Diversity and spatial clustering of shade trees affect cacao yield and pathogen pressure in Costa Rican agroforests

The importance of the spatial organisation of individuals in explaining species coexistence within a community is widely recognised. However, few analyses of spatial structure have been performed on tropical agroforests. The main objective of this study was to highlight the links between spatial organisation of shade trees on the one hand, and shade tree species richness and cacao yield on the other, using data from 29 cacao agroforests in Costa Rica. A method of spatial statistics, Ripley's K-function, was used to analyse the spatial organisation of shade and cacao trees in the study plots. For each stand, the X and Y coordinates of >= 2.5-m-tall trees were recorded. In each plot we also assessed shade tree species richness and cacao yield (with total number of pods = number of pods damaged by frosty pod rot + number of healthy pods). Three types of stands were identified: the first was characterised by significant clustering of shade trees, the highest shade tree species richness (S = 6), and the highest number of damaged pods (139 pods ha(-1) year(-1)). The second type was characterised by random spatial organisation of shade trees. The third type showed a trend towards regular organisation. Species richness of shade trees did not differ significantly between the last two types (S = 4 for both), nor did the number of damaged pods (56 pods ha(-1) year(-1) and 67 pods ha(-1) year(-1) respectively). Although the trends were not statistically significant for all the variables in our data set, the clustered spatial structure appears to favour a synergy between environmental (tree species richness), and provisioning (cacao production) services.