Based on the large dataset from Jianfengling National Forest Ecological Station, Research Institute of Tropical Forestry (RITF), Chinese Academy of Forestry (CAF), the species spatial distribution along climate and edaphic gradients were analyzed and some important findings were reported as follows.
A large data set consisting of 580 tropical tree species sampled in 163 25×25 m quadrats along an altitudinal gradient covering an area of 160 km2 of tropical rain forest in Jianfengling reserve (Hainan Island, China). For each plot, the data include tree species and abundance, altitude and six soil properties from which a two dimensional environmental space was constructed.
With this extensive data set, we tested the hypothesis that different combination of environmental factors can generate multiple hotspots on three axes of diversity: species richness, Shannon-equivalent species richness and habitat preference, a measure of evenness in the distribution of individuals across an environmental gradient.
We found that humid and cool areas with more nitrogen availability were occupied by richer and more diverse communities of wide range species. Rare (in terms of number of individuals) and range-restricted species instead, tended to prefer minor habitats, generally warmer with high potassium, calcium, magnesium and, in particular, phosphorous. As a result, wide and range-restricted species were segregated across the environmental space.
These findings indicate rare species tend to occur more frequently where common species are less abundant. A clear pattern of species richness and diversity was driven by a combination of several environmental factors (soil properties and climate). The complexity of the environment not only explains the different species distribution along each habitat, but also determines the relative abundance of each species in the entire community. Although some habitats have low species richness and diversity, they are highly preferred by rare species; therefore, biodiversity conservation efforts should consider protecting these fragile ecosystems.
This paper was published in Journal of Ecology. The reviewers regarded that this paper solved the question about the relationship between species distribution and habitat utilization in ecology, biogeography and conservation biology. It is important and can be used to define the boundary of natural protection area, manage the natural protection area and forest park, and promote the recovery of tropical natural forests.
Reference：Han Xu, Matteo Detto, Suqin Fang, Yide Li*, Runguo Zang and Shirong Liu. 2015. Habitat hotspots of common and rare tropical species along climatic and edaphic gradients. Journal of Ecology, DOI: 10.1111/1365-2745.12442.
Fig. 1. Correlation matrix between species distributions along range sizes. Colours indicate Pearson coefficient of correlation. Range size is normalized from zero to one. As an example, the distributions of large (h = 0.75) and small range species (h0 = 0.15) and their scatter plots are also shown.
Fig. 2. (a–c) Species richness, Shannon-equivalent species richness and habitat preference distributions along the first two principal components. (d–f) Expectations of null model 1: random habitat association. (g–i) Expectations of null model 2: Mid-domain effect. (j–l) Expectations of null model 3: given richness–habitat association, diversity is independent on habitat. Contour lines are defined by the 95% confidence regions of 1000 randomizations with black (red) lines indicating lower (higher) than expected.