Spatial Distribution Patterns And Correlation Between Species Diversity And Functional Diversity In Tropical Montane Rain Forest Of Jianfengling,Hainan Island

Biodiversity had always attracted the interests of ecologists.In recent years the research about functional diversity was also paid more and more attention to explain the relationship between biodiversity and ecosystem function.In this study,based on the 60 hm2 dynamic plot dataset of tropical mountain rain forest in Jianfengling,Hainan Island,we tried to illustrate the effects of spatial scales on species and functional diversity patterns.We also analyzed the relationship between functional diversity and species diversity as well as aboveground biomass.It is aim to explore the plant spatial distribution patterns and coexistence mechanisms and provide more scientific proofs to understand the relationship between biodiversity and ecosystem function in tropical forest areas.The main results are listed as follows:(1)Results showed that spatial distribution patterns of species richness,species abundance and Shannon-Wiener index are more significantly influenced by the spatial scales than Simpson and Pielou's evenness indices.Variance of the species richness with the increase of spatial scales was unimodal,and had the maximum value at the 20 m × 20 m scale.Variance of the species abundance showed a linear relationship with the increase of spatial scales.The species-area curves of the four different subplots indicated that high species richness observed at a particular spatial scale may become low at another spatial scale.Our results suggested that it is necessary to take the location of the subplot and spatial scale into account when analyzing the species biodiversity.(2)For the eight plant functional traits,LCC has the minimum variation coefficient value(8.30 %)and LKC has the maximum variation coefficient value(65.32 %)at the species level.In general,variation coefficients of LCC and pH were less than 10.0 % and SLA,LNC and LPC were between 25.0 % and 40.0 %.These variations of functional traits were prossibily correlated with the local climate in Jianfengling,which were formed gradually in long-term evolution history.Variation coefficient of functional richness with the increase of spatial scales was unimodal,and had the maximum value at 10 m × 10 m scale.Variation coefficient of functional evenness decreased at first and then increased.And the minimum value appeared at the 20 m × 20 m scale.Variation coefficient of the functional divergence and functional dispersion showed a negative linear relationship with the increase of spatial scales.(3)At different sampling scales,species richness was significant positive with functional richness.With the increasing of species richness,the probability of extreme functional traits in community will also increase and this will eventually lead to the increasing of functional richness.At the same time,this phenomenon could also be explained by the niche theory.We believe that there are no any two species with the exactly same functional characteristics in a community,and so the greater the number of species are,the larger the change in functional characteristics and the greater the functional richness are.For the species-area curve of the 60 hm2 plot in Jianfengling,the species accumulation rate became flat at the 7.5 hm2 sampling area scale,but for the FRic-area curve,the FRic accumulation rate became flat at the 40 hm2 sampling area scale.Minimum sampling area reflected the species composition and the basic structure of the community.Compared to the species-area curve,the FRic-area curve could fully and objectively reflected the role of species and ecosystem function.(4)At the 20 m × 20 m scale,there were no correlation between FEve,FDiv and AGB and significant negative correlation between FRic,FDis and AGB based on eight functional traits,though both of them had low coefficients of determination(R~2).Based on the 22 functional groups we found that: FRic,FEve could explain AGB well when calculated by the functional groups of SLA,LCC and LPC,with each of the coefficient of determination 0.0188 and 0.0307,respectively.FDiv,FDis could explain AGB well when calculated by the functional groups of WD,LCC and LPC,with each of the coefficient of determination 0.012 and 0.0775,respectively.Compared to functional diversity indexes,no matter based on the relative species abundance or the relative basal area,CWM could explain AGB better.Because it not only took the difference of plant functional traits into account,but also considered the positions and roles(such as relative abundance and relative dominance)of species in community.