干旱区生态环境知识资源中心

The worldwide loss of utilisable rangeland in (semi-) and areas results in huge economic and social costs. Only adaptive management strategies are able to cope with these systems, which are mainly driven by unpredictable, and stochastic rainfall. The aim of the study was to investigate the relevance of rest periods as part of the management scheme in these non-equilibrium rangeland systems. The starting point of the analysis is an approved management system - the Karakul sheep-breeding Gamis-Farm (Namibia). The farmer applies a flexible strategy, which combines short-term adaptation of the stocking rate to the available forage and long-term adaptation by resting a third of the paddocks in years with sufficient rainfall.

We developed a simulation model that focuses on the key dynamics of this non-equilibrium system. Beginning with the strategy used by the Gamis-Farm, a set of alternative grazing strategies was defined, all adapted to the available forage but differing in whether and when resting is granted for a part of the pasture. The effectiveness of these strategies was compared according to the long-term productivity of the pasture and the farmer’s livelihood.

Our results reveal ecological settings during which resting is essential for the recovery of the vegetation in a fluctuating environment, as well as those during which it is not. The growth rates of both the vegetation and of the livestock are demonstrated to be highly influential. Rests during wet years are crucial for the regeneration of the pasture. We conclude that even though a non-equilibrium rangeland system is assumed, the application of pure opportunistic strategies - destocking in times of drought and fast post-drought restocking - are not always adequate to maintain the long-term productivity of the pasture. Rest periods are indispensable when vegetation has a low regeneration potential. On an applied level, the study emphasises that improved farming conditions (supplementary feeding, unrestricted options to purchase livestock) may run the risk of ecological as well as economic damages. (c) 2006 Elsevier Ltd. All rights reserved.