Exploring ecosystem functioning, service provision and management options on grazed semi-arid rangelands and climate change

Abstract

Global environmental change is increasingly recognised as a concern particularly in dry rangelands where livelihoods rely heavily on ecosystem services from climate sensitive agriculture. Current models predict rising temperatures and decreasing precipitation with high variability, and increasing frequencies of droughts in these ecosystems. For African rangelands, livestock numbers are also expected to increase in response to increasing human population. These changes, in combination, are expected to impact negatively on ecosystem function and service provision with devastating effects particularly in Africa due to the high rural populations that have a low adaptive capacity. Thus it is critical to advance ecological understanding of these systems’ response and resistance to the effects of drought and grazing. Motivated by both the current condition and predicted changes in local rangelands, the overall aim of this study was to explore ecosystem function and service provision in grazed semi-arid rangelands and ultimately suggest viable management options and adaptation strategies. Four questions were asked in this regard; 1) Do existing drought and grazing literature adequately address predicted future climate change scenarios particularly in the context of southern African rangelands? 2) How can the knowledge gaps in drought and grazing (combined) studies be overcome in these ecosystems? 3) How does ranch-scale grazing management influence herbaceous and woody vegetation dynamics? and 4) How does herbivore impact and recovery periods influence rangeland dynamics in climatically variable semi-arid ecosystems? In order to answer the above questions, this study carried out a critical review of drought and grazing literature to evaluate the relevance of conventional grazing research in the wake of global environmental change. The study also established a large-scale experiment that combines precipitation manipulations with cattle grazing and possible management options in the rangelands of Limpopo Province – South Africa to complement plot-scale clipping and ranch-scale grazing case studies conducted in semi arid rangelands of southwestern Botswana. The review revealed that grazing-related research is well documented in Africa, although it lacks certain levels of realism in terms of scales (temporal and spatial), study designs (e.g. common garden vs in situ manipulations) and the nature of treatments (i.e. clipping vs grazing), whereas relevant drought (and / or grazing) research is lacking. Furthermore, from the review, it was demonstrated that dry rangelands are complex ecosystems that require multifactorial standardized experimental approaches to study individual and interactive effects of several ecosystem drivers simultaneously. The study also established a novel experimental approach that combines real grazing with extreme drought (according to the standardized precipitation index specific to the area) and found that optimal vegetation performance in previously undegraded veld, should be obtained by growing-season resting for durations not exceeding one season, even under extreme drought conditions, to avoid an accumulation of standing dead material that inhibits new growth. From ranch-scale case studies of grazing management, the results demonstrated that rangeland health (as measured by cover of palatable perennial grasses) in degraded areas (i.e. areas with increased woody vegetation cover and less herbage) may not be achieved by destocking alone – as is usually common practice, but through adaptive management at appropriate scales that involve uniform grazing and adequate season-long grazing – an approach that may be achieved on rotationally grazed (RG) rather than continuously grazed (CG) systems. Furthermore, the study found through clipping experiments that while broad-leaved palatable perennial grasses such as Brachiaria nigropedata are able to overcompensate under clipping, their productivity decreases with increasing previous season clipping frequency – an example of lagged effects of grazing history, whereas less palatable needle-leaved species such Stipagrostis uniplumis are less resistant to grazing. In addition, the importance of mechanistic components of herbivory (defoliation, trampling and nutrient deposition) in clipping experiments were highlighted where the interaction of clipping, dung and trampling increased grass cover, contrary to the neutral response from their individual effects. In conclusion, this PhD has advanced scientific knowledge on grazed ecosystems and how they may be impacted by predicted global environmental changes. The study also has important implications for theory, management and policy particularly with respect to drought mitigation and adaptation strategies. Suggestions are made for grazing management as well as feeding strategies during and after drought years to help the veld to recover. Thus, rangelands with a long evolutionary history of grazing, such as those in southern African savannas, are expected to show some convergent responses to grazing and extreme drought conditions, the strength and direction of which will be determined by the underlying grazing management. The findings of this thesis may be applied to alleviate problems of grazing-related degradation in semi-arid rangelands of Limpopo and elsewhere, and also to develop adaptation strategies for predicted future global change challenges.