Relative effects of herbivory by sheep, rabbits, goats and kangaroos on recruitment and regeneration of shrubs and trees in eastern South Australia.

Published online
04 Oct 1997
Content type
Journal article
Journal title
Journal of Applied Ecology
DOI
10.2307/2405281

Author(s)
Tiver, F. & Andrew, M. H.

Publication language
English
Location
Australia & South Australia

Abstract

Short-term recruitment and long-term regeneration of 18 species of shrubs and trees in over 600 populations in eastern South Australia were determined by surveying membership in life history classes. The populations were in predominantly shrubland communities dominated by chenopods (with Atriplex vesicaria and Maireana pyramidata common in the south and ranges, and M. astrotricha to the north), and with overstorey, where present, of Eucalyptus in the south, Dodonaea, Eremophila or Senna on the ranges, and Acacia aneura in the north. Sites were stratified using current herbivory levels by sheep, rabbits (European rabbits, Oryctolagus cuniculus, introduced to Australia for hunting), goats (Capra hircus, originally domestic but now feral) and kangaroos (Macropus rufus), past herbivory by sheep, and edaphic (soil) variables. Ten species showed significant negative effects of present or past herbivory, either on recruitment, regeneration or both. Sheep grazing was the most important variable and the other herbivores frequently did not have independent effects on regeneration. There was a poor relationship within a plant species between suppression of recruitment and suppression of regeneration by herbivores. The negative effects of herbivory on recruited juveniles must exceed natural thinning before overall regeneration is affected. Some species showed major disruption of regeneration at sites with a history of heavy grazing during the 19th century regardless of their present responses to grazing. This evidence supports non-equilibrium models of vegetation dynamics and indicates the importance of taking historical and other non-equilibrium effects into account when designing studies of plant populations. If many plant species are to avoid local extinction, it will not be sufficient to rely on controlling goats and rabbits as a conservation measure while ignoring the overriding effects of sheep grazing. The pattern of grazing use of the area will have to be modified to secure long-term conservation. A regional network of small and large reserves is suggested.

Key words