Seasonal and spatial heterogeneity in food intake by Serengeti lions

Background/Question/Methods

One of the key uncertainties in community ecology is the degree to which heterogeneity in prey spatial distribution and prey species composition buffers the tendency for predator-prey fluctuations. Here we use a 7 year dataset from Serengeti National Park on the monthly spatial  distribution of 8 herbivore species most heavily utilized by lions to address this question. Herbivore data were gathered by counting herbivore herds encountered within 200m wide transects driven along 8 tracks that form a network over a mixture of open plains and wooded savannah grasslands. These data were then lumped into hexagonal cells for subsequent spatial analysis. Rainfall data were obtained from rain gauges scattered throughout this study area and linked with monthly variation in NDVI integrated over the same hexagonal cells as the herbivore data as well as the Southern Oscillation Index, a key climatic indicator in East Africa. These data were fed into a lion functional response model based on encounters with herds rather than prey individuals and derived entirely by direct observation of behavioral and physiological parameters rather than fitting to lion kills and density data. The functional response predictions were then compared with opportunistic records of lion diets recorded over the same 7 year period.

Results/Conclusions

Our results suggest 2-fold temporal variation in lion food intake rates (kg/day) due to seasonal population fluxes of migratory prey (wildebeest, zebra, and Thomson's gazelles primarily) and climatic variation measured by NDVI, rainfall and the Southern Oscillation Index. Model predictions of daily food intake nicely bracketed the small set of direct field observations that are available, validating the efficacy of the group-dependent multi-species functional response derived from first principles. There was little indication that lions specialized to any appreciable degree on any prey species, which is consistent with expectations based on optimal diet choice models derived from the multi-species functional response. Our conclusion is that spatial and temporal heterogeneity in prey diversity and abundance buffer the potential for unstable trophic interactions one might expect for systems with a single species of predator feeding on a single species of prey. This could help explain the relative stability of predator-prey dynamics in the Serengeti ecosystem, as demonstrated by long-term census data.