Regressive dynamics of coffee agroforestry systems: determinants and effects on system composition and structure in Cameroon

The aim of this study was to analyse the structure and diversity of coffee agroforestry systems, with an emphasis on the spatial dynamics of land use in the different coffee production basins and agro-ecological zones and their associated determinants in Cameroon. Satellite image data, field inventories in eighty-one (81) 40 × 40 m plots, as well as surveys of 240 producers, were conducted. The results showed that the largest conversions and changes in land use occurred during the period 2001–2019, with a reduction in agroforestry and forest cover and an expansion in the area covered by other crops. In total, 49 woody species divided into 24 families, were inventoried in the coffee agroforestry systems. The importance value index reveals that Elaeis guineensis (28.5%), Dacryodes edulis (25.9%), Persea americana (17.5%), Leuceana leucocephala (15.8%), and Mangifera indica (12.4%) were the most important. The diversity indices show that the systems in the forest zones (Mungo and Haut-Nyong) were more diversified than those in the humid highland savannah (Noun); however, the density of woody plants is higher. The diametric structure of coffee trees, characterized by the limited presence of young coffee trees, reflects the low level of renewal in the coffee agroforests in all the production basins, where some farmers replace dead or unproductive coffee trees with other crops, such as Theobroma cacao, Elaeis guineensis, Musa spp. and fruit trees. This is due to a number of constraints faced by the producers, and the analysis indicated that the main variables that significantly predicted the susceptibility of the farmers to adopt an intervention strategy of preservation or conversion of the agroforestry system were farm size and farm age. Taking these variables into consideration is necessary for decision-making and for the development of coffee agroforestry systems in similar contexts.