Factors influencing tree composition in agricultural landscape

The understanding of issues surrounding the relationship between farmer and farm characteristics and on-farm tree composition is necessary to support biodiversity conservation in agricultural landscapes. This research was conducted to enhance the understanding of issues surrounding on-farm tree composition and management in central Ghana. 108 respondents were randomly selected and interviewed using semi-structured questionnaire followed by field visit. Tree species diversity and richness was expressed in terms of Shannon-Weiner (H') diversity and Margalef richness (S) indices. A generalized linear regression model (Poisson) and Kruskal-walis tests were used to study the relationship between farmer and farm characteristics and tree composition. In all 16,764 trees belonging to 37 species and from 19 families were enumerated. The dominant species identified were: Tectona grandis Anacardium occidentale, Mangifera indica and Elaeis guineensis. There was a significant difference in on-farm tree species richness in the various land use types (p-value <0.05). Likewise, farm size, land ownership and land use type were found to influence tree diversity and richness (p <0.05).