Thesis (PhD)--University of Stellenbosch, 2010. / The aim of the study was to assess the importance of spatial and temporal variation in diet quality and abundance for determining sustainable stocking rates on commercial, communal and game ranches in a semi-arid savanna, with the ultimate goal of avoiding land degradation in the long term, to provide sustainable livelihoods in rangelands and to make policy that will help in managing the available natural resources in the rangelands. Thus, firstly the effects of grazing, fire, nitrogen and water availability on nutritional quality of grass in semi-arid savanna was assessed. Secondly, spatial and temporal variation in plant quantity and quality among management (commercial, communal and game) types and habitat types (open savanna, rocky, shrubby and pans) and stocking rates in different management types were determined. Thirdly, the quality and quantity of variation inside and outside herbivore exclosures among commercial, communal and game management and habitat types in the semi arid savanna were estimated. Fourthly, faecal profiling was used to assess the effects of different management types on diet quality in semi-arid savanna. Lastly, policy based on the results of the present study was formulated.

The objective of this study was to determine sheep production practices, constraints, flock dynamics, body condition and weight variation in two ecologically different resource-poor communal farming systems of the Eastern Cape Province of South Africa. Mean sheep flock sizes per household were not significantly different between the two ecologically different areas (villages); Gaga (19.0±3.10) and Sompondo (18.3±3.10). Shortage of feed, disease and parasite were reported the most important constraints across the two villages. In both villages, sheep housing was poorly constructed using acacia brushwoods. Fewer farmers owned rams: the rams to ewes ratio for the two villages were 1:20, 1:19 for Gaga and Sompondo, respectively. The low ram: ewe ratios reported suggest that inbreeding might have been reducing productivity of their flocks. There was also uncontrolled breeding due to undefined and mating seasons. Gall sickness, heart water and footrot caused most of the sheep mortalities. Dohne Merinos were the common genotypes in the two villages. Total entrances for each flock were higher (p < 0.05) in hot-dry season and hot-wet season than in other seasons. Most of the entrances were lambs and were born in hot-dry season (September) and cool-dry season (June) for larger flocks (10.90 ± 3.02) and for small flocks (3.65 ± 3.02). High lamb mortalities were experienced in the post-rainy (April) and hot-wet (December) season. There was a significant interaction between season and flock size. Most of the sales occurred in the hot-wet season. Ecological area had significant effect on sheep production potential (p < 0.05) in both flock classes. The average sheep production efficiency (SPE) value for Gaga and Sompondo were 0.50 ± 0.116 and 0.50 ± 0.096 respectively. The SPE for large flock was higher (p < 0.05) by season and flock size. Large flocks had a higher (p < 0.05) SPE values and the SPE ranged from 1.11 ± 0.193 in April, a post-rainy season month to 1.55 ± 0.193 in December, a hot-wet season month. Lamb mortalities constituted the greater part of outflows. High lamb mortalities occurred in hot-wet (December), hot-dry (September) and post-rainy (April) seasons. There was a significant interaction between season and age of sheep on body weight of sheep. Highest (p < 0.05) body weights were recorded in the post-rainy and autumn season in both lambs and ewes. It is therefore very important to come up with affordable interventions which take into play ecological differences of the areas for improved nutritional status of sheep in communal areas which will lead to improved sheep productivity and the poor-resourced farmer human nutritional and livelihood.