Search

Vegetation monitoring, currently complemented by climatic data in particular rainfall areas, does not always supply sufficient information to provide reliable guidelines for appropriate stocking densities on ranch or reserve scale. This paper considers how the nutrient status of animals may be used to improve the setting of management guidelines by reflecting how wild herbivores perform under differing environmental and management conditions. Faecal nitrogen concentrations of between 13 and 16 g kg-1 are above a threshold level that indicates a dietary deficiency which may precipitate nutritional stress in animals (such as happened during the 1991/1992 drought). Faecal phosphorous concentrations of between 1.9 and 2.0 g kg-1 over a long period of time would indicate a deficiency that may lead to low reproductive rates. Faecal analysis could therefore be used to assess whether applied stocking densities will allow grazers, and probably mixed feeders such as impala too, to select for a diet of sufficient quality for maintenance.

Methods of determining stocking rate based on metabolic mass, animal type and biomass are reviewed in the context of equilibrial/disequilibrial paradigms. In South Africa, the calculation of 'carrying capacity' is based on conversion of animal species to metabolic mass equivalents. This assumes homogenous systems that tend to some point of equilibrium. It is applied widely in commercial livestock systems involving one or two species. Examination of a case study in the Lowveld of the Northern Province, South Africa, showed that the determination of 'stocking rate' in this multi-herbivore and heterogenous system, overestimated the 'carrying capacity' of the reserve over 20 years. The actual animal numbers in the system dropped by approximately 4 000 kg km-2 after a drought in the early period of the study into the bounds as determined by a model incorporating rainfall and animal biomass. An approach to determine stocking density using animal type, biomass, rainfall and vegetation parameters is suggested. The development of this as a coarse-scale (regional) and ranch-specific model to cover a range of scales and heterogeneity in key resources is advocated.