Clear-cutting has been a widespread commercial logging practice, causing substantial changes of biodiversity in many forests throughout the world. Forest recovery is a complex ecological process, and examining the recovery process after clear-cutting is important for forest conservation and management. In the present study, we established fourteen 20 m × 20 m plots in three recovery stages (20-year-old second growth, 35-year-old second growth and old growth) and explored the changes in evergreen and deciduous species diversity after clear-cutting in a subtropical evergreen-deciduous broadleaved mixed forest in central China. The results showed that total species richness was highest at the intermediate recovery stage. The species richness and stem abundance of evergreen species increased, while total and deciduous species stem abundance decreased with forest recovery. The basal area of both total and evergreen species increased, while that of the deciduous species showed a unimodal pattern. The abiotic conditions varied with the recovery process. Changes in species compositions were generally correlated with soil pH, total phosphorus, and CO. Our results suggest that deciduous species richness and stem abundance can recover after 20–35 years, but evergreen species need more time to recover following clear-cutting.

Biogeographic patterns of soil fungal diversity have been well documented in forest ecosystems, but the underlying mechanisms and processes that shape these patterns remain relatively unknown. This study took soil samples from 300 forest plots spanning six forest types along a latitudinal gradient in eastern China, which ranges from tropical rainforest to boreal forest ecosystems. A null-model analysis was used to compare the observed soil fungal beta diversity (β-diversity) with the β-diversity expected from random sampling of each local species pool. We also compared the relative importance of environmental and spatial variables on soil fungal β-diversity among forest types along the latitudinal gradient. Our results found that observed β-diversity was greater than expected β-diversity in all six forest types, which means that species tend to be more aggregated than expected. We determined that this species aggregation resulted from both environmental filtering and species dispersal limitations. Further, environmental variables had stronger influences on β-diversity than spatial dispersions. Additionally, the co-occurrence network showed that more species interactions occurred in the mid-latitude forests which lead to decreased soil fungal β-diversity and low interpretations of environmental and spatial variables. Study of these processes in different forest types along latitudinal gradients will provide important insights that local differences in the relative importance of different community assembly processes creates different gradients in global biodiversity.

Variation patterns in plant functional traits and their interrelationships play a crucial role in understanding species coexistence mechanisms and ecological differentiation within local plant communities. However, the dynamic patterns of plant functional traits across different forest successional stages remain insufficiently understood. Here, we investigated the woody species composition of subtropical evergreen–deciduous broadleaved mixed forest across 75 plots, representing three successional stages (20-year-old secondary forest, 35-year-old secondary forest, and old-growth forest (>80 years)), in Xingdoushan and Mulinzi National Nature Reserves, Hubei Province, Central China. We measured four functional traits of woody plants: leaf area (LA), specific leaf area (SLA), leaf dry matter content (LDMC), and wood density (WD). For each different age plant community, we calculated (1) species abundance-weighted mean community trait values, and (2) species-level mean trait values. We applied trait gradient analysis to partition and assess correlations of four functional traits across communities of different successional stages, separating within-community (α components) and between-community (β components) variation. To quantify the extent to which environmental constraints influence trait expression, we used the ecological constraint index (C_i). The results revealed significant variation in the four functional traits across communities at different successional stages. Community-level mean LA and SLA decreased significantly with age, WD increased significantly with age, and there was no significant relationship between LDMC and age. The α trait components consistently varied more widely than β components at different successional stages, indicating that biological competition dominates the assembly of local forest communities across various successional stages. Correlations between the four functional traits were dynamically adjusted with the study scale (community-level and species-level) and forest age. The ecological constraints on the four functional traits varied significantly across forest successional stages, with SLA being subject to the strongest constraints. Our findings reveal that biotic competition predominantly shapes community assembly during the succession of subtropical evergreen–deciduous broadleaved mixed forests, while stronger ecological filtering in old-growth stands underscores their role in maintaining ecosystem stability. These insights support more effective conservation and restoration strategies.