Long-term vegetation dynamics (40 yr) in the succulent Karoo, South Africa : effects of rainfall and grazing

QUESTIONS : Vegetation change in arid regions with a coefficient of rainfall exceeding 33% usually displays non-equilibrium dynamics, where abiotic factors override internal biotic controls. Irreversible changes have nonetheless also been described for arid regions. What are the contributions of internal/equilibriumvs external/non-equilibriumfactors to vegetation dynamics and can degradation due to overstocking be reversed after removal of livestock? LOCATION : Goegap Nature Reserve, Namaqualand, South Africa. METHODS : The descending point method was conducted annually from 1974 at two transects. Vegetation change was assessed in terms of vegetation cover, species composition, life-form composition, range condition, species richness and diversity. Principal coordinates analysis was used to illustrate the trajectories in floristic data, and the effects of stocking density and rainfall were examined with redundancy analysis. RESULTS : Vegetation cover, species richness and Shannon-Wiener index of diversity showed an increase and range condition improved with time. These positive changes could be related to the removal of high numbers of livestock and low wildlife numbers in the first years of survey. A gradual decline in the rate of increase in some of these parameters could be related to high grazing pressure during the later monitored years. There was a notable increase in nonsucculent chamaephytes, but the initial increase in succulent chamaephytes was not sustained. The directional change evident in perennial species composition, supports the equilibrium concept, whereby the negative changes induced by heavy grazing were partially reversed. Within the directional change, four quasi-stable states could be distinguished, which could be reconciled with the state-and-transition model. The annual component showed no directional change, but displayed event-driven, non-equilibrium dynamics by fluctuating in reaction to the timing and quantity of rainfall. CONCLUSIONS : The vegetation change displayed elements of both equilibrium and non-equilibrium dynamics, and demonstrated that the effects of heavy grazing in the Succulent Karoo were reversible. Overall, the recovery process proceeded slowly and was primarily detected in the perennial component of the vegetation. The increase in wildlife numbers in the later studied years and decline in perennial vegetation cover stress the need for active management of animal numbers to avoid vegetation degradation.