Bibliography: pages 122-129. / Soil erosion is regarded as a serious problem throughout the world. Erosion is caused by both water and wind. Although the two usually occur together, wind erosion has received little attention with the exception of the problems associated with croplands. Wind erosion can, however, also be a serious problem in natural grazing lands. In this research project an attempt is made to determine the areas of potential wind erosion in the Cape Province through the use of two different models. The models used were developed and applied in semi-arid areas and thus were considered to be applicable in South Africa. The models used are: The Wind Erosion Equation developed by Chepil, Woodruff and Siddoway in the United States; and Lynch and Edward's Model for the Analysis of Limited Climatic Data, developed in Australia. There are two aspects to soil erosion by wind - the erodibility of the soil as determined by moisture, grains size, aggregates, plant cover and surface topography; and soil erosivity as determined by wind strength and duration. Methods to control wind erosion are based on decreasing erosivity through the establishment of shelterbelts and by decreasing erodibility through improving plant cover, aggregate stability and moisture retention properties. Efforts at wind erosion measurement are generally ineffective. A number of models have been developed to overcome these difficulties and to allow for prediction of soil loss. Two of these models are applied to conditions in the Cape Province. This area covers a wind range of climatic, soil and agricultural conditions and as such provides an appropriate area for their application. It is, however, concluded that neither of these models can be directly applied to conditions in the Cape Province. The seasonal rainfall distribution and the uneven distribution of the data points contribute to the ineffectiveness of the models. The greatest problem, however, is the importance of management in determining whether or not wind erosion occurs. As a result, although the models illustrate the general climatic trends affecting the susceptibility of an area to wind erosion, the lack of a management factor accounts for the lack of detail.

Thesis (MSc (Earth Sciences))--Stellenbosch University, 2008. / This study aims to provide a practical tool for the prediction and management of dust generated by the activities of an opencast mining operation. The study was conducted on opencast gypsum mines in the semi-arid environment of the Bushmanland, 90 km north of Loeriesfontein in the Northern Cape Province from April 2000 to October 2007. The vertical and horizontal components of wind transported sediment were sampled and a dust settling model was designed to predict the settling pattern of dust generated by opencast mining operations. The model was applied to soil samples collected from an area surrounding a mine. The influence sphere of the mining operation was predicted by the application of the model and then mapped. Once the influence sphere is mapped, the dust influence can be managed with the aid of an onsite weather station. By further applying the predictions based on climatic data, the influence sphere can be modelled. The model is not only applicable to the planning phase of an opencast mine to plan the position of dust sensitive areas like the living quarters, office buildings and workshops etc., but also to indicate the historical impact that a mining operation had once a quarry on an active mine is worked out and rehabilitated or a mine is closed. The model application can also aid with the explanation and visual or graphic representation of the predicted impact of planned mining operations on communities or neighbouring activities to them and thus avoid later penalties.