Abstract
M.Sc.
The aim of the present study is to determine the influence of human activities on fluvial
erosion in the Nsikazi District (Mpumalanga). Recommendations for the optimal use of
available resources in the study area are made after considering the soil erosion potential of the area, and by comparing the Nsikazi District with the nearby conservation control area, the Kruger National Park (KNP). Soil erosion can be optimally prevented by employing the most suitable management practice; therefore reliable information is required concerning the location, causes and extent of soil erosion. During the present study, satellite remote sensing is evaluated as a cost effective and timely source of information to fulfill this requirement. Due to the fact that soil erosion is a natural process, which can be amplified by certain natural physical factors, it is necessary to determine the natural soil erosion potential of an area before any assumptions can be made regarding the cause of soil erosion. In this study the integration of the natural soil erosion potential map and bare soil map resulted in the establishment of the overall soil erosion potential map for the area. Landuse can be described as the most significant contributing factor in the occurrence of bare soil (devoid of vegetation), and therefore this is an indicator not only of soil degradation, but also of human impact. The data used during the present study consist of digital satellite images (Landsat TM and SPOT) and aerial photographs provided by the Institute for Soil, Climate and Water (ISCW). Additional information was also collected from pre-existing soil and topographical maps of the area. The ILWIS 2.1 computer programme was then employed for image processing and GIS analysis of the data. The study was carried out in two analytical phases. A data processing phase, which was carried out as follows: - The physical erosion factors, which determine soil erosion potential, namely soil erodibility, slope, slope length, slope form and slope aspect as well as plant cover, were converted to GIS data layers and mapped , followed by
- the identification and mapping of the presence of bare soil and landuse patterns.
A data intergration phase was included, whereby the above mentioned data layers were
integrated to determine the natural and overall soil erosion potential, as well as the
evaluation of landuse, thereby indicating the soil cultivation potential for the study
area.