Accurate vegetation phenology monitoring is essential for understanding ecosystem responses to climate change. Here, we systematically evaluated five NDVI datasets, Daily NDVI (1 day), GIMMS-3G+ (15 day), MOD13C2 (1 month), PKU GIMMS (15 day), and SPOT (10 days), for estimating the start (SOS), end (EOS), and length (GSL) of the growing season across China from 2000 to 2020, validated with ground observations from 99 phenological stations and 20 flux towers. The Daily NDVI product exhibited the strongest agreement with both validation sources, achieving the highest correlation (R2 > 0.37 with ground observations; R2 > 0.45 with flux tower data), the smallest bias (|PB| < 1%), and the lowest root mean square error (10–22 days). Statistical tests confirmed that only the Daily NDVI-derived phenology showed no significant difference from ground observations, while both Daily NDVI and PKU GIMMS NDVI showed no significant difference from flux tower estimates. All products captured the broad-scale spatial patterns, SOS delayed and EOS advanced with increasing latitude and altitude, resulting in a clear south-north gradient in GSL. However, long-term trends diverged substantially. While forest SOS trends were consistent across products (showing advancement), grassland SOS trends were highly variable. More notably, EOS and GSL trends showed pronounced inter-product divergence, particularly in grasslands. Overall, this study demonstrates that temporal resolution critically impacts phenological accuracy, and the gap-free Daily NDVI product captures fine-scale vegetation dynamics, yielding more reliable phenological metrics essential for quantifying ecosystem responses to climate change.

The seasonal variation characteristics of vegetation gross primary productivity (GPP) and its response mechanisms to climatic and vegetation structural factors represent a key scientific issue for assessing regional carbon sink functions. Based on MODIS GPP, leaf area index (LAI), and meteorological data from 2003 to 2020, we comprehensively employed trend analysis, partial correlation analysis, and multiple linear regression to investigate the spatiotemporal variations of GPP in spring, summer, and autumn at the seasonal scale in the Qinling-Daba Mountains. We further explored the synergistic driving mechanisms of vegetation structure and climatic factors. The results showed that, from 2003 to 2020, GPP in spring, summer, and autumn all exhibited significant increasing trends, with the magnitude ranked as summer > spring > autumn, and the proportions of areas with significant increasing trends were 82.9%, 64.1%, and 20.6%, respectively. LAI increased significantly in all the three seasons, with the proportion of increasing areas exceeding 86.9%. In spring, precipitation showed an overall fluctuating upward trend, vapor pressure deficit (VPD) displayed a downward trend, while temperature remained relatively stable. In summer, both precipitation and VPD exhibited overall downward trends, while maximum temperature showed a significant upward trend, indicating a certain warming-drying trend in the climate. In autumn, precipita-tion increased slightly, while temperature and VPD remained relatively stable. Regarding the temporal effects of climatic factors, there were significant seasonal differences. In spring, precipitation was mainly characterized by lag effects, temperature by synchronous effects, and VPD by cumulative effects. In summer, all climatic factors were mainly characterized by synchronous effects. In autumn, precipitation, maximum temperature, and VPD were domi-nated by synchronous effects, while minimum temperature was mainly controlled by cumulative effects. LAI was the primary positive driving factor for the seasonal-scale increase in GPP. Temperature factors predominantly exerted promoting effects. Precipitation showed inhibiting effects in spring and autumn but promoting effects in summer. VPD exhibited inhibiting effects in both spring and summer, but a weak promoting effect in autumn. At the seasonal scale, we revealed the dominant driving mechanism of improved vegetation canopy structure on GPP enhancement in the Qinling-Daba Mountains, providing scientific basis for evaluating the carbon sink function of regional ecosystems.

Climate warming is exacerbating soil drought and precipitation events, as well as different land use types lead varying nutrient inputs, driving substantial shifts in microbial communities that may influence soil respiration. Microbial thermal compensation mechanisms serve as critical biological regulators, alleviating the warming-induced enhancement of soil respiration. However, the effects of soil moisture and land use types on the thermal responses of microbial respiration remain insufficiently understood and constrained. Here, we collected soil from four distinct sites, each comprising both farmland and forest, and conducted indoor experiments to simulate drought and rewetting events. We then assayed the thermal responses of microbial respiration rates at 40

The ecological niche is crucial for revealing and predicting the impacts of environmental changes on communities. However, the niche differentiation mechanism of soil microbial communities on the nutrient limitation shaped by abandoned agricultural succession remains unclear. Here, we assessed the microbial niche breadth along a long-term post-agricultural succession sequence (1- to 150-year-old stages) and linked it with microbial nutrient limitations by combining the eco-enzymatic stoichiometry model with high-throughput sequencing of 16S rRNA and ITS genes. We found that bacterial communities exhibited a broader environmental niche breadth than fungal communities across the abandoned agricultural successional gradient, with a significantly negative correlation between the bacterial and fungal niche breadth. Vegetation succession enhanced the bacterial niche breadth but decreased the fungal niche breadth. Critically, bacterial niche breadth expanded under phosphorus (P) limitation but contracted under carbon (C) limitation, whereas fungal niche breadth showed the opposite pattern. Using network modularity analysis, we identified that specific microbial modules dominated by Acidobacteria and Proteobacteria (bacteria), along with Ascomycota and Basidiomycota (fungi), drove these niche breadth dynamics, with module abundances tightly coupled to nutrient limitations. Partial least squares path model further demonstrated that soil nutrients regulate microbial niche breadths indirectly by altering community complexity, structure, and metabolic activities. Together, our findings reveal that nutrient limitation acts as a key environmental filter that differentially shapes bacterial and fungal ecological strategies, driving their niche differentiation through both physiological constraints and biotic interactions. This mechanistic understanding offers critical insights for harnessing soil microbial communities as promising tools to support restoration practices and biodiversity conservation in abandoned agricultural systems.

The nutrient cycling mediated by soil microorganisms is crucial for maintaining forest ecosystems' stability and sustainable development. However, a comprehensive understanding of how microbial community assembly processes affect multiple nutrient cycling (MNC) remains unclear. In this study, rhizosphere soils were collected from different seedling ages (2-, 3-, and 6-year-old) under varying canopy densities, including forest gap (FG), forest edge (FE), medium canopy density 0.4–0.6 (MCD), and high canopy density 0.7–0.9 (HCD). The aims were to evaluate changes in MNC, the assembly processes of bacterial and fungal communities, and the main factors influencing MNC during early succession. The results demonstrated that MNC indices in FG and FE were significantly higher than those in MCD and HCD (P < 0.05). Moreover, stochastic processes dominated the bacterial community, while deterministic processes dominated the fungal community during the early succession of the secondary forest. Co-occurrence networks suggested that the bacterial community was more complex in HCD, whereas the fungal community exhibited greater complexity in FG. Random forest (RF) indicated that soil organic carbon (SOC) and total phosphorus (TP) were the primary nutrient predictors, and structural equation modeling (SEM) suggested that microbial community assembly processes were also the main indicators affecting MNC. Specifically, the assembly processes of the bacterial community (stochastic processes) were positively correlated with MNC, while those of the fungal community (deterministic processes) showed a negative correlation with MNC. Overall, the bacterial and fungal communities exhibited distinct assembly processes in early succession, with different effects on MNC. The findings enhanced our understanding of the potential links between soil microbial communities and multiple nutrient cycling.

Insects and pathogens interact with plant species, influencing the natural regeneration of plant populations. However, the mechanisms by which they mediate the early regeneration process remain unclear. Herein, we evaluated the pest and disease status of 380 Quercus aliena var. acuteserrata seedlings and its effects on their photosynthetic and physiological properties in the Qinling Mountains, China. Results indicated that 243 seedlings were affected by pests and diseases, mainly in the 0−20 cm height range, and the affected leaf area was mainly in the range of 0−60%. Biology and soil factors were the primary drivers of these conditions, with canopy density being a largely influencing factor. Disease significantly decreased the chlorophyll content (Chl), net‐photosynthesis ( P n ), stomatal conductance ( G s ), intercellular CO 2 concentration ( C i ), and transpiration rate ( T r ) in seedlings but increased malondialdehyde (MDA) and soluble sugar content. The light‐response curve varied among pests, diseases, and healthy seedlings. Pest stress prompted the seedlings to maintain high levels of photosynthesis unlike the effects seen with diseases. Our findings highlight the impact of biotic factors on forest natural regeneration and contribute to enhancing healthy forest management and promoting sustainable development.

Forest canopy density can change the soil function and quality by altering the understory microenvironment. A comprehensive assessment of soil quality is essential for the development of sustainable forest management practices. In this study, we aimed to evaluate the effects of different canopy densities including forest gap (FG), forest edge (FE), medium-canopy density 0.4–0.6 (MCD), and high-canopy density 0.7–0.9 (HCD) on soil properties and soil quality index (SQI), and further to determine the main indicators that influence soil quality. The SQI was calculated using principal component analysis (PCA) and minimum data set (MDS) for indicator selection and weight assignment. Our results showed that pH decreased significantly with increasing canopy density (P < 0.05). In addition, soil organic carbon (SOC) and total phosphorus (TP) were significantly improved, and soil microbial phosphorus limitation gradually changed to nitrogen limitation with increasing canopy density. The SQI values ranged from 0.37 to 0.62, with the highest value in MCD. The MDS including leucine aminopeptidase (LAP; weight 0.27), soil available potassium (AK; weight 0.25), soil total nitrogen (TN; weight 0.23), and soil available phosphorus (SAP; weight 0.26) accounted for 80.79

AimsPlant functional traits (PFTs) and rhizosphere soil metabolites (RSMs) play crucial roles in the connection between plants and soil environments, influencing plant ecological adaptation processes. However, the strategies and driving factors underlying their variation in different habitats are poorly understood.MethodsIn this study, we investigated P.tabuliformis population traits and RSMs in two habitats, assessed trait and metabolite variations, and analyzed how traits and metabolites were associated with plant strategies.ResultsThe first principal component analysis (PCA) axis of the PFTs reflected trade-offs in traits linked to habitat: ridge habitat Chinese pine populations had high leaf dry matter content (LDMC) and fine root tissue density (FRTD) trait values that increased resource conservation, and slope habitat populations had high specific leaf area (SLA) and fine root specific root length (FRSRL) values that improved the efficiency of resource acquisition. The results of the metabolic pathway analysis of differential metabolites showed that in ridge habitats, adversity resistance-related metabolic pathways were significantly upregulated. Random forest model (RFM) analysis showed that the first PCA axis of PFTs and differential metabolites was significantly related to soil nitrogen and phosphorus content, indicating that soil nutrients are the primary factors driving the variation in P.tabuliformis population ecological adaptation strategies in different habitats.ConclusionsThus, the trade-offs between PFTs and the regulation of rhizosphere soil metabolism shape the population distribution of P.tabuliformis in different habitats. Along the terrain gradient, soil nutrients are the primary factors driving trait and metabolite regulation-based strategies.

Microbial functional genes serve as the core genetic foundation driving microbial ecological functions; however, its microbial functional gene composition across varied habitats and its ecological adaptation interplay with plants remain understudied. In this study, we investigated the P. tabuliformis rhizosphere microbial functional genes which are related to N and P cycles across ridge and slope habitats between different elevational gradients, analyzed their composition and abundance, and analyzed their responses to environmental factors. Results showed that slope habitats had a significantly greater abundance of N and P cycling functional genes compared to those of ridge counterparts (p < 0.05). Specifically, slope environments showed an enhanced gene abundance associated with denitrification, nitrogen fixation, nitrification, assimilatory/dissimilatory nitrate reduction, and nitrogen transport processes, along with the superior expression of genes related to inorganic/organic phosphorus metabolism, phosphorus transport, and regulatory gene expression. These nutrient cycling gene levels were positively correlated with soil nutrient availability. Our findings revealed distinct ecological strategies: Ridge communities employ resource-conservative tactics, minimizing microbial investments to endure nutrient scarcity, whereas slope populations adopt competitive strategies through enriched high-efficiency metabolic genes and symbiotic microbial recruitment to withstand resource competition.

Appropriate vegetation restoration measures are beneficial to ecosystem restoration and nutrient retention in ecologically fragile areas. However, the high water consumption of planted forests and the increasing frequency of drought events may reshape or complicate this ecological process. The effects of forest types and drought stress on nutrient limitation remain unclear. In this study, we selected five different vegetation restoration types on the Loess Plateau, China, and applied three drought levels to assess their effects on extracellular enzyme activity, soil microbial biomass, and soil nutrient limitations.We measured the activities of carbon-, nitrogen-, and phosphorus-acquiring enzymes and investigated the relationships among enzyme activity, microbial biomass, and nutrient limitations under drought conditions. Our results showed that vegetation types and drought significantly influenced soil enzymatic activity and stoichiometry. Mixed forests demonstrated higher enzyme activity and nutrient content compared to pure forests, indicating greater resilience under drought conditions. Short-term drought significantly reduced soil enzyme activity and microbial biomass, whereas mild drought stimulated enzyme activity, and moderate drought promoted microbial biomass. Drought markedly decreased microbial carbon and nitrogen content but increased the microbial carbon-to-nitrogen ratio. Furthermore, drought enhanced the correlation between microbial biomass carbon and carbon-acquiring enzymes, but there was no correlation between microbial biomass nitrogen and nitrogen-acquiring enzymes under drought. All vegetation types exhibited nitrogen limitation, and a negative correlation was observed between nitrogen and carbon limitations under drought conditions. Drought significantly exacerbated nitrogen limitation, while its impact on carbon limitation varied with drought severity and vegetation type. Overall, plant communities exhibited distinct nutrient acquisition strategies under drought stress, resulting in complex changes in soil enzyme activities and microbial biomass. This study advances our understanding of microbial nutrient limitations and enzymatic activities under varying vegetation restoration patterns and drought conditions, providing critical insights for enhancing soil resilience and nutrient cycling under climate change.

The process of vegetation restoration is often accompanied by significant changes in aboveground plant diversity. To explore the driving mechanism of litter nutrient-soil nutrient-enzyme activity stoichiometry on aboveground vegetation change is of great importance for maintaining regional biodiversity conservation and ecological stability. Taking typical abandoned farmland of different restoration years （1， 8， 16， 31， and 50 a） in the Qinling Mountains as the research object， the variation characteristics of plant community diversity during vegetation restoration were analyzed through field investigation. Litter nutrients， soil nutrients， and the activities of five extracellular enzymes， including β-1，4-glucosidase （BG）， cellobiohydrolase （CBH）， β-1，4-N-acetylglucosaminidase （NAG）， leucine aminopeptidase （LAP）， and acid phosphatase （AP）， were determined. The characteristics of litter nutrients， soil nutrients， and enzyme stoichiometric ratios during vegetation restoration and the driving mechanism of plant diversity changes were discussed. The results showed that the plant community diversity index firstly decreased and then increased with the increase in vegetation restoration years， and the minimum was reached at 16 years after restoration. The results of principal component analysis showed that there were significant differences between total plant community diversity index and litter-soil-enzyme stoichiometric characteristics in different years of vegetation restoration. The plant community diversity index had a strong positive correlation with litter C∶P ratio and litter N∶P ratio but had a negative correlation with soil enzyme C∶P ratio （EEA C∶P）. The results of redundancy analysis showed that soil EEA C∶P had the highest explanation rate of plant diversity changes during vegetation restoration （25.93%）， followed by soil TP （5.94%）， which was the key factor regulating plant diversity changes. In conclusion， plant species and quantity increased significantly in abandoned farmland in the middle part of the Qinling Mountains at the late stage of vegetation restoration. Changes in the soil environment affected microbial metabolic activities and thus changed enzyme activities. Litter-soil-soil extracellular enzymes affected the community environment and plant diversity through feedback and regulation. EEA C∶P and TP were the main driving factors of aboveground plant diversity change during vegetation restoration.

Forest ecosystems are important carbon (C) pools in terrestrial ecosystems, and the decomposition of soil organic matter depends on soil microbial metabolism. This study aimed to determine how forest thinning affects microbial metabolic limitation and microbial carbon use efficiency (CUE). We determined microbial metabolic limitations via extracellular enzymatic stoichiometry and microbial CUE by using a biogeochemical equilibrium model under different thinning intensities in the Qinling Mountains. We also analyzed the relationships among microbial metabolic processes and soil enzymes, soil properties (i.e., pH, soil nutrition content, and stoichiometry), and soil microbial biomass (i.e., microbial biomass and stoichiometry). Soil microbial metabolism was limited by C and phosphorus under the thinning intensities, and mild and moderate thinning intensities reduced the soil microbial C limitation. The effect of thinning intensity on CUE in the subsoil was greater than that in the topsoil. The pathways of CUE in the topsoil and subsoil differed, and microbial biomass and soil nutrients were the main factors affecting CUE in the topsoil and subsoil, respectively. Thinning changes the CUE by directly affecting microbial biomass in the topsoil, and thinning affects soil microbial biomass indirectly by affecting soil nutrients (i.e., total phosphorus), changing CUE in the subsoil. Our results show that moderate thinning can alleviate microbial metabolic limitation, and this finding has implications for understanding how microbial metabolism affects soil C dynamics under forest thinning.

Soil microorganisms are often limited by nutrients, representing an important control of heterotrophic metabolic processes. However, how nutrient limitations relate to microbial community structure and stability remains unclear, which creates a knowledge gap to understanding microbial biogeography and community changes during forest restoration. Here, we combined an eco-enzymatic stoichiometry model and high-throughput DNA sequencing to assess the potential roles of nutrient limitation on microbial community structure, assembly, and stability along a forest restoration sequence in the Qinling Mountains, China. Results showed that nutrient limitations tended to decrease during the oak forest restoration. Carbon and phosphorus limitations enhanced community dissimilarity and significantly increased bacterial alpha diversity, but not fungal diversity. Stochastic assembly processes primarily structured both bacterial (average contribution of 74.73 % and 74.17 % in bulk and rhizosheath soils, respectively) and fungal (average contribution of 77.23 % and 72.04 % in bulk and rhizosheath soils, respectively) communities during forest restoration, with nutrient limitation also contributing to the importance of stochastic processes in the bacterial communities. The migration rate (m) for bacteria was 0.19 and 0.23, respectively in both bulk soil and rhizosheath soil, and was greater than that for the fungi (m was 1.19 and 1.41, respectively), indicating a stronger dispersal limitation for fungal communities. Finally, nutrient limitations significantly affected bacterial and fungal co-occurrence with more interconnections occurring among weakly nutrient-limited microbial taxa and nutrient limitations reducing community stability when nutrient availability changed during forest restoration. Our findings highlight the fundamental effects of nutrient limitations on microbial communities and their self-regulation under changing environmental resources.

Forest ecosystem nutrient cycling functions are the basis for the survival and development of organisms, and play an important role in maintaining the forest structural and functional stability. However, the response of forest nutrient cycling functions at the ecosystem level to whole-tree harvesting remains unclear. Herein, we calculated the ecosystem nitrogen (N), phosphorus (P), and potassium (K) absorption, utilization, retention, cycle, surplus, accumulation, productivity, turnover and return parameters and constructed N, P, and K cycling function indexes to identify the changes in ecosystem N, P, and K cycling functions in a secondary forest in the Qinling Mountains after 5 years of five different thinning intensities (0% (CK), 15%, 30%, 45%, and 60%). We showed that the ecosystem’s N, P, and K cycling parameters varied significantly and responded differently to thinning treatments. As the thinning intensity increased, the N, P, and K cycling function indexes increased by 5%~232%, 32%~195%, and 104%~233% compared with CK. Whole-tree harvesting promoted ecosystem N and P cycling functions through two pathways: (a) directly regulated litter biomass, indirectly affected soil nutrient characteristics, and then regulated ecosystem N and P cycling functions; (b) directly regulated plant productivity, indirectly affected plant and soil nutrient characteristics, and then regulated ecosystem N and P cycling functions. In contrast, whole-tree harvesting mainly indirectly affected the plant and soil nutrient characteristics by directly adjusting the plant productivity, and promoting the ecosystem K cycling function. Furthermore, N and P cycling functions were mainly regulated by understory plant productivity while tree and herb nutrient characteristics were key driving factors for K cycling functions. These findings indicated that whole-tree harvesting significantly improved the ecosystem N, P and K cycling functions, and reveals varied regulatory mechanisms, which may aid in formulating effective measures for sustainable forest ecosystem nutrient management.

该研究以退耕还林工程的典型试验区延安城郊森林为研究对象,探讨两种主导造林树种刺槐和侧柏的不同配置模式对生态服务功能的影响.以课题组前期筛选的 8 个功能和结构较优的配置模式为典型林分,建立了24 个相对固定的监测样地,参照《森林生态系统服务功能评估规范》(GB/T 38582-2020),进行相关指标测算,通过比较分析分别得到了刺槐和侧柏这两种优势树种的不同生态服务功能的较优配置模式和服务功能簇的组成情况,以期为城市森林高质量发展提供参考.

Forest soil microbial community is easily modified by changes in plant composition. Therefore, differences in the assembly of soil microbial community within different forest types should be evaluated to gain a deeper understanding of microbial biodiversity and forest system functioning. In the present study, pine, oak, and pine-oak mixed forests at different developmental stages (young, immature, and mature forests) were evaluated to explore the assembly, co-occurrence network, and influencing factors of bacterial and fungal community. The assembly of bacterial community was dominated by deterministic processes (homogeneous selection), whereas the assembly of fungal community was dominated by stochastic processes (dispersion limited). At the young forest stage, the proportion of deterministic processes in bacterial community assembly among the three forest types differed less. In contrast, the proportion of deterministic processes at the immature and mature forest stages in the bacterial community assembly in the pine-oak mixed forest was larger than that in the pure forests. The proportion of stochastic processes in the assembly of fungal community in the pine-oak mixed forest was greater than that in the pure forests and was observed at all forest stages. Network analysis showed that the links and average degree of bacteria in the pine-oak mixed forest were lower than that of the pure forests, whereas the network topological features of fungi in the pine-oak mixed forest were higher than that of the pure forests. The stochastic processes increased microbial co-occurrence. The results of the random forest indicated that Pielou's evenness index (E) of plant was the most important predictor of fungal community assembly, and Shannon's diversity index (H) of plant also was the critical factor influencing bacterial community assembly. Our study demonstrated the importance of plant diversity in the assembly of forest soil microbial community.

The Chinese pine (Pinus tabuliformis) community on the ridge is one of the most important zonal forest communities on the southern slope of the mid-Qinling Mountains. This study aimed to investigate the driving factors of Chinese pine population distribution in the ridge habitats and its adaptability characteristics. Population age structure and the relationship between regeneration dynamics and environmental factors were investigated in 32 plots in the Huoditang Forest region. The results showed that the niche of Chinese pine was wide but overlapped greatly with that of Quercus aliena var. acutiserrat, an oak species. The population in the ridge habitats exhibited an expansion trend, while that inhabiting slope habitats was declining. Seedling density in ridge habitats was much higher than that of the understory in the slope habitats. Still, the seedling growth rate in both ridge and understory habitats was much lower than that characteristic of gap habitats. Seedling density positively correlated with understory solar conditions, while growth positively correlated with soil fertility, indicating that environmental factors significantly influence the regeneration process. Thus, light conditions and intrinsic biological traits of Pinus tabuliformis influence its distribution. In ridge habitats, sufficient light conditions promote Pinus tabuliformis regeneration and recruitment of larger classes, but poor soil conditions also limit its growth.

Background and aims Soil seed bank plays a significant role in the natural recovery and the succession of forest. A large number of natural secondary forests are distributed in the Qining Mountains, China. However, it remains unclear how the soil seed banks develop and influence the recovery of secondary forests. Methods We explored the aboveground vegetation and soil seed bank at 27 plots along forest developmental stages of pine and oak forests in the Qinling Mountains. Species composition and abundance of aboveground vegetation as well as seed bank composition and density were surveyed and their correlations with forest developmental stages were determined. Results The results showed that the species richness of seed banks was the highest in middle-aged forests, as well as the aboveground vegetation. The seed density and abundance increased with forest developmental stages and varied among different forest types. The similarity between seed banks at different developmental stages of the same forest type was high, as well as aboveground vegetation. The seed bank and aboveground vegetation showed low similarity across the three developmental stages of both forest types, and the persistent soil seed bank did not reflect the changes in aboveground vegetation. Conclusion Our findings demonstrate that developmental stages play an important role in the composition of seed banks and aboveground vegetation. Our results also highlight the persistent soil seed bank contributes less to the forest recovery. We can assume that the natural recovery of disturbed natural secondary forests is largely dependent on transient soil seed bank and seed dispersal.

Plant litter decomposition is a crucial pathway for carbon (C) and nutrient cycling, and controls the net primary productivity in ecosystems worldwide. However, little is known about how multi-type litter (leaf and different diameter fine roots) decomposition rates and nutrient release change at the community level following whole-tree harvesting (WTH). In the present study, we followed decomposition of leaf and different diameter fine root (∅ < 0.5 mm, 0.5-1 mm, 1-2 mm) litters at plot level over 2 years in a secondary forest in the Qinling Mountains after 5 years of five different thinning treatments (0%, 15%, 30%, 45%, and 60%). Our results demonstrated that WTH had no effects on leaf and different diameter fine root litter decomposition at the plot level. Leaves had significantly higher decomposition rate than different diameter fine roots. There were significant positive correlations between decomposition rate of different diameter fine roots, but not related to leaf litter decomposition rate. WTH did not affect the nutrient release of leaf and different diameter fine root litters at the plot level. The nitrogen (N), phosphorous (P) and potassium (K) mass remaining in leaf litters were significantly higher than different diameter fine roots after 2 years decomposition, while different diameter fine roots had higher C mass remaining. Leaf and fine root litter decomposition rates were mainly influenced by stand and litter quality attributes. Nutrient release of leaf and fine root N, P and K were mainly predicted by litter quality characteristics, while there were no consistent driving factors for C release. Our results suggested that WTH had no effects on multi-type litter decomposition and nutrient release at plot level after 5 years of recovery. Moreover, leaf litters had excellent N, P and K nutrient preservation mechanisms, and C conservation in fine root litters.

Clarifying the characteristics of soil microbial nutrient limitation and its driving mechanisms during vegetation restoration after farmland abandonment has important implications for revealing soil nutrient cycling and maintaining ecosystem stability. To determine the limitation of soil microbial nutrients and its relationship with soil properties along a chronosequence of abandoned farmland in the middle of the Qinling Mountains, the soil physicochemical properties and five enzyme activities (β-1,4-glucosidase (BG), cellobiohydrolase (CBH), β-1,4-N-acetylglucosaminidase (NAG), leucine aminopeptidase (LAP), and acid phosphatase (AP)) were measured, and models of extracellular enzymatic activity were applied. The results showed that the activities of BG, CBH, NAG, LAP, and AP were significantly increased following farmland abandonment. With the increasing years of abandonment, the ratios of (BG+CBH)/(NAG+LAP) and (BG+CBH)/AP significantly decreased, whereas the ratio of (NAG+LAP)/AP increased. Correlation analysis showed that most soil physicochemical properties were significantly correlated with extracellular enzyme activities and extracellular enzymatic stoichiometry. The vector length of extracellular enzymatic stoichiometry decreased with the increase in abandonment years, indicating that the limitation of soil microorganisms on carbon (C) was reduced. Moreover, the vector angles (>45°) showed a decreasing trend, indicating that microbial metabolisms were limited by phosphorus (P) and gradually decreased. Regression analysis showed that the C and P limitations were significantly related to total nutrients, available nutrients, nutrient ratio, and soil physical properties. Partial least squares path modeling (PLS-PM) revealed that the C and P limitations were directly regulated by nutrient ratio. PLS-PM further showed that soil total nutrients indirectly affected soil microbial C and P limitations by affecting nutrient ratio, and nutrient ratio affected the soil metabolism limitation via available nutrients and pH. Our study suggests that the characteristics of microbial metabolism during the vegetation restoration process reflect the mechanism of microorganism-mediated soil nutrient cycling, which provides a theoretical basis for revealing the community dynamics and stability during the vegetation restoration process and maintaining the regional ecological environment security in the Qinling Mountains.