The Relationship Between Bite Rate and Local Forage Abundance in Wild Thomson's Gazelles

Abstract

The foraging of female Thomson's gazelles (Gazella thomsoni) on shortgrass plains was monitored over one annual cycle in southwestern Kenya. Sward dry green biomasses and protein densities were estimated regularly throughout the study site. Changes in protein densities with season and locale were strongly correlated with underlying changes in grass physiognomy: sward height and dry green bulk biomass density were particularly important and were found to vary inversely. The relationship between bite rates and underlying sward parameters varied with season: gazelle bite rates in the dry season were positively correlated with underlying dry green biomass and protein densities, as predicted by either the Process 1 or Process 2 foraging model of Spalinger and Hobbs. Nonlinear regressions of within—bout bite rates on these model equations significantly explained 21.8 and 23.7% of the dry season variance, respectively. In contrast, bite rates in the early wet season showed significant negative correlations with underlying protein densities: the fit of the within—bout bite rate data to Spalinger and Hobbs' Process 3 model explained 18.4% of the overall variation. The late wet season showed a flat (insignificant) relationship between bite rates and protein levels and was thus intermediate between early wet— and later dry—season patterns. Logistic regression of the type of correlation between bite rate and protein density (positive, flat, negative) on two principal components of grass physiognomy suggested that a component heavily weighting sward height was the major correlate of foraging process, whereas a second major component heavily weighting bulk density and other grass quality measures was less critical. At least during this single annual cycle, shorter swards were associated with Processes 1 or 2, whereas taller swards showed Process 3 foraging. One interpretation of these results is that sward height modulates bite mass, which in turn plays the major role in controlling foraging process. Whether the switching point remains the same in subsequent years or not, the results make it clear that the direction of the bite rate vs. foraging density relationship can change markedly with season, as predicted by the Spalinger and Hobbs models.