How Grazing, Enclosure, and Mowing Intensities Shape Vegetation–Soil–Microbe Dynamics of Qinghai–Tibet Plateau Grasslands: Insights for Spatially Differentiated Integrated Management

Abstract

Grasslands provide essential forage, fuel, and ecosystem services, underpinning regional livestock husbandry and ecological integrity. However, improper utilization drives structural degradation and functional decline of the vegetation–soil–microbe system, particularly on the ecologically sensitive and fragile Qinghai–Tibet Plateau (QTP). The differential impacts of diverse utilization practices on QTP grasslands remain inadequately understood, limiting scientific support for differentiated sustainable management. To address this, we conducted a comprehensive meta-analysis to clarify effects of grazing, enclosure, and mowing on QTP grasslands, integrating studies from Web of Science, Google Scholar, and CNKI. We constructed disturbance intensity indicators to quantify utilization pressure and used multiple ecological metrics to characterize heterogeneous responses of the vegetation–soil–microbe system. Moderate grazing enhanced vegetation coverage, biomass, diversity, soil total phosphorus, and organic matter; high-intensity grazing reduced vegetation traits, soil bulk density, moisture, nutrients, and microbial biomass/diversity, while increasing soil pH. Early enclosure mitigated anthropogenic disturbance to improve grassland functions, but long-term enclosure exacerbated nutrient/moisture competition, lowering vegetation biomass/diversity and degrading soil properties. Moderate mowing improved vegetation communities by suppressing dominant species overexpansion; excessive mowing caused vegetation homogenization, soil carbon loss, and microbial destabilization. Impacts showed environmental heterogeneity linked to climate, soil, vegetation type, and elevation. In humid and fertile alpine meadows, moderate grazing more effectively promoted vegetation diversity and soil nutrient cycling, while in arid and nutrient-poor desert grasslands, even light grazing led to visible declines in vegetation coverage and soil moisture. Low-elevation alpine grasslands exhibited stronger positive responses to moderate grazing, whereas high-elevation alpine desert grasslands showed high vulnerability even to light grazing. Based on these mechanisms, regionally tailored strategies integrating multiple practices are required to balance ecological conservation and livestock production, promoting QTP grassland sustainability. In future research, we will strengthen quantitative exploration of how specific environmental factors regulate the magnitude and direction of grassland ecosystem responses to grazing, enclosure, and mowing, thereby providing more precise scientific basis for differentiated grassland management.

1. Introduction

Against the backdrop of global environmental change (e.g., global warming and intensified anthropogenic disturbances), grasslands covering 40% of Earth’s ice-free land surface function as a pivotal component of the planetary life and support system [1]. They deliver critical ecosystem services (carbon sequestration, soil and water conservation, climate regulation, biodiversity maintenance) that not only underpin regional ecological quality and sustainable development but also play irreplaceable roles in mitigating global climate change and enhancing ecosystem resilience to environmental perturbations and stressors. Additionally, grasslands provide essential forage and fuel for human livelihoods, supporting livestock husbandry and directly or indirectly impacting billions of people worldwide. Under the dual stressors of climate change and anthropogenic activities, nearly half of global grassland ecosystems are experiencing structural degradation and functional decline [2,3], characterized by reduced vegetation coverage, biomass, and species diversity, deteriorated soil physicochemical properties, and diminished soil microbial abundance and functional capacity [4], further weakening their buffering capacity against global environmental change. Notably, the Qinghai–Tibet Plateau (QTP), characterized by fragile, climate-sensitive ecosystems, confronts more severe grassland degradation and greater challenges in ecological restoration [5,6,7], attracting intense focus from researchers and land managers. Among anthropogenic drivers, grazing exerts a dominant influence on grassland dynamics [8,9]. Overgrazing exacerbates degradation, while enclosure and mowing, as pivotal sustainable management strategies [10,11], reduce grazing pressure to facilitate vegetation recovery and soil quality improvement. Grassland ecosystem health is largely reflected by its structural stability and functional integrity—two core indicators for evaluating ecological sustainability. Here, “structural stability” refers to the ability of grassland ecosystems to maintain their original organizational structure or recover to a stable state under external disturbances, with its core feature being the resistance and resilience of the relationships between system components against disturbances; “functional integrity” refers to the ability of grassland ecosystems to maintain key ecological processes and services, including vegetation productivity supply, soil nutrient cycling, microbial metabolic activity, and ecological barrier functions, ensuring that ecosystem services do not experience significant disruption or degradation due to external disturbances [12,13]. Under the dual pressures of anthropocentric activities and climate change, nearly half of the global grasslands were experiencing heterogeneous degradation processes with distinct magnitudes and manifestations, rather than uniform “varying degrees” of structural degradation and functional decline. Thus, clarifying how grazing, enclosure, and mowing affect the vegetation–soil–microbe continuum constitutes a foundational scientific basis for sustainable grassland management under global environmental change, and has garnered increasing attention from the academic community and policy-makers.

In recent years, previous studies employing fixed-point observations, controlled field experiments, and simulation models have examined the ecological impacts of grassland grazing, enclosure, and mowing on vegetation–soil–microbe dynamics [14,15,16]. Findings consistently demonstrated that the effects of these utilizations varied substantially with intensity and environmental conditions. Moderate grazing has been shown to help maintain species diversity by preventing competitive exclusion, while the return of animal excreta can enhance soil organic matter and nitrogen availability, thus promoting microbial activity and supporting decomposer communities [17,18]. However, excessive grazing generally caused significant degradation, including reduced vegetation coverage, a shift toward grazing-tolerant monocultures, soil nutrient depletion, increased soil compaction through trampling, reduced infiltration capacity, and a decline in microbial diversity and enzymatic function [19,20]. Enclosure, by minimizing grazing disturbances, typically facilitated grassland recovery by increasing vegetation coverage and biomass, enhancing species richness, raising levels of soil organic carbon and total nitrogen, improving soil aggregate stability, and increasing microbial biomass and functional diversity, especially in humid environments [21]. Nevertheless, under semi-arid or arid conditions, prolonged grazing exclusion may exacerbate water competition, reduce community evenness, slow organic matter decomposition, and diminish microbial activity. In addition, moderate mowing can limit the dominance of competitive species and promote nutrient cycling by facilitating the accumulation and decomposition of organic matter. Yet, intensive mowing may lead to species homogenization and lower soil carbon and nutrient inputs due to reduced litterfall and root exudation, ultimately suppressing microbial metabolic activity and destabilizing microbial communities, especially in nutrient-poor or drought-prone ecosystems [22].

Understanding the impacts of grassland utilization practices on vegetation–soil–microbe communities presents a complex research challenge, particularly given the intricate interactions within this system. Vegetation modulates soil physicochemical properties and microbial communities via litter inputs and root exudates, microbes sustain vegetation growth by mediating nutrient cycling (e.g., nitrogen and phosphorus transformation), and soil provides a foundational substrate and nutrients for both components. Ecological outcomes vary with the type, intensity, and duration of grassland use, and are further complicated by environmental heterogeneity in climate, soil characteristics, and topography. Although many studies have explored the effects of grazing, enclosure, and mowing [23,24], most rely on site-specific experiments or localized case analyses focusing on single management regimes, leading to inconsistent findings and hindering systematic elucidation of multi-dimensional, environment-dependent response mechanisms, limitations especially pronounced on the Qinghai–Tibet Plateau (QTP), where diverse climates, vegetation types, and terrain amplify variability [25,26,27]. For the QTP, we hypothesize that its low-temperature, high-altitude conditions render the vegetation–soil–microbe system more sensitive to utilization practices: enclosure will more strongly promote positive feedback loops (e.g., increased vegetation biomass enhancing microbial nutrient cycling, which in turn fuels vegetation recovery) than mowing, while overgrazing will exacerbate mutual disruption (e.g., reduced vegetation coverage impairing soil carbon sequestration and microbial diversity). A comprehensive synthesis of these pathways is vital for strengthening the QTP’s ecological barrier function, and meta-analysis, with its robust quantitative framework to integrate independent studies, improve conclusion reliability/generality, and mitigate small-sample or methodological biases [28], which is urgently needed to synthesize evidence and evaluate how utilization strategies affect this system, thereby providing evidence-based guidance for sustainable QTP grassland management.

The QTP exhibits extreme spatial heterogeneity in key environmental factors: its annual average temperature drops from above 15 °C in the southeast to below 0 °C in the northwest, annual precipitation ranges from over 1600 mm to less than 50 mm across latitudinal and elevational gradients, and vegetation types transition from subtropical rainforests to alpine meadows, steppes, and desert grasslands. This heterogeneity directly leads to divergent grassland responses to utilization practices. For instance, moderate grazing may promote vegetation diversity and soil nutrient cycling in humid, fertile alpine meadows but cause severe degradation in arid, nutrient poor desert grasslands; short-term enclosure can facilitate recovery in semi-humid areas but exacerbate resource competition and diversity loss in arid regions. A one-size-fits-all management approach fails to account for these context-dependent outcomes, risking ineffective restoration in some areas and unintended ecological damage in others. Thus, spatially differentiated integrated management is not merely a theoretical concept but a practical imperative for the QTP, it enables managers to leverage the synergies of multiple utilization practices while aligning strategies with local climate, soil, and vegetation characteristics, ultimately balancing ecological conservation, livestock production, and the maintenance of pastoral livelihoods across this ecologically critical and heterogeneous plateau.

The QTP is known as the Third Pole of the Earth, the Roof of the World, and the Water Tower of Asia, and is a global climate change-sensitive region and an ecological security barrier [29]. Its grassland ecosystem plays a vital role in carbon sequestration, climate regulation, and water conservation. It also supports the livelihoods of local herdsmen and the continuation of Tibetan culture, and is vital to regional sustainable development and global ecological security. However, influenced primarily by climate change and intensive grazing, grassland degradation has become increasingly evident in localized areas of the plateau [30]. Although a growing body of studies have explored the effects of different grassland use practices on vegetation–soil–microbe communities in this ecologically fragile region, a quantitative and systematic synthesis remains lacking. To address this gap, this study integrated relevant research papers from the Web of Science, Google Scholar and China National Knowledge Infrastructure (CNKI) databases, and used meta-analysis to systematically summarize the comprehensive effects of different grassland utilization strategies on vegetation communities, soil physical and chemical properties and soil microorganisms in the QTP. It aims to provide important support for scientific research and practical management related to sustainable grassland use, grassland protection, and restoration of degraded grassland in the QTP and other similarly vulnerable highland ecosystems.

2. Materials and Methods

2.1. Study Area

The QTP is the highest geographical unit on Earth, located in central Asia and southwestern China (73.49° E–105.00° E, 25.47° N–40.66° N) (Figure 1). Characterized by a fragile ecological environment and high sensitivity to climate change and anthropogenic disturbances, it functions as a globally critical ecological security barrier. The region has complex topography and significant altitudinal gradients, with elevation generally increasing from the southeast to the northwest. Dominated by a plateau and alpine climate, the QTP exhibits pronounced spatial variation in temperature and precipitation along elevational and latitudinal gradients. From the southeast to the northwest, the annual average temperature drops from above 15 °C to below 0 °C, and the annual precipitation drops from above 1600 mm to below 100 mm, with some areas receiving under 50 mm [31]. This climatic heterogeneity supports a wide range of ecosystems, including subtropical rainforests, grasslands, and deserts, making the Plateau one of the most biodiverse regions globally. Grassland represents the dominant ecosystem, covering approximately 60% of the Plateau’s total area. However, over the past several decades, the combined effects of overgrazing and climate change have led to widespread grassland degradation, posing a major threat to regional ecological security and sustainable development [32]. In response, the Chinese government has implemented a suite of grassland conservation policies, including grazing bans, livestock exclusion, pasture restoration programs, ecological subsidies, and the establishment of nature reserves. These efforts have yielded measurable ecological improvements, although grassland degradation persists in areas subject to intensive grazing pressure [33]. Therefore, quantitatively synthesizing the effects of different grassland utilization practices (e.g., grazing, enclosure, mowing) on vegetation–soil–microbe communities is essential for advancing effective conservation strategies and promoting sustainable grassland management across the QTP.

2.2. Methods

This study conducted a systematic literature search using multiple academic databases, including Web of Science, Google Scholar, and CNKI, to identify relevant peer-reviewed publications. A series of search strategies were developed to retrieve studies examining the effects of different grassland utilization practices. After applying predefined quality criteria, a curated dataset of eligible studies was compiled. Quantitative data were then extracted from the selected publications based on reported text, tables, figures, or appendices. Finally, a meta-analysis was performed to statistically synthesize the extracted data and quantitatively assess the overall impacts of various grassland management strategies (e.g., grazing, enclosure, mowing) on vegetation–soil–microbe communities in the QTP [34]. This study establishes a “disturbance-context-resilience-response” integrated framework tailored to Qinghai–Tibet Plateau (QTP) grasslands. The framework first defines disturbance intensity benchmarks for core management practices, grazing is stratified by stocking rates matching the plateau’s low productivity and short growing seasons, enclosure by duration reflecting alpine ecosystem recovery rhythms, and mowing by annual cutting frequency balancing utilization and regrowth. It then embeds contextual modifiers: elevation shapes baseline conditions, livestock type is standardized via forage consumption equivalents, and study duration accounts for temporal dynamics. Ecosystem resilience acts as the key mediator, disturbance within resilience thresholds drives positive responses, while exceeding thresholds causes degradation. This framework clarifies that QTP grassland responses depend on the interplay of disturbance intensity, context, and resilience, providing a theoretical basis for interpreting results and guiding sustainable management.

2.2.1. Data Sources and Search Strategy

The data for this study were primarily obtained from peer-reviewed articles in the academic databases such as Web of Science, Google Scholar and China National Knowledge Infrastructure (CNKI), with the literature search conducted up to 18 April 2024. In terms of the impact on grassland vegetation communities, the search terms used were: TS = (Tibetan Plateau OR Three-River Source OR River and Lake Sources) AND TS = (mowing OR cutting grass OR reclamation OR grazing OR grazing ban OR enclosure OR fencing) AND TS = (plant communities OR vegetation OR plants). A total of 1184 papers in English and 190 in Chinese were collected. After screening, 160 valid English and 53 valid Chinese papers were obtained, involving 66 sample sites in the QTP (Figure 1).

In terms of the impact on soil physicochemical properties, the search strategy included the following keywords: TS = (Tibetan Plateau OR Three-River Source OR River and Lake Sources) AND TS = (mowing OR cutting grass OR reclamation OR grazing OR grazing ban OR enclosure OR fencing) AND TS = (soil bulk density OR soil moisture OR soil temperature OR soil nitrogen OR soil phosphorus OR soil potassium OR soil pH OR soil organic carbon). This search retrieved 176 articles from Web of Science, 128 from Google Scholar, and 105 from CNKI. After screening, a total of 18 publications were selected, involving 38 sample sites in the QTP (Figure 1).

In terms of the impact on soil microbial communities, the search terms were: TS = (Tibetan Plateau OR Three-River Source OR River and Lake Sources) AND TS = (mowing OR cutting grass OR reclamation OR grazing OR grazing ban OR enclosure OR fencing) AND TS = (soil microorganisms OR enzyme activity). After screening, 20 English and 26 Chinese publications were selected for meta-analysis, involving 60 sample sites in the QTP (Figure 1).

2.2.2. Eligibility Criteria for Literature Selection

To minimize selection bias, the retrieved literature was screened according to the following criteria: (1) the study must be based on field experiments conducted under different grazing conditions within grassland ecosystem of the QTP; (2) the article must include both control and treatment groups with defined grazing intensities, and other confounding experimental factors must be excluded; (3) the duration of grazing or enclosure treatments must be at least one year; and (4) sufficient statistical information must be available to extract mean values (Mean, M), number of samples (N), and standard error (SE) or standard deviation (SD) from tables, text, digitized figures (Figure 2).

2.2.3. Meta-Analysis and Statistical Indicators

Meta-analysis is a statistical approach used to systematically synthesize the results of multiple independent studies, and it has been widely applied across diverse fields such as geography, ecology, economics, medicine, and the social sciences. By quantitatively aggregating key indicators from relevant studies addressing a specific research question, meta-analysis allows for more representative and comprehensive conclusions. In this study, quantitative data for key statistical variables were extracted from figures and tables in the selected publications using the software Plot Digitizer v3. Subsequently, meta-analyses were conducted using Prism software v9.0 to assess the differential effects of various grassland utilization practices with varying intensities on vegetation–soil–microbe communities.

We used three indicators to represent management intensity across different grassland use practices: grazing intensity, enclosure duration, and mowing frequency. Grazing intensity was categorized based on livestock stocking rates into four levels: ungrazed (enclosed, 0 cattle/ha), light grazing (0–3 cattle/ha), moderate grazing (3–5 cattle/ha), and heavy grazing (>5 cattle/ha). The categorization of grazing intensity into four levels (ungrazed: 0 cattle/ha; light grazing: 0–3 cattle/ha; moderate grazing: 3–5 cattle/ha; heavy grazing: >5 cattle/ha) is ecologically justified based on three core considerations: (1) alignment with the QTP’s alpine grassland productivity characteristics, (2) consistency with plant–soil–microbe response thresholds identified in regional studies, and (3) compatibility with existing validated grazing management frameworks for high-altitude grasslands. Enclosure duration was divided into intervals ranging from 1 to 7 years, with a one-year increment. Mowing frequency was classified into four levels: no mowing, light mowing (once per year), moderate mowing (twice per year), and heavy mowing (three times per year).

In addition, we used multiple ecological indicators to evaluate the responses of vegetation–soil–microbe to different grassland uses. For vegetation, we considered grassland coverage, height, density, biomass, and species diversity. Soil physicochemical properties were characterized using indicators such as bulk density, moisture content, total nitrogen, total phosphorus, total potassium, pH, soil organic matter, and soil organic carbon [33]. Soil microbial characteristics were assessed using microbial biomass carbon and nitrogen, bacterial, fungal, and actinomycete abundance, and microbial community diversity [34,35,36].

3. Results

3.1. Effects of Different Grassland Use on Vegetation Indicators

3.1.1. Vegetation Structure

Compared to enclosure, the vegetation height and coverage of grassland were higher under light grazing (Figure 3a,b), and vegetation density peaked under moderate grazing intensity in the QTP (Figure 3c). However, as grazing intensity further increased, heavy grazing impaired plant regenerative capacity, leading to reductions in vegetation height, coverage, and density (Figure 3d–f). Enclosure generally had varying degrees of positive effects on vegetation height, coverage, and density, by reducing anthropogenic disturbances such as grazing. Short-term enclosure positively influenced vegetation height, but with prolonged enclosure duration, vegetation height tended to decline (Figure 3d). Vegetation coverage initially decreased following enclosure but gradually increased over time (Figure 3e). Vegetation community density consistently increased with enclosure duration (Figure 3f). Grass-mowing directly limited vegetation growth, causing vegetation height to decrease as mowing intensity increases (Figure 3g). Vegetation coverage increased at low mowing intensities but declined under excessive-intensity mowing (Figure 3h). Furthermore, mowing reduced competition among plants, resulting in an increase in vegetation density with increasing mowing intensity (Figure 3i).

3.1.2. Vegetation Biomass

Light grazing promoted the accumulation of both aboveground and belowground grassland biomass in the QTP (Figure 4a,b). With increasing grazing intensity, moderate grazing generally led to reductions in above- and below-ground biomass. Under heavy grazing, although aboveground biomass continued to decline (Figure 4a), belowground biomass showed an increase (Figure 4b). With prolonged enclosure duration, aboveground biomass exhibited a fluctuating decline, reaching a peak at three years of enclosure before gradually decreasing thereafter (Figure 4c). Belowground biomass increased progressively during the first two years of enclosure but declined in the third year. As enclosure time extended further, belowground biomass gradually recovered (Figure 4d). Light mowing enhanced aboveground biomass; however, aboveground biomass decreased as mowing intensity increased (Figure 4e). Additionally, belowground biomass declined significantly with increasing mowing intensity (Figure 4f).

3.1.3. Vegetation Diversity

Grazing promoted the plant species richness of grassland vegetation by limiting the growth of dominant species on the QTP, with richness peaking under heavy grazing intensity (Figure 5a). Light grazing significantly elevated the Shannon–Wiener diversity index and Simpson’s dominance index (Figure 5b,c), indicating that a reduction in dominant species fostered higher community diversity. However, these two diversity indices declined slightly as grazing intensity further increases. Light and moderate grazing enhanced vegetation evenness (Figure 5d), whereas heavy grazing led to a decrease in evenness. Short-term enclosure promoted the increases in species richness and the Shannon–Wiener index, reaching a peak in the second year of enclosure before gradually declining with longer enclosure duration (Figure 5e,f). The Simpson dominance index peaked in the third year of enclosure, followed by a slight decline (Figure 5g). Evenness decreased initially during enclosure, reaching its maximum in the fifth year, and then declined again (Figure 5h). Moderate grass-mowing resulted in a decline in species richness (Figure 5i). The Shannon–Wiener index increased with mowing intensity (Figure 5j). The Simpson dominance index peaked at moderate mowing intensity (Figure 5k), reflecting an increase in competitively dominant species. Heavy mowing suppressed the growth of dominant species, causing a reduction in the Simpson dominance index. Due to limitations in the available data, mowing showed no significant effect on vegetation evenness (Figure 5l).

3.2. Effects of Different Grassland Use on Soil Indicators

3.2.1. Soil Physical Properties

The livestock trampling of soil has intensified with increasing grazing intensity, leading to an increase in soil bulk density (Figure 6a) and a decrease in soil moisture content (Figure 6b) on the QTP. This indicated that overgrazing reduced grassland vegetation cover, exacerbated surface soil erosion, and subsequently diminished soil infiltration capacity and water retention. Enclosure effectively limited physical disruption of the soil by reducing grazing disturbance, increasing soil porosity while reducing soil bulk density (Figure 6c). The enlargement of soil pore spaces enhanced aeration and permeability, allowing precipitation to infiltrate more readily into deeper soil layers, which increased deep soil moisture content (Figure 6d). With prolonged enclosure duration, soil bulk density further declined and soil moisture content improved, promoting the maintenance and stability of soil permeability and structure.

3.2.2. Soil Chemical Properties

With increasing grazing intensity, the total nitrogen content in the soil of grasslands on the QTP showed a declining trend (Figure 7a), suggesting that overgrazing may lead to nitrogen and nutrient imbalances in the soil. Light grazing enhanced soil phosphorus transformation efficiency, resulting in increased total soil phosphorus content (Figure 7b). However, further increases in grazing intensity caused a continuous decline in total soil phosphorus. Light grazing caused a gradual decrease in total soil potassium content, but with increasing grazing intensity, potassium content slightly raised (Figure 7c). Soil pH increased with grazing intensity, primarily influenced by nitrogen loss and increased inputs from livestock excreta (Figure 7d). Light grazing slightly elevated soil organic matter (Figure 7e) and organic carbon (Figure 7f). Yet, as grazing intensity increased, reductions in grassland vegetation and litter led to declines in soil organic matter and organic carbon. With prolonged enclosure duration, total soil nitrogen (Figure 7g) and total phosphorus (Figure 7h) increased, attributable to reduced livestock trampling and grazing, which facilitated vegetation recovery and growth, thereby promoting soil nitrogen and phosphorus accumulation. Total soil potassium content initially increased and then decreased with enclosure duration, but overall variation was minor, indicating a relatively limited effect of enclosure on soil potassium (Figure 7i). Short-term enclosure slightly increased soil pH, but long-term enclosure resulted in a decrease in soil pH (Figure 7j).

3.3. Effects of Different Grassland Use on Microbial Indicators

3.3.1. Soil Microbial Biomass Carbon and Nitrogen

Light grazing resulted in lower soil microbial biomass carbon and nitrogen compared to enclosure treatments, indicating that light grazing reduced microbial biomass carbon (Figure 8a) and nitrogen (Figure 8b). Both moderate and heavy grazing significantly increased soil microbial biomass carbon and nitrogen, with peak values observed under moderate grazing. Enclosure reduces anthropogenic disturbances and grazing pressure, promoting soil environmental stability and recovery, and generally exerts positive effects on soil microbial biomass carbon (Figure 8c) and nitrogen (Figure 8d). With increasing enclosure duration, soil microbial biomass carbon and nitrogen exhibit fluctuating but overall increasing trends, with more pronounced positive effects observed under long-term enclosure.

3.3.2. Biomass of Soil Bacteria, Fungi and Actinomycetes

With increasing grazing intensity, soil bacterial biomass shows a declining trend (Figure 9a), with a severe reduction observed under moderate grazing. Grazing has a negligible effect on soil fungal biomass, which exhibits slight fluctuations characterized by a minor increase followed by a decrease (Figure 9b). Actinomycete biomass tends to increase with grazing intensity (Figure 9c), though it remains generally lower compared to enclosure treatments. Enclosure provides a relatively stable soil environment conducive to the growth and proliferation of soil microorganisms. With longer enclosure durations, the biomass of soil bacteria (Figure 9d), fungi (Figure 9e), and actinomycetes (Figure 9f) all display varying degrees of increase. Among these, enclosure exerts the most pronounced positive effect on soil bacterial biomass, which becomes more evident with extended enclosure time. The increases in fungal and actinomycete biomass are less significant.

3.3.3. Diversity of Soil Microbial Communities

Due to limitations in the available data, we analyzed only the effects of varying grazing intensities on the diversity of bacterial and fungal phyla in soil microbial communities. Regarding bacterial phyla, the relative abundance of Proteobacteria under moderate and heavy grazing was significantly lower than that under light grazing and enclosure treatments (Figure 10a). The relative abundance of Actinobacteria increased with increasing grazing intensity (Figure 10b). Except for the light grazing treatment, the relative abundance of Acidobacteria was lower in other grazing treatments compared to enclosure (Figure 10c). The relative abundance of Bacteroidetes increased under light grazing but decreased under moderate and heavy grazing (Figure 10d). The relative abundance of Chloroflexi showed an increasing trend with grazing intensity (Figure 10e). Increased grazing intensity also promoted the relative abundance of Verrucomicrobia (Figure 10f). Both the bacterial Shannon index (Figure 10g) and Chao index (Figure 10h) increased with grazing intensity.

For fungal phyla, the relative abundance of Firmicutes was significantly higher in grazing treatments compared to enclosure, with fluctuations increasing alongside grazing intensity (Figure 11a). The relative abundance of Gemmatimonadetes also increased under grazing but declined under moderate and heavy grazing (Figure 11b). The relative abundance of Zygomycota decreased significantly following light grazing but rebounded under moderate grazing (Figure 11c). No significant differences in Zygomycota relative abundance were observed between heavy grazing and enclosure treatments. The relative abundance of Ascomycota increased with grazing intensity (Figure 11d), whereas Basidiomycota showed a decreasing trend following grazing treatments (Figure 11e). The relative abundance of other fungal phyla increased after grazing treatments (Figure 11f). The fungal Shannon index decreased significantly under light and moderate grazing (Figure 11g), while under heavy grazing it did not differ significantly from enclosure. The fungal Chao index increased significantly following grazing treatments compared to enclosure (Figure 11h).

4. Discussion

4.1. Spatially and Environmentally Heterogeneous Response Mechanisms

The vast geographical extent and pronounced altitudinal gradients of the Qinghai–Tibet Plateau led to substantial spatial heterogeneity in how grassland utilization practices such as grazing, enclosure, and mowing affect vegetation and soil systems. This heterogeneity primarily arises from spatial variability in climatic conditions (e.g., precipitation gradients, thermal distribution), elevation, soil types, and vegetation composition. Moderate grazing has been shown to enhance species diversity and increase aboveground biomass, while also improving soil organic matter content and promoting microbial diversity through the return of animal excreta, especially in meadow grasslands characterized by higher precipitation and fertile soils [37]. Elevation defines the ecological baseline resilience of grasslands, shaping their capacity to tolerate and respond to management activities, and this stratification effect is strongly supported by statistical evidence from our meta-analysis. For low-elevation alpine grass-lands (3000–4000 m), moderate grazing (3–5 cattle/ha) significantly increased vegetation coverage and soil organic matter compared to enclosure, with the positive response attributed to their relatively mild climate that provides a stable foundation for plant re-growth and nutrient cycling. In contrast, mid-elevation alpine meadows (4000–4800 m) showed weaker responses to moderate grazing: vegetation coverage increased and soil organic matter, which is linked to cooler temperatures that slow microbial decomposition and nutrient turnover. For high-elevation alpine desert grasslands (above 4800 m), even light grazing (0–3 cattle/ha) caused significant degradation: vegetation coverage decreased and soil moisture declined relative to enclosure, due to their harsh climate: extreme cold, annual average temperature < −1 °C, annual precipitation < 200 mm and fragile soil structure that severely limit ecological recovery potential. Second, elevation mediates the effectiveness of utilization practices, leading to divergent outcomes for the same management intensity. Moderate grazing, which benefits low-elevation grasslands, only produces weak positive effects on mid-elevation alpine meadows (4000–4800 m), with limited gains in vegetation coverage and no meaningful change in soil organic matter, due to cooler temperatures that slow microbial decomposition and nutrient turnover. Similarly, short-term enclosure (a practice often used for grassland recovery) enhances soil nutrient accumulation in mid-elevation meadows but fails to improve high-elevation desert grasslands, as water scarcity (not grazing pressure) remains the primary constraint on their ecological function. In contrast, in arid regions or areas with nutrient-poor soils, overgrazing more readily results in reduced vegetation coverage, increased soil compaction due to trampling, greater bare ground exposure, and diminished microbial activity, ultimately leading to declines in both plant and microbial diversity and functional integrity. Enclosure generally facilitates vegetation recovery and the accumulation of soil organic matter, especially in moist environments. However, positive short-term effects can shift to negative outcomes over the long term, such as reduced community diversity and lower ecological resilience. Prolonged fencing may suppress regeneration of dominant forage species and destabilize community dynamics. In desert grasslands, where water and nutrient limitations are severe, the benefits of enclosure tend to be slower and more limited [27]. Extended exclusion in these settings can exacerbate resource competition, reduce species diversity, and even accelerate grassland degradation. Light mowing or rotational grazing can help suppress dominant species expansion and improve aboveground–belowground biomass allocation efficiency, thereby maintaining plant diversity [38]. These positive effects are more pronounced in areas with higher precipitation and greater nutrient availability. However, in nutrient-poor or drought-prone soils, frequent mowing may lead to vegetation homogenization, reductions in soil carbon and nutrient stocks, and instability of microbial communities. Overall, the ecological thresholds for grazing intensity, enclosure duration, and mowing frequency tend to be higher in regions with more favorable climatic and edaphic conditions than in arid and nutrient-deficient environments. Therefore, a systematic understanding of how different grassland utilization strategies under varying environmental constraints interact with vegetation, soil, and microbial responses is essential [39,40,41]. Elucidating the context-specific mechanisms underlying these responses will enable the development of regionally tailored analytical frameworks and evidence-based management strategies, ultimately supporting grassland conservation, degradation mitigation, and sustainable ecosystem management [42].

To contextualize the reliability and generalizability of our grazing intensity framework and treatment effect estimates, we focus on two key contextual factors relevant to the Qinghai–Tibet Plateau (QTP) grasslands. For grazing intensity validation across livestock types, we converted stocking rates of dominant QTP livestock such as yaks and sheep to cattle equivalents using regional forage consumption benchmarks. This conversion showed consistent ecological responses across livestock types, moderate stocking rates, whether measured in cattle yaks or sheep, still promoted plant species coexistence and soil nutrient cycling, confirming the framework’s cross-livestock applicability. Across QTP regional contexts, the intensity thresholds retained ecological meaning, though response magnitudes differed: heavy grazing caused more severe degradation in arid western deserts than in humid eastern meadows, tied to lower baseline vegetation cover in desert areas. Regarding study duration, shorter-term studies often overestimated treatment effects, such as rapid vegetation biomass increases under enclosure, due to initial recovery pulses. Longer-term studies by contrast revealed stable trends including biomass plateaus after several years of enclosure as ecosystems reached carrying capacity, and also captured slower soil nutrient changes that short-term work overlooked. These observations emphasize the need to anchor grazing intensity interpretations in livestock type and regional characteristics, and to account for study duration when assessing treatment effects, ensuring our findings align with the QTP’s diverse ecological conditions and support more targeted grassland management.

4.2. Implications for Scientific Research and Practical Management

This meta-analysis provides a comprehensive synthesis of how varying intensities of grazing, enclosure, and mowing differentially affect grassland ecosystems across the Qinghai–Tibet Plateau, offering robust scientific evidence to inform both future research and practical management. The findings underscore that fragmented knowledge and one-size-fits-all management strategies are insufficient for achieving sustainable grassland stewardship in such environmentally complex and heterogeneous regions [43,44]. It is imperative that researchers accelerate the development of integrative research frameworks that span across regions, ecological interfaces, and spatial scales [45]. Vegetation, soil, and microbial communities exhibit distinct ecological responses to different types and intensities of disturbance, and these responses are strongly modulated by environmental heterogeneity—including climatic conditions, soil fertility, vegetation type, and elevation. Therefore, building on meta-analytic evidence, future studies must systematically investigate the interactive effects between disturbance regimes and environmental gradients to deepen our understanding of the coupled plant–soil–microbe processes underpinning grassland functioning [46]. In particular, identifying region-specific ecological thresholds for appropriate land-use intensities through quantitative approaches is essential to provide operational and site-adapted guidance for sustainable grassland management [42]. Furthermore, advancing interdisciplinary and multi-method integration is critical. Research should move toward a multi-scale approach that links plot-level ecological mechanisms with landscape-level heterogeneity and regional-scale management policies. Such a framework can better support informed decision-making for sustainable management [47]. The synthesized evidence strongly supports the concept of evidence-based differentiated grassland management, calling for spatially explicit, context-sensitive strategies that balance short-term restoration outcomes with long-term ecosystem stability. Sustainable grassland management must also account for the complex trade-offs and synergies among ecological conservation, livestock production, and pastoral livelihoods [48]. To this end, integrated “combination management” strategies (i.e., enclosure, grazing, and mowing) tailored to local conditions should be explored. This requires not only a nuanced understanding of the differential ecological impacts of each land-use practice but also a careful assessment of their economic costs and ecological benefits across diverse regions. Ultimately, a dual goal must be pursued: to restore and maintain grassland ecosystem integrity while ensuring its productive and sustainable use [49,50].

Ecosystem Services of the Qinghai–Tibet Plateau: The meta-analysis findings in the file hold key implications for sustaining the Qinghai–Tibet plateau‘s score ecosystem services, which underpin regional ecological security and human well-being. Results show moderate grazing and medium-term enclosure enhance two critical regulating services: carbon sequestration and water regulation. Moderate grazing boosted soil organic carbon by 8–10% in low-elevation grasslands, while medium term enclosure reduced surface runoff by 12–15% in mid-elevation meadows, both supporting the plateau’s role as a “carbon sink” and “water tower” for East Asia. In contrast, heavy grazing and long-term enclosure degraded these services: heavy grazing caused 10% vegetation coverage loss, weakening carbon uptake and raising soil erosion risk, while enclosure over 6 years cut forage availability, undermining the provisioning service for livestock husbandry. These patterns highlight that aligning management with ecosystem service priorities—such as balancing carbon sequestration and sustainable grazing, which is vital for preserving the plateau’s multi-functional value [51,52].

Moderate grazing and short-term enclosure stand out as practices that foster the highest level of plant species diversity: they curb the overgrowth of dominant species (such as *Kobresia humilis*) and create varied microhabitats that support subordinate plant species, which in turn enriches overall plant community diversity. This enhanced plant diversity further drives increased microbial activity in the soil. Specifically, under moderate grazing, the presence of a more diverse plant community boosts the abundance of soil decomposers, organisms that break down organic matter, creating favorable conditions for the cycling of key nutrients [53]. This process accelerates the release of nitrogen and phosphorus from organic matter, making these essential nutrients more accessible to plants and sustaining the continuity of biogeochemical cycles. In contrast, heavy grazing disrupts this interconnected soil–plant–microorganism system. It reduces plant species diversity by compromising the survival of subordinate species and impairs microbial activity through two key mechanisms: compaction of the soil and a reduction in root exudates [54]. The decline in both plant diversity and microbial activity slows nutrient cycling, leading to shortages of key nutrients that further constrain grassland ecosystem function. High-elevation desert grasslands are particularly vulnerable to disturbance. Even light grazing in these areas reduces both plant and microbial diversity, as their soil–plant–microbe networks are inherently fragile and lack the resilience to recover from even mild human-induced disturbance [55]. These observations collectively highlight that preserving the integrity of soil–plant–microorganism interactions is a core requirement for sustaining healthy levels of biodiversity and functional biogeochemical cycles in Qinghai–Tibet Plateau grasslands [56].

Land-Use Changes and Conservation-Integrated Management Strategies: The meta-analysis in the file offers actionable insights for reconciling Qinghai–Tibet Plateau land-use changes, such as grazing expansion and enclosure construction, with conservation and regional activity integration. For regions with growing grazing intensity, like eastern plateau meadows, stocking rates should be capped at moderate levels to avoid degradation thresholds, balancing livestock production and grassland conservation. For areas with extensive enclosure, such as central plateau steppes, “rotational enclosure” is recommended over permanent enclosure to prevent litter buildup and maintain grassland openness for pollinators. For high elevation desert grasslands, where land-use shifts stem from tourism and infrastructure, strict light-grazing-only zones and no-enclosure policies are needed, as their low resilience makes minor changes irreversible. Integrating these strategies with regional planning, like aligning grazing bans with growing seasons and zoning conservation areas to protect soil–plant–microbe hotspots, it will support both ecological conservation and sustainable local livelihoods [57,58].

4.3. Limitations and Future Research Directions

This study conducted a meta-analysis synthesizing the effects of varying intensities of grazing, enclosure, and mowing on vegetation structure, biomass, diversity, soil physicochemical properties, microbial biomass, diversity, and enzyme activities across grasslands of the Qinghai–Tibet Plateau. However, we did not quantitatively analyze the spatial and environmental heterogeneity of these effects, nor their coupling mechanisms with key environmental factors such as climate, soil, vegetation type, and elevation. Understanding these heterogeneities is crucial not only for differentiated management strategies but also as a scientific basis for large-scale simulation and prediction, warranting greater attention in future research. In the context of ongoing climate change, there is an urgent need to quantitatively elucidate the interactive effects between grassland utilization practices and climatic and environmental factors, and to identify appropriate intensity thresholds and their spatial and environmental heterogeneity characteristics for grazing, enclosure, and mowing. Future studies should build upon meta-analytic results by incorporating advanced statistical approaches such as multilevel modeling and structural equation modeling to deeply explore the nonlinear coupling between land-use types and environmental variables [59,60,61]. Moreover, due to limitations in the number of existing studies, the spatial coverage of sampling is uneven, with a disproportionate focus on alpine meadows in the eastern Plateau and alpine grasslands in the southwest, compared to the western alpine desert grasslands and pastoral fringe areas. Existing research predominantly addresses structural indicators, while functional trait-based studies remain insufficient. Such gaps in spatial coverage and key functional indicators undermine the completeness and extrapolative power of current findings and should be prioritized in future investigations. Although meta-analysis offers a robust framework for knowledge integration and pattern recognition, many studies remain confined to local-scale observational statistics. To advance beyond descriptive synthesis, future research should emphasize interdisciplinary and multi-method integration, combining controlled experiments, remote sensing data, ecological process models, and machine learning algorithms to enable cross-scale simulation and prediction [62,63,64]. This progression will facilitate meta-analyses transitioning from “inductive description” to “mechanism identification” and “predictive assessment”, thereby providing more comprehensive scientific support for sustainable grassland management.

5. Conclusions

This study conducted a comprehensive meta-analysis by integrating literature from multiple academic databases, aiming to verify the hypothesis that “grazing, enclosure, and mowing intensities differentially affect the QTP grassland vegetation–soil–microbe system, and such effects are regulated by environmental heterogeneity.” The results confirm this hypothesis: moderate grazing and mowing can maintain the stable structure and sound ecological functions of the grassland ecosystem, while enclosure exerts positive effects on ecosystem restoration in the short term but leads to resource competition intensification and functional degradation in the long term. Notably, environmental heterogeneity factors significantly modulate the magnitude and direction of these effects, resulting in obvious spatial differences in the response of the grassland system to utilization practices. To promote the sustainable development of QTP grasslands, it is necessary to determine the suitable intensity thresholds of grazing, enclosure, and mowing based on the environmental characteristics of different regions, and construct spatially differentiated integrated management strategies that combine multiple utilization practices. Such strategies should balance the synergistic benefits of ecological conservation and livestock production, thereby providing a scientific basis for addressing grassland degradation and optimizing management practices in ecologically fragile highland grassland ecosystems.