COS 76-8
Dominant bunchgrasses occur in non-random spatial patterns in pine savanna groundcover plant communities

Wednesday, August 13, 2014: 4:00 PM
Regency Blrm A, Hyatt Regency Hotel
Katherine A. Hovanes, Department of Biological Sciences, Louisiana State University, Baton Rouge, LA
Kyle E. Harms, Department of Biological Sciences, Louisiana State University, Baton Rouge, LA
Background/Question/Methods

Dominant species are important drivers of community assembly in ecosystems characterized by both high species diversity and predominance of dominant species. In communities of sessile organisms, individuals are most likely to interact with other individuals in their immediate neighborhood. Therefore dominant species likely have frequent intra- and interspecific interactions. Individuals of dominant species may occur in non-random spatial patterns at the local scale. The spatial pattern of dominant species indicates which pair-wise interactions are likely to occur most frequently, and thus, which interactions likely affect community assembly. Longleaf pine savannas of the southeastern United States have hyperdiverse groundcover plant communities dominated by several bunch-forming grass species and present a rich opportunity to address questions of dominant species interactions. We wished to determine the spatial pattern of dominant bunchgrasses at the plant-neighborhood scale in pine savannas to gain insights into species interactions governing community assembly. We plotted the size and location of mature bunchgrasses (> 5.0 cm basal diameter) in three 9m2 plots in each of six pine savanna sites across the southeastern United States (18 plots total). Each site was dominated by one of three bunchgrass species. Using Programita, we characterized the spatial pattern of bunchgrasses at multiple spatial scales.

Results/Conclusions

Mature bunchgrasses occurred in an overdispersed (significantly regular) pattern for all sites and species. The scale at which the non-random overdispersed pattern occurred varied ranging from a radius of 0.4 – 1.0 m around focal bunches. The overdispersed pattern was only significant at scales near bunches; when the radius was expanded beyond 1.0 m from the focal bunch the spatial pattern we observed was indistinguishable from random patterns generated by Programita. Variation in the scale of overdispersion was independent of site and bunchgrass species. These results suggest that mechanisms are operating to prevent individual bunches from growing directly adjacent to one another. Furthermore, these results indicate significant spatial structure at the scale of near-neighbor interactions. This suggests that dominant bunchgrasses are likely frequently interacting with the subordinate species that occupy space in between regularly spaced bunches. The ubiquity of non-random, overdispersed spatial patterns of dominant bunchgrasses indicates that these dominant bunchgrasses most likely play an important role in community assembly and the maintenance of diversity of groundcover plant communities in high diversity pine savannas.