Detangling the Effects of Environmental Filtering and Dispersal Limitation on Aggregated Distributions of Tree and Shrub Species: Life Stage Matters.

The pervasive pattern of aggregated tree distributions in natural communities is commonly explained by the joint effect of two clustering processes: environmental filtering and dispersal limitation, yet little consensus remains on the relative importance of the two clustering processes on tree aggregations. Different life stages of examined species were thought to be one possible explanation of this disagreement, because the effect of environmental filtering and dispersal limitation are expected to increase and decrease with tree life stages, respectively. However, few studies have explicitly tested these expectations. In this study, we evaluated these expectations by three different methods (species-habitat association test based on Poisson Clustering model and spatial point pattern analyses based on Heterogeneous Poisson model and the jointly modeling approach) using 36 species in a 20-ha subtropical forest plot. Our results showed that the percentage of species with significant habitat association increased with life stages, and there were fewer species affected by dispersal limitation in later life stages compared with those in earlier stages. Percentage of variance explained by the environmental filtering and dispersal limitation also increases and decreases with life stages. These results provided a promising alternative explanation on the existing mixed results about the relative importance of the two clustering processes. These findings also highlighted the importance of plant life stages for fully understanding species distributions and species coexistence.

[Topographic distribution patterns of forest gap within an evergreen broad-leaved forest in Tiantong region of Zhejiang Province, Eastern China].

To explore the effects of multi-dimensional topographic factors on forest gap distribution, the forest gaps in a 20 hm2 dynamic monitoring plot of an evergreen broad-leaved forest in Tiantong region of Zhejiang were taken as the objects to study the distribution patterns of the gap fraction, gap density, and gap area under the effects of altitude, slope degree, slope shape, slope aspect, and slope position by using a geographic information system (GIS) software. In the plot, the gap fraction was 13.1% , gap density was 9.5 ind.hm-2, and average gap area was 137.82 m2. Because of the greater intensity of typhoon disturbance at high altitudes, the gap fraction and gap density at the high altitude (> or =500 m) sections were significantly larger than those at the medium and low altitude (<500 m) sections. The heavy precipitation produced by typhoon could easily cause small scale landslide, and thus, lead to the gap fraction and gap density being larger in valley area than in side-slope and ridge. It was suggested that typhoon and its produced heavy precipitation could be the main causes of the significant differences in the forest gaps along the gradients of altitude and slope position.

[Spatial heterogeneity of soil properties and its relationships with terrain factors in broad-leaved forest in Tiantong of Zhejiang Province, East China].

By using geostatistical methods, this paper studied the spatial heterogeneity and distribution patterns of soil pH, total carbon, total nitrogen, and total phosphorus in an evergreen broad-leaved forest in Tiantong of Zhejiang Province, and the effects of terrain factors (elevation, convexity, and slope) on the soil properties were quantified based on RDA ordination and partial regression analysis. The coefficient of variation for the soil pH, total carbon, total nitrogen, and total phosphorus was 5.18%, 42.98%, 36.55%, and 46.27%, respectively, and the spatial dependence of the soil properties was at a scale of 81.6-54.5 m. The soil pH, total carbon, and total nitrogen had moderate spatial autocorrelation, while the soil total phosphorus had a strong spatial autocorrelation. The soil pH, total carbon, and total nitrogen showed scattered spatial distribution, while the soil total phosphorus presented banded type. Among the terrain factors, convexity had the strongest effects on the soil pH and total phosphorus, both of which had negative correlation with convexity, and the convexity could explain 21.24% and 14.62% of the spatial variability of soil pH and total phosphorus, respectively. Elevation had the most powerful effects on the soil total carbon and total nitrogen, both of which had positive correlation with elevation, and the elevation could explain 10.54% and 10.60% of the spatial variability of soil total carbon and total nitrogen, respectively. There existed differences in the effects of different terrain factors on the spatial variability of the soil properties, which was related to the effects of terrain factors on the distribution of acidic rainfall in the region and on the local soil moisture content and air temperature.