OOS 37-6 - Interactive effects of drought grazing and fire on grasslands community dynamics: a cross-continental comparison

Thursday, August 11, 2011: 9:50 AM
14, Austin Convention Center
Sally E. Koerner, Nicolas School for the Environment, Duke University, Durham, NC and Scott L. Collins, Department of Biology, University of New Mexico, Albuquerque, NM

Climate, in particular water availability, is thought to be the primary determinant of species composition in grasslands, with other factors like grazing and fire operating within the constraints imposed by different moisture conditions. Little information exists on how climate variation may interact with trophic controls or fire on annual net primary production, and even less is known about how these interactions will affect community structure. This study was designed to determine the interactive effects of the three main drivers of plant community structure in savanna grasslands - precipitation, grazing, and fire. Community composition and structure were measured in Kruger National Park, South Africa (SA) and Konza Prairie LTER, Kansas (NA). A clipping and rainfall manipulation experiment was established within ongoing long term fire manipulation studies in both sites.  Rainout shelters that reduce ambient precipitation by 30% were constructed and left in place for three years. In NA, seven shelters were constructed in each of three burn treatments - unburned, 4-year, and annually burned – while SA only unburned and annually burned sites were used. A 1x1 m plot was established under each rainout shelter and in a corresponding open area, and the grasses in half of each plot were clipped to simulate grazing. In all plots, cover of each species was estimated twice during each growing season, and stem density counts, ANPP, and root biomass were determined during the last growing season.  


In unburned grasslands, NA and SA respond quite differently to grazing and climate manipulations. In NA the grass richness and cover remain stable across all treatments.  Drought caused a decrease in forb cover whereas grazing increased forb cover. When exposed to both drought and grazing, the positive effects of grazing on forb cover were lessened by drought. In Kruger the forb richness and cover were unaffected by the treatments. Drought decreased grass cover significantly with a stronger response seen in the grazed plots. In annually burned grasslands, drought and grazing caused no significant effects on either grasses or forbs. These results suggest that annual burning overrides the effects of drought and grazing, but that community composition in unburned sites is influenced by grazing and soil moisture.  Water availability across a large scale may be the primary determinant of vegetation structure, but within sites fire mediates the influence of soil moisture on the plant community.

Copyright © . All rights reserved.
Banner photo by Flickr user greg westfall.