Minneapolis Convention Center
Daniel L. Hernandez, Department of Biology, Carleton College, Northfield, MN
Mark J. McKone, Department of Biology, Carleton College, Northfield, MN
Emily Rogers, Biology Department, Carleton College, Northfield, MN
Background/Question/Methods

Selective browsing by herbivores may have significant impacts on targeted plant species, with potential consequences for plant fitness and community composition.  In restored tallgrass prairie, white-tailed deer (Odocoileus virginianus) are a common herbivore and are highly selective herbivores on a few legume species.  We used deer exclosures established in 2010 in the Carleton Arboretum (Northfield, MN, USA) to determine the effects of deer browsing on the size, flowering phenology, and seed production of three preferred legume species (Desmodium canadense, Dalea candida, and Dalea purpurea).  Inside each exclosure, we identified all individuals of the target species and randomly placed them into two treatments – unbrowsed (“unbrowsed inside”; UBI) and experimentally clipped, in which the top 10cm of each stem of the plant was removed.  Outside the exclosures, we randomly selected individuals from those that had been browsed and those that were unbrowsed by deer (“unbrowsed outside”; UBO).  We followed each individual (n = 202) for seven weeks beginning on 13 June 2012, taking weekly measurements of flowering phenology and plant size.  At the end of the experiment, we determined seed production for Dalea candida and D. purpurea in each treatment.

Results/Conclusions

Both deer browsing and clipping delayed caused a 50-80% reduction in the percent of flowers that completed flowering by the end of the experiment when compared to unbrowsed plants.  There was also a significant delay in flowering phenology in UBO plants when compared to the UBI treatment for D. canadense, with UBO flowering 1-2 weeks later than UBI plants.  UBI plants were also larger (both in maximum plant height and the number of stems per plant) than UBO plants.  Despite significant differences in plant size and the percent of flowering completed, there was no significant difference in seed production in D. candida and D. purpurea in response to browsing or clipping.  Again though, UBI plants produced more seeds than UBO plants, despite the fact that browsing did not occur in either treatment.  Our results suggest that deer selectively browse larger individuals of all three target species and that both the direct effects of browsing and indirect effects of shifting the population toward smaller plants has significant effects on flowering phenology and seed production.