Forest cover dynamics are studied on a routine basis, but how changes in forest cover impact forest fragmentation has rarely been studied over a long time period resolution. This is, however, important because forest fragmentation critically impacts ecosystem services, such as biodiversity and cooling effects. Here, we apply a long time series of Landsat images from 1986–2018 and study how forest fragmentation has changed along with forest cover dynamics in southern China. Furthermore, we attribute drivers and study the impact on local air temperature changes. The region is particularly relevant as it was largely deforested three decades ago, and most of the current forests are the result of protection and forestation measures. We found a reduction in the forest fragmentation index FFI (−34.4%) from 1986 to 2018. In 81.2% of the area, forest cover increased and fragmentation decreased, while 18.5% of the area showed increases in both forest cover and fragmentation. The contribution of human activities to forest fragmentation increased by 9%, with a distinct spatial correlation between areas of increasing forest fragmentation and high levels of human disturbance. Furthermore, we found that the average level of cooling effects in areas with increased forest cover of less than 40% is heavily dominated by forest fragmentation, whereas the cooling effects are primarily controlled by changes in forest cover. These findings underscore the role of human disturbance in driving forest fragmentation, which in turn affects the functioning of forest ecosystems. The results emphasize the need for integrated land management strategies that balance forest restoration with the mitigation of human-induced fragmentation to sustain ecosystem services in the face of ongoing environmental change.

Mining of sand, gravel, and stone, collectively referred to as aggregates, has increased in recent years due to growing usage of these materials in developing urban landscapes and increasing living standards, especially in African countries that are pursuing rapid economic growth. The rise in demand, however, has wide-ranging consequences for the environment from which the material is extracted. Nevertheless, our knowledge of where the material is extracted from remains insufficient, which hinders assessments of the overall sustainability of extraction. Here, we provide a nationwide distribution of sand extraction in Rwanda by applying deep learning on highly refined resolution aerial imagery. We show that mining sites increased by 18% to 3,751 between 2008 and 2023, which are spatially linked to river valleys and urban development at the sub-watershed level. We further reveal that the expansion of sand extraction over the past decade often encroached into agricultural areas. Our approach offers a scalable way to monitor sand extraction activities to assess their implications on the natural and built environment.

Shrubland expansion is a globally occurring phenomenon under global change and has caused a wide range of ecological consequences. However, due to the visual similarity between shrubland and grassland, the accuracy of shrubland interpretation and its spatial distribution varies across different products, making shrub encroachment on the Tibetan Plateau (TP) uncertain. To address these challenges, we developed a phenology-based and pixelwise method utilizing the Landsat, Sentinel-1, and Sentinel-2 image archives to map shrubland distribution from 1990 to 2022 across the TP. We also investigated the factors affecting shrubland distribution. Using the Random Forest (RF) model, we achieved moderate to high accuracies (Kappa = 0.70-0.81) in predicting shrubland distributions, and we found that shrubland primarily occupies transitional zones between forest and grassland. In the southeast TP, solar radiation intensity is the dominant factor explaining the spatial distribution of shrubland, whereas in arid regions, water availability is the most important. From 1990 to 2022, the shrubland area slightly increased from 3.40% to 4.71%, with expansion showing a clustered pattern, mainly in the shaded aspects of arid regions. The shrubland identification method proposed here shows potential applicability in other areas with similar environmental conditions, such as arid landscapes or high-altitude ecosystems with pronounced seasonal vegetation dynamics. It also plays a crucial role in evaluating vegetation responses to climatic changes.

Vegetation growth may adapt to climate warming by adjusting the relationship between photosynthetic capacity and temperature. However, changes in the optimal temperature for ecosystem productivity during recent decades of warming remain uncertain. Here we provide empirical evidence that global optimal temperature increased at a rate of 0.017 ± 0.002 °C y−1 from 1982 to 2016, using multiple datasets of satellite-derived productivity and climate variables. Model simulations show that the optimal temperature will increase by 0.027 ± 0.001 °C y−1 until the end of 21st century. The global increasing optimal temperature is consistent with increasing mean air temperatures and model simulations further confirm the key role of temperature in regulating changes in optimal temperature, while being co-regulated by other factors, such as CO2 and precipitation. These results suggest that vegetation is acclimating to warming and that the negative impacts of climate change on ecosystem productivity may be less severe than previously thought. Climate change may have less negative impact on terrestrial ecosystem productivity as vegetation growth adapts to climatic warming, increasing optimal temperature from 1982 to 2016, according to evidence from satellite-derived ecosystem productivity, climate variables, and ecosystem simulations.

AbstractAmerican drylands account for circa 20% of the global drylands and form a critical part of the global ecosystems. This study comprehensively assessed the ecology and socio-economic status of American drylands by analyzing original and published data. The research findings reveal that North and South American drylands have more differences than commonness. In terms of commonness, both North and South American drylands have higher productivity and soil fertility than other drylands of the globe. Under this high ecosystem productivity context, North American drylands are the high agricultural productivity regions and South America is the largest beef exporter in the world. There are several aspects of differences between North and South American drylands. North American drylands possess an ecosystem productivity twice that of South American drylands. Precipitation has significantly decreased in North America drylands, while South American drylands have become wetting over the past three decades. Population in both North and South American drylands have increased. Vegetation coverage trends exhibit a weak rising trend in South America, while North America drylands have become significantly greener, mainly due to croplands irrigation. The driving forces on land use change and ecosystem productivity in North American drylands comprise a variety of factors, while those on South American drylands are relatively simpler, mostly caused by one driving agent. In dealing with the dual pressures of climate change and socio-economic developments, countries in both North and South America have implemented a series of drylands ecosystem protection measures, such as setting national park and conservation agriculture. These efficient and successful experiences can be examples for other dryland ecosystem protection around the world.

As the largest and highest plateau in the world, ecosystems on the Tibetan Plateau (TP) imply fundamental ecological significance to the globe. Among the variety, alpine grassland ecosystem on the TP forms a critical part of the global ecosystem and its soil carbon accounts over nine tenths of ecosystem carbon. Revealing soil carbon dynamics and the underlying driving forces is vital for clarifying ecosystem carbon sequestration capacity on the TP. By selecting northern TP, the core region of the TP, this study investigates spatiotemporal dynamics of soil total carbon and the driving forces based on two phases of soil sampling data from the 2010s and the 2020s. The research findings show that soil total carbon density (STCD) in total-surface (0-30 cm) in the 2010s (8.85 ± 3.08 kg C m-2) significantly decreased to the 2020s (7.15 ± 2.90 kg C m-2), with a decreasing rate (ΔSTCD) of -0.17 ± 0.39 kg C m-2 yr-1. Moreover, in both periods, STCD exhibited a gradual increase with soil depth deepening, while ΔSTCD loss was more apparent in top-surface and mid-surface than in sub-surface. Spatially, ΔSTCD loss in alpine desert grassland was -0.41 ± 0.48 kg C m-2 yr-1, which is significantly higher than that in alpine grassland (-0.11 ± 0.31 kg C m-2 yr-1) or alpine meadow (-0.04 ± 0.28 kg C m-2 yr-1). The STCD in 2010s explained >30 % of variances in ΔSTCD among the set of covariates. Moreover, rising temperature aggravates ΔSTCD loss in alpine desert grassland, while enhanced precipitation alleviates ΔSTCD loss in alpine meadow. This study sheds light on the influences of climate and background carbon on soil total carbon loss, which can be benchmark for predicting carbon dynamics under future climate change scenarios.

Recently, whether drylands of Northwest China (NW) have become wetting has been attracting surging attentions. By comparing the Standard Precipitation Evapotranspiration Indices (SPEI) derived from two different potential evapotranspiration estimates, i.e., the Thornthwaite algorithm (SPEI_th) and the Penman-Monteith equation (SPEI_pm), we try to resolve the controversy. The analysis indicated that air temperature has been warming significantly at a rate of 0.4°C decade−1 in the last five decades and the more arid areas are more prone to becoming warmer. Annual precipitation of the entire study area increased insignificantly by 3.6 mm decade−1 from 1970 to 2019 but NW presented significantly increasing trends. Further, the SPEI_th and SPEI_pm demonstrated similar wetting-drying-wetting trends (three phases) in China’s drylands during 1970–2019. The common periodical signals in the middle phase were identified both by SPEI_th and SPEI_pm wavelet analysis. Analysis with different temporal intervals can lead to divergent or even opposite results. The attribution analysis revealed that precipitation is the main climatic factor driving the drought trend transition. This study hints that the wetting trend’s direction and magnitude hinge on the targeted temporal periods and regions.

The knock-on effects between earlier vegetation activities and summer droughts may have important consequences for broad ecological processes. To date, little is known about how the chained effects drive the carbon and water cycles on the Tibetan Plateau (TP). Using the naturally occurring above-mentioned sequential events in spring and summer in 2015 and 2017, we applied the observations at the site, landscape, and regional scales to evaluate the chained effects on the TP. Our findings indicated that higher spring vegetation productivity is caused by early vegetation activities, partially compensated for summer drought-induced loss. Concurrently, increased spring evapotranspiration induced by earlier spring may drain soil water resources earlier, exacerbating summer water restrictions caused mainly by sparse precipitation. This lagged effect of early spring, accompanied by summer drought, significantly increased summer sensible heat flux by 23.2%. Remarkably, the mean air temperature (Ta) was lower than the baseline during drought. This decrease was contributed mainly by lower nighttime Ta, indicating that the region-specific characteristics of the TP could offset the heating effects as mentioned above. The characteristics of high altitude, low air pressure, and thin air could strongly weaken the cloud insulations. More substantial decreases in cloud amount during drought further decreased atmospheric counter radiations, leading to lower mean/nighttime Ta. The simulation results showed that lower mean Ta alleviated the decreases in gross primary productivity by 4.3% through reducing vapor pressure deficit by 5.1%. In conclusion, the present study highlighted the need to comprehensively consider the buffering effects of lower temperature during summer drought to precisely assess the chained effects on the TP.

Aim Ecosystem carbon use efficiency (CUEe) is a core parameter of ecosystem process models, but its relationships with climate are still uncertain, especially for ecosystems with harsh environments. Large inconsistencies in climate impacts on the CUEe have been reported among various spatial scales. The goal of this study was to examine whether warming promotes or restricts the CUEe and whether the CUEe responds to a warming gradient in a linear or nonlinear manner. Location Tibetan Plateau. Time period 2000-2018. Major taxa studied Alpine grassland ecosystem. Methods We integrated multiple-source data of carbon fluxes and CUEe, including warming experiments at a site scale, eddy covariance observations at a landscape scale and synthesized warming experiments and ecosystem process models at a regional scale. Next, we deployed a statistical model to examine the warming impacts on the CUEe across scales; the effects of biotic and abiotic factors on the CUEe and its components were summarized based on the results of standardized major axis tests and routines, structural equation modelling and nonlinear models. Results This study reported a suppressive warming impact on the CUEe, which followed a nonlinear curve with severe inhibition in the high-level warming treatment. With a warming threshold of 1.5-2.0 degrees C, CUEe response patterns transitioned from no change to a significant decrease. The restriction effects can be ascribed to the joint adverse and asymmetric effects of warming on CUEe components under multiple-level warming. Warming-modified relationships among CUEe components and the nonlinear effects of biotic and abiotic factors led to the nonlinear responses of CUEe to warming. Main conclusions This study revealed suppressive and nonlinear effects of warming on the CUEe, including especially dramatic CUEe decreases with high-level warming. These findings are critical for optimizing model parameters and improving predictions of the carbon sequestration capacity of alpine grasslands.

Global warming has impacted Northern Hemisphere (NH) grassland ecosystems to a great extent. Vegetation growing season length (GSL) has been extended by concurrent advances in spring green-up and postponements in autumn dormancy. However, the driving mechanisms of phenology are unclear as limited factors have been considered so far. Therefore, it is still elusive to what extent phenological changes shaped GSL. In this study, we used remote sensing normalized difference vegetation index (NDVI) to extract spring and autumn phenology of NH grasslands, and further explored the contribution of each phenophase to GSL through the coefficient of variation (CV) and contribution coefficient (CntC). We found that 65% of NH grasslands exhibited advanced start-of-season (SOS) and circa 58% showed delayed end-of-season (EOS) in the three decades. Changes in GSL was regulated more by EOS changes than by SOS changes, as evidenced by their respective 52 vs. 48% CntC. As for the relationship between phenology and environmental elements, the causing factor analysis revealed that climatic factors (temperature, precipitation, and their interactions) played a dominant role in SOS variations, while environmental and internal factors exerted dominant effects on EOS. Also, interactions of temperature and precipitation contributed a higher variation of SOS than either of them individually. The differentiated factors controlling the two bounding ends of the growing season suggested that it is impossible for GSL to continue to extend without limits under global warming.

Atmospheric water demand is practically characterized as vapor pressure deficit (VPD) and has been identified as a critical driver of ecosystem function, by affecting plant mortality, wildfires, and carbon loss. In this study, we used daily eddy covariance data across Chinese forest, grassland and shrubland ecosystems, in combination with remote sensing data, to investigate the impacts of VPD on gross primary productivity (GPP). We found divergent VPD impacts on GPP among grassland, shrubland and forest ecosystems. The VPD yielded substantial inhibitory impacts on GPP in grassland ecosystems and this suppressing impact was regulated by soil water content (SWC), showing that GPP declined with VPD under dry conditions but increased with VPD under wet conditions. This GPP variance was attributed to VPD, SWC and their interactions. More than 50% of the variability in GPP was explained by SWC in grassland ecosystems and by VPD in forest and shrubland ecosystems. Partial correlation analysis, random forest, and multiple linear regression revealed similar results when temperature, radiation and SWC were considered. Compared with shrubland and forest ecosystems, grassland has drier environmental conditions and poorer soil water-holding capacity that led to lower SWC and stronger negative impacts on VPD. Thus, GPP was susceptible to the negative impacts of higher VPD, especially under dryness stress. Our results highlight the need to comprehensively consider divergent VPD impacts for different ecosystems to more accurately assess climate impacts on ecosystem function.

Aim To study global patterns and temporal changes in the seasonal dynamics (quantity and seasonal distribution) of terrestrial gross carbon uptake in response to global environmental change. Location Global. Time period 2000-2016. Major taxa studied Terrestrial ecosystems. Methods Following a phenology-based definition of photosynthetic seasonality, we decompose gross primary production (GPP) into three periods, green-up, maturity and senescence, and derive their corresponding GPP (GPPgp, GPPmp and GPPsp, respectively) from a newly developed time series of satellite-based global GPP to study spatio-temporal dynamics of seasonal GPP. Results We find that the global fraction of GPPsp (19.8%) is larger than GPPgp (14.3%), indicating a globally asymmetric seasonal distribution of gross carbon uptake by terrestrial ecosystems. Globally, GPPmp plays a dominant role in shaping spatial patterns and increasing/decreasing trends in GPP, while GPPgp/GPPsp contributes to increasing GPP at the regional scale. Higher fractions of GPPgp/GPPmp (lower of GPPsp), as well as the co-occurrence of increasing GPP and non-tree vegetation cover in major croplands, are likely to be caused by agricultural intensification. Global changes in GPPgp and GPPsp are closely related to changes in their seasonal distributions (R = .86/.8, respectively), whereas this relationship is weaker for GPPmp (R = .53). Finally, high correlations are observed between changes in GPPgp and GPPsp and changes in their durations (R = .78/.78, respectively), while GPPmp shows a relatively lower correlation with its duration (R = .67). Main conclusions The asymmetric spatio-temporal patterns in the seasonal dynamics of global terrestrial gross carbon uptake found here have been substantially reshaped by anthropogenic land-use/cover changes and changes in photosynthetic phenology. Compared to calendar-based meteorological seasons more suitable for temperate/subpolar ecosystems, our phenology-based approach is expected to provide an alternative starting point for a better understanding of global spatio-temporal changes in the seasonal dynamics of terrestrial ecosystem processes and functioning under accelerating global change.

The Tibetan Plateau is the highest and largest plateau in the world, hosting unique alpine grassland and having a much higher snow cover than any other region at the same latitude, thus representing a "climate change hot-spot". Land surface phenology characterizes the timing of vegetation seasonality at the per-pixel level using remote sensing systems. The impact of seasonal snow cover variations on land surface phenology has drawn much attention; however, there is still no consensus on how the remote sensing estimated start of season (SOS) is biased by the presence of preseason snow cover. Here, we analyzed SOS assessments from time series of satellite derived vegetation indices and solar-induced chlorophyll fluorescence (SIF) during 2003-2016 for the Tibetan Plateau. We evaluated satellite-based SOS with field observations and gross primary production (GPP) from eddy covariance for both snow-free and snow covered sites. SOS derived from SIF was highly correlated with field data (R2 = 0.83) and also the normalized difference phenology index (NDPI) performed well for both snow free (R2 = 0.77) and snow covered sites (R2 = 0.73). On the contrary, normalized difference vegetation index (NDVI) correlates only weakly with field data (R2 = 0.35 for snow free and R2 = 0.15 for snow covered sites). We further found that an earlier end of the snow season caused an earlier estimate of SOS for the Tibetan Plateau from NDVI as compared to NDPI. Our research therefore adds new evidence to the ongoing debate supporting the view that the claimed advance in land surface SOS over the Tibetan Plateau is an artifact from snow cover changes. These findings improve our understanding of the impact of snow on land surface phenology in alpine ecosystems, which can further improve remote sensing based land surface phenology assessments in snow-influenced ecosystems.

Global changes are profoundly affecting the global terrestrial ecosystems, especially for the vegetation. Simultaneously, the affected vegetation gives feedback to the climates. The Tibetan Plateau (TP), one of the most sensitive areas to global changes, has undergone extraordinary changes on its ecosystem processes. In the multitudinous land surface ecosystem processes affecting the climate, the process of land surface energy balance affecting by vegetation activity is one of the most important and still has not been well recognized. The spatial and temporal patterns of the broadband emissivity (BBE) on the TP and its relations to the vegetation activity and land surface temperature were examined in this research. We find that elevated BBE is regulated by increasing vegetation activity for grasslands over the TP from 2000 to 2015. The spatial patterns of BBE and its interannual changes are highly correlated with vegetation activity. The BBE changing rate generally declines along rising elevation, due to the shrunk effects from vegetation activity. A greater sensitivity of BBE to vegetation activity occurs in the sparse vegetation area or high elevation zone than in the dense vegetation area or low elevation zone. Increasing BBE has a cooling effect on the land surface, especially at night. This cooling effect is related to wind speed. The growing season BBE trend as regulated by vegetation activity highlights the importance to take mounting notice of the growing season long-wave energy fluxes of surface energy balance studies in the future.

Climate change critically affects the carbon cycle of terrestrial ecosystems and thereafter the climate-carbon feedback. Previous studies have addressed the changes in temporal and spatial distribution of precipitation, termed as altered precipitation regimes. However, the response of terrestrial carbon cycle to the changes remain unclear. In this study, we used three process-based models to investigate the effect of changes in precipitation regimes on interannual variability (IAV) of gross primary productivity (GPP), net ecosystem productivity (NEP) and ecosystem respiration (RE). Annual sum of precipitation (PRE) and its monthly distribution, including its uniformity (PCI, precipitation concentration index) and timing (PCT, precipitation center time index), were used to describe the changes in precipitation regimes. Our results showed that the anomalies of both GPP and NEP in most biomes can be dominantly explained by anomalies of PRE. In addition, we also found that PCI and PCT were responsible for carbon dynamics in some specific biomes. For example, PCI dominated anomalies of both GPP and NEP in evergreen broadleaf forests (EBF), and a lower PCI (more uniform) is conducive to the increase of GPP and NEP. PCT is critical in driving the anomaly of GPP in mix forest (MF), grassland (GRA) and barren vegetation (Barren), and more precipitation distributed in early growing season benefits the increase of GPP. The sensibility of carbon cycle to changes in precipitation is biome specific, being likely due to the mismatch between water use ratio (WUR, defined as the ratio of transpiration to evapotranspiration) and precipitation distribution, implying that changes in distribution of precipitation among seasons/months may change the water availability. Our findings highlighted the necessity to pay more attention to the effects of altered precipitation distribution on changes in ecosystem multifunctionality related to carbon cycle.

Temperate and high-latitude forests are carbon sinks and play pivotal roles in offsetting greenhouse gas emissions of CO 2 . However, uncertainty still exists for subtropical forests, especially in monsoon-prevalent eastern Asia. Earlier studies have depended on remote sensing, ecosystem modeling, carbon fluxes, or single period forest surveys to estimate carbon sequestration capacities, and the results vary significantly. This study was designed to utilize multi-period forest survey data to explore spatial-dynamics of biomass storage in subtropical forests of China. Jiangxi province, a region with over 60% subtropical forest cover, was selected as the case study site and is located in central east China. Based on forest inventory data 1984–2013, and the stock-difference and biomass expansion factor methods, the carbon storage and density, of arboreal forests, economic forests, bamboo forests, woodlands and shrubberies were estimated. The results show that carbon storage increased from 159.1 Tg C in 1988 to 276.1 TgC in 2013, making up 3.1–3.8% of carbon stored throughout China. Among the four types of forests, the amount of carbon stored was as follows: arboreal forest > economic forest > bamboo forest > woodland and shrubbery. Arboreal forests accounted for 64.0–79.4% of the total. Forest carbon density increased from 21.2 Mg C ha −1 in 1984 to 26.2 Mg C ha −1 in 2013, equal to 61.2–70.2% of the average carbon density of China’s forests in the same period. Forest carbon storage in Jiangxi will reach 355.5 Tg C and 535.8 Tg C in 2020 and 2030, respectively, and forest carbon density is predicted to be 31.9 Mg C ha −1 and 46.4 Mg C ha −1 , respectively. As one of the few studies using multi-period data tracking biomass dynamics in Jiangxi province, the findings of this study may be used as a reference for other research. Using Jiangxi as a case study underlies the fact that subtropical forests in China have great carbon sequestration potential and have fundamental significance to offset global environmental change effects.

One-third of the global fossil fuel CO2 emissions is offset by carbon uptake of terrestrial ecosystems, while its strength is highly sensitive to drought events. It is predicted that frequencies of drought events would increase under a changing climate, which entails improving our understanding about their effects. Here, we combined direct observations at plot (experiment sites) and landscape (eddy-covariance, EC) scales with remote sensing observations at a regional scale, and evaluated the linkages between ecological resistance (summer drought loss, SDL) and resilience (post-drought regrowth, PDR). The study was conducted for an alpine grassland ecosystem on the Tibetan Plateau, which is highly vulnerable to climate changes. The results showed that alpine grasslands possess low resistance to drought. A summer drought of 2015 reduced net primary productivity (NPP) or net ecosystem productivity (NEP) by 25.4 g C m (- 2), 48.6 g C m (- 2) and 14.2 Tg C at the plot, landscape and regional scale relative to the baseline, respectively. In another dry summer of 2017, NEP was 11.0 g C m (- 2) and 7.5 g C m (- 2) lower than the baseline at the landscape and plot scale, respectively. To be noted, NEP or NPP completely recovered and exceeded the baseline due to rewetting induced PDR, compensating for the prior SDL to a certain extent. In 2015, the SDL of NEP or NPP was compensated by 39.0%, 17.3% and 10.6% due to the PDR effects at the plot, landscape and regional scale, respectively. The PDR of NEP in 2017 offset 23.6% and 70.7% of the prior SDL at the landscape and plot scale, respectively. Overall, these results demonstrated that weakened ecosystem function due to drought (e.g., low resistance) does not preclude rapid ecosystem recovery and regrowth (e.g., high resilience), which compensates for the prior loss to a certain degree.

Snow cover occupies large percentage of land surface in Tibetan Plateau. Snow cover duration (SCD) during non-growing seasons plays a critical role in regulating alpine vegetation’s phenology by affecting the energy budgets of land surface and soil moisture conditions. Different period’s snow cover during non-growing season may have distinct effect on the vegetation’s phenology. Start of season (SOS) has been observed advanced under the ongoing climate change in the plateau, but it still remains unclear how the SCD alters the SOS. This study attempts to answer the following questions: (i) What is the pattern of spatial and temporal variations for SCD and grassland SOS? (ii) Which period’s SCD plays a critical role in grassland’s SOS? The remote sensing datasets from the Moderate Resolution Imaging Spectroradiometer (MODIS) were utilized to compute the SOS and SCD on the Tibetan Plateau over 2003–15. The Asymmetric Gaussian function was applied to extract SOS. We also explored the spatial pattern and temporal variation of SOS and SCD. Then, by using linear correlation coefficients, we investigated the driving effects of different period’s non-growing season SCD on SOS. The non-growing season SCD slightly decreased during 2003–15, while SOS exhibited an overall advancing trend. Advanced trends in SOS were observed in the eastern plateau, and the delayed trends were mainly founded in western plateau. Snow cover area exhibited two separate peaks during autumn and late winter over the plateau. Extended SCD regions mainly distributed in middle-east of the plateau, while shrunken SCD distributed in other regions of the plateau. SCD of different seasons caused distinct effects on vegetation SOS. Lengthened autumn SCD advanced SOS over the eastern plateau. The slightly lengthened SCD postponed SOS over the western plateau. In the wet meadow regions, advanced SOS was positively associated with SCD during the entire non-growing season, whereas for the dry steppe, SCD over the preseason played a more dominant role. The SCD of previous autumn and winter also showed lag effect on SOS over meadow regions to a certain extent. This study confirmed the importance of SCD to phenological processes at the beginning of growing season and further suggested that role of SCD should be discriminated for different periods and for different heat-water conditions. With the lag effects and SCD’s distinct effect of different seasons considered, predictions on the Tibetan Plateau’s spring phenology could be improved.

青藏高原气候严酷,陆地表层生态系统脆弱,其高寒植物群落特征及生态系统生产力对气候变化的响应极其敏感。利用开顶箱（OTCs,Open Top Chambers）式装置在藏北高山嵩草（Kobresia pygmaea）草甸设置不同增温梯度实验（W1、W2、W3、W4）,探究增温对高寒草甸植物群落特征及地上生产力的影响。研究结果表明:1)与对照样地相比,增温减少了植物群落总盖度（2015年,W1、W2、W3、W4分别显著减少了28%、23%、59%、60%; 2016年,W4显著减少了83%）和高山嵩草盖度（2015年,W1、W2、W3、W4分别显著减少了26%、33%、681%、64%; 2016年,W4显著减少了85%）,而低幅度增温（W1、W2）对委陵菜属植物盖度无显著影响,高幅度增温（W3、W4）显著减少了委陵菜属植物盖度（2015年,W3、W4分别显著减少了58%和60%;2016年,W4显著减少了71%）; 2)对整个植物群落而言,增温幅度较低时,增温对群落的生长和生物量的积累有促进作用,当温度升高超过一定程度,这种促进作用会逐渐减弱甚至变成抑制作用（2015年,W4显著减少了地上生物量69%; 2016年,W4显著减少了地上生物量82%）; 3)高山嵩草盖度和其他物种总盖度存在显著的年际差异,而委陵菜属植物盖度无明显的年际变化。研究结果预示着,一定程度的升温会促进高寒草甸植物群落的生长,但温度升高超过一定幅度时,会导致草地生产力下降,草地退化加剧,同时当地群落中委陵菜属植物在全球变化背景下相对稳定,这类物种在未来气候变暖的背景下可能具有更强的竞争力。