Stand spatial structure and microbial diversity are key drivers of soil multifunctionality during secondary succession in degraded karst forests.

Studying the relationship between biodiversity and ecosystem multifunctionality (the ability of ecosystems to provide multiple ecosystem functions) (BEMF) is a current hotspot in ecology research. Previous studies on BEMF emphasized the role of plant and microbial diversity but rarely mention stand spatial structure. To investigate the effect of stand spatial structure on BEMF, this study established 30 forest dynamic plots in three natural restoration stages (shrubbery, secondary growth forest, and old-growth forest) in Maolan National Nature Reserve, Guizhou province, China. A positive response in soil multifunctionality (SMF), plant species diversity, stand spatial structure, and fungal ß diversity (p < 0.05) followed natural restoration. However, bacterial ß diversity showed a negative response (p < 0.05), while microbial α diversity remained unchanged (p > 0.05). These results based on a structural equation model showed that plant species diversity had no direct or indirect effect on SMF, soil microbial diversity was the only direct driver of SMF, and stand spatial structure indirectly affected SMF through soil microbial diversity. The random forest model showed that soil microbial ß diversity and the Shannon-Wiener index of the diameter at breast height for woody plant species were the optimal variables to characterize SMF and soil microbial diversity, respectively. These results suggested that natural restoration promoted SMF, and microbial diversity had a direct positive effect on SMF. In the meantime, stand spatial structure had a significant indirect effect on SMF, while plant species diversity did not. Future work on degraded karst forest restoration should direct more attention to the role of the stand spatial structure and emphasize the importance of biodiversity.

Variations in species diversity patterns and community assembly rules among vegetation types in the karst landscape.

The various vegetation types in the karst landscape have been considered the results of heterogeneous habitats. However, the lack of a comprehensive understanding of regional biodiversity patterns and the underlying ecological processes limits further research on ecological management. This study established forest dynamic plots (FDPs) of the dominant vegetation types (shrubland, SL; mixed tree and shrub forest, MTSF; coniferous forest, CF; coniferous broadleaf mixed forest, CBMF; and broadleaf forest, BF) in the karst landscape and quantified the species diversity patterns and potential ecological processes. The results showed that in terms of diversity patterns, the evenness and species richness of the CF community were significantly lower than other vegetation types, while the BF community had the highest species richness. The other three vegetation types showed no significant variation in species richness and evenness. However, when controlling the number of individuals of FDPs, the rarefied species richness showed significant differences and ranked as BF > SL > MTSF > CBMF > CF, highlighting the importance of considering the impacts of abundance. Additionally, the community assembly of climax communities (CF or BF) was dominated by stochastic processes such as species dispersal or species formation, whereas deterministic processes (habitat filtering) dominated the secondary forests (SL, MTSF, and CBMF). These findings proved that community assembly differs mainly between the climax community and other communities. Hence, it is crucial to consider the biodiversity and of the potential underlying ecological processes together when studying regional ecology and management, particularly in heterogeneous ecosystems.

How do species richness and its component dependence vary along the natural restoration in extremely heterogeneous forest ecosystems?

Giant habitat heterogeneity is an important factor contributing to the high species richness (SR) in karst forests. Yet, the driving factor behind the alterations in SR patterns during natural restoration remains unclear. In this study, we established the forest dynamics plots along the natural restoration sequence (including shrub-tree mixed forest stage (SC), secondary forest stage (SG) and old-growth forest sage (OG)) in degraded karst forests to compare the SR and the dependence on its components (including total community abundance, species abundance distribution (SAD), and conspecific spatial aggregation (CSA)) among stages of natural restoration. By evaluating the degree of contribution of the components to local SR and rarefied SR, we found that the SG exhibited the highest local SR, while the rarefied SR remained increasing along the restoration sequence after controlling the sample size. At SC-SG stage, SAD and CSA contributed negatively to the differences in SR, while abundance made a positive contribution to SR differences. At SG-OG, abundance contributed positively to the difference in SR at all scales, while SAD contributed negatively at small scales. No significant contribution of CSA was found at observed scales. In addition, local SR varied more significantly with PIE than with abundance. Our research emphasizes the importance of eliminating the influence of abundance on species richness in forest ecology and management, as well as the significance of separately evaluating the components that shape the diversity patterns.

Dominant influence of plants on soil microbial carbon cycling functions during natural restoration of degraded karst vegetation.

The impacts of natural restoration projects on soil microbial carbon (C) cycling functions have not been well recognized despite their wide implementation in the degraded karst areas of southwest China. In this study, metagenomic sequencing assays were conducted on functional genes and microorganisms related to soil C-cycling at three natural restoration stages (shrubbery, TG; secondary forest, SG; old-growth forest, OG) in the southeast of Guizhou Province, China. The aims were to investigate the changes in microbial potentials responsible for soil C cycling and the underlying driving forces. The natural restoration resulted in vegetation establishment at all three restoration stages, rendering alterations of soil microbial C cycle functions as indicated by metagenomic gene assays. When TG was restored into OG, the number and diversity of genes and microorganisms involved in soil C cycling remained unchanged, but their composition underwent significant shifts. Specifically, microbial potentials for soil C decomposition exhibited an increase driven by the collaborative efforts of plants and soils, while microbial potentials for soil C biosynthesis displayed an initial upswing followed by a subsequent decline which was primarily influenced by plants alone. In comparison to soil nutrients, it was determined that plant diversities served as the primary driving factor for the alterations in microbial carbon cycle potentials. Soil microbial communities involved in C cycling were predominantly attributed to Proteobacteria (31.87%-40.25%) and Actinobacteria (11.29%-26.07%), although their contributions varied across the three restoration stages. The natural restoration of degraded karst vegetation thus influences soil microbial C cycle functions by enhancing C decomposition potentials and displaying a nuanced pattern of biosynthesis potentials, primarily influenced by above-ground plants. These results provide valuable new insights into the regulation of soil C cycling during the restoration of degraded karst vegetation from genetic and microbial perspectives.

Soil capacity of intercepting different rainfalls across subtropical plantation: Distinct effects of plant and soil properties.

Forest management practices play an important role in soil water conservation. However, the soil water-holding capacity and associated drivers under different management practices remain uncertain, especially when the precipitation varies substantially at the regional scale. Here, we used hydrogen stable isotope to explore the contribution of rainfall to soil water (CRSW) under light, moderate and heavy precipitation in Pinus massoniana plantations with multiple management practices (pure stand, mixed stand, understory removal, light-intensity thinning and high-intensity thinning) in subtropical China. We further used variation partitioning analysis and structural equation modeling to identify the dominant driver affecting CRSW. Our results showed that after light rainfall, the highest CRSW (28.7%) was found in the high-intensity thinning plantation. However, after heavy rainfall, the high-intensity thinning plantation received the lowest CRSW (43.3%), while the mixed stand showed the highest CRSW (67.1%). These results demonstrated that the mixed stand of P. massoniana had a stronger capacity for soil water conservation, whereas high-intensity thinning showed poorer capacity. Furthermore, our results revealed that plant properties (i.e., tree, root and litter biomass) were the dominant controls of the CRSW under light rainfall, while soil properties (i.e., bulk density, total porosity, field capacity) were the primary drivers under moderate and heavy rainfall, indicating that the determinants influencing plantation capacity for intercepting rainfall vary with different levels of precipitation. These results highlight the importance of the level of precipitation in determining the dominant driver of CRSW. More importantly, these results suggest that the mixed stand, rather than high-intensity thinning, was better forest management since the former has a higher capacity for intercepting heavy rainfall.

[Effects of water vapor source and local environmental factors on the hydrogen and oxygen isotopic compositions of precipitation in Huitong, Hunan Province, China].

Deuterium (D) and oxygen-18 (18O) are common environmental tracers in water. Understanding the isotopic compositions of precipitation is necessary for further studies on local and global water cycling processes. To reveal the mechanism of isotopic compositions of precipitation in subtropical monsoon region in response to environmental changes, we collected 49 precipitation samples and recorded related environmental factors from May 2017 to August 2019 in Huitong field station of Chinese Academy of Sciences in Hunan Province. We analyzed the temporal variations in D and 18O values in precipitation and analyzed the influence of water vapor source and local environmental factor on stable isotopic compositions of precipitation. The local meteoric water line was established as δD=(7.45±0.17)δ18O+(10.10±1.25) (R2=0.93, P<0.01), the slope of which was slightly lower than China's meteoric water line and the global meteoric line. The D and 18O values of precipitation samples were closely coupled with local meteorological conditions and dominant moisture sources. The 18O and D contents were depleted during summer monsoon season but enriched during winter monsoon season. During the summer monsoon and post-monsoon seasons, precipitation in this area came mainly from the Bay of Bengal, the South China Sea, and the West Pacific at three different types of air pressure, which lead to the similarity of the D-excess value of the precipitation to global average. The lower intercept of meteoric water line and the higher D-excess value for precipitation during the winter monsoon season resulted from moisture from remote westerly air masses, degenerated tropical marine air masses from the Bay of Bengal, and inland moisture in the pre-monsoon period, which were also affected by local environmental factors.