RESULTS

Beta diversity values

As shown in Table 2, the expected β-diversity of each plot at the region, zone, area, district, and plot levels ranged between 0.332 and 0.684, 0.107 and 0.664, 0.046 and 0.694, 0.042 and 0.695, and 0.000 and 0.537, respectively. The mean values of the expected β-diversity decreased as the pooling size decreasing from the regional to the plot level. In addition, the observed β-diversity of each plot ranged between 0.000 and 0.737, and the mean value was found to be higher than the expected β-diversity at the plot level.

Effects of the environmental filtering and spatial aggregation processes

As shown in Table 3, the effects of the environmental filtering at the region-zone, zone-area, area-district, and district-plot scales were 0.077, 0.027, 0.019, and 0.189, respectively. The environmental filtering effects at each spatial scale were all observed to be significant. The effect of the spatial aggregation was 0.074 and found to be significant at the within-plot scale. The change in the β-diversity patterns (Figure 4) indicated that the magnitude of ecological effects was the strongest at the district-plot scale, followed by that at the region-zone and within-plot scales. The comparison of ecological effects along the latitudinal gradients (Figure 5) showed that the environmental filtering effects at the region-zone and area-district scales had significantly increased with increasing latitude. In addition, the environmental filtering effects at the district-plot scale, and the spatial aggregation effects at the within-plot scale, decreased significantly with increasing latitudes.

Key factors of the environmental filtering processes

At the region-zone, zone-area, area-district, and district-plot scales, approximately 74%, 22%, 20%, and 33% of the environmental filtering effects, respectively, could be explained by the climate variables and plot attributes (Table 4). At all four scales, both the temperature and precipitation factors significantly contributed to the environmental filtering. Furthermore, the other climate variables and plot attributes had contributed to the environmental filtering effects, including the wind speed, elevation, and soil depth at the region-zone scale; solar radiation, wind speed and litter thickness at the zone-area scale; solar radiation and litter thickness at the area-district scale; and slope, aspect and litter thickness at the district-plot scale.