A thesis submitted to the Faculty of Science, University of the Witwatersrand, Johannesburg, in fulfilment of the requirements for the degree of Doctor of Philosophy Johannesburg 29 September 2017 / As in other developing countries in Africa, unsustainable harvesting of fuelwood in South Africa has become a matter of concern among policy makers and researchers. This is despite the country having the highest electrification access in the continent, because the majority of rural households in communal woodland areas cannot afford to cook on electric stoves, due to high levels of poverty and unemployment,. Moreover, relevant institutions do not have the capacity to monitor compliance with harvesting rules, resulting in resource users harvesting livewood due to the scarcity of deadwood with almost impunity, despite this constituting an illegal activity under the traditional management rules. Other interventions such as woodlots have largely been unsuccessful in addressing the unsustainable harvesting problems. Thus, researchers have recommended that rotational coppice harvesting be implemented by the local institutions in partnership with local resource users as a potential strategy to sustainably provide fuelwood, while addressing incapacity problems in terms of the shortage of physical and human resources in the relevant institutions. The majority of South Africans using fuelwood reside in the savanna biome which covers 34% of the country. Savanna tree species generally coppice (resprout) easily and this provides an opportunity for rotational harvesting strategies, involving local resource users. Although ecological data suggest that rotational harvesting of coppice regrowth can be sustainable, rotational harvesting of coppice has not gained popularity in South Africa, because of tenurial and institutional challenges in communal areas. Thus, whether or not rotational harvesting strategies are feasible in communal woodland is yet to be established. This thesis reports the findings of a study that assessed the implementation of a rotational harvesting scheme in selected rural communities. In order to investigate the ecological and socio-economic feasibility of community-based coppice management (CBCM), four communities across Limpopo and Mpumalanga Provinces (Thorndale, Peninghotsa, Homu and Makhuva) in South Africa were selected based on a set of criteria, and involved in a coppice management trial. The harvesting trial was implemented over a one year period by the local traditional leadership in partnership with local resource users and relevant government institutions, facilitated by the researcher. The trial was successfully piloted through one iteration of the adaptive management cycle in the first two villages. The intervention entailed rotational harvesting of specified stem size classes and species in blocks (coupes), according to consensus-based rules that were enforced by local institutions in cooperation with resource users. The study period was not long enough to monitor the rotational harvesting of coppice per se, but enabled the creation of coppice stools and a community-based system for rotational harvesting of wood resources. Both qualitative and quantitative methods were used in this action research study to assess fuelwood harvesting and use patterns, ecological sustainability, social feasibility, and the adequacy of local resource-governance systems to implement CBCM, over the course of the first year of the intervention. Data such as energy mix used in the households, time spent acquiring fuelwood, compliance with harvesting rules, perceived cooperation between institutions, and socio-economic information were collected using questionnaires. Data were collected in two survey periods, the first occurring in early January 2014 before the trial started, and the second one from April 2014, after it had started, to January 2015. The fuelwood resource base, coppicing, and wood harvesting patterns in the village communal woodlands were assessed before and during the trial in fixed 100 m2 circular plots. Allometric equations were used to convert the observed density of stems and coppice shoots into wood biomass. The measured growth rates of coppice shoots were used to estimate the amount of time that would be required for shoots to reach the allowable harvest diameter of 4–9 cm, and hence the number of years required for one harvest rotation between coupes. These data, along with the measured rates of self-thinning of shoots and village wood demand, were used to make projections of fuelwood supply by coppice and unharvested stems in the village coupes over a number of harvest rotations. The effectiveness of management under traditional management rules and that of CBCM were assessed in focus group discussions, interviews and questionnaires with the local traditional leaders and key informant interviews with government rangers before and during the trial. Levels of compliance by resource users with the agreed CBCM rules were high in Thorndale and Peninghotsa because they were given an opportunity to decide and agree on harvesting rules that were conducive to their harvesting practices. Although the intervention did not change the energy use patterns in the households in these villages, local resource users spent significantly less time acquiring fuelwood than they did prior to the intervention, thus reducing the opportunity costs of wood harvesting. There was a noticeable decline in the incidence of illegal harvesting of fuelwood by “outsiders” due to greater vigilance by resource users and community-based monitors as a result of the successful implementation of CBCM in these villages. Based on the growth rate data averaged across species, coppice shoots would reach the harvestable diameter of 4 cm in 3 years in Thorndale and 4 years in Penninghotsa. The difference may be explained by species differences in these villages, e.g. Terminalia sericea which coppices vigorously was one of the predominant species recorded in Thorndale. The successful implementation of the trial in Thorndale and Pennighotsa in the first year of the intervention is attributed to a number of socio-economic factors, including relatively small and homogenous populations, innovative and adaptive local TAs which were flexible and had encouraged community participation and supportive resource users who had positive perceptions of their leadership and the project. This is in contrast to Homu 14B and Makhuva which were characterised by larger, more heterogeneous populations, inflexible leadership who were incapacitated by lack of physical resources, lack of transparency by the leadership, and negative attitudes and perceptions of resources users. This study demonstrates that community-based coppice management is potentially applicable and feasible for sustainable provision of fuelwood in communal areas, and identified under what conditions it is most likely to succeed. / MT2018