Tuesday, August 5, 2008

PS 24-100: The potential for a seed-driven invasion of Phragmites australis in developed vs. forested watersheds of the Chesapeake Bay

Karin M. Kettenring1, Melissa K. McCormick1, Heather M. Baron2, Andrew H. Baldwin3, and Dennis F. Whigham1. (1) Smithsonian Environmental Research Center, (2) Oregon State University, (3) University of Maryland

Background/Question/Methods

An exotic haplotype of Phragmites australis is rapidly invading wetlands across North America where it can form large monotypic stands and out-compete native vegetation. In the Chesapeake Bay, Phragmites is more abundant in subestuaries in developed watersheds. Little is known, however, about how this plant spreads and what factors may contribute to its success. Our study addressed the role of sexual reproduction and potential for spread by seed in the invasion of Phragmites in the Chesapeake Bay. We evaluated viable seed production in 45 Phragmites patches from 9 subestuaries with different levels of watershed development (3 forested, 3 mixed-developed, 3 developed) to determine the potential impact of land-use on reproductive output. We also assessed patch level genetic diversity using 8 microsatellite markers to determine the contribution of genetic diversity to viable seed output. Finally, we focused on four of these patches to determine the contribution of viable seed output to Phragmites seed bank densities.
Results/Conclusions

We found that Phragmites viable seed production was highly variable within and among the 9 subestuaries of the Chesapeake Bay. However, viable seed output from patches in forested watersheds was generally very low (often <1%) while seed viability was >50% for some patches in mixed-developed and developed watersheds. Viable seed production was positively related to patch level genetic diversity; patches with greater genetic diversity produced more viable seeds. Not surprisingly then, patches within forested watersheds had the lowest levels of genetic diversity. We found a direct consequence of viable seed production for Phragmites seed banks: 25-60X as many Phragmites seedlings emerged from the seed bank from patches known to produce abundant viable seed. Our study illustrates the potential importance of a seed-driven invasion of Phragmites and the critical role of land-use and patch level genetic diversity in this invasion. A cascading effect can occur whereby Phragmites patches in developed watersheds are more likely to have greater genetic diversity and produce more viable seed to form large seed banks. As Phragmites spreads by seed it can increase local genetic diversity and thus, viable seed production.