Community assembly during secondary forest succession in a Chinese subtropical forest

Subtropical broad-leaved forests in southeastern China support a high diversity of woody plants. Using a comparative study design with 30 × 30 m plots (n = 27) from five successional stages (<20, <40, <60, <80, and ≥80 yr), we investigated how the gradient in species composition reflects underlying processes of community assembly. In particular, we tested whether species richness of adult trees and shrubs decreased or increased and assessed to which degree this pattern was caused by negative density dependence or continuous immigration over time. Furthermore, we tested whether rare species were increasingly enriched and the species composition of adult trees and shrubs became more similar to species composition of seedlings during the course of succession. We counted the individuals of all adult species and shrubs >1 m in height in each plot and counted all woody recruits (bank of all seedlings ≤1 m in height) in each central 10 × 10 m quadrant of each plot. In addition, we measured a number of environmental variables (elevation, slope, aspect, soil moisture, pH, C, N, and C/N ratio) and biotic structural variables (height and cover of layers). Adult species richness varied from 25 to 69 species per plot, and in total 148 woody species from 46 families were recorded. There was a clear successional gradient in species composition as revealed by nonmetric multidimensional scaling (NMDS), but only a poor differentiation of different successional stages with respect to particular species. Adult richness per 100 individuals (rarefaction method) increased with successional stage. None of the measured abiotic variables were significantly correlated with adult species richness. We found no evidence that rare species were responsible for the increasing adult species richness, as richness of rare species among both adults and recruits was independent of the successional stage. Furthermore, the similarity between established adults and recruits did not increase with successional stage. There was a constant number of recruit species and also of exclusive recruit species, i.e., those that had not been present as adult individuals, across all successional stages, suggesting a continuous random immigration over time.