Although arbuscular mycorrhizal fungi (AMF) have been assumed to facilitate soil organic carbon (SOC) sequestration, they also regulate SOC decomposition via specific interactions with saprotrophs. We tested AMF hyphal impacts on SOC dynamics (e.g. labile C vs persistent C) under monoculture conditions with different grasses (e.g. Leymus chinensis and Stipa grandis) using an in-growth core method. Although the total SOC pool was unaffected by the presence of AM fungal hyphae, the proportional composition of labile and persistent C within SOC pools differed significantly between treatments with and without hyphal access. The presence of AM fungal hyphae from L. chinensis was associated with increased abundances of actinomycetes and Gram-positive bacteria, alongside the higher activity of polyphenol oxidase that breaks down persistent soil C, leading to a higher proportion of labile C in the SOC pool. Under S. grandis, however, hyphal presence corresponded with a greater abundance of Gram-negative bacteria that often can degrade labile soil C, resulting in a higher proportion of persistent C in the SOC pool. The influence of AM fungal hyphae on SOC depends on the identity of host plants and thus shifts in plant community composition may strongly alter SOC dynamics in grasslands.

ToxiPlantNet: A UAV-based deep learning framework for toxic plant detection and quantitative analysis of spatiotemporal dynamics in grasslands

Alpine grasslands on the Qinghai-Tibet Plateau and temperate grasslands on the Mongolian Plateau are key components of the global carbon cycle but differ markedly in their responses to climate change. To investigate the spatiotemporal variations in Gross Primary Productivity (GPP) and their response to climate change in these two types of grasslands, we developed a novel Random Forest Regression-Light Use Efficiency-Solar Induced Fluorescence (RFR-LUE-SIF) model that integrates machine-learning regression with physiological efficiency principles and satellite-derived SIF observations. This framework bridges tower-based GPP observations with large-scale remote-sensing estimates, improving model interpretability and accuracy. The model reproduced observed GPP with high fidelity (R 2 = 0.91), identifying EVI, NIRv, and GOSIF as the most influential predictors. Spatially, alpine grassland GPP decreases from southeast to northwest, while temperate grassland GPP declines from northeast to southwest. From 2001 to 2023, both grassland types exhibited increasing GPP trends, with temperate grasslands showing a faster rise, indicating stronger climatic sensitivity. Further, partial correlation analysis and Structural Equation Modeling (SEM) reveal that alpine grassland productivity is generally more sensitive to temperature, particularly under adequate moisture conditions, whereas temperate grasslands exhibited stronger dependence on precipitation and vapor pressure deficit (VPD). The proposed RFR-LUE-SIF model provides a scalable, data-driven, and physiologically consistent approach for assessing grassland carbon dynamics and their hierarchical climatic responses across contrasting ecosystems.

Lifespan and dormancy type affect post-fire seed germination strategies: evidence from a multispecies experiment in a temperate grassland.

Under the pronounced warming–wetting trend in Northwest China, understanding vegetation responses to the redistribution of hydrothermal resources is essential for interpreting regional ecohydrological processes. Here, we developed a bivariate Long Short-Term Memory (LSTM) model to simulate leaf area index (LAI) dynamics for four representative vegetation types (cold temperate forest, shrubland, grassland, and cropland), using air temperature and soil moisture as predictors. The model reproduces seasonal vegetation phenology well across vegetation types (R2 > 0.9), indicating that LSTM effectively captures the cumulative and lagged effects of hydrothermal drivers. However, its performance diverges at the interannual scale. Interannual variability in grasslands in water-limited environments is reasonably represented (R2 = 0.31), consistent with their sensitivity to short-term hydroclimatic variability under warming–wetting conditions. In contrast, the model fails to reproduce the observed long-term greening trend in forests when driven solely by hydrothermal variables. This contrast suggests distinct underlying mechanisms across ecosystem types. Grassland dynamics are closely linked to high-frequency hydroclimatic variability, whereas forest growth appears to be governed by slower processes and low-frequency drivers, including CO2 fertilization, nitrogen deposition, and ecological inertia. As a result, hydrothermal variables alone are insufficient to explain long-term forest dynamics. Overall, these findings highlight a transition from water-limited to energy- and process-limited controls across vegetation types and underscore the limitations of purely climate-driven models. Integrating biogeochemical processes or process-based constraints into machine learning frameworks may therefore be necessary to improve predictions of long-term vegetation change under climate change.

Large herbivores and abiotic drivers jointly shape spatiotemporal grassland dynamics in a subalpine ecosystem.

Alternative vegetation states and limited reversibility: Insights from a novel grazing experiment.

Grasslands are widely recognized as important sinks of methane (CH4). However, their CH4 uptake capacity has been increasingly weakened driven by human activities and changes in precipitation regimes. In particular, the rapid expansion of livestock grazing can lead to substantial increases in CH4 emissions. For the Eurasian grasslands that have been subject to long-term grazing, how well they currently function and will continue to function as CH4 sinks under the changing precipitation regimes remains uncertain. Here we conducted an 11-year grazing-gradient experiment to assess the combined effects of grazing intensity and precipitation variability on grassland CH4 fluxes. We measured soil CH4 uptake in a manipulative grazing experiment in both wet and dry years, estimated livestock-derived CH4 emissions, and examined a range of biotic and abiotic drivers, including vegetation attributes, soil properties, and microbial biomass, for short- (prior to the onset of the lagged response) and long-term (the entire experimental period) influences on the dynamics of CH4 uptake in temperate meadow grasslands. Over the long term, CH4 fluxes are jointly regulated by precipitation and grazing, with overgrazing amplifying the suppressive effect of rainfall on CH4 uptake. Soil CH4 uptake showed a lagged response to grazing, which may have been triggered by extreme rainfall events. While moderate grazing sustained the long-term CH4 uptake, it fails to offset livestock-derived CH4 emissions. In the short term, CH4 fluxes are primarily governed by grazing, whereas in the long term they are jointly regulated by precipitation and grazing intensity. These results offer a new perspective for understanding the source-sink CH4 dynamics of grasslands in the context of ongoing climate change.

Abstract Accurate estimation of aboveground biomass (AGB) in alpine ecosystems is challenging because of strong phenological variability, heterogeneous canopy structure and complex spectral–biomass relationships. Using 2 years (2023–2024) of monthly ground‐based hyperspectral observations collected during the growing season (May–September), this study examined alpine meadow, degraded alpine meadow and alpine shrub communities on the northeastern Qinghai–Tibetan Plateau. Four regression approaches—elastic net, random forest (RF), support vector regression with a radial basis function kernel and extreme gradient boosting (XGBoost)—were compared with an emphasis on temporal transferability. A two‐stage feature selection framework combining correlation screening and LASSO regression was applied to smoothed, first‐derivative and continuum‐removed spectra, reducing more than 1000 spectral predictors to approximately 20 physiologically meaningful features. Tree‐based ensemble models, particularly RF and XGBoost, consistently achieved the highest predictive performance and showed strong robustness across vegetation types. Model accuracy exhibited clear seasonal dependence, with lower performance early in the growing season and marked improvement during mid and late season periods. The most informative predictors were concentrated in the red‐edge, near‐infrared and shortwave‐infrared regions, and models based on these optimized features matched or exceeded full‐spectrum performance. The results demonstrate that combining targeted spectral features with ensemble learning provides a robust framework for seasonal AGB estimation in heterogeneous alpine grasslands.

Abstract In this study, we examined the impact of the regularity of low-intensity grazing on vegetation dynamics of dry grasslands through 24 years monitoring of 127 permanent plots. The regularity of grazing in each permanent plot was defined as the ratio of the number of years in which grazing occurred to the number of years in which the plot was studied. From a nature conservation perspective, it was positive that regular grazing led to a reduction in woody plant cover. However, the cover of Red List (threatened) species significantly decreased with increasing grazing regularity. A negative correlation was found between monocotyledonous plant cover, especially grasses, and both the cover and number of Red List species. Although grazing regularity had no statistically significant influence on monocotyledonous plant cover, it was linked to reduced cover of dicotyledonous species. Additionally, no relationship was detected between the grazing regularity and cover of the target dry grassland Festuco-Brometea species or ruderal species. No direct relationships were identified between grazing regularity and overall vegetation conditions, as summarized by correlation coefficients for diversity, stability, and synchrony, but dry grassland community stability decreased with increasing synchrony, and synchrony was positively influenced by community diversity or species richness. At some sites, there was a very close positive correlation between community stability and total community cover and between diversity and M-Godron instability. Our findings show that while grazing regularity does not consistently dictate vegetation conditions, it significantly influences specific plant groups and interacts with community stability in complex ways. Low-intensity grazing appears to be an effective management strategy for dry grasslands, especially when combined with multi-year rest periods or occasional mowing, while continuous grazing generally suppresses dicotyledonous plants, including a number of threatened species.

Grassland degradation is a critical ecological problem worldwide that threatens ecosystem integrity and functional services. Although previous studies have documented the drivers of climate change, overgrazing, and anthropogenic perturbation, research concerning the impact of invasive alien plants on grassland ecosystems remains limited. The present study, integrating pairwise field investigation of Ageratina adenophora invasion and non-invasion plots across heterogeneous grassland types (tropical grasslands [TG]; tropical shrub-grasslands [TS]; warm-temperate grasslands [WG]; and warm-temperate shrub-grasslands [WS]) and A. adenophora indigenous plants phytotoxicity bioassay, aims to assess the invasibility and resilience of heterogeneous grassland landscapes to A. adenophora invasion. The field investigation demonstrated the greater vulnerability of TG and TS to A. adenophora invasion, whereas WG and WS possessed higher resilience. In addition, regression analysis revealed significant reductions of the Shannon-Wiener index and the Pielou index as the A. adenophora's important value reached the threshold 0.36. Bioassay showed that A. adenophora aqueous extracts inhibit seed germination and seedling growth of recipient plants, with Saccharum arundinaceum exhibiting the highest tolerance to A. adenophora stress. In summary, our findings not only highlight the flora communities' dynamics and invasibility of diverse grasslands driven by A. adenophora invasion in subtropical regions but also verify S. arundinaceum's potential for A. adenophora replacement management.

Modeling grazing effects on carbon dynamics in alpine grasslands of the Qinghai-Tibetan Plateau using the Biome-BGCMuSo model

Abstract Mountain grasslands are vital ecosystems providing critical services such as carbon sequestration, water regulation, and biodiversity conservation. However, these ecosystems are increasingly threatened by climate change and human activities. This study evaluates vegetation dynamics in global mountain grasslands (2000–2021) using remote sensing data, CMIP6 climate projections, and human modification indices. By means of machine learning models (Random Forest, eXtreme Gradient Boosting (XGBoost), and Long Short‐Term Memory (LSTM)) we identified dewpoint temperature, total evaporation, and human modification as dominant predictors of vegetation change, while soil water content and latent heat flux exhibited region‐specific impacts. Results indicate that 35.1% of grasslands remained stable, 32.1% improved, and 32.7% degraded, with degradation hotspots identified in the Tibetan Plateau, Ethiopian Highlands, Rocky Mountains, and Andes, while the Middle Eastern Mountain Ranges showed signs of improvement. Future projections using a hybrid XGBoost–LSTM model indicate limited global‐scale vegetation shifts by 2050 across SSP1‐2.6, SSP2‐4.5 and SSP5‐8.5 scenarios. However, regional differences are notable: the Tibetan Plateau shows a substantial increase in vegetation cover, while the Andes, Rocky Mountains, and parts of East Africa exhibit slight changes. Distributional shifts, especially under SSP5‐8.5, suggest increasing spatial heterogeneity in grassland responses. These findings underscore the importance of regional‐scale strategies to support grassland resilience under future climate and land‐use pressures.

Abstract. In semi-arid regions, grasses and shrubs often form spatially heterogeneous patterns interspersed with bare soil as a strategy to optimize resource use and maximise productivity. Accurately representing the matrix of vegetation and bare soil in global land surface models is essential for advancing the understanding of the carbon, water, and dust cycles. This study focuses on grasslands using the land surface model ORCHIDEE (ORganizing Carbon and Hydrology In Dynamic EcosystEms), which originally assumes a globally fixed grassland density representing a fixed number of individuals per unit of land. This assumption, referred to as the fixed density approach, limits the model's ability to capture grassland responses to environmental changes, resulting in unsustainable productivity and unrealistically frequent mortality events, particularly in resource-limited regions. To address these limitations, we introduced a dynamic density approach that simulates grassland density based on indicators of vegetation growth, such as reserve and labile carbon content in the grass. The simulated grassland density was consistent with field-based estimates from five regional case studies and showed a better representation of bare soil in grasslands than the fixed density approach. The emerging positive correlation between precipitation and simulated grassland density supported the validity of the approach. Compared to the fixed density approach, the dynamic density approach substantially reduced simulated mortality events, raised the aridity threshold for frequent mortality, improved the simulated leaf area index (LAI) both globally and in key semi-arid regions, and maintained realistic grassland productivity in regions where the presence of grassland is confirmed by remotely sensed LAI. This study not only demonstrates that simulating grassland density as a function of carbon availability improves ORCHIDEE's capacity to capture grassland dynamics under environmental variability, but also provides a promising foundation for investigating dust dynamics and subsequent land–atmosphere feedbacks in (semi-)arid regions.

Examining the long-term spatiotemporal distribution of grassland types and their transitions is crucial for better understanding regional and global changes. Most research in this field has examined the spatial distribution, temporal dynamics of grasslands, and their causes as a unified entity. This study predicted the distribution of nine major grassland types in Xinjiang under three climate change scenarios from 2041 to 2100 based on 1980s grassland maps, field data in 2023, and 28 factors. The total area of the nine grassland types showed a decreasing trend from 2041 to 2100. The lowland meadow (LM), temperate meadow steppe (TMS), temperate steppe desert (TSD), temperate desert steppe (TDS), and mountain meadow (MM) expanded, while significant declines occurred in alpine meadow (AM), alpine steppe (AS), temperate desert (TD), and temperate steppe (TS). Among cumulative contribution rate of the 28 factors examined in this study, NDVI, vegetation type, slope, elevation, soil_symbol, soil_ph, Bio1, Bio5, Bio8, Bio9, Bio10, Bio12, Bio13, Bio15, and Bio18 played important roles in most grassland types. LM, TD, and AS grassland were found to be more sensitive to E (environment), while AM, TDS, and TSD were more influenced by T (temperature). The distributions of MM and TMS are significantly influenced by the combined effects of all three categories of factors. For TS, the impacts of both temperature and environmental factors are substantial. These findings provided a robust foundation for conservation planning and the sustainable management of grassland ecosystems in temperate and alpine regions.

Resilience of soil organic carbon under precipitation variability: Insights from carbon-nitrogen dynamics in semi-arid grasslands

The aboveground live biomass in the grasslands of Manipur varied signifi cantly, ranging from 220.74 g m -2 in January to 812.89 g m -2 in September.The growth patterns of seven dominant species observed at these study sites exhibited distinct variations in their peak biomass values.The amount of standing dead material was recorded as lowest in February, measuring 321.23 g m -2 , and peaked in November at 864.14 g m -2 .Litter biomass fluctuated between 78.46 g m -2 in August and 205.56 g m -2 in December, while belowground biomass ranged from 1,074.00 g m -2 in June to 1,622.62 g m -2 in November throughout the study period.Notably, the highest concentration of belowground biomass was found in the 0-10 cm soil layer, with variations ranging from 73.24% in February to 83.36% in May.The root-to-shoot ratio was observed between 0.73 and 2.05.The analysis of variance for live biomass, standing dead material, litter, and belowground biomass indicated highly signifi cant differences across the various months.

Underground hydrocarbon pipelines represent a growing source of disturbance in arid and cold ecosystems of southern South America, yet their long-term ecological impacts remain poorly understood. We evaluated the effects of pipeline construction on vegetation and soil in Festuca gracillima (coironal) grasslands of the Magellanic steppe, Chile. Fourteen pipelines of varying ages (3–14 years) were surveyed using 442 vegetation plots (221 along the pipeline corridor, 221 in reference areas), combined with soil sampling (0–15 cm) for physical and chemical analyses. Vegetation was strongly affected. Species richness declined significantly in the pipeline corridor (19.6 ± 9.9 vs. 31.0 ± 9.6 species), as did total cover (36.2 ± 19.6% vs. 74.0 ± 6.0%), while bare soil increased markedly. Shannon diversity was slightly higher in disturbed sites (H' = 2.06 ± 0.37 vs. 1.75 ± 0.45), reflecting the proliferation of ruderal and introduced species. SIMPER analysis showed that the decline of native dominants such as Festuca gracillima and Empetrum rubrum, together with the expansion of Poa pratensis, Rumex acetosella, and Taraxacum officinale, explained more than 50% of community dissimilarity. The invasive Hieracium pilosella was unexpectedly more abundant in reference areas, suggesting a preference for intermediate disturbance regimes. Soil responses were more subtle but not negligible. Most physical properties (bulk density, porosity, water holding capacity) showed no consistent differences between treatments. However, several chemical parameters were significantly altered in disturbed soils, including increases in total nitrogen, available phosphorus, manganese, calcium, magnesium, and nitrate, and a decrease in exchangeable aluminum and clay content, although most changes faded out at short term. Canonical Correspondence Analysis revealed clear floristic segregation along gradients of bare soil, sand content, organic matter, and pH, indicating that soil properties, although moderately affected, shape successional trajectories. Overall, our results demonstrate that underground pipelines cause long-lasting changes in vegetation structure and composition, even when soil alterations appear modest or transient. Natural regeneration progresses slowly, leading to simplified communities dominated by opportunistic species with protective but low functional value. These findings highlight the need to critically evaluate restoration strategies based on agronomic sowing, and to consider assisted natural regeneration as a more sustainable option for conserving biodiversity, ecosystem function, and soil carbon stocks in arid grasslands.

Aims Nitrogen (N) deposition has emerged as a major driver of ecological change in alpine grasslands of the Qinghai-Tibetan Plateau under global climate change. To predict the ecological consequences of increasing nitrogen deposition, nitrogen addition experiments have been widely employed as a key methodological approach to simulate this process. However, the effects of nitrogen addition—considering its rate, duration, and form—on carbon (C) dynamics in these ecosystems remain inconsistent across studies. Understanding these effects is critical for predicting global carbon stocks and guiding sustainable grassland management. Methods We conducted a meta-analysis of 57 peer-reviewed studies (794 observations) to quantify the response of alpine grassland C dynamics to N addition. Results N addition significantly increased plant-derived carbon inputs, increasing aboveground biomass by 42.7%, belowground biomass by 16.2%, and dissolved organic carbon (DOC) by 10.7%. The soil organic carbon (SOC) content increased by 3.6% overall. Conversely, soil respiration decreased by 5.1%, whereas the microbial respiration rate increased by 21.9%. The addition of nitrogen decreased the soil pH by 0.20 units and the soil C/N ratio by 1.7%. The soil ammonium (NH4+) and nitrate (NO3-) contents decreased by 20.1% and 52.1%, respectively. The microbial biomass nitrogen (MBN) increased by 14.5%, whereas the microbial biomass carbon (MBC) decreased by 2.8%. The soil fungal-to-bacterial ratio (F/B) decreased by 31.0%. Conclusion These results indicate that shifts in microbial community structure drive SOC dynamics in alpine grasslands. Short-term N addition (≤5 years; ≤30 kg N ha -1 yr -1 ) enhances SOC through increased plant biomass and microbial C sequestration. However, long-term additions promote C loss via soil acidification and a critical shift in the microbial community, notably a decreased fungal-to-bacterial ratio. To sustain alpine ecosystem function, N addition rates should not exceed 10 kg N ha -1 yr -1 . Future research should prioritize interactions between N deposition status and soil acidification/microbial function in high-altitude regions.

The spatiotemporal patterns and dynamics of grassland established in the US conservation reserve program (CRP)