In the context of unbalanced regional development posing severe challenges to the coordinated advancement of the "dual carbon" goals， it is of critical scientific significance to explore the spatio-temporal evolution characteristics and driving factors of carbon emission and economic growth decoupling at the provincial level in China. Based on panel data from 30 provincial regions from 2003 to 2022， this study systematically analyzes the spatio-temporal evolution characteristics and influencing factors by integrating the Tapio decoupling model， exploratory spatio-temporal data analysis （ESTDA）， and LMDI decomposition model. The results show： ① The carbon decoupling status exhibited a significant regional gradient pattern of "eastern leading， central and western regions catching up." ② The carbon decoupling effect in provinces showed significant spatial dependence， with 83.3% of provinces exhibiting path dependence characteristics， and a prominent spatial club convergence effect forming between neighboring provinces. ③ Economic growth was the most essential positive driving factor for the increase in carbon emissions， with a cumulative contribution of 531 843 tons from the east， central， and western regions. The reduction of energy intensity was the primary negative inhibiting factor， with a cumulative contribution of -320 516 tons from the east， central， and western regions. Carbon emission intensity had turned into a significant negative suppression in the eastern region， while showing fluctuating positive driving in the central and western regions. The suppressive effect of industrial structure optimization primarily appeared in the later stage in the eastern region， and its impact on carbon emissions， along with population size， showed significant regional heterogeneity. ④ The efforts towards decoupling showed phase evolution and regional gradient differentiation； improvements in energy intensity were the only continuous positive factor promoting decoupling. This research provides important scientific evidence for formulating precise and coordinated regional emission reduction strategies.
Excessive exploitation of land resources has driven global land-use changes that diminish ecological spaces， posing significant challenges to biodiversity conservation. To synergize ecological protection with high-quality development in the Yellow River Basin while balancing biodiversity conservation and socioeconomic needs in Shaanxi Province， this study designed three scenarios-Baseline （natural evolution）， Beautiful Shaanxi （ecological priority）， and Harmonious Shaanxi （coordinated development）-based on intrinsic linkages among land-use types， disturbance intensity， and biodiversity. Using the GeoSoS-FLUS and FLUS-Biodiversity models， we simulated land-use patterns and spatial heterogeneity of Mean Species Abundance （MSA） for 2035 and 2050. XGBoost-SHAP modeling further deciphered MSA drivers. Key findings reveal that： ① Structural shifts in most land-use types occurred， with moderately disturbed forests expanding significantly across all scenarios while dryland areas decreased most markedly. ② Under natural trends， Shaanxi's MSA rose from 0.533 （2020） to 0.539 （2050）， whereas the Beautiful and Harmonious scenarios elevated MSA to 0.546 and 0.544 by 2050， confirming their positive biodiversity effects. ③ MSA exhibited pronounced regional disparities-lower in densely populated economic zones and higher in natural resource-rich areas. ④ Driver contributions to MSA exhibited significant heterogeneity： Population density was the dominant negative factor， while annual potential evapotranspiration and temperature acted as primary positive drivers. This study demonstrates that scientifically coordinating land-use changes with biodiversity conservation， implementing targeted protection policies， and enhancing governance efficacy constitute effective pathways to achieve biodiversity targets.
Under the "dual carbon" strategy， this study systematically analyzes the spatiotemporal patterns， regional disparities， and dynamic evolution of the synergy between carbon reduction and pollution control in the Chengdu-Chongqing Economic Circle. Based on panel data of 144 districts and counties from 2000 to 2022， empirical research was conducted using methods such as the coupling coordination model， standard deviational ellipse （SDE）， spatial autocorrelation analysis， Dagum Gini coefficient decomposition， and both traditional and spatial Markov chain models. The results show that： ① The synergistic effect demonstrated a phased evolution of "rapid improvement-high-level fluctuation-differentiation adjustment." The coupling coordination degree rose from 0.256 to a peak of 0.323 before falling back to 0.262. The spatial autocorrelation was significant （Moran's I > 0.556）， with high-value areas concentrated in the Chengdu Plain and the main urban area of Chongqing. ② The average Dagum Gini coefficient was 0.214， and the contribution of intra-regional disparities exceeded 60%， with the imbalance being higher on the Chongqing side than on the Sichuan side. ③ Dynamic evolution analysis revealed that the distribution of synergistic effects had evolved from a single peak to a double peak. The synergy exhibited significant "state stability" （the minimum probability of maintaining the original state was 86.1%） and a "state locking" effect influenced by spatial neighborhoods. This resulted in limited upward transition possibilities and made "leapfrog" jumps difficult， indicating a gradual and stable evolutionary process. The synergy of carbon reduction and pollution control in the Chengdu-Chongqing Economic Circle exhibited significant spatiotemporal differentiation and polarization trends and demonstrated a stable， progressive evolutionary characteristic， providing a theoretical and practical basis for the region's green transition.
Nanhua County in central Yunnan is the largest and most important wild mushroom production area in Southwest China. To precisely quantify the origins of soil heavy metals and evaluate their associated health risks， thereby establishing a robust theoretical foundation for the prevention of soil heavy metal contamination and the safeguarding of human health， 571 surface soil samples were systematically collected from Nanhua County. To elucidate the origins of soil heavy metals， we employed a combination of multivariate statistical analysis and the UNMIX model， which facilitated the identification and quantification of distinct heavy metal sources within the soil matrix. The pollution levels of soil heavy metals in the study area and their potential health risks to humans were evaluated using the geoaccumulation index method and a health risk model based on Monte Carlo simulation. The findings reveal that： ① The levels of As and Cr in the surface soil of the investigated region markedly exceeded the regional background values for Yunnan Province. The spatial pattern of the eight heavy metals was largely influenced by the geological background， yet it exhibited considerable deviations attributable to external influences. ② The primary sources of soil heavy metals in the study area were high geological background sources closely associated with gold-arsenic deposits （17.08%）， atmospheric deposition sources related to coal mining and combustion （17.87%）， mixed sources of transportation and agriculture （12.78%）， industrial and mining activity sources closely linked to lead-zinc deposits （12.66%）， and natural parent material sources （39.61%）. ③ The geoaccumulation index （Igeo） evaluation revealed that the pollution levels of Cd， Cr， Cu， Ni， Pb， and Zn were relatively low， while As and Hg exhibited varying degrees of contamination. ④ The non-carcinogenic health risks of soil heavy metals to the local population were negligible. However， there were notable carcinogenic health risks， particularly for children， with As being the dominant carcinogenic factor. The heavy metal element As and ingestion frequency （IngR） were identified as the main factors influencing human health risks in the study area. High geological background sources closely related to gold and arsenic deposits and natural parent material sources were significant contributors to human health risks and should be prioritized for control， with enhanced monitoring and regulation.
To reveal the evolutionary patterns of regional ecological quality under human-nature interactions， we analyzed Henan Province （from 2001 to 2023） using MODIS data， Google Earth Engine （GEE）， and a vegetation ecological quality index （EQI）. Integrated with Theil-Sen trends， Mann-Kendall tests， spatial autocorrelation， and optimal parameter Geodetector （OPGD）， the results showed that： ① EQI increased significantly （0.21%·a-1， P < 0.05； mean： 36.67 to 41.34）， with spatial divergence （30.99% western improvement vs. 13.51% eastern degradation） reducing global autocorrelation （Moran's I：0.84 to 0.74）. ② Drivers transitioned from natural dominance （temperature∩elevation， q=0.30， 2001） to human-nature coupling （nighttime light ∩ elevation， q=0.409， 2012）， and finally land use-human synergy （land type ∩ nighttime light， q=0.5045， 2023）. ③ Policy effectiveness was evident in stable periods （CV < 33%， 2013 to 2023）， while eastern high-variability areas （CV > 35%） indicated compounding anthropogenic and natural pressures. This study validates a "natural→human→integrated" driver progression and proposes actionable zoning strategies， with OPGD-based methods offering broader applicability.
The Zhuozhang River， known as the "mother river" of the Changzhi region， holds significant importance for water conservation and industrial/agricultural safety due to its sediment heavy metal pollution status and source investigation. This study analyzed 57 sediment samples from the northern， southern， western， and main stream sections of the Zhuozhang River to examine the occurrence and spatial distribution of six metals： As， Pb， Cr， Cu， Hg， and Cd. Metal pollution severity was assessed using the point source index （PI） and pollution load index （PLI） methods， while potential ecological risk was evaluated through the Nemerow integrated pollution toxicity （PN-T） and potential ecological risk index （RI） approaches. Principal component analysis （PCA） and correlation analysis were employed to trace metal sources. The results revealed that Pb， Hg， Cr， and Cd were diffuse pollutants， with 80.7%-100% of sampling sites exceeding Shanxi Province's soil element background values. Cu and As were point source pollutants， exceeding standards in 35.1% and 15.8% of samples， respectively. Total metal concentrations showed a gradient trend： southern source>main stream>western source>northern source， with variation rates ranking as southern source （34.8%）>western source （22.6%）>northern source （19.2%）>main stream （15.6%）. Mass fraction concentrations followed a gradient of Cr>Pb>Cu>As>Hg>Cd， exhibiting spatial heterogeneity： The northern source had higher proportions of Cr， Pb， and Cd； the western source showed elevated Cu and Hg； and the southern source demonstrated higher As levels. The watershed was classified as mildly polluted （1.0<PI≤3.0， 1<PLI≤2） with low （1.0<PN-T≤2.0， RI<150） and moderate （2.0<PN-T≤3.0， 150≤RI<300） risk levels， primarily attributed to Cd （40≤EI<160） and Hg （80≤EI<320） contamination. The risk gradient showed a west-to-east progression： wcstern source>northern source>main stream>southern source. Contributions from transportation sources （Pb， Cu， Cr， As， and Cd）， agricultural sources （As， Cu， and Cd）， and industrial sources （Hg， Cd， As， and Cu） accounted for 40.4%， 25.1%， and 16.1%， respectively.
As a crucial ecological barrier in China， the fragile ecosystems in Northwest China exhibit high sensitivity to land use changes. Therefore， investigating the relationship between land use/cover change （LUCC） and ecosystem service value （ESV） is of great significance for regional ecological protection and sustainable development. Using Northwest China as the study area， this study systematically analyzed the spatiotemporal evolution patterns of land use based on land use data from 2008 to 2023 using spatial statistical methods. The ESV was assessed using a modified equivalent factor method and geodetector to reveal its spatiotemporal differentiation characteristics and driving factors. A boosted regression tree （BRT） model was employed to quantitatively identify the nonlinear impact mechanisms of proportional changes in different land use types on the ESV and determine their critical threshold effects. The results indicated the following： ① During the study period， the overall land use structure in Northwest China remained relatively stable. Construction land showed the most significant increase， expanding by 9 941.85 km2， and was primarily converted from cultivated land and grassland. ② ESV exhibited an initial decline followed by a rising trend over time. Spatially， it displayed a pattern of "higher values in the southeast and lower values in the northwest，" closely correlated with land use types， with different ESV grades showing clustered distribution. ③ The spatial heterogeneity of ESV was jointly driven by natural and socio-economic factors. Land use intensity contributed the most among the drivers. All factor interactions exhibited enhancing effects， with the interaction between land use intensity and other factors being particularly prominent in explanatory power. ④ Forest land， grassland， and unused land were identified as the key land use types influencing ESV in the mountain system. Unused land， water bodies， and cultivated land were the key factors affecting the ESV in the oasis and desert systems. ⑤ The proportion of key land use types exhibited a significant threshold effect on ecosystem service value （ESV）. To maintain ESV at a medium or higher level， the land use proportions in each system must meet the following conditions： For the mountain system， forest land should account for 21.56%-98.99%， grassland for 61.07%-98.99%， and unused land for 3.03%-11.39%； for the oasis system， water area should account for 3.75%-98.99%， unused land for 1.01%-28.01%， and cultivated land for 3.03%-14.22%； and for the desert system， water area should account for 0.97%-98.99%， unused land for 1.01%-74.39%， and cultivated land for 0.00%-4.03%.
Shanxi Province is a typical resource-based region and a major energy hub in China. Facing critical challenges of high carbon emissions and air pollution， this province urgently needs to advance synergistic control of pollution-carbon reduction. Utilizing the low emissions analysis platform （LEAP） with energy consumption as the system boundary and 2020 as the base year， this study predicts energy consumption， greenhouse gas （GHG）， and air pollutant emissions from 2021 to 2060 under three scenarios： baseline， low-carbon， and carbon neutrality. The predicted CO2 emissions for 2021-2023 were cross-validated against the actual CO2 emissions from energy consumption to verify the accuracy of the model. Subsequently， based on these predictions， it employed decoupling analysis， the synergy elasticity coefficient of pollution-carbon reduction， and the LMDI model to quantitatively assess synergistic effects and driving mechanisms. The results indicate that the baseline scenario failed to achieve the carbon peak target， with air pollutant emissions projected to increase until approximately 2045. In contrast， the low-carbon and carbon neutrality scenarios would achieve carbon peaking in 2029 （534.78 ×106 t） and 2025 （492.45×106 t）， respectively. Under the carbon neutrality scenario， GHG emissions were projected to drop below 200 ×106 t by 2060， and atmospheric pollutant emissions would begin a sustained decline around 2030. GHG and air pollutant emissions exhibited correlated emission patterns and synergistic mitigation potential； compared to the baseline scenario， both optimized scenarios advanced synergistic control by 15-20 years， characterized by GHG-dominant synergy. Analysis of driving factors identified economic development as the primary driver （contributing 904.98%）， while energy structure optimization and energy intensity reduction were key constraints. Control measures aligned with dual-carbon objectives effectively facilitated synergistic governance. Industrial transformation， stricter environmental policies and controls， and optimization of the energy consumption structure will enable Shanxi Province to achieve carbon reduction and air quality improvement targets sooner.
Microplastics， as an emerging pollutant， pose significant potential risks to environmental media due to their persistent nature. In this study， 6，008 soil samples were selected from 71 articles published between 2018 and 2024. Data on microplastic abundance， morphological features， and polymer types were collected to investigate the characteristics of microplastic pollution across different land use types in China， the correlations between their properties， and the variations in ecological risks. The results revealed that the order of microplastic abundance was agricultural soils （average abundance： 3 856.834×103 items·kg-1） > unused lands （1 002.34×103 items·kg-1） > construction lands （461.21×103 items·kg-1）. The most prevalent microplastic shapes in Chinese soils were fragments， fibers， and films， with sizes predominantly concentrated in small particles （0-0.5 mm）. Transparent， black， and white were the most frequent colors observed， while polypropylene （PP） and polyethylene （PE） emerged as the dominant polymer types. Moderate correlations were found between microplastic abundance and color/shape in agricultural soils， whereas strong correlations existed between abundance and polymer types in construction land soils. Soil microplastic pollution in China arises from the combined effects of multiple factors， including economic development， the tertiary sector， agricultural practices， and human activities. Studies have shown that agricultural lands exhibit the highest ecological risk， while unused lands present relatively lower risks.
Exploring the changes of coastal carbon storage and its driving factors is of great significance for assessing regional carbon sink potential and formulating ecological sink enhancement strategies. Taking the coastal zone of the Yellow River Delta as the research area， based on the annual land use/cover （LULC） data of the coastal zone of the Yellow River Delta from 2000 to 2023， the linear regression， Theil-Sen trend analysis， Mann-Kendall test， coefficient of variation， Hurst index， land use transfer matrix， and geographical detector model were used to systematically reveal the spatial and temporal dynamic characteristics of carbon storage in the coastal zone of the Yellow River Delta， the driving effect of LULC transformation， and the driving mechanism of spatial and temporal differentiation. The results show that： ① From 2000 to 2023， the carbon storage in the coastal zone of the Yellow River Delta decreased from inland to coastal areas， with a total net loss of approximately 1.54 Tg （calculated by C）， and showed a phased evolution of high-level fluctuation-stepwise decline-low-value platform. ②The overall stability of carbon storage was high， 80.69 % of the region remained stable， but 2.66 % of the region had mutated， and the carbon loss area accounted for 14.06 %. In the future， 16.26 % of the region will continue to face the risk of carbon loss， and only 5.25 % of the region will be expected to recover. ③ Carbon loss was mainly driven by human activities， and reclamation， urbanization， and the expansion of aquaculture ponds and salt pans were the core driving paths. Land use intensity and NDVI were the two core driving factors， and their explanatory power changed with time. The interaction of all factors showed two-factor and non-linear enhancement. The results of single factor detection and interactive detection showed that human activities had a significant leading role in the spatial and temporal differentiation of carbon storage and showed an increasing trend.
Understanding the spatiotemporal evolution and driving mechanisms of anthropogenic carbon emissions in urban agglomerations is critical for formulating regional carbon reduction policies. Based on the carbon emission factor method and multi-source remote sensing data， this study systematically analyzed the spatiotemporal evolution characteristics and driving mechanisms of anthropogenic carbon emissions in the Beijing-Tianjin-Hebei Urban Agglomeration from 2000 to 2020. The results showed that： ① The total anthropogenic carbon emissions （in terms of C） in the Beijing-Tianjin-Hebei Urban Agglomeration exhibited a fluctuating upward trend， increasing from 105.90 Tg in 2000 to 358.58 Tg in 2020. Hebei Province was the main contributor， showing significant growth. Beijing municipality's anthropogenic carbon emissions peaked in 2010 and then exhibited a downward trend， while Tianjin municipality showed a fluctuating trend of initial increase， subsequent decrease， and renewed increase. ② Regarding anthropogenic emission structure， energy consumption was the dominant contributor， while emissions from industrial processes and waste showed a clear increasing trend， and agricultural emissions remained relatively stable. ③ Spatially， anthropogenic carbon emissions displayed a clear "core-periphery" structure. Megacities and industrial-transportation corridors were the main concentration areas， and spatial polarization trends intensified over time. ④ Analysis of driving mechanisms revealed that economic development， urbanization， and anthropogenic carbon emission intensity were the primary influencing factors across all three areas. However， population scale and industrial structure showed regional differences： Tianjin municipality and Hebei Province were notably influenced by population growth and industrialization， while Beijing municipality experienced more significant industrial transformation. Energy intensity showed substantial emission reduction effects in Tianjin municipality and Hebei Province， but Beijing municipality's emission reduction potential was relatively lower. Additionally， the emission-reducing effect of technological innovation had not yet been effectively realized， contributing little overall to reducing anthropogenic carbon emissions. The extended STIRPAT-ridge regression model constructed in this study demonstrated strong stability and explanatory power for analyzing the driving mechanisms of regional anthropogenic carbon emissions， thus providing scientific support for precise emission reduction and differentiated carbon management strategies in the Beijing-Tianjin-Hebei Urban Agglomeration.
Located in the transitional zone of three major natural regions in China， Gansu Province serves as a critical area for investigating vegetation dynamics in response to climate change and human activities， providing a scientific basis for regional vegetation conservation and ecological engineering policies. Based on normalized difference vegetation index （NDVI） data and influencing factors datasets from 2001 to 2021， this study employed Theil-Sen Median trend analysis， partial correlation analysis， spatial autocorrelation analysis， and residual analysis to systematically analyze the spatiotemporal variations of vegetation NDVI and their driving mechanisms in Gansu Province. The results indicate that： ① The mean NDVI value during the study period was 0.352， exhibiting a distinct "high in the southeast， low in the northwest" spatial pattern. Both the entire province and individual geomorphic units showed significant upward trends， with 74.25% of the area experiencing vegetation improvement， particularly pronounced in the Loess Plateau of Longzhong and the southern mountainous regions. ② Climatic factor analysis revealed that temperature and precipitation exerted positive effects on vegetation NDVI in over 66% of the study area， with precipitation demonstrating a stronger influence than temperature. ③ Human activities predominantly enhanced vegetation NDVI， with stable grassland and cropland areas contributing the most （cumulatively 60%）. The change rates of nighttime light （NTL） and NDVI exhibited a weak positive spatial correlation， displaying "low-low" clustering in the Gannan Plateau， Hexi Corridor， Qilian Mountains， and Beishan Mountains， while "low-high" clustering characterized the southern mountainous area of Gansu and the Loess Plateau in central Gansu. ④ Contribution analysis highlighted that human activities （70.61%） had a greater overall impact on vegetation changes than climate change （29.39%）， though spatial heterogeneity was evident-the Gannan Plateau was the only geomorphic unit where climate change dominated. These findings elucidate the differential driving mechanisms of vegetation dynamics in transitional zones and offer critical insights for optimizing regional ecological management strategies.
Objectively assessing the spatiotemporal evolution and future trends of the ecological environmental quality （EEQ） of the urban agglomeration around Poyang Lake is crucial for achieving a balance between economic development and ecological preservation in the region. Utilizing the Google Earth Engine （GEE） cloud platform， this study filtered MODIS remote sensing images from 2000 to 2020 to extract ecological indicators. The remote sensing ecological index （RSEI） model was constructed via principal component analysis （PCA） to characterize EEQ. The spatiotemporal changes in EEQ within the urban agglomeration around Poyang Lake were analyzed using Theil-Sen slope estimation， coefficient of variation （CV）， and Mann-Kendall tests. Furthermore， the Hurst exponent was introduced to analyze future trends in EEQ. Finally， the CA-Markov model was applied to simulate and predict the EEQ for the year 2030. The results indicate that： The EEQ in the urban agglomeration around Poyang Lake exhibited a fluctuating upward trend from 2000 to 2014， followed by a slight decline thereafter. Over the 21-year period， the overall EEQ remained at a moderate to high level， showing an overall improvement trend. Areas classified as "excellent" were primarily concentrated in the northeastern and western regions， while the central and southern areas were predominantly characterized by "poor" and "fair" grades. The dominant trend of EEQ change was non-significant improvement， covering the largest areal proportion. The centers of gravity for all EEQ grades were located within Nanchang City and its surrounding areas. Over the past five years， these centers exhibited a minor shift predominantly towards the southeast. EEQ demonstrated overall stability， characterized mainly by low and relatively low fluctuations. Future changes are projected to consist primarily of persistent improvement and anti-persistent improvement. By 2030， the EEQ of the urban agglomeration around Poyang Lake is predicted to show some improvement； however， areas classified as "Poor" and "Fair" are expected to remain concentrated in the central， southwestern， and southeastern parts of the study area. This study provides a methodological foundation for rapidly and accurately evaluating regional EEQ， investigating its long-term dynamics， and forecasting its future trajectory. It is essential for reconciling regional development with the natural environment.
Accurately estimating the carbon emissions from land use at the county scale is an important prerequisite for rationally formulating carbon reduction measures. Limited by the acquisition of energy statistics， previous studies on land use carbon emissions at the county scale mainly estimated energy consumption carbon emissions using night light data. The integration of multi-source remote sensing data to estimate land use carbon emissions remains to be further explored. Therefore， based on multi-source remote sensing data such as energy statistics， nighttime light data， and XCO2 data， this study adopted methods such as the energy consumption carbon emission model， standard deviation ellipse analysis， and exploratory spatiotemporal data analysis to depict the spatiotemporal characteristics of carbon emissions from county land use in China from 2010 to 2020. Furthermore， the geographic detector model was adopted to explore the influencing factors of carbon emissions from land use in counties of China. The results show that： ① From 2010 to 2020， the carbon emissions from land use in China's counties decreased， with significant regional differences. The carbon emissions in the eastern region were the highest， followed by the central and northeastern regions， and the carbon emissions in the western region were relatively low. ② The carbon emissions from land use in China's counties showed a trend of dispersion along the "northeast-southwest" direction and aggregation along the "northwest-southeast" direction， with obvious directionality and the center of gravity shifting towards the northeast. Its spatial distribution was significantly positively correlated， with obvious spatial heterogeneity and agglomeration， and the spatial correlation was relatively stable. ③ The carbon emissions from land use in counties of China were mainly affected by urbanization and economic development， and the intensity of the interaction between the two was higher than that of a single influencing factor. The results of this study can provide a reference for the formulation of carbon emission reduction targets in China.
The sources and composition of dissolved organic matter （DOM） in water bodies exhibit high spatiotemporal heterogeneity. Taking the Quanmin Reservoir watershed in Sichuan Province as the study area， water samples were collected from four inflow rivers and nearby pollution sources such as farmland drainage and sewage treatment plant effluents during the wet and dry seasons. Three-dimensional excitation-emission matrix spectra （3D-EEMs） combined with parallel factor analysis （PARAFAC） was used to analyze the composition， distribution characteristics， and sources of DOM in the study area. The correlation between water quality parameters and DOM parameters was further used to clarify the sources of DOM. The results showed that： ① The average concentrations of TP， NH4+-N， NO3--N， TN， and DOC in the Quanmin Reservoir watershed were all lower in the wet season than in the dry season， and TN was the main pollutant in the watershed. ② The PARAFAC analysis identified three components in the DOM of the Quanmin Reservoir watershed， including terrestrial humic-like C1 （Ex/Em = 360/425 nm）， visible light region fulvic-like C2 （Ex/Em = 400/480 nm）， and microbial metabolic products C3 （Ex/Em = 335/417 nm）. ③ The total fluorescence intensity of DOM was higher in the wet season than in the dry season. The relative abundance of terrestrial humic-like C1 was the largest in both periods， accounting for 37.40% and 37.52%， respectively. The average maximum fluorescence intensity of DOM in the two periods was in the order of Xiongjiagou River > Guxian River > Xixi River > Motan River. ④ Based on fluorescence spectral parameters and multi-parameter correlation analysis， the three components of DOM in the Quanmin Reservoir watershed demonstrated homogeneity， and DOM was mainly of terrestrial origin， showing weak humification characteristics. Water quality parameters in different periods had different degrees of correlation with DOM components and fluorescence parameters. The research results can provide a basis for the source tracing of nitrogen and phosphorus pollutants and water environment management in the Quanmin Reservoir watershed.
As one of the core economic circles in China， the Beijing-Tianjin-Hebei Region is confronted with the challenge of a sharp increase in the demand for ecosystem services during the rapid urbanization process. Urban expansion， population concentration， and industrial upgrading have led to a continuous increase in the pressure on ecological resources such as water resources， carbon sink capacity， soil retention， and biological habitats. The imbalance in regional ecological management and the contradiction between sustainable development have become increasingly prominent. Scientifically assessing and accurately predicting the dynamics of supply and demand for ecosystem services has become a key breakthrough in solving the problems of regional ecological security and sustainable development. The SD-PLUS coupling model was used to predict the land use demand in Hebei Province in the future. The supply and demand evolution laws of water conservation （WY）， carbon storage （CS）， soil conservation （SC）， and habitat quality （HQ） in the historical and future periods of Hebei Province were evaluated based on methods such as the InVEST model and spatial assignment method. The spatio-temporal patterns of supply and demand matching of four types of ecosystem services in Hebei Province from 2000 to 2070 were revealed by using the supply-demand ratio of ecosystem services， and suggestions for optimizing the balance of supply and demand of ecosystem services in Hebei Province were put forward. The results showed that there were significant differences in land use demand in Hebei Province under different SSP-RCP scenarios. However， the demand for cultivated land and water areas decreased， while the demand for construction land increased significantly. Over the past two decades， both the supply and demand of water conservation in Hebei Province have declined， while the supply and demand of soil conservation have shown an increasing trend. The supply of carbon storage has decreased， while the demand has increased. The supply and demand of habitat quality have remained relatively stable. In the future， the changes in the supply and demand of the four ecosystem services under the three different scenarios will basically continue the trends of the past two decades. Between 2000 and 2020， Hebei Province exhibited spatio-temporal heterogeneity in the matching of ecosystem service provision and requirement， accompanied by a significant supply-demand mismatch. The imbalance of supply and demand matching will intensify in different scenarios in the future.
Land use change stands as a pivotal element influencing the carbon storage capacity of terrestrial ecosystems. Conducting a thorough investigation into the inherent influencing mechanism through which land use change impacts the carbon storage of terrestrial ecosystems holds paramount importance for the rational planning of land use patterns and the effective realization of carbon peak and carbon neutrality objectives. Leveraging land use data of the Yihe River Basin spanning the period from 1980 to 2020， this study employed the PLUS model to simulate land use change across four scenarios， namely natural development， cultivated protection， urban development， and ecological protection. Subsequently， the InVEST model was applied to evaluate carbon storage under various scenarios from 1980 to 2020 and for the year 2030. Additionally， the Geodetector was adopted to identify and analyze the driving forces behind the spatial variability of carbon storage. The results show that： ① From 1980 to 2020， the Yihe River Basin witnessed the largest reduction in cultivated land area and the greatest increase in construction land area. The predominant forms of land transformation primarily occurred between cultivated land and grassland， as well as between cultivated land and construction land. Under different scenarios in 2030， the acreage of cultivated land is projected to persist in its declining trend， whereas the area allocated to construction land is anticipated to maintain a continuous expansion. ② Over the period spanning from 1980 to 2020， the carbon storage within the Yihe River Basin exhibited a general downward trend. Carbon storage in 2030 will decline to different degrees under different scenarios， with the smallest decline under the cultivated land protection scenario and the largest decline under the urban development scenario. In general， the spatial distribution of carbon storage aligned closely with the spatial distribution of land use types. ③ The conversion of different land use types had a significant impact on carbon storage changes， among which the transformation of cultivated land into construction land had the most obvious effect on carbon storage. ④ The dominant factor influencing the spatial differentiation of carbon storage in the Yihe River Basin from 1980 to 2020 was land use type， followed by elevation and NDVI. The driving factors mainly exhibited nonlinear enhancement and dual-factor enhancement. The research results can provide scientific references for optimizing the allocation of land resources in the Yihe River Basin and promoting the region's sustainable development.
Coupling ecosystem service value （ESV） with landscape ecological risk （LER） for ecological zoning in Gansu Province is of significant importance for accurately identifying ecological and environmental problems arising from rapid urbanization， enhancing regional ecological security， and achieving sustainable development goals. The PLUS model was used to simulate the land use pattern of Gansu Province in 2030 under the scenarios of natural development， ecological protection， economic development， and construction planning. We adopted the modified value equivalent method to evaluate ESV， combined landscape disturbance and vulnerability models to calculate LER， and identified ecological zones in Gansu Province through empirical Bayesian modified spatial autocorrelation analysis. The results indicate that： ① Ecological protection and construction planning scenarios increased ESV by 1.28% and 1.54%， respectively. Notably， the increase in water resource supply function under the construction planning scenario was the highest， reaching 3.19%. ② LER presented a spatial differentiation feature of "low in the northwest and high in the southeast，" with high-risk areas concentrated on the Loess Plateau in eastern Gansu and the edge of oases in Hexi. The degree of landscape fragmentation was significantly positively correlated with human activity intensity. ③ The spatial clustering of ESV and LER in Gansu Province was mainly based on low value low risk in the northwest and high value high risk in the southeast. Research has found that integrating ecological protection planning and development strategies can effectively coordinate the contradiction between ecological service supply and risk carrying in Gansu Province， providing effective support for its ecological spatial planning.
Under the "Dual-Carbon" strategy， assessing the spatial distribution of carbon sinks in typical regions represented by terrestrial ecosystems is essential for accurately characterizing regional carbon sink functions. Net ecosystem exchange （NEE） serves as a key indicator of ecosystem carbon fluxes. To accurately simulate the spatial patterns of regional NEE， commonly used methods include the inventory method， process-based models， and atmospheric inversion techniques. In recent years， machine learning has emerged as a powerful data-driven approach for spatial NEE modeling， offering distinct advantages in capturing complex nonlinear relationships. This study reviews the main methods applied in NEE spatial distribution modeling， elaborates on the principles and current applications of machine learning-based approaches， analyzes key feature selection and algorithm choices across different ecosystem types， and discusses the potential of applying machine learning to investigate NEE spatial characteristics in complex environments such as urban ecosystems.
Investigating and clarifying the mechanisms by which climate change influences ecological security patterns is an effective measure for maintaining ecological security. Based on this， the study region selected is the Tianshan North Slope Economic Belt， utilizing three typical climate scenarios from the sixth coupled model intercomparison project （CMIP6）： SSP1-1.9， SSP2-4.5， and SSP5-8.5. First， the patch-generated land use change simulation model （PLUS model） was used to simulate land use evolution under different climate scenarios by 2060. Based on this， the integrated assessment model for ecosystem services and trade-offs （InVEST model） was applied to assess four key ecosystem services： water production， soil conservation， habitat quality， and carbon storage. Additionally， morphological spatial pattern analysis （MSPA） and the patch importance index （DPC） were employed to identify ecological source. Next， the analytic hierarchy process （AHP） was used to determine the weights of seven indicators， including land use， topography， habitat quality， and climate factors， to construct an ecological resistance surface. Finally， ecological corridors were extracted based on circuit theory. The results indicate： ① Ecological source areas were spatially distributed with more in the south and fewer in the north， denser in the south and sparser in the north， concentrated in the forested and grassland areas north of the Tianshan Mountains. In 2020， the area was 76 807 square kilometers， and in the SSP245 scenario by 2060， it slightly decreased to 75 979 square kilometers， showing the highest stability. Under the SSP585 scenario， the source area decreased to 66 904 square kilometers， with the highest degree of fragmentation. Under the SSP119 scenario， it expanded to 80 031 square kilometers with the best connectivity. ② Under the SSP245 scenario， the total length of ecological corridors reached its maximum， increasing to 4 509 kilometers. Under the SSP585 scenario， degradation was severe， with the total length of corridors decreasing by 21.7%， while the number of corridors increased by 10， resulting in a fragmented ecological network characterized by "increased quantity but decreased quality." Under the SSP119 scenario， the total length of corridors was 3 592 kilometers， a slight increase of 5.4%， with the number of corridors decreasing to 81. The distribution structure of ecological corridors was optimized， with the most significant improvement in connectivity， highlighting the effectiveness of ecological conservation efforts. ③ Ecological pinch points increased in all three scenarios， with the most dramatic increase occurring in the SSP585 scenario， in which the number of hotspots increased by 235%， and the area increased by 15.91 square kilometers. The area of ecological barriers decreased in all three scenarios but remained large in the SSP585 scenario. Through SSP scenario comparisons， it was found that under the low-carbon scenario， source areas and corridor networks were expanded and strengthened， indicating that low emissions are beneficial for ecological stability； however， under the high-emission scenario， source area sizes decreased， ecological resistance increased， corridors became fragmented， and the vulnerability of the ecological security pattern rose.
