Study On Dynamics Of Tree Species Composition And Spatial Pattern In The Tropical Lowland Rainforest On Hainan Island, China

A major challenge in studies of tropical forest successional dynamics is to reveal therelative importance of deterministic versus stochastic processes affecting speciescomposition, spatial distributions, and their rates of change. In this thesis, we explored thespecies composition, tree population patterns and interspecific spatial associations along achronosequence of secondary forest plots, which were ranging from15years to60yearssince abandonment after shifting cultivation, and as two old-growth forest plots forcomparison. The results were listed as followings:We evaluated the species composition of different size classes (seedlings, saplings andadult trees) of communities. The results showed that both species diversity for all sizeclasses and the species similarity among size classes in old-growth forests exhibitedsignificantly higher than those in secondary forests, but the proportion of species dominancein old-growth community was lower than that of secondary forests. Species similaritybetween secondary forests and old-growth forests increased with forest recovery, supportingthe view of equilibrium succession dynamics. In each recovery stage, the number ofindividuals, species richness and abundance-based coverage estimator (ACE) in seedlingcommunity were lower than those both sapling communities and adult communities.Moreover, the species compositions for seedling communities of secondary forests differedsignificantly from that of both sapling communities and adult communities, suggesting thatthe recruitments into the seedling community may be a rather unpredictable process. Ourresults suggest that the community assembly processes during secondary forest recovery aredriven by both stochastic and deterministic processes.We used spatial point pattern analysis to elucidate the changes of spatial patterns of treespecies across forest ages, and applied the random-labelling null model within a case-controldesign to estimate the mutual effects of density-dependent. The results showed that most species were aggregated in plots of different ages and the small-scale aggregated speciesdecreased during succession. The number of species that juveniles as ‘cases’ showingadditional aggregation pattern was significantly lower than that of saplings, which suggestthat the extent of gathering around the adult tress from saplings to juveniles are declining.The strength of density dependence at scale of0–5m was decreased during secondarysuccession. Furthermore, the distance of spatial correlation between dead trees and adulttrees were different across forest ages. Our findings highlight density dependence as aprevalent mechanism for regulating the population spatial structure in successionalcommunities, and indicate that there are vary effect of density dependent by which dictatedthe variance of species abundance and life history in different successional stages.Habitatheterogeneity, seeddispersal limitation and interspecific competition could be confoundingfactors in studies that seek evidence for density dependence.We used techniques of spatial point pattern analysis to conduct a comprehensive test ofthe independence assertion by analyzing the species associations in successionalcommunities. We hypothesize that stochastic dilution effects owing to increasing speciesrichness overpower signals of species associations, thereby yielding approximate speciesindependence. The results showed that the proportion of species pairs with no signifcantinterspecifc association increased with during secondary succession and that the proportionof species pairs showing segregation decreased during succession. Moreover, the proportionof species pairs with non-signifcant interspecifc small-scale interactions should increaseduring succession. This suggests that independence may indeed be a good approximation inthe limit of very species-rich communities. Our findings were a step towards a betterunderstanding of factors governing species-rich communities and we proposed a hypothesisto explain why species placement in species-rich communities approximates independence.