Manipulating surface water availability to manage herbivore distributions in the Kruger National Park, South Africa

A prevalent hypothesis in the management of African savanna ecosystems is that surface water availability can be manipulated to influence herbivore distributions. This hypothesis rests on two assumptions: (1) surface water availability is a primary determinant of herbivore distributions; and (2) surface water availability can be manipulated at scales that impact herbivore distributions. My research uses data collected from 1981–1993 in the Kruger National Park (KNP), South Africa, to analyze these two assumptions.; To evaluate determinants of KNP herbivore distributions, logistic regression is used to model dry season distance-to-water distributions for eight herbivore species in soil substrate-defined landscapes and in wet versus dry years. The soil substrate-defined landscapes represent a potential surrogate for forage quality and wet versus dry years represent a potential surrogate for forage quantity. The results indicate that although significant relationships exist between species' distribution patterns and distance-to-water, variability occurs between landscapes and years. This variability is interpreted using an optimal foraging approach designated “limiting factor mitigation” (LFM). Application of LFM analysis to the distance-to-water distributions of KNP herbivores suggests that grazers may be limited by forage quality and/or quantity and suggests future research programs that can expand our understanding of relationships between forage characteristics and distance-to-water distributions for browsers and mixed feeders.; To determine whether surface water availability in the KNP can be manipulated at scales that impact herbivore distributions, rainfall-related heterogeneity in the location of dry season water sources is analyzed. These analyses suggest that our capacity to manipulate surface water availability over large scales is limited if water sources are characterized by high temporal and spatial heterogeneity and if the area of interest is characterized by high rainfall. In general, this research suggests that distance-to-water and forage characteristics may influence the distribution patterns of KNP herbivores. Additionally, this research suggests the importance of determining how herbivores respond to ephemeral water sources created by dry season rainstorms. If forage or dry season rainstorms exert a substantial influence on herbivore distributions, it is not likely that we can manage distribution patterns of large, mobile herbivores by manipulating surface water availability in the KNP.