Clustering of Leaf Characteristics in Secondary Dry Dipterocarp Forest based on Particle Swarm Optimization

The study of the relationship between forest and environment especially in climate change is needed for forest management. Seasonally dry tropical forest is one type of the tropical forests that is a home of biodiversity hotspots. The species in the deciduous forest has the various response to climate variables. The optimization approach called Particle Swarm optimization (PSO) is a technique that has been recognized in a lot of research as an effective approach to analyze data. This approach has been applied to solve many problems. In this research, we applied PSO to cluster the leaf characteristic patterns in secondary dry dipterocarp forest. The results showed the main patterns of leaf characteristics in the forest can be extracted by PSO. Nevertheless, it cannot separate the small different characteristics of leaf phenology as complete and incomplete in shorten period during the usual dry season.