Winter rotational cover crops (WRCC) are often used to boost soil fertility and plant nutrition. However, selection and use of WRCC for soil physical improvement is usually overlooked. The objective of this study was to determine the effects of WRCC on soil strength, aggregate stability and water conservation of a hardsetting soil. The soil physical properties were determined after four rotations of growing monocultures of vetch (Vicia dasycarpa cv. Max), lupin (Lupinus angustifolius cv. Tanjil) and oats (Avena sativa cv. Sederberg) and after two rotations of growing bicultures of oats (Avena sativa cv. Pallinup) and vetch (Vicia dasycarpa cv. Max) across two soil layers, 0 to 15 cm and 15 to 30 cm. The individual WRCC and a weedy fallow constituted the treatments in the monoculture study whilst in the biculture study the various combinations of WRCC namely; 90% oat plus 10% vetch (O90V10); 70% oat plus 30% vetch (O70V30) and 50% oat plus 50% vetch (O50V50) and a weedy fallow constituted the treatments. After four rotations with cover crop monocultures, oats significantly (P ≤ 0.05) reduced penetration resistance (PR) whilst vetch increased PR in both soil layers compared to the weedy fallow control. The effect of the biculture treatments was only experienced within the 15 to 30 cm depth. The treatments O50V50 and O70V30 increased the PR compared to the control. The WRCC in monoculture significantly increased the soil aggregate stability relative to the control in both soil layers. Vetch, lupin and oats resulted in a 41.7%; 20.4% and 15.7% increase in MWD in the 0 to 15 cm soil layer and 47.2%; 44.2% and 39.7% in the 15 to 30 cm depth, respectively. An increase in aggregate stability was associated with increased macro-aggregation. Under the biculture, WRCC slightly increased, non- significantly, the aggregate stability. Both hot water and dilute acid extractable polysaccharides showed no significant correlation with aggregate stability in the two studies. Oats monoculture resulted in a significant difference (P ≤ 0.05) on cumulative infiltration compared to the control. However, after 2 h vetch and lupin showed no significant difference from the control on cumulative infiltration. Oats resulted in a 7.8% increase in final infiltration rate (FIR) whilst vetch and lupin reduced FIR by 9% and 16.7% respectively, compared to the control. Bicultures of oats and vetch significantly (P ≤ 0.05) increased cumulative infiltration compared to the weedy fallow control. A similar significant increase in FIR was also observed under bicultures. The treatments O50V50; O90V10 and O70V30 resulted in a 163.3%; 113.3% and 105.4% increase in FIR respectively, compared to the control. Cover crop monocultures significantly (P ≤ 0.05) increased plant available water (PAW) compared to the weedy fallow, with vetch, oats and lupin resulting in a 28.3%; 22% and 23.9% increase respectively, in PAW. However, no significant differences were observed on PAW after two rotations with bicultures. Compared with winter weedy fallow, WRCC improved most of the soil physical properties under study, with the most suitable results expected under bicultures compared to monocultures. Under CA, selection of WRCC like oats, vetch and lupin, one should therefore take into consideration their effects on soil physical properties as a selection criterion and not biomass and fertility alone.

Low cover crop biomass production is a major obstacle to the success of conservation agriculture currently promoted as panacea to the inherent problem of soil erosion and loss of soil productivity in the Eastern Cape (EC). Therefore, this study evaluated cover crop management strategies for optimizing biomass production for better soil cover, soil nitrogen (N) and phosphorus (P) fertility, weed control and maize yields. The strategies tested are cover crop bicultures, selection of an adapted lupin cultivar and seeding rate, and the feasibility of rain fed winter cover cropping. The cover crop experiments were carried in rotation with summer maize between the winter of 2009 and summer of 2010/2011. Biculture trial was carried out by seeding oat (Avena sativa) and vetch (Vicia dasycarpa) at three mixture ratios and as sole crops under irrigation. On a separate irrigated trial, two lupin cultivars (Lupinus angustifolius var Mandelup & Qualinock) were seeded to a range of seeding rate, 40 to 220 kg ha-1. To study the feasibility of rain fed winter cover cropping, oat, vetch, rye (Lolium multiflorum), barley (Hordeum vulgare), radish (Raphanus sativa) and triticale (Triticale secale) were relayed into a maize crop in February, March and April of 2010. The irrigated trials were followed with SC701 maize cultivar, whilst the rain fed trial was followed with DKC61-25 maize cultivar. Bicultures gave higher cover crop biomass than sole vetch, increasing with an increase in the oat component of the mixture. Increased N and P uptake was observed with bicultures compared to sole oat, however, the levels were comparable to sole vetch. Sole vetch increased soil inorganic N and P at maize planting, whilst the slow decomposition by sole oat residue resulted in mineral lock up. Bicultured cover crop residues had intermediate decomposition rates and resulted in optimum levels of inorganic N and P for prolonged periods compared to sole crops. Weed suppression by the bicultures was comparable to sole cover crops. Biculturing technology significantly (P<0.05) increased maize grain yield compared to sole oat and the yields were comparable to those from sole vetch. For lupins, 206 kg ha-1 seeding rate gave the optimum biomass yield. Weed dry weights in both cover crop and maize crop decreased with an increase in lupin biomass. Comparable soil total N and inorganic P values at maize planting, were observed from plots planted to 120, 180 and 220 kg ha-1. Maize grain yield increased with an increase in lupin seeding rate. The study on rain fed winter cover cropping had most cover crop species’ biomass decreasing with each delay in planting except for radish, which increased. Vetch produced the highest amounts of biomass from February and March planting whilst radish had the highest biomass in April planting. The two species resulted in the greatest N improvement compared to the other species. Regardless of the grazing, the grass specie residues managed to persist to the next cropping season and the residue remaining were comparable to that of radish and vetch. Late-planted cover crops had the greatest residue remaining than early-planted, as a result, April planted cover crops provided better weed suppression than March and April planted. Vetch provided the highest maize grain yield (4005 kg ha-1) whilst all other species tested had comparable grain yields. The results suggested that bicultures could be grown to give sufficient biomass for both weed suppression and soil fertility improvement. Furthermore, increasing lupin plant densities improve its function as a cover crop with respect to weed suppression, soil fertility improvement and maize yields. The study also showed that for dry land systems, February and March planted vetch and April planted radish can provide the greatest biomass and maize yield improvement.

The study was conducted in Mbozi and Ciko villages in Mbhashe Local Municipality of the Eastern Cape Province of South Africa. Two irrigation projects in the area were studied. Consumers and agricultural commodity traders in Willowvale Town, Dutywa, Butterworth and East London were also interviewed. The major objective of the study is to profile and map cabbage, maize and cattle food value chains broadly, and to understand their nature, constraints and opportunities in smallholder agriculture. A multi-stage random sampling procedure was used in which the first stage involved selecting the local government areas. This was followed by the selection of the district and then the respondents. A total of 168 participants were sampled in the proportion of 82 smallholder farmers, 41 consumers, 26 hawkers and 20 agricultural commodity traders. Focus group discussions and key informant interviews were also used during the data collection process. Value Chain mapping was done using the commodity based approach. All value chains under study indicated that they are short and commodities were transacted in unprocessed form. As cabbages and maize move from the farm to retail outlets, value addition start to take place through transportation to the market and processing in supermarkets. The cattle value chain however does not have a forward linkage beyond the two administrative boundaries of the two communities. Less than 3% of the farmers traded livestock, and this was mostly through private sales to neighbours. The farmers‘ major goal in agricultural production is assumed to be an important aspect in lengthening the value chain. As such, results of a Pearson‘s correlation exercise indicated that there is a significant relationship at 0.05% level between goals of the farmers and the village of origin. Some factors that showed significance (p=0.05) in influencing farmers‘ goals are membership of an irrigation project and household sources of income. An analysis of determinants of technical efficiency at farm level was performed using the stochastic frontier model for cabbage, maize and cattle enterprises. The results showed that rainfall adequacy, input costs, market channels and quantity sold are important determinants of cabbage production efficiency. On the other hand, maize production efficiency is positively determined by market price, area under production and rainfall adequacy. Market related variables are major drivers of the cattle value chain efficiency and these include cattle prices, market satisfaction, market channel and farm labour.

For decades, there has been significant investment in the development of agricultural technologies that aim to increase productivity of smallholder farmers in Africa. But farm output and productivity have stagnated and poverty rates have remained high and even increasing in some areas. At the same time, increases in human population levels have resulted in rising demand for food as well as for arable land. The growing intensification of farming has been accompanied by degradation of wild lands, including tropical forests and wetlands, at an alarming rate. Further pressure on fragile land has come from associated urbanization, leading to agricultural land being converted to residential and industrial uses with serious consequences for agricultural production and food supply. The recent increases in food prices across the globe as well as South Africa have drawn attention to this problem even more strongly. The main objective of the current study was to investigate farmer’s perception of the relative importance of crop-livestock integration in the small holder farming systems. Data were collected from 70 emerging and smallholder farmers selected by stratified random sampling in the communities surrounding Alice, Middledrift and the Seymour- Balfour area of Nkonkobe municipality of the Eastern Cape Province. To collect the data, a semi-structure questionnaire was administered to the respondents through face-to-face interviews. Descriptive statistics were used to characterize the farmers in terms of their socio-economic and demographic backgrounds. A series of multiple linear regression models and a binary logistic regression equation were fitted to determine the factors influencing farmers’ perception and how these in turn contribute to the decision to adopt or not to adopt crop-livestock integration. The results of the study reveal that small farmers in the Nkonkobe municipality have the possibility of realizing immense benefits from the integrated systems which also have the potential to lead to substantial improvements of the physical, chemical and biological soil properties. There is clear evidence of widespread interest to experiment with the practices based on the strong positive perceptions that a majority of the survey farmers exhibited during the course of the survey. But the farmers are facing challenges in coping with the associated complexities of competition on land, and management skill which are often in limited supply. That in most cases is not enough and efficient even to manage one of these two enterprises alone and reduction in crops yield due to use of manure as a substitute of fertilizer. Constraints to integrating crops and livestock include the competition for resources, especially land. Managing two types of farming on the same farm was perceived as difficult and many respondents held the view that use of waste of one enterprise as input to the other enterprise can reduce productivity. For example some farmers considered that the use of manure to improve soil fertility may not lead to output growth to the same extent as the use of fertilizer. A number of farmers (86%) pointed out that they only market their produce after deducting their consumption share, highlighting the crucial role of food security as a motivation for crop-livestock integration. Complementation of inputs rather than substituting inputs is required to render the system more productive and sustainable as costs are minimized and output is boosted. Associations of grain and livestock producers are useful for filling these gaps which include limited access to credit, technology and knowledge and can promote the adoption of a crop-livestock system.

The problem that is researched in this study relates to the effectiveness of the extension model applied in the High Value Crop programme in improving rural livelihoods in the Eastern Cape. The extent of poverty in rural areas in the Eastern Cape motivated for the intervention of the Is‘Baya through the introduction of Integrated Village Renewal Programme (IVRP). The need to improve the quality of life of rural households through the promotion of agriculture and industry gave rise to the collaborative effort between the Is‘Baya Development Trust and Agricultural Research Council (ARC) in South Africa. The HVC programme was established to mitigate the effects of food shortage and poverty through the production of fruit trees, herbs and vegetables at household level. Evaluation of the effectiveness of this collaborative effort was carried out in four local municipalities of the OR Tambo District and these included: King Sabata Dalindyebo municipality (Zangci); Nyadeni municipality (Hluleka); Port St Johns municipality (Noqhekwana) and Ngquza Hill municipality (Hombe). An equal number of villages where Siyazondla was implemented by DRDAR were visited for the survey to relate the extension model applied with HVC. The villages visited included: OR Tambo (Mhlontlo local municipality) Xhokonxa village; Amathole (Amahlathi local municipality) Ndakana village, (Ngqushwa local municipality) Mgababa village and (Mbhashe local municipality) Mbanyana village. The broad objective of the study therefore, was to evaluate the effectiveness of the extension model applied on the HVC programme as well as to identify the roles and responsibilities of different stakeholders involved in the model. The specific objectives of the study were to describe the HVC based extension model as it is currently organized and implemented in the Eastern Cape Province of South Africa in relation to the erstwhile Siyazondla programme, to establish the effectiveness of the extension model in terms of skills transfer and capacity development, to determine the socio-economic impact of the extension model on the livelihood of involved households, to establish the extent to which the extension model has empowered women and youths and to study the factors that has contributed to the sustainability of the extension model. In carrying out this research and in line with the practice of Agricultural Research for Development (ARD), qualitative and quantitative methods of information gathering were applied. Group approach such as the Rural Rapid Appraisal (RRA) and the Participatory Rural Appraisal (PRA) methods of data collection were used. These tools include semi-structured interviews, focus-group discussions, transect walks, seasonal calendars, key-informant interviews, resource maps and secondary data. A sample of 149 respondents was selected from eight villages in the OR Tambo and Amathole districts. The HVC model was analysed using qualitative approach, descriptive statistics and inferential statistics. The Is‘Baya/ARC extension model is thus said to be effective in meeting the objectives of the HVC programme due to the efficient transfer of skills and technology and its positive impact on the livelihood of the people. There was also an improvement of the income and food security status of the rural dwellers in O.R Tambo district. The regression analysis model was applied and the results of the model were significant to the highest income earned. Out of 22 explanatory variables fitted in the regression model, 12 were significant. The R² and adjusted R² are 73 percent and 68 percent respectively which shows the significance of the fitted variables in the model. The very high F value of 15.427 shows strong significance of the fitted variables to the model. The study therefore concluded that the HVC based extension model implemented by Is‘Baya and ARC was very effective in improving rural livelihoods. The study further recommended investment in infrastructure, market linkages, value adding, public-private partnerships, creation of tenure security, investigation of different funding sources, investment on agricultural research, extension of skills provided and implementation of similar model by the public sector.