Fire, grazing, and patch structure: Do South African and American grasslands respond in the same way

Background/Question/Methods
The interaction between fire and grazing can have significant impacts on small scale patch structure and dynamics in savanna grasslands. This study was designed to determine the effects of fire and grazing on small-scale grass-forb interaction. Patch composition and structure were measured in long-term fire and grazing experiments in Kruger National Park, South Africa and Konza Prairie LTER, Kansas. Vegetation transects were performed in areas with varying fire (annual, unburned) and grazing (ungrazed, single grazer, and multiple grazer) manipulations. To measure patch structure and forb richness, we established transects of 128 0.25m² contiguous quadrats at each site. Cover of each plant species rooted in each quadrat was visually estimated. Spatial variation and patch structure in each treatment was quantified with semi-variance analysis in order to determine the patch structure of dominant grasses and to determine if the patch structure of grasses affects the distribution and abundance of forb species. We predicted that (1) the patch structure of each of the dominant grasses in Kruger and Konza would respond similarly to fire and grazing, and (2) that forb richness is correlated to grass patch structure.

Results/Conclusions

At Konza, grazing increased small scale total species richness in both burned and unburned grassland. In the absence of grazing, fire did not play a dominant role in maintaining species richness; in the presence of grazing fire reduced total richness. At Kruger, grazing without fire increased richness. However, in annually burned sites, grazing by Cape buffalo alone led to higher richness than sites open to multiple grazers. All semi-variograms had relatively large ratios of spatially structured variance to sill (overall mean: 0.796), indicating that much of the variation is found at scales greater than 0.25m². The unburned site with multiple grazers exhibited the largest ranges for two dominant grasses indicating a larger patch structure at this site compared to the other treatments. At neither site was forb cover and richness related to grass cover. Large scale patch structure at Konza was similar to small scale structure; however, patch structure at Kruger responded differently depending on the scale of analysis. These analyses indicate that North American and South African grasslands respond differently to fire and grazing, and that these differences are a function of system-specific differences in patch size.