Agroforestry and Forestry in Sulawesi series: Modelling tree production based on farmers’ knowledge: case for kapok (Ceiba pentandra) and candlenut (Aleurites mollucana) under various agroforestry scenarios

The role and value of species in agroforestry systems are often well known, however, less is known of the actual growth and production of many of these species, particularly, for underutilized species. This limited information hinders the assessment of growth and production of the species. Candlenut (Aleurites moluccana) and kapok (Ceiba pentandra) are two underutilized agroforestry species for which limited information is available. Growth and production (yield) predictions can be used to identify the best strategies for managing agroforestry systems and analysing the costs and benefits of various scenarios.
Production and growth estimation of candlenut and kapok in this study combined measurement of growth indicators (diameter, crown width) with farmers’ local knowledge. The growth prediction model, Spatially Explicit Individual-based Forest Simulator (SExI-FS), was used to explore the yield dynamics of candlenut and kapok under various agroforestry scenarios. Farmers’ perceptions were shown to be reliable for estimating fruit yields. Perceptions were more reliable from farmers who directly harvested the fruit. The inclusion of Crown Position and Crown Form indexes improved the estimation correlation of yields with tree diameters. The simulation model in a mixed kapok and candlenut scenario showed that candlenut was more competitive than kapok if planted at more than 30% of the total plant density.