Quantitative assessment of multispecies spatial pattern with high species diversity

This paper developed statistical methods for quantitatively assessing spatial pattern of communities with abundant species richness such as in tropical forests. Based on multivariate point processes, species richness and the Simpson's diversity index can be extended to the functions illustrating basic characteristics of multispecies spatial pattern. The spatial extension of the species richness is a sum of the detectabilities of constituent species within a given distance. By means of calculating the functions for individuals belonging to specific size class, the size structure of multispecies spatial pattern can also be examined. Therefore, comprehensive analysis of species, size, and spatial pattern together with the roles of each species in a community can be conducted. This paper demonstrated its descriptive utility in exploratory analysis by applications to four subtropical forest tree communities of different ages in Okinawa Island, southern Japan, all of which have abundant species richness. The results quantitatively revealed the contrasts among the four stands ranging from a young secondary stand to an old-growth forest as well as changes of relative positions of species in communities depending on their ecologic properties.