Long-term monitoring of vegetation dynamics is essential for assessing ecosystems resilience and responses to climate change. The China Ecosystem Research Network (CERN) is a national long-term observation network that provides an ecological baseline across China’s major ecosystems. This study analyzed vegetation trends and their driving factors from 2000 to 2024 across 34 CERN field stations and their surrounding areas. An intercomparison of multiple NDVI products revealed substantial inconsistencies. MODIS NDVI exhibited superior temporal stability and spatial coherence and was therefore selected for long-term analysis. The mean NDVI at CERN stations (0.419) was 37.8% higher than the national average (0.304), and showed significantly faster greening trends (0.022/10 a) compared to suburban (0.013/10 a) and rural (0.017/10 a) areas, reflecting effective vegetation restoration and stable ecosystem management. Vegetation changes at CERN field stations were predominantly governed by climatic rather than anthropogenic factors. Among different ecosystem types, air temperature (AT), sunshine duration (SD), and relative humility (RH) were the dominant drivers in farmland, forest, and wetland. For grassland, AT and SD were the main drivers, whereas AT and RH exerted the strongest influence in the desert ecosystem. These findings confirm the representativeness of CERN stations and underscore the predominant role of climatic factors in shaping long-term vegetation trajectories across China’s diverse ecosystems.
Rapid urbanization threatens eco-environment quality (EEQ), in which the urban landscape pattern (ULP) plays a key role. However, the multi-scale mechanisms behind their spatial nonstationarity and nonlinear interactions remain underexplored. This study constructed an EEQ index, incorporating urban impervious surface and air pollution, based on urban ecosystem elements. Using a novel framework that integrates MGWR with a random forest model interpreted by SHAP, we explored the spatially nonstationary and nonlinear association between ULP and the EEQ index, and further quantified the contributions and thresholds of ULP both globally and locally. The results indicate that: (1) EEQ index in China’s coastal areas (CNCA) averaged 0.64, indicating a generally favorable level, with lower values in urban agglomeration distribution; (2) Urban built-up area percentage (UAP), urban built-up height (UBH), Shape, Contagion, and patch Euclidean Nearest-Neighbor Mean Distance index were the top contributors; (3) UAP and UBH have bidirectional effects on EEQ index in CNCA, reflecting characteristics of the Environmental Kuznets Curve. However, patterns were inconsistent for urban agglomerations within CNCA, indicating significant spatial discrepancies and nonlinear features, and growing inequality among urban agglomerations has heightened the complexity of their thresholds. Overall, prioritizing urban landscapes with order, strong cohesion, and effective connectivity in landscape planning can guide future urban expansion.
Despite the reported “warming-wetting” trend, Central Asia faces severe water insecurity due to climate shifts and anthropogenic activities. This study integrates multi-source remote sensing data (GRACE, TRMM, MODIS) with machine learning to analyze drought dynamics from 2003 to 2022 using the Water Storage Deficit Index (WSDI). Results reveal significant declines in terrestrial water storage (TWS) particularly in the Tianshan Mountains (-10.20 mm/yr) and the Central Desert (-6.19 mm/yr). Drought severity has intensified since 2014, with 67% of subregions transitioning to moderate drought. Random Forest modeling indicates that drought is no longer solely climate-driven but is increasingly dominated by anthropogenic factors (GDP, urbanization, cropland), which explain over 85% of the variability. Furthermore, the WSDI outperformed traditional indices by identifying 12 major drought events linked to deep aquifer depletion—a “hidden” structural deficit often overlooked by surface-based metrics. These findings challenge the optimistic “warming-wetting” narrative, highlighting the urgent need for storage-based management strategies to address anthropogenic groundwater depletion.
As crucial water quality indices, the magnitudes, variation trends, and influencing factors of riverine dissolved oxygen (DO) for different geographic features of rivers (e.g., stream order and location characteristics) are regulated by interacting natural drivers in time and space; however, these interactions remain largely unclear. Based on data obtained from 3293 monitoring sites across China during 2020-2022, we constructed random forest models and structural equation models to explore the drivers of riverine DO associated with meteorological and geographical factors. The results revealed higher DO in large rivers (8.612 mg L−1) and lake/reservoir outlets (9.216 mg L−1) and lower DO in headwater streams (8.397 mg L−1) and estuaries (8.278 mg L−1). Temperature was identified as a key DO driver, followed by latitude and shortwave solar radiation for most stream orders and location characteristics. Diverse and complex drivers of DO for different geographic features of rivers emphasize that China’s water quality would further benefit from more flexible strategies for maintaining water security and ecosystem health.
Glaciers in the Mount Qomolangma region are sensitive indicators of climate change. Despite extensive studies on modern glacier-climate interactions, quantitative reconstructions of Holocene glacier dynamics and climate conditions remain limited. This study comparatively reconstructed the Holocene extents for the Khumbu and Rongbuk glaciers on the southern and northern slopes of Mount Qomolangma, respectively. Based on well-preserved glacial landforms and with chronological data, this paper, using PalaeoIce model, quantitively reconstructed the Holocene glacier extents and applied the P-T model that links equilibrium line altitude (ELA) variations to climatic parameters to infer past summer temperature variations, with precipitation being constrained by lake pollen records. The results show that the Rongbuk Glacier experienced larger ELA variations and greater summer temperature changes than the Khumbu Glacier. The disparities could be attributed to an interplay of multiple factors, encompassing glacier geometry, valley morphology, and distinct precipitation patterns across the northern and southern slopes. The ELA-based temperature estimates are broadly consistent with pollen-based reconstructions, highlighting the potential of this method in inferring climatic conditions at high-altitude regions.
The Dongting Lake Basin, a key ecological barrier in the Yangtze River Economic Belt, is under increasing pressure from climate change and human activities. We quantified the ecological environment quality (EEQ) from 2001 to 2021 using the Remote Sensing Ecological Index (RSEI) in the Google Earth Engine (GEE). We analyzed the spatiotemporal variations of EEQ and major natural and anthropogenic factors and applied the Geodetector model to estimate their individual contributions and interactive effects. The results showed that the mean RSEI increased by 0.0045/yr, indicating overall improvement in EEQ across the basin. Restoration occurred over 86.38% of the basin, whereas degradation was limited to 9.84%. Anthropogenic-driven vegetation improvement (Normalized Difference Vegetation Index, NDVI), temperature variations, and land-use/land-cover change (LUCC) were identified as the primary drivers of this phenomenon. Interactive effects between NDVI and LUCC and NDVI and temperature were stronger than single-factor effects. Moderate increases in temperature associated with climate change can enhance EEQ, whereas excessive warming can threaten ecosystem stability. Ecological restoration projects improved EEQ by increasing the vegetation cover and altering LUCC, but rapid urbanization and unsuitable afforestation in fragile karst regions led to localized degradation. These findings highlight the management trade-offs and support sustainable, site-specific strategies to enhance basin resilience under continued warming.
Ecological management zoning is essential for balancing conservation and development in ecologically fragile regions. However, research is still needed to improve the effectiveness of ecological management zoning. This study develops an integrated framework that couples landscape ecological risk (LER) and ecological resilience (ER) to delineate functional zones on the Loess Plateau (LP), China. LER was quantified using the ecosystem service supply-demand ratio and landscape disturbance metrics, while ER was assessed through resistance, adaptability, and recovery indices. Spatial dynamics of LER and ER from 2010 to 2022 were analyzed, and five management zones—protection, control, prevention, restoration, and adaptation zones—were identified. Results indicate that (1) LER increased continuously, expanding mainly in the central eastern LP due to intensified land use and fragmentation; (2) ER improved overall, shifting from a pattern of clustered high values to a more dispersed distribution, suggesting enhanced system resilience but greater spatial heterogeneity; (3) the proportions of ecological control and restoration zones increased, whereas protection, prevention, and adaptation zones all showed slight contractions, indicating a general trend of decoupling between LER and ER; and (4) geographical detector analysis showed that precipitation, temperature, population density, and per capita cropland area are the main determinants of spatial differentiation of LER and ER. The proposed LER-ER coupling framework in this study refines ecological management zoning and provides practical guidance for adaptive land use governance in semi-arid landscapes.
Controlling key drivers is essential to align the supply and demand of ecosystem services (ES) for sustainable development. However, spatial heterogeneity and nonlinearity are often overlooked in studying ES mismatch. Hence, this paper evaluates the supply-demand balance of bundled ecosystem services in eastern China, revealing the spatial characteristics of the supply-demand relationship of ecosystem services. We used the InVEST model to assess ecosystem services, applied a Gaussian mixture model to identify distinct ES supply-demand bundles, and constructed a synergy network among bundles via weighted network analysis. We then employed redundancy analysis and generalized additive models to examine drivers. The main findings of this study are as follows: Ecosystem service demand increased by approximately 18.25%, showing a clear trend toward supply-demand mismatch. The results showed strong synergies among different supply-demand clusters, and the supply-demand relationships of the same ecosystem service were not entirely consistent across clusters. The NDVI emerged as the most influential factor affecting ES supply-demand relationships (exceeding 10%), but social factors had a greater average effect. This study emphasizes the importance of the spatial relationship and driving factors of the supply-demand states of services mismatch in formulating effective spatial strategies.
The periodic impacts of land reclamation and Spartina alterniflora invasion on habitat suitability under shifting management policies remain unquantified. Under business- as-usual (BAU) and ecological protection (EP) scenarios, we assessed how coastal reclamation and S. alterniflora distribution affected shorebird habitat suitability in the Yellow Sea across three periods (2000-2010, 2010-2020, and 2020-BAU). Habitat suitability declined during 2000-2010 and 2010-2020 under the BAU scenario, but increased under the EP scenario. The mean habitat suitability declined by 0.26 (0.26 ± 0.17) due to reclamation during 2000-2010, which was nearly double the decline observed during 2010-2020 (0.15 ± 0.09). Meanwhile, the area of decreased habitat suitability caused by S. alterniflora invasion between 2010 and 2020 was 1.2 times greater than that between 2000 and 2010. Between 2010 and 2020, habitat degradation was more severe inside protected areas (PAs) than outside (T= -3.21, df=76, p<0.01). The increase in invaded habitat loss inside PAs from 2000-2010 to 2010-2020 was twice that outside. Meanwhile, the proportion of habitat loss from reclamation inside PAs declined in 2010-2020 compared to 2000-2010. These results suggest that S. alterniflora invasion mitigates the PAs’ positive effect.
Disentangling the driving mechanisms of urban impervious surface (UIS) spatial changes is critical for developing effective urban growth planning. However, previous studies often overlooked the indirect effects of multiple factors especially inter-urban spatial interaction and policy factors on urban expansion in urban clusters. Here, we used structural equation modeling and geographically weighted regression to quantify the spatiotemporal patterns of direct and indirect effects of socioeconomic factors, geographical environment, inter-urban spatial interaction, and policy factors on urban expansion in the Beijing-Tianjin-Hebei (BTH) region from 1990 to 2020. The findings showed that urban population, inter-urban spatial interaction, tertiary industry, road density, and policies were the main drivers of UIS changes. Among them, inter-urban spatial interaction primarily had an indirect positive effect on urban spatial patterns by increasing urban population and optimizing industrial structures, with the strongest impact in southern BTH. Regional planning policies such as development zones promoted urban expansion by stimulating industrial development and attracting urban population, with their influence escalating from 0.41 in 1990 to 0.57 in 2010. These findings highlight the importance of strategically guiding inter-urban spatial interactions and optimizing industrial layouts to foster compact urban development and sustainable land use in the BTH region.
China’s population age structure (PAS) has already entered a modern phase, with the effects of aging and declining birth rates becoming increasingly severe. These trends pose significant challenges to demographic and socioeconomic stability. In this study, the scale nesting theory was used to analyze the population census data for 2000‒2020 from prefecture-level cities. Using a multiscale geographically weighted regression model, this study explores the spatial characteristics of China’s age structure and quantitatively examines the driving factors within this nested framework. The evolution of PAS in the prefecture-level cities in China exhibited distinct stages, while depicting a rapid transition toward an aging model. In the nested space, single-scale nesting was primarily synchronous, while double-scale nesting was characterized by simultaneous synchronization. The Northeast region exhibited the highest degree of synchronous development in the PAS within the nested space. The regression coefficients indicated that the population system and per capita gross domestic product (GDP) were the primary and secondary factors, respectively. Topographical variation exerted an influence on only the synchronous-advance type, while PM2.5 exhibited a significant association with the advance-lag type. This study provides scientific support for high-quality development across regions in the context of heterogeneous population age structures.
Terrain viewshed analysis, which is a core spatial analysis technology, holds an important position in GISs. Viewsheds are typically represented as field models of discrete sets of visible scattered points in which the local clustering and spatial structural characteristics of the points are neglected. As a branch of terrain viewsheds, communication viewsheds are used to analyze signal reception with signal towers as observation points. In this study, a structured representation method in which communication viewsheds are used as examples, is proposed. First, each terrain point is treated as a base station, and the communication viewsheds are obtained. Second, each communication viewshed is segmented into subregions using feature recognition, and seven parameters are defined to describe the characteristics of each subregion. Finally, consistency and application verification are used to evaluate the effectiveness of the proposed methods. The experimental results show that the method can effectively reveal the spatial distribution characteristics of communication signal coverage, thereby providing theoretical support for quickly obtaining solutions to base station site planning problems. Furthermore, the structured representation method is also a lossy data compression method. This method is compared with the existing visibility data compression method, and the results indicate that the compression ratio can reach hundreds and increases exponentially with increasing terrain complexity.
