Long-term Vegetation Change Before and After Converting from Livestock farming to Game Ranching in Asante Sana Game Reserve, South Africa

Master Thesis

2018

Permanent link to this Item
Authors
Journal Title
Link to Journal
Journal ISSN
Volume Title
Publisher
Publisher

University of Cape Town

License
Series
Abstract
Although wildlife production is widely considered beneficial for semi-arid environments, few studies have reported on the long-term environmental effects of converting from livestock production to game ranching. Asante Sana Game Reserve in the Eastern Cape has centuries old land use history, during which it was cultivated and heavily overgrazed by domestic livestock with associated loss in vegetation productivity and subsequent soil erosion. After 1996 game ranching was adopted in the reserve, with observed positive results on vegetation productivity. This thesis investigates the long-term (1987-2017) spatial and temporal change in vegetation in the reserve. It documents the change in vegetation types and cover using Landsat Top of Atmosphere (TOA) reflectance multispectral data and Soil Adjusted Vegetation Index (SAVI). Correlative relationships between vegetation cover and different drivers (e.g. rainfall, fire and stocking density) are explored using generalized linear mixed models and the implications of the findings for reserve management are discussed. The results show that the relative area of Thicket has increased over time at the expense of Grassland and Shrubland while Bare-ground has expanded into Shrubland and Thicket. A ground-truthing exercise revealed a significant (p<0.01, R2 = 0.6) positive relationship between the vegetation cover estimated on the ground and satellite derived SAVI values, suggesting that SAVI can be used as a proxy for vegetation cover. Overall vegetation productivity increased over time, with the greatest increases in Thicket and Cultivated land and the lowest in Shrubland and Bare-ground. Grassland and Riverine thicket experienced surprisingly small increases in productivity, which can be explained by high prevalence of grazing ungulates and elephants in areas of Grassland and Riverine thicket respectively. Rainfall, burning and stocking numbers all had an effect on productivity in the reserve. Rainfall had clearly the strongest influence, supporting the non-equilibrium theory for semi-arid rangelands. The management can undertake restoration actions such as tree thinning, erosion control and prevention as well as fencing off affected Grassland and Riverine thicket. A long-term ecological monitoring programme should be established for the reserve for improved understanding of the vegetation dynamics so that effective evidence-based management decisions can be undertaken.
Description

Reference:

Collections