SEARCH

SEARCH BY CITATION

Keywords:

  • Acacia savanna;
  • Aepyceros melampus;
  • dik-dik;
  • elephant;
  • Laikipia;
  • Loxodonta africana;
  • Kenya;
  • Madoqua kirkii;
  • shrub encroachment;
  • wildlife conservation

Abstract

  • 1
    Herbivores, edaphic features and fire are primary factors regulating the balance between woody and herbaceous vegetation in savannas. Many observational studies have evaluated the potential effects of browsing herbivores on woody plant dynamics in African savannas, but few experimental studies have compared the dynamics of African savannas with and without browsers.
  • 2
    A replicated herbivore exclusion experiment was used to assess the role that native ungulates play in regulating woody plant dynamics on commercial rangeland in central Kenya, where the indigenous fauna have been allowed to coexist with cattle.
  • 3
    Exclusion of native browsing ungulates for just 3 years showed that they have dramatic effects at every scale from individual twig growth rates to overall rates of woody biomass accumulation in the ecosystem.
  • 4
    At the scale of individual Acacia twigs, browsers significantly reduced leaf density, leaf biomass and growth rates of twigs < 50 cm above-ground (within the foraging height of dik-diks Madoqua kirkii), but browsers had no effects on twig leaf density or leaf biomass at a height of 1·0–1·25 m above-ground.
  • 5
    Reductions in the growth rate of twigs within the foraging height of dik-diks was associated with a 6-fold reduction in the rate at which shrubs < 0·5-m tall were recruited into the 0·5–1·5 m height class. This reduced recruitment combined with measured rates of shrub mortality in larger height classes shows that browsers reduced the rate of increase in shrub density nearly to zero (7·1 ± 10·2 shrubs ha−1 year−1) compared to the rapid rate of increase in the absence of browsers (136·9 ± 13·6 shrubs ha−1 year−1).
  • 6
    Damage to shrub canopies by elephants Loxodonta africana caused large, significant reductions in cover of A. mellifera and Grewia tenax, but lesser reductions in cover of A. etbaica. For Acacias, elephant damage was focused on shrubs > 2·5 m tall, such that Acacias in intermediate height classes (0·5–2·5 m) experienced minimal browser impacts. Elephants influenced shrubland dynamics by altering shrub height-class distributions, shifting species composition from broad-leaved Grewia tenax to fine-leaved Acacia species, and suppressing woody biomass accumulation; but elephants had little influence on changes in shrub density.
  • 7
    Synthesis and applications. Our results suggest that a community of native browsers that includes both small, selective species (e.g. dik-diks) and large, bulk-feeding species (elephants) can provide an important ecosystem service by suppressing shrub encroachment on commercial rangeland.