Grazing systems, covering about half of the terrestrial surface, tend to be either equilibrial or non-equilibrial in nature, largely depending on the environmental stochasticity.The equilibrium model perspective stresses the importance of biotic feedbacks between herbivores and their resource, while the non-equilibrium model perspective stresses stochastic abiotic factors as the primary drivers of vegetation and herbivore dynamics.In semi-arid and arid tropical systems, environmental stochasticity is rather high, making the systems essentially non-equilibrial in nature, suggesting that feedback between livestock and vegetation is absent or at least severely attenuated for much of the time. In southern Africa, range and livestock management however, has been built around the concept of range condition class and the practices of determining carrying capacities and manipulating livestock numbers and grazing seasons to influence range condition. This management approach is derived from the equilibrium or climax concept of Clementsian succession. The erratic and variable rainfall in many pastoral areas of Africa poses a fundamental challenge to this conventional notion of carrying capacity in range management. This realization has caused a shift towards models that embrace non-equilibrium dynamics in ecosystems. The main concern is that application of the range model may contribute to mismanagement and degradation of some rangeland ecosystems. However, only a few studies in rangelands have empirically tested the non-equilibrium hypothesis leading to the debate on rangeland dynamics remaining unresolved.
Across the savannas of Africa, grasslands are being changed into cultivation due to increasing human population, at the expense of decreasing wildlife populations. African savannas however, still contain pockets of wilderness surviving as protected areas, but even there, species richness of large mammals is decreasing. The inevitable result is the loss of most of the wild plants and animals that occupy these natural habitats, at the same time threatening the well-being of the inhabitants of these savannas. Hence, to facilitate the management of arid and semi-arid savannas for both biological conservation and sustainable use (improving human welfare) an improved understanding of the complex dynamics of these savannas is critical. Furthermore, it is widely recognized that a high level of uncertainty typifies the lives of rural farmers in developing countries.Non-equilibrium dynamics bring additional uncertainty and risk to the system.However, attempts to understand efficient and sustainable ways to improve biodiversity and human welfare in systems showing non-equilibrium dynamics have been rare.The behaviour of non-equilibrium systems is characterised as more dynamic and less predictable than equilibrium systems. Therefore, non-equilibrium dynamics in dryland ecosystems present a different kind of management problem for both livestock and wildlife systems since their management has been dictated by the equilibrium assumption. Additionally, loss of biodiversity is regarded today as one of the great unsolved environmental problems.Faced with this biodiversity crisis, the challenge is to find ways to respond in a flexible way to deal with uncertainty and surprises brought about by non-equilibrium dynamics.
In this thesis I use a bioeconomic approach in analyzing the implications of non-equilibrium dynamics for the efficient and sustainable management of wildlife and livestock in dryland grazing systems. The study area for this thesis is southeastern lowveld of Zimbabwe.
In chapter 2, I investigate the role of abiotic and biotic factors in determining plant species composition. While early studies emphasized the importance of edaphic and environmental controls on plant species distribution and spatial variation in vegetation composition, recent studies have documented the importance of both natural and anthropogenic disturbances in this respect. At a regional scale vegetation structure (i.e., grass/tree ratio) and species composition in savannas is largely determined by precipitation, whereas at the nested landscape-scale vegetation structure and composition is more prominently determined by geologic substrate, topography, fire and herbivory. Chapter 2, shows that at the landscape scale, abiotic variables such as rainfall and soil fertility override the effect of humans and livestock on the herbaceous and the woody plant composition.
Then, in Chapter 3, I ask the question whether there is something like non-equilibrium and what are the impacts of such dynamics on cattle herd dynamics? I studied the relevance of non-equilibrium theory to my study area by testing whether annual changes in cattle numbers showed the presence of crashes and if so, what were the factors best explaining those crashes and what age and sex classes of cattle were most vulnerable to such crashes? Chapter 3 showed that crashes in annual cattle numbers were evident and were best explained by rainfall and NDVI and their lags. Immigration i.e., movement in of animals was also an important factor in years when rainfall was below the threshold and so it was a possible source of cattle recovery after a crash together with high calving rates. In years when rainfall was above the rainfall threshold, NDVI explained more variation in annual changes of livestock. The impacts of crashes were greater on calves than other cattle age categories thus explaining why there are legacy effects (lags) in cattle numbers that can only partly be offset by cattle purchases from elsewhere because of poverty or lack of surplus stock elsewhere. These findings make the southeastern lowveld system to be dominated by non-equilibrium dynamics.
The welfare of local people is the issue that I focused on in my economic section of this thesis (Chapters 4 and 5). I addressed the question of how risks of fluctuations in household income can be managed in order to improve human welfare. The expectation was that in systems exhibiting non-equilibrium dynamics people can improve their welfare by exploiting a combination of wildlife and agricultural activities (livestock and cropping) in their attempts to reduce fluctuations in their annual welfare. This would be possible if the risks in wildlife and agro-pastoral systems were sufficiently different. Exploiting different sources of income requires efficient allocation of resources. The most prominent resource is land and land varies spatially in quality and ecological resources require spatial connectivity. Therefore the spatial dimension is important in this allocation.
In Chapter 4 I asked the question: To what extent can wildlife income buffer rural households’ incomes against fluctuations in rainfall? I studied the extent to which wildlife derived income can buffer local households’ income against fluctuations due to rainfall. The addition of wildlife as an asset for rural farmers’ portfolio of assets showed that wildlife can be used as a hedge asset to offset risk from agricultural production without compromising on return. However, the power of diversification using wildlife is limited because revenues from agriculture and wildlife assets were positively correlated. However, the correlation was very weak (only 0.4 and the explained variance thus only be 16 %) which gives ample scope for buffering. Therefore, revenues from wildlife have potential to reduce household income fluctuations due to drought, but only to a limited extent.
In Chapter 5 the question was: From a theoretical perspective, can wildlife income have an insurance value to local people? I used a modelling approach to study the extent to which wildlife income offers an insurance value to local people against fluctuating annual rainfall. Findings did not support the common assertion that wildlife can offer insurance to local people against income fluctuations due to rainfall fluctuations. The failure by wildlife income to offer insurance value to local people could be explained by high costs of harvesting the wildlife resource and high densities of both human and livestock populations in southeastern lowveld.As corollary I draw the conclusion that wildlife cannot pay its way in these rangelands as long as there are high densities of people as shown in Chapter 5. Definitely wildlife income becomes insufficient if long-term sustainability of wildlife resources is considered.
Chapter 6, finally synthesizes the conclusions that can be drawn from the preceding chapters and puts the issues addressed in a broader context. In summary, this thesis shows evidence of non-equilibrium dynamics in semi-arid grazing systems. Furthermore, the small contribution of wildlife income to local people’s welfare goes to show the widely shared view that financial rewards generated through integrated conservation and development programmes such as CAMPFIRE have generally been seen as insufficient. This led me to suggest that if we have a moral or ethical obligation to protect wildlife species, then an important way for people to meet their aspirations economically was suggested by Malthus.