Abstract:
Tropical savannas occur prevalently in Africa and they are important for wildlife conservation and human livelihoods. Wild and domestic ungulates often co-occur and share forage and water resources in these savannas. This situation is prevalent in areas with minimal coverage of protected areas like Laikipia. Habitat sharing can result to varied ecological interactions, including competition and facilitation depending on season. This study assessed the responses of wild ungulates to livestock grazing in three livestock-based land-use types in Laikipia savanna ecosystem, and whether these responses are influenced by seasonal changes. Census data were obtained from the Department of Resource Survey and Remote Sensing (DRSRS) for censuses conducted in February 2001, February 2003 and February 2005 (dry periods) and June 1997, February 2010 and November 2012 (wet periods). Wild ungulates abundance was estimated using Jolly’s method 2 while species diversity (α-diversity) was calculated using Simpson’s diversity index in each census grid. Mean diversity index (ᾱ-diversity) was also calculated for each land-use type. The β and γ diversity were also calculated. Analysis of variance (ANOVA) and Kruskal-Wallis were used to test for differences in wild ungulates abundance and species diversity. Regression analysis was used to assess the relationship between livestock densities and wild ungulates abundance and diversity across land-use types in different seasons. Significant differences were accepted at p < 0.05. Wild ungulates-cattle spatial relationship was analyzed using Ripley’s bivariate K12 function. Livestock driven land-use type significantly affected nine ungulate species, four wild ungulate guilds and both the α and β-diversity. Seasonality had minimal effect on wild ungulates. Waterbucks and Gerenuk were neither significantly affected by land-use type nor by seasonality. Seasonality did not affect the α and β diversity significantly. Cattle and various wild herbivore guilds exhibited different spatial relationships (attraction or repulsion) at various scales of distance. The departure of L12 function from complete spatial randomness (CSR) was generally low in PR than in TRL and PGA; implying better co-existence in PR than in TRL and PGA. For effective wildlife conservation in Laikipia and similar savanna landscapes, restoration of native vegetation in degraded areas especially in TRL and PGA is important. More importantly, the local communities need to be sensitized on the need to maintain the proper livestock density which may not negatively impact on native wild herbivore. Research should be done to establish the correct stocking density that if exceeded, it can impact negatively on the native wild ungulates.
