Land degradation and soil nutrient depletion have become serious threats to agricultural productivity in sub-Saharan Africa. Soil fertility depletion in smallholder areas has been cited as the fundamental biophysical cause of declining per-capita food production in Africa. Manure application is a well established and known practice, but not effectively used among South African smallholders. This study investigated the practice, constraints and perceptions of improving soil quality through manure application through a case study of three smallholder farmer groups. Three groups from rural areas of KwaZulu-Natal (Mkhambatini, Mooi River and Richmond) were selected to participate in the study. Participatory methodologies were used to identify and clarify the study problem. Three participatory focus group discussions, one per area, were conducted with farmers at the study sites to discuss farming methods, experience and perceptions of manure use, manure management practices and constraints farmers experience with manure use. Force Field Analysis was used for each group to explore for forces against and in support for manure use. Random soil and manure samples were collected for laboratory analysis to determine fertility levels. Some farmers indicated that soil fertility was low. However, half the sample perceived the land to be productive to some extent. The study showed that 40 per cent of farmers reported improved soil fertility following the application of manure. Due to the limited availability of livestock manure, farmers prefer to use both livestock manure and commercial fertilisers. Furthermore, the study found that except for young farmers (20 per cent of the sample), farmers had not received formal training and very limited extension advice on composting and manure use and management. The study participants were aware of the consequences of declining soil fertility and were attempting to improve soil quality. However, low livestock numbers and poor management led to inadequate amounts of manure, and, limited access to information on manure and compost use. Unless better knowledge of optimal soil nutrient management practice is acquired by the farmers, soil fertility levels will continue to decline, further reducing production potential and rural household food security. Government needs to revisit extension support to meet the needs of smallholders and offer training on sound soil management, sustainable production methods, composting and livestock management. A handbook with graphic detail should be accompanied to provide smallholders with information and advice on how to manage soil fertility. / Thesis (M.Agric.)-University of KwaZulu-Natal, Pietermaritzburg, 2009.

The physical and chemical properties of soils can greatly influence the vegetation patterns in a landscape. This is especially so through the effect that particular characteristics of soils have on the water balance and nutrient cycling in savanna ecosystems. Areas in the savanna environment found in Hluhluwe iMfolozi Park have experienced a number of changes in the vegetation patterns observed. This study, therefore, looks at the effect that soil characteristics may have on the vegetation growth in this area and on the changes that have taken place over time. Fixed-point photographs, taken every four years, were used to choose fourteen sites in the Park, which showed either a ‘change’ or ‘no-change’ in vegetation from 1974 to 1997. The sites consisted of four which had ‘no-change’ in vegetation, two sites with a slight increase (5- 20%) in tree density, three sites with a greater increase in tree density (>20%), two sites with a slight decrease in tree density (5-20%), and three sites with a greater decrease in tree density (>20%). Transects were then carried out at each site, in which the soil was classified to the form and family level. Each horizon was then sampled and the field texture, structure, Munsell colour and depth of each horizon and profile recorded. The data recorded in the field were statistically analysed through a principal component analysis (PCA). The type of horizon, horizon boundary, structure type, colour group and depth for the top and subsoil were included in the models and were analysed with the number given to each site for each of the three sections of the Park, namely Hluhluwe, the Corridor and iMfolozi. The most prominent textures at all sites were sandy loam, loam, clay loam and silt loam for both the top and subsoil for all site categories. The texture classes were also compared across the Hluhluwe, Corridor and iMfolozi sections. The dominant textures in the Hluhluwe and Corridor sections are loam, clay loam and silt loam for both top and subsoils. Sites sampled in the iMfolozi section appear to have textures mainly associated with the clay loam and sandy loam classes. The structure classes of the soil including sub-angular blocky, granular and crumb which are associated with a moderate structure appear to be the most dominant type in all categories for the topsoil; single-grain and sub-angular blocky classes the main types for the subsoil. Generally the colour of the soil at all the sites sampled was yellower than 2.5YR and the values and chromas mostly fell within the range of 3-5 and 2-6, respectively. This is also shown in the PCA results obtained, which associate particular soil characteristics with the various sites sampled for the different vegetation change categories investigated. The samples collected were also analysed in the laboratory after being air-dried. The laboratory analysis included measurements of pH, exchangeable acidity, organic carbon, extractable phosphorus, particle size distribution and cation exchange capacity (CEC). The data recorded in the laboratory were also analysed by PCA. This was used to determine which soil properties are associated with the particular sites investigated. The pH of the soil, in all areas, fell within a wide range. The pH is influenced by the rainfall in the area and thus sites sampled in the Hluhluwe section are more acidic than those sampled in the Corridor and iMfolozi sections. The topsoils had a higher pH for all the samples and were in the range between 5 and 7. The exchangeable acidity measurements were low, although they were higher in the subsoil as opposed to the topsoil. The nutrient contents did not appear to vary greatly between the different sites in the Park. Generally extractable phosphorus, CEC and organic carbon were low across the Park. The particle size analysis showed that the clay percentage increases between the top and subsoil for all the sites sampled. The silt and various fractions of sand percentages vary across all sites and are lower than the clay percentage at all sites except the A horizon of the ‘slight increase’ sites. The ‘no-change’, and ‘increase’ sites have a higher percentage of clay as compared to the silt and sand fraction for both the A and B horizon. The ‘slight increase’ sites have a higher percentage of sand in the A and B horizon, the ‘slight decrease’ sites have a more equal percentage between the sand, silt and clay fractions in the A horizon and a greater percentage of clay in the B horizon. The ‘decrease’ sites have a greater percentage of clay and silt in the A and B horizon. While certain soil properties have a definite effect on the plant growth, no relationship between specific soil properties and vegetation changes was shown. However, it is likely that the soil structure and texture affect the vegetation patterns, through their influences on the water and nutrient holding capacity. With an increase in the clay percentage and more strongly structured soils, plants can access more water and nutrients and this will increase the tree density in an area. However, the recent changes in the vegetation patterns observed in the Park appear to be more associated with other environmental factors. The soil properties analysed would have generally been more constant at the sites sampled, particularly over the relatively short period of time in this study. Therefore, the changes which were recorded in the fixed-point photographs would have been enhanced by other factors experienced in the Park, including fire and the effect that grazers and browsers have on the vegetation. / Thesis (M.Sc.)-University of KwaZulu-Natal, Pietermaritzburg, 2009.