The definition of the principle of effective stress and its implications have been examined for saturated and partly saturated soils. The results of a series of oedometer and all round compression tests of partly saturated and fully saturated soils are presented. These results together with additional experimental data indicate that most soils, from sands right through to clays, exhibit behaviour which, below a critical degree of saturation, cannot be accounted for by the effective stress principle. In sands the critioal degree of saturation appears to be below 50%. In clays, however, the critical degree of saturation is upwards of 85%. An explanation for the observed behaviour of partly saturated soils is offered. It is apparent that structural changes resulting from a change of pressure deficiency in a soil are very different from those resulting from an equivalent change in applied stress. The investigation as a whole indicates that, below the critical degree of saturation, the concept of effective stress in a partly saturated soil is not valid. It is suggested that the term 'intergranular stress' is more suitable than the term 'effective stress' since its use does not imply the validity of the principle of effective stress. The practical significance of the investigation is discussed briefly and the lines along which further research would prove profitable are indicated.

No description available.

Accelerated soil erosion holds strong links to excessive land degradation, socioeconomic problems and accelerated climate change, particularly in developing countries such as South Africa. An analysis of the properties of severely eroded soils is undertaken to determine which characteristic properties influence the erodibility of the soils at several gully and badland sites in three catchments of the Ngqushwa rural municipality, which is tagged as one of the area’s most severely affected by soil erosion and degradation in the Eastern Cape. Physical, biological and chemical properties of the soils were analysed, including aggregate stability, soil texture, organic matter- and carbon content, clay mineralogy, sesquioxide content as Fe2O3 and cation composition. The distribution of a number of these properties was also determined, particularly between erosion sites, along hillslope profiles and with depth. The study area comprises three catchments, namely Mgwalana, Bira and Gqora which share common soil parent materials, land use history, exhibit similar topography and advanced degree of erosion. A total of 63 soil samples were strategically collected from identified erosion zones and site controls devoid of active erosion; where stability was evident by means of vegetation cover and soil surface condition. Surface samples were collected above 30cm depth and subsurface samples at depths greater than 30cm. To determine the fate of carbon additional samples were collected from runon and sink zones at the Mgwalana catchment along a transect from top to bottom slope. Laboratory analyses was performed to determine the soil properties, whereby C content was measured by the dry combustion method, Fe2O3- and cation content by Atomic Absorption (AA) Spectrometer, texture by the separation method described by Schlichting, Blume & Stahr (1995), clay mineralogy by X-Ray Diffraction and organic matter content by conversion of total C. In addition to these properties being used to infer aggregate stability, the wet sieving method was also used for this purpose and for validation of the other soil properties. The results revealed that no one soil property has a greater influence on aggregate stability than the others, but rather that the stability of the soil is dependent on the combinations of these properties in the soil and the interactions that occur between them. Based on this, the results indicated greater stability for runon, sink and control samples, mainly in the topsoil, which were characterised by a relatively higher C- and organic matter content, loam texture and higher concentrations of Ca2+. Furthermore, the less stable eroded samples were characterised by a more clay rich texture particularly in the subsoil, relatively lower C- and organic matter contents and a greater sodicity due to higher concentrations of Na+. The least stable soils belonged to the Gqora catchment, which were found to consist of higher silt content in the topsoil and a higher Fe2O3 content in the subsoil. The clay mineralogy was relatively uniform across all catchments, comprising of primarily illite and secondly kaolinite. The sample with the highest sodicity in terms of ESP contained smectite clay in addition to illite and kaolinite which is assumed to contribute to this increased Na+ concentration. These findings aid in the conclusion that the chemical characteristics of the soils, in association with biological and, to a less extent, physical properties of the soil exacerbate the erosion problem initiated by the extrinsic contributors, such as climate and topography. Investigation into the fate of carbon on eroded lands revealed a topographically driven dynamic whereby the total carbon content was found to be greater at the top and middle slope positions as well as in the sink zone. These hillslope sections were found to have a lower slope gradient and slightly more pronounced concave shape to those sections with lower C values. These topographic variables influence the degree of gully erosion taking place at different hillslope sections, which tends to be greater where slope angle is increased and convexity exists, resulting in the removal of soil C at these positions and its deposition in areas of accumulation, namely the runon and sink zones of low angle concave slope sections. The findings of this research may be used to develop restoration and management strategies with the ultimate goal to reduce the soils vulnerability to erosion by enhancing those soil properties conducive to greater aggregate stability as determined in the present study.

No description available.