Large cities around the world are faced with severe challenges that make life difficult for the growing population. These problems include air pollution, flooding, urban heat island effect and windstorms. Now there is a strong emphasis on resilience. Urban forestry, generally defined as the management of trees in cities for human well-being, has been suggested as a strategy that can help cities manage these problems. However, urban forestry especially in developing countries faces several problems and research gaps that must be addressed before the benefits can be realized. Across Ghana, a crisis of urban forest decline prevails. Increasingly, problems in urban forestry are related to land access and exclusion, unclear institutional framework for governance and challenges related to low public support and participation in decision making processes and the use of urban forests resources. It was against these backgrounds that the study was conducted. The specific objectives were to: (1) assess public perceptions and attitudes toward urban forestry, (2) analyse the institutional framework for urban forest governance and (3) explore how access to land affects urban forestry development. These are crucial but little studied issues in Ghana to date. Drawing on a case study approach with mixed methods in the city of Kumasi, Ghana, the overall goal of the dissertation was to contribute to policy recommendations for improving urban forestry in the study area as well as to contribute to scientific literature on access and institutions. Data for the study were collected over a period of six months using both quantitative and qualitative data collection methods and techniques. Sources of secondary data included journal articles, monographs, archival records, Ghanaian newspaper publications as well as published and unpublished official reports of government organizations in Ghana. Primary data sources included a survey of 396 randomly selected urban residents from three different residential classes in Kumasi using questionnaires and Likert scales. It also involved interviews with 57 experts and key informants who were selected on the basis of their level of knowledge and involvement on the subject matter. Specifically, for objective one, that public perception and attitude toward urban forestry were being assessed, the study used stratified random sampling technique to interview as many as 132 respondents per each residential class in Kumasi. For objectives two and three, institutional and access analysis involved interviews with experts using purposive sampling technique. Data collected in the study was analysed using content analysis, descriptive statistics and two types of statistical tests; (a) 'the Pearson’s Chi-square Test of independence (χ2) and (b) Kruskal Wallis H test' of differences between responses of respondents. For objective one, the results showed that citizens in Kumasi genuinely care about urban forest and trees and found them important but were not satisfied with the current state of management. The respondents found issues related to lack of safety in public parks and gardens, destruction of utility lines and buildings by urban trees among the most pressing problems associated with urban forestry in Kumasi. There were more similarities in opinion than differences between respondents in terms of perception. Moreover, socioeconomic and demographic factors explained perceptions only to some extent. Many respondents were willing to pay money for urban forestry development. The results revealed that the age of respondents, education and income were not significant factors affecting people’s perception and attitude towards urban forestry in the study area. For objective two, the study found out that institutions in terms of rules and regulations affect urban forestry outcomes. In spite of the benefits derived from urban forestry, Ghana's urban forestry management efforts are being undermined by institutional challenges such as lack of tenure security for parks and gardens, overconcentration of land property rights to chiefs and traditional authorities, overreliance on common fund for urban forestry development and ineffective decentralization system in Ghana. For urban forestry institutions to succeed, six action situations including financing, planning, policy and strategic direction, conflict resolution as well as monitoring needed to be considered. For objective three, the results from access analysis showed that legal means (right based way) constitute only one set of mechanisms amongst many other factors by which access to land affects urban forestry development. Other means such as violent evictions and theft (rights-denied mechanisms of access) also characterized how access to land affects urban forestry development in the case study area. Against these backgrounds, policy makers in Ghana should consider a mix of recommendations to help improve urban forestry in the study area. These include access regulations and creation of land banks for urban forestry development; security of tenure and simplification of land allocation procedures for establishing parks and gardens; adequate financing and budgeting for urban forestry development; institutionalization of urban forestry activities in spatial and development plans; active public and private sector engagement as well as selection of tree species that will provide continuous ecosystem services such as shade provision and reduction of temperature in the city. In addition, the institutional arrangement for urban forestry in Ghana is not well structured and needs a proper restructuring.