Abstract. Long-term, high-resolution canopy cover data are essential for understanding grassland ecosystem dynamics and informing sustainable management. However, existing products are largely limited to coarse spatial resolutions, constraining their utility for high-precision, large-scale analyses. In this study, we collected over 16,000 drone image tiles (30 m × 30 m) from 2,144 sites across China and developed a machine learning model to produce a spatially seamless, 30 m annual dataset of national grassland canopy cover from 1990 to 2023 by integrating drone and Landsat-series imagery. The model achieves high predictive accuracy (R2 = 0.73, RMSE = 18.4 %) and robust temporal transferability (R2 = 0.68, RMSE = 20.6 %). Comparisons with existing large-scale products demonstrated significantly improved accuracy and reduced residual artifacts, underscoring the robustness of our approach across diverse grassland types and time periods. Spatiotemporal analysis indicated a multi-decadal mean canopy cover of 43.80 ± 18.69 % across China’s grasslands. Over the 34-year period, 41.76 % of grasslands exhibited significant increases, 57.16 % showed nonsignificant change, and 1.08 % experienced significant declines. Climatic factors—including drought, precipitation, and temperature—emerged as the dominant drivers of canopy cover dynamics at the national scale, although their effects exhibited pronounced spatial heterogeneity. In contrast, anthropogenic pressures played a secondary role overall but could override climatic influences at local scales. Collectively, these findings, together with the long-term, high-resolution canopy cover dataset developed in this study, provide an essential basis for advancing the understanding of grassland ecosystem dynamics and for supporting evidence-based conservation and sustainable management strategies, particularly under intensifying climate change and increasing frequency of extreme events. The national grassland canopy cover dataset generated in this study is archived on Zenodo and can be freely downloaded from https://doi.org/10.5281/zenodo.20301123 (Jiang et al., 2026).

It remains unresolved why biodiversity experiments consistently yield positive biodiversity–ecosystem functioning relationships, with species performing better, on average, in species-rich than in species-poor communities 1–5 , whereas natural communities often show mixed patterns 6,7 . To address this gap, we developed a framework applicable to both systems that partitions ecosystem functioning into average and covariance effects arising from proportionate and abundance-dependent disproportionate species contributions, respectively. Across both systems, ecosystem functioning was positively related to both effects, and species diversity was positively associated with the average effect but negatively associated with the covariance effect. In natural communities, a strong covariance effect generated mixed or sometimes negative biodiversity–ecosystem functioning relationships, including under nitrogen addition 8–11 , whereas in biodiversity experiments, its low relative importance, resulting from the assembly of species with controlled abundances 12–14 , maintained positive relationships. This framework reveals shared underlying mechanisms and reconciles discrepant biodiversity–ecosystem functioning relationships across the two systems.

Ecosystem multifunctionality (EMF), integrating the multifaceted nature of ecosystem functioning, is increasingly used as a key indicator for assessing grassland health and guiding restoration strategies. It remains unclear whether biodiversity consistently enhances EMF across aridity gradients in water-limited ecosystems. Based on a 445-site survey spanning a broad aridity gradient in the temperate grasslands of northern China, we revealed threshold-type responses of EMF to aridity, with a critical value of 0.565. Below this threshold, EMF was strongly driven by water availability and declined sharply with increasing aridity. Above the threshold, intensified environmental filtering for drought-tolerant species weakened the influence of aridity but amplified the role of plant species diversity, likely due to their unique and irreplaceable contributions to EMF. Grazing exerted relatively weak direct effects on EMF compared with aridity. However, greater grazing intensity reduced plant species diversity, thus reducing EMF below the threshold, while above the threshold, it increased soil pH, reducing both plant diversity and EMF. These findings reveal the vulnerability of dryland ecosystem functioning to ongoing climatic drying and the strengthening influence of biodiversity in sustaining it, underscoring the necessity of conserving biodiversity to maintain ecosystem multifunctionality and resilience in arid regions.

Biodiversity can stabilize ecosystem functioning, yet this relationship is often weakened under environmental change such as nutrient enrichment. The mechanisms underlying this weakening remain unclear, particularly whether it arises from shifts in species that consistently contribute to community dynamics or from increased species turnover. Species that persist through time are expected to underpin stability by maintaining species stability and asynchronous dynamics, whereas transient species, characterized by low temporal occupancy, primarily contribute to compositional turnover with limited stabilizing effects. Disentangling the roles of these two groups may therefore explain how nutrient enrichment alters biodiversity-stability relationships. Using data from 49 grasslands with and without fertilization, we partitioned communities into persistent and transient species based on temporal occupancy. We quantified their respective contributions to community temporal stability and evaluated how fertilization altered the relationships between species richness, species asynchrony, species stability and community stability using structural equation models. Community stability was primarily driven by persistent species: communities with more stable and asynchronous persistent species exhibited higher temporal stability. In unfertilized conditions, species richness was positively associated with the stability and asynchrony of persistent species, as well as with the asynchrony between persistent and transient species, which were linked to higher community stability. Under fertilization, these pathways were weakened, as fertilization reduced persistent species stability and persistent-transient asynchrony, resulting in lower community stability. Transient species contributed little to stability in either condition and may obscure biodiversity-stability relationships when not explicitly accounted for. Synthesis. These results indicate that biodiversity-stability relationships are mediated predominantly by species that persist through time, consistent with theoretical expectations that invariability emerges from stable and asynchronous population dynamics. Nutrient enrichment weakens these relationships by disrupting the stabilizing role of persistent species rather than by increasing the contribution of transient species. Accounting for species temporal persistence thus provides a mechanistic basis for predicting how global change alters ecosystem stability.

The desertified ecological restoration vegetation of Wuzhumuqin grassland plays an important role in the ecological restoration and protection of the region. However, there are few studies on the monitoring of the changes in ecological restoration vegetation in grassland sandy land in the past. In order to improve the low efficiency of ecological restoration vegetation monitoring, this study used Gaofen-6 (GF-6) remote sensing data to calculate the kernel Normalized Difference Vegetation Index (kNDVI) and vegetation coverage of ecological restoration vegetation and analyze their spatial and temporal trends. At the same time, a transform three-branch network structure based on deep learning is proposed to extract visual features. The kernel Normalized Difference Vegetation Index-position-temporal awareness transformer (kNDVI-PT-Former) model monitoring method based on two-phase remote sensing image features combined with kNDVI for spatio-temporal feature extraction can accurately obtain the vegetation changes in desertification ecological restoration in Wuzhumuqin grassland. The results show that the kNDVI of the study area shows an increasing trend from 2019 to 2024. The kNDVI value is 0.4086 in 2019 and 0.4927 in 2024. From the perspective of the change trend of vegetation coverage, the overall vegetation coverage of the Wuzhumuqin desertification restoration study area showed a gradual increase trend from 2019 to 2024, and the vegetation coverage increased by 19% in 2024 compared with 2019. The transformation of vegetation coverage from low level to high level in the study area is more prominent. Based on the self-built monitoring dataset of more than 5.2 million pairs of grassland vegetation changes, through model comparison and analysis, the kNDVI-PT-Former model obtains that the Class Pixel Accuracy (CPA) is 0.7295, the Intersection over Union (IoU) is 0.7228, and the overall monitoring accuracy of the model is improved by 11%. Furthermore, the stability of the model’s performance was confirmed through evaluation with five-fold cross-validation.

The biogeochemical niche (BN) hypothesis posits that each species has a specific elemental composition. However, the BN of roots and its interaction with leaf BN have largely been neglected until now across diverse environmental conditions. We investigated the relationships between the elemental compositions of leaves and roots, phylogeny, and environmental variables, as well as the connection between leaf and root BN. We analyzed the concentrations of carbon, nitrogen, phosphorus, potassium, calcium, and magnesium in the leaves and roots of 12 394 individuals from 1238 species. Consistent with the BN hypothesis, despite significant differences in elemental concentrations and their ratios between leaves and roots, we observed strong legacy (phylogenetic + species) signals in the species-specific elemental compositions. This finding confirms that the elemental compositions of leaves and roots can contribute to identifying species niches. Our study revealed a higher phylogenetic conservatism for BN in leaves than in roots and provided evidence of a tight association between the species-specific BN of leaves and roots. Our results underscore the broad applicability of the BN hypothesis across diverse species and biomes and demonstrate the critical role of evolutionary legacy in driving coordinated dynamics in both above- and belowground ecological niches.

The biogeochemical niche (BN) hypothesis posits that each species performs species-specific functions linked to its unique elemental composition. However, the relationship between BN and plant functions-such as biomass-remains to be fully understood. In our study, we assessed the shoot and root biomass, as well as the concentrations of 10 elements in the leaves and roots of 368 individuals from 18 species within the Potentillinae across northern grasslands in China. We tested the BN hypothesis for both leaves and roots and investigated the relationships between the BN and biomass in the above- and below-ground in these grasslands. Our analysis revealed that legacy (phylogenetic + species) effects explained a significant proportion of the variation in leaf and root elemental composition among species within a narrower taxonomic range, and leaf elemental composition exhibited higher phylogenetic conservatism compared to roots. Moreover, we also identified a strong connection between species B and N that estimated by leaves and roots, along with positive associations between them and the shoot and root biomass. Synthesis. Our results underscore the strong connections between the BN size and plant biomass above- and below-ground. We suggest that the BN could serve as a valuable framework for linking elemental composition to ecosystem functions in grasslands. (sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(BN)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic).(sic)(sic),BN(sic)(sic)(sic)(sic)(sic)((sic)(sic)(sic)(sic))(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic). (sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)18(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)368(sic)(sic)(sic)(sic)(sic)(sic),(sic)(sic)(sic)(sic)(sic)/(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)10(sic)(sic)(sic)(sic)(sic)(sic).(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)BN(sic)(sic),(sic)(sic)(sic)BN(sic)(sic)(sic)(sic)(sic)/(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic). (sic)(sic)(sic)(sic),(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic),(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)((sic)(sic)(sic)(sic)+(sic)(sic))(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic),(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic).(sic)(sic)(sic)(sic)(sic),(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)BN(sic)(sic)(sic)(sic)(sic)(sic)(sic),(sic)(sic)(sic)(sic)(sic)(sic)/(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic). (sic)(sic)(sic)(sic)(sic)(sic)(sic)BN(sic)(sic)(sic)(sic)(sic)(sic)(sic)/(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic),(sic)(sic)BN(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic).

As global temperatures rise, extreme climate events are becoming more intense, frequent, and prolonged, profoundly altering forest ecosystems, particularly in arid and semi-arid regions. This study employs dendrochronological methods to examine the relationship between the radial growth of Picea schrenkiana (P. schrenkiana) and Juniperus jarkendensis (J. jarkendensis) and extreme climate events in the Eastern Pamirs. The findings reveal that: P. schrenkiana is more sensitive to extreme temperatures fluctuations, whereas J. jarkendensis exhibits a stronger response to extreme precipitation. In the context of global warming, P. schrenkiana and J. jarkendensis exhibit a gradual trend of shifting from negative to positive responses to mean temperature (T) and extreme minimum temperature (TNn). Following climatic abruptions, both species display significant positive correlations with T and TNn, reflecting their adaptive adjustments to a warming climate and indicating their ability to leverage more favorable temperature conditions to promote growth. These results suggest that global warming has significantly altered the growth dynamics of these tree species. The interaction of multiple climate factors, rather than a single variable, drives tree growth. Consequently, targeted management and conservation strategies are essential to mitigate the impacts of extreme climate events on different tree species.

Isolated individual processes of ecosystem carbon (C) cycles have largely shaped our understanding of C cycle processes under environmental change. Yet, in reality, C cycle processes are inter-related and hierarchical. How these processes respond to warming and grazing has rarely been investigated in a single manipulative experiment. Moreover, biodiversity loss is a major driver of ecosystem change under environmental change, but whether these responses are mechanistically linked to biodiversity remains unclear. Here, we performed a 5-year field manipulative warming with seasonal grazing experiment in an alpine meadow on the Qinghai-Tibetan Plateau. Our results showed that both warming and moderate grazing decreased net ecosystem productivity (NEP) by 42.1% and 38.3%, and their interaction decreased it by 56.2% during the summer grazing period. However, they had no significant effects on NEP during the winter grazing period. Overall, annual gross primary productivity (GPP) and ecosystem respiration (Re) were mainly determined by aboveground rather than belowground processes, and Re variation which was mainly controlled by aboveground respiration explained 50% of the variation in annual NEP under warming and grazing. Moreover, lower species richness induced by warming and grazing caused smaller NEP with smaller net primary productivity and higher aboveground respiration. The responses of aboveground C cycle processes were greater than that of belowground C cycle processes, suggesting asymmetric above- and belowground responses to warming and grazing. Therefore, our findings suggested that there were higher GPP and Re with lower C sequestration (‘two high with one low patterns’) under warming and moderate grazing. Plant diversity modulated the responses of soil C sequestration to warming and grazing. It is essential to understand the underlying mechanisms of the effects of biodiversity on hierarchical C cycle processes under combined warming and grazing in the future.

The increasing prevalence of drought events in grasslands and shrublands worldwide potentially has impacts on soil organic carbon (SOC). We leveraged the International Drought Experiment to study how SOC, including particulate organic carbon (POC) and mineral-associated organic carbon (MAOC) concentrations, responds to extreme drought treatments (1-in-100-year) for 1 to 5 years at 19 sites worldwide. In more mesic areas (aridity index > 0.65), SOC and POC concentrations decreased by 7.9% (±3.9) and 15.9% (±6.2) with drought, respectively, but there were no impacts on MAOC concentrations. However, drought had no impact on SOC, POC, or MAOC concentrations in drylands (aridity index < 0.65). The response of SOC to drought varied along an aridity gradient, concomitant with interannual precipitation variability and standing SOC concentration gradients. These findings highlight the differing response magnitudes of POC and MAOC concentrations to drought and the key regulating role of aridity.

The steppes on the Mongolian Plateau cover a large land area of Mongolia (MG) and Inner Mongolia of China (IM). They share similar biophysical conditions, but have experienced divergent land use systems over the past several decades. However, no systematic field study has been done across different climate-vegetation zones for a comparison of the patterns of these steppes subject to different land use systems. We performed a field survey on 65 sites along four climatic transects, two in MG and two in IM, across desert steppe, typical steppe and meadow steppe zones, and investigated plant species composition, diversity and productivity. We compared these vegetation features between two regions, and attributed their variation patterns to climate and human activities. The results showed that plant species or plant functional group composition were similar between the two regions. However, plant productivity was significantly higher, whereas plant diversity indices were lower, in IM than MG. Aridity is the primary factor driving the spatial variation in plant diversity and productivity of the steppes across the Mongolian Plateau, while land use also has profound effects. The contrast vegetation pattern between MG and IM may be related with their difference in overgrazing history and grassland management strategies. These results suggest that overgrazing induced decline in plant productivity and its recovery is faster than the species diversity loss and recovery. Our results indicate that land use policies are urgently needed to promote destocking for preventing biodiversity loss in Mongolia, and that practical measures need to be developed to facilitate the recovery of species diversity in Inner Mongolia.

Root-borne microbial necromass carbon (MNC) is an under-investigated contributor to soil organic carbon (SOC) in grasslands. Here we conduct a benchmark assessment of root-borne MNC based on amino sugars in the fresh roots of 27 dominant species and mixed roots (including dead ones) of mixed species in Inner Mongolian grasslands. We find that mixed roots contain 9.1-62.4 mg g-1 MNC, which may contribute similar to 10% of MNC in surface soils (0-10 cm) if all enters the soil without degradation. Root (and arbuscular mycorrhizal fungi; AMF) turnover rather than AMF biomass controls root-borne MNC accumulation. Based on a microcosm decomposition experiment of mixed roots in model soils under optimal conditions (similar to 22 degrees C; 60% of maximum water holding capacity), we further estimate that 43%-75% of root-borne MNC remains after 2-year decomposition, implying that root-borne MNC is relatively stable. Hence, root-borne MNC may be an overlooked potential source of soil MNC and SOC in grasslands.

Canopy cover is a crucial indicator for assessing grassland health and ecosystem services. However, achieving accurate high-resolution estimates of grassland canopy cover at a large spatial scale remains challenging due to the limited spatial coverage of field measurements and the scale mismatch between field measurements and satellite imagery. In this study, we addressed these challenges by proposing a regression-based approach to estimate large-scale grassland canopy cover, leveraging the integration of drone imagery and multisource remote sensing data. Specifically, over 90,000 10 x 10 m drone image tiles were collected at 1,255 sites across China. All drone image tiles were classified into grass and non-grass pixels to generate ground-truth canopy cover estimates. These estimates were then temporally aligned with satellite imagery-derived features to build a random forest regression model to map the grassland canopy cover distribution of China. Our results revealed that a single classification model can effectively distinguish between grass and non-grass pixels in drone images collected across diverse grassland types and large spatial scales, with multilayer perceptron demonstrating superior classification accuracy compared to Canopeo, support vector machine, random forest, and pyramid scene parsing network. The integration of extensive drone imagery successfully addressed the scale-mismatch issue between traditional ground measurements and satellite imagery, contributing significantly to enhancing mapping accuracy. The national canopy cover map of China generated for the year 2021 exhibited a spatial pattern of increasing canopy cover from northwest to southeast, with an average value of 56 % and a standard deviation of 26 %. Moreover, it demonstrated high accuracy, with a coefficient of determination of 0.89 and a root-mean- squared error of 12.38 %. The resulting high-resolution canopy cover map of China holds great potential in advancing our comprehension of grassland ecosystem processes and advocating for the sustainable management of grassland resources.

A central question concerning biodiversity loss is how it impacts ecosystem functions and services. Experiments manipulating species diversity often show the complementarity effect, stemming from niche differentiation or facilitation among species, contributes dominantly to ecosystem functions. The selection effect, resulting from the increased likelihood of species-diverse communities containing high-performance species, plays a limited role. However, the applicability of these findings to natural ecosystems remains unclear. By partitioning the two effects and further separating them into dominant-species and subordinate-species components, we found that the selection effect was significant in enhancing natural grassland functions (plant biomass production and community coverage), and better predicted functions than the complementarity effect. In natural grasslands, the selection effect was largely driven by dominant species and independent from community-wide species diversity, while the complementarity effect was largely driven by subordinate species and positively associated with species diversity. Our results suggest that the selection effect may play a more important role in driving the functioning of natural ecosystems, which may be underestimated in biodiversity experiments.

Fertilization and mowing are commonly used grassland management measures. However, an understanding of how plant, soil bacterial, and soil fungal diversity responds to these two management measures are limited. We ran a fertilization and mowing experiment using a completely randomized design for 7 years (from 2014 to 2020) to explore the effects of fertilization and mowing on plant, soil bacterial, and soil fungal diversity. We used one-way and two-way Analysis of Variance (ANOVA) to compare how fertilization and fertilization + mowing impacted plant, soil bacterial and fungal diversity. We also used a structural equation model to analyze the maintenance mechanisms about fertilization and mowing affecting plant, soil bacterial, and soil fungal diversity. Our ANOVA results showed that when nitrogen addition level was beyond 8 GNM-2YR-1, fertilization signifi-cantly reduced plant taxonomic and functional diversity. However, the fertilization + mowing significantly increased plant diversity compared to fertilization. Fertilization and fertilization + mowing had no significant effect on soil bacterial and fungal diversity, but could change soil bacterial and fungal community structure by altering soil pH. Our structural equation model showed that fertilization reduced plant diversity mainly by increasing light competition in plant community and causing soil acidification. These results confirm light competition and soil acidification as maintenance mechanisms of fertilization and mowing affecting grassland plant diversity. Our findings highlight mowing can mitigate the adverse effects of fertilization on grassland plant diversity. However, the application of fertilization + mowing in grassland management needs to consider fertilization concentration.

Anthropogenic eutrophication is known to impair the stability of aboveground net primary productivity (ANPP), but its effects on the stability of belowground (BNPP) and total (TNPP) net primary productivity remain poorly understood. Based on a nitrogen and phosphorus addition experiment in a Tibetan alpine grassland, we show that nitrogen addition had little impact on the temporal stability of ANPP, BNPP, and TNPP, whereas phosphorus addition reduced the temporal stability of BNPP and TNPP, but not ANPP. Significant interactive effects of nitrogen and phosphorus addition were observed on the stability of ANPP because of the opposite phosphorus effects under ambient and enriched nitrogen conditions. We found that the stability of TNPP was primarily driven by that of BNPP rather than that of ANPP. The responses of BNPP stability cannot be predicted by those of ANPP stability, as the variations in responses of ANPP and BNPP to enriched nutrient, with ANPP increased while BNPP remained unaffected, resulted in asymmetric responses in their stability. The dynamics of grasses, the most abundant plant functional group, instead of community species diversity, largely contributed to the ANPP stability. Under the enriched nutrient condition, the synchronization of grasses reduced the grass stability, while the latter had a significant but weak negative impact on the BNPP stability. These findings challenge the prevalent view that species diversity regulates the responses of ecosystem stability to nutrient enrichment. Our findings also suggest that the ecological consequences of nutrient enrichment on ecosystem stability cannot be accurately predicted from the responses of aboveground components and highlight the need for a better understanding of the belowground ecosystem dynamics.