PARENT SESSION
Symposium 8: Old-Growth Savannas as Models for Management of Upland Coastal Systems: The Role of Disturbance in a Time of Global Change
Organized by: F Gilliam and W Platt
Tuesday, August 5. 1:30 PM to 5:00 PM, SITCC Chatham Ballroom B.
Are pine savannas strongly or weakly structured by competition? A two year removal experiment and its implications.
Roth, Agatha-Marie1, Keddy, Paul*,1, Campbell, Daniel1, Dozier, Hallie2, Montz, Glen1, 1 Southeastern Louisiana University, Hammond, LA2 Louisiana State University, Baton Rouge, LA
ABSTRACT- Plant communities can be strongly or weakly structured, that is with large numbers of non-zero entries or with many zero elements in the community matrix. Many removal studies show large numbers of non-zero elements, and species tend to be organized into competitive hierarchies. Here we ask to what degree competition organizes plant communities in wet longleaf pine savannas. These savannas have exceptional numbers of plant species including groups with Madrean, Caribbean and even South American biogeographic affinities. The Poaceae, Cyperaceae and Asteraceae show particularly large diversity, while families such as the Eriocaulaceae, Haemodoraceae, Melastomataceae, and Xyridaceae suggest biogeographic linkages to the Guyana Highlands of South America. The greatest diversity of carnivorous plants in North America also occurs in pine savannas. We describe diversity patterns in a Louisiana pine savanna and explore the degree to which this vegetation is structured by competitive interactions. Species-area curves were calculated from nested plots up to 200 m2 using the Arrhenius equation [log Species = c + zlog(Area)]. The slopes were high (overall z = 0.30) and did not change with time since burn (6, 18 or 30 months) although shrubs increased in prevalence. A removal study was performed for two years in two different habitat types (wet, dry) to evaluate the competitive effects of the dominant grass (Andropogon virginicus L.) upon the rest of the herbaceous flora. Mantel tests found that the removal of A. virginicus had minimal effects upon the composition of other species when compared to control plots. Pair-wise tests for removal effects showed no effects for any single species or functional groups. The lack of a conspicuous competitive effect from an abundant species is unexpected, and suggests that these communities are weakly rather than strongly interactive. This could contribute to the high levels of biodiversity by reducing rates of competitive exclusion.
Key words: competition, longleaf pine savanna, biodiversity