Edaphic, Nutritive, and Species Assemblage Differences between Hotspots and Matrix Vegetation: Two African Case Studies

Areas of locally intense and frequent grazing, or 'hotspots', are pervasive features in tropical grasslands and savannas. In some ecosystems, hotspot presence is clearly associated with edaphic factors (e. g., high clay content and elevated soil fertility), such as those that develop in abandoned cattle bomas. Studies in a range of other savanna ecosystems, however, have failed to find intrinsic soil differences between hotspots and the surrounding matrix. Also, it remains unclear to what extent hotspots are associated with specific assemblages of nutrient-rich plant species, as opposed to being a manifestation of intraspecific variation in nutritive quality. We conducted simultaneous studies in Kruger (South Africa) and Serengeti (Tanzania) National Parks to re-evaluate the role of edaphic correlates of hotspot occurrence and to test whether intraspecific variation in plant quality occurs across hotspot-matrix boundaries. We sampled soils and plants in paired hotspot and matrix plots at multiple sites within each ecosystem to test our a priori hypothesis that hotspots would be associated with distinct species assemblages and differences in soil fertility. We found clear hotspot-matrix differences in foliar N, particularly within species, despite finding no differences in any soil or plant-soil variables, including N mineralization potential and mycorrhizal inoculation levels. We found only weak differences in community composition across the boundary, suggesting that intraspecific variation in foliar N rather than species turnover is mainly responsible for the enhanced nutritive value of hotspot vegetation. We propose that grazer-plant interactions may be stronger drivers of hotspot maintenance in these systems than plant-soil interactions.