Vegetation gradients, soil microbial pools and carbon sequestration in cactus-invaded semi-arid rangelands of western Free State

Thesis (D. Tech. Health and Environmental Sciences: Agriculture) - Central University of Technology, Free State, 2019

This study was done in the Xhariep district, western Free State region. The western Free State rangeland is classified as Nama Karoo biome, and is characterized by dwarf Karoo shrubs mixed with grasses. Eriocephalus eriocoides, Eriocephalus spinescens, Pentzia spp., Pteronia spp., Euryops asparagoides, Lycium spp. and Asparagus spp. are dominant shrubs. The grass stands are dominated by Aristida spp., Eragrostis spp. and Stipagrostis spp. However, cacti (Opuntia humifusa) and companion invaders are overrunning the rangelands. The study therefore assessed the effects of invasive creeping prickly pear on the composition of forage grasses and shrubs in rangelands continuously grazed by sheep and cattle herds. The objective of study 1 was to characterize the vegetation profiles in sheep- and cattle-grazed areas. The rangeland was stratified based on utilization and density of cacti. The newly invaded zone (NI) resembled the typical western Upper Karoo vegetation type, where disturbance indicator (DI) grasses constituted 86%, intermediate grasses (Aristida adscensionis and Aristida congesta) constituted 10%, and shrubs were 4% of the vegetation. The dominance of DI grasses suggests that grazing management, climate and competition for nutrient resources exerted by cactus (O. humifusa) invasions were affecting vegetation characteristics. The objective of study 2 was to (a) determine the relationship between vegetation cover type on soil carbon, microbial populations and minerals in degraded rangelands, and (b) to characterize above-ground forage quality in the degraded rangelands. Four cover types were defined as cactus cover (CC), shrub cover (SC), grass cover (GC) and bare area (BA). The soil in bare area (BA) had the highest N content as nitrate (NO3-) and ammonium (NH4+) forms. Bacterial and fungal populations were the lowest in the BA. The bacterial count was 30%; and the fungal count 15% less than in areas covered by invasive species are critical in validation of global vegetation models that predict the response of fragile rangelands to climate change. In study 3, interactions of rainfall and invasive species on rangeland grazing capacity were assessed. Semi-structured questionnaires were administered on 40 cattle and sheep farms within the Eastern Parkland vegetation type and the Transitional Zone. There was an increase in the grass component in both zones; however, the former had a higher proportion of DI. Shrub density increased by 16% over the past 30 years. Subclimax grass increased by 15.4%, and by 7.1% in the Eastern Parkland and Transitional Zone, respectively. There were no differences in change of palatable species in both zones. However, a small increment of 5.1% was noted with unpalatable shrubs in the Transitional Zone, and 1.7% in the Eastern Parkland. The Eastern Parkland’s vegetation type was generally more stable and less sensitive to annual rainfall changes, compared to the Transitional Zone. It is uncertain whether vegetation patterns in the Koffiefontein rangeland (Eastern Parklands, Eragrostis-Karoo shrub and Transitional stage between Eastern Parkland and Eragrostis karoo shrub vegetation type) represent short-term cycles or permanent retrogression. There is a need for further analysis of the interactions of moisture, invasive and carrying capacity, which most likely will shape Karoo vegetation and utilization patterns in the next half century. It is also critical to assess the soil organic matter changes and carbon sequestration attributable to cacti invasion, and to ascertain if improvements in carbon accumulation would provide ecosystem services that compensate for losses in palatable forages in the next half century.