Search Results (5,114)

Urban tourism has evolved into a complex adaptive system, where unregulated expansion disrupts the ecological balance and intensifies resource stress. Understanding the dynamic evolution and resilience mechanisms of the tourism ecological health network (TEHN) is essential for supporting sustainable urban tourism as a coupled human–natural system. Using Shanghai as a case study, we applied the "vigor–organization–resilience–services" (VORS) framework to evaluate ecosystem health, which served as a constraint for constructing the TEHN, using the minimum cumulative resistance (MCR) model for the period from 2001 to 2023. A resilience framework integrating structural and functional dimensions was further developed to assess spatiotemporal evolution and guide targeted enhancement strategies. The results indicated that as ecosystem health degraded, particularly in peripheral areas, the urban TEHN in Shanghai shifted from a dispersed to a centralized structure, with limited connectivity in the periphery. The resilience of the TEHN continued to grow, with structural resilience remaining at a high level, while functional resilience still required enhancement. Specifically, the low integration and limited choice between the tourism network and the transportation system hindered tourists from selecting routes with higher ecosystem health indices. Enhancing functional resilience, while sustaining structural resilience, is essential for transforming the TEHN into a multi-centered, multi-level system that promotes efficient connectivity, ecological sustainability, and long-term adaptability. The results contribute to a systems-level understanding of tourism–ecology interactions and support the development of adaptive strategies for balancing network efficiency and environmental integrity. Full article

Accurate prediction of key ocean elements (e.g., chlorophyll-a concentration, sea surface temperature, etc.) is imperative for maintaining marine ecological balance, responding to marine disaster pollution, and promoting the sustainable use of marine resources. Existing spatio-temporal prediction models primarily rely on either physical or data-driven approaches. Physical models are constrained by modeling complexity and parameterization errors, while data-driven models lack interpretability and depend on high-quality data. To address these challenges, this study proposes OKG-ConvGRU, a domain knowledge-guided remote sensing prediction framework for ocean elements. This framework integrates knowledge graphs with the ConvGRU network, leveraging prior knowledge from marine science to enhance the prediction performance of ocean elements in remotely sensed images. Firstly, we construct a spatio-temporal knowledge graph for ocean elements (OKG), followed by semantic embedding representation for its spatial and temporal dimensions. Subsequently, a cross-attention-based feature fusion module (CAFM) is designed to efficiently integrate spatio-temporal multimodal features. Finally, these fused features are incorporated into an enhanced ConvGRU network. For multi-step prediction, we adopt a Seq2Seq architecture combined with a multi-step rolling strategy. Prediction experiments for chlorophyll-a concentration in the eastern seas of China validate the effectiveness of the proposed framework. The results show that, compared to baseline models, OKG-ConvGRU exhibits significant advantages in prediction accuracy, long-term stability, data utilization efficiency, and robustness. This study provides a scientific foundation and technical support for the precise monitoring and sustainable development of marine ecological environments. Full article

To address the limitations of the existing cotton seedling growth prediction methods in field environments, specifically, poor representation of spatiotemporal features and low visual fidelity in texture rendering, this paper proposes an algorithm for the prediction of cotton seedling growth from images based on FCA-STNet. The model leverages historical sequences of cotton seedling RGB images to generate an image of the predicted growth at time t + 1 and extracts 37 phenotypic traits from the predicted image. A novel STNet structure is designed to enhance the representation of spatiotemporal dependencies, while an Adaptive Fine-Grained Channel Attention (FCA) module is integrated to capture both global and local feature information. This attention mechanism focuses on individual cotton plants and their textural characteristics, effectively reducing the interference from common field-related challenges such as insufficient lighting, leaf fluttering, and wind disturbances. The experimental results demonstrate that the predicted images achieved an MSE of 0.0086, MAE of 0.0321, SSIM of 0.8339, and PSNR of 20.7011 on the test set, representing improvements of 2.27%, 0.31%, 4.73%, and 11.20%, respectively, over the baseline STNet. The method outperforms several mainstream spatiotemporal prediction models. Furthermore, the majority of the predicted phenotypic traits exhibited correlations with actual measurements with coefficients above 0.8, indicating high prediction accuracy. The proposed FCA-STNet model enables visually realistic prediction of cotton seedling growth in open-field conditions, offering a new perspective for research in growth prediction. Full article

Understanding the interplay between lightning activity and hydrometeor distribution is crucial for advancing knowledge of thunderstorm electrification processes. Using three-dimensional lightning mapping and dual-polarization radar observations, this study investigates the spatiotemporal correlations between low-frequency (LF) lightning sources and hydrometeors during a severe thunderstorm on 11 June 2014, in North Carolina, USA. The results reveal that lightning sources are predominantly observed above 6 km (near the −10 °C isotherm) and stabilize into a dual-peak vertical distribution as the storm progresses into its mature stage, with peaks located at 6–7 km (−10 °C to −15 °C) and 10–11 km (approximately −40 °C). Low-density graupel (LDG) and aggregates (AGs) dominate at lightning locations. Stronger updrafts lead to higher proportions of LDG and high-density graupel (HDG), and lower proportions of AG. LDG exhibits the strongest positive correlation with LF lightning sources, with a peak correlation coefficient of 0.65 at 9 km. During the vigorous development stage, HDG and hail (Ha) also show positive correlations with LF lightning sources, with peak correlation coefficients of 0.52 at 7 km and 0.42 at 8 km, respectively. As the storm reaches its mature phase, the correlation between LDG and lightning sources also displays a dual-peak vertical distribution, with peaks at 7–8 km and 13–14 km. Both the peak correlation coefficient and its corresponding height increase with the strengthening of updrafts, underscoring the critical role of updrafts in microphysical characteristics and driving electrification processes. Full article

Digital technologies could facilitate green development by enhancing energy efficiency. However, existing research on coupling coordination between digital infrastructure and green development remains scarce. To fill this research gap, this study analyzes the spatio-temporal variations and barriers of coupling coordination. An evaluation index system is established and then the coupling relationship and the barrier factors between digital infrastructure and green development are analyzed. A provincial analysis is conducted by using data from China. The results in the study indicate (1) coupling coordination between digital infrastructure and green development exhibits a relatively low state, characterized by an overall upward trend; (2) noteworthy disparities are observed in the spatio-temporal pattern of the coupling coordination degree, reflecting the overall evolutionary trend from low to high coupling coordination, along with the characteristics of positive spatial correlation and high spatial concentration; and (3) obstacle factors are analyzed from the aspects of digital infrastructure and green development, emphasizing the construction of mobile phone base stations and investment in pollution control, among other aspects. This study contributes valuable insights for improvement paths for digital infrastructure and green development, offering recommendations for optimizing strategies to promote their coupled development. Full article

This study proposes an intelligent patrol planning framework that leverages reinforcement learning, spatiotemporal crime forecasting, and simulated sensor telemetry to optimize autonomous vehicle (AV) navigation in urban environments. Crime incidents from Washington DC (2024–2025) and Seattle (2008–2024) are modeled as a dynamic spatiotemporal graph, capturing the evolving intensity and distribution of criminal activity across neighborhoods and time windows. The agent’s state space incorporates synthetic AV sensor inputs—including fuel level, visual anomaly detection, and threat signals—to reflect real-world operational constraints. We evaluate and compare three learning strategies: Deep Q-Network (DQN), Double Deep Q-Network (DDQN), and Proximal Policy Optimization (PPO). Experimental results show that DDQN outperforms DQN in convergence speed and reward accumulation, while PPO demonstrates greater adaptability in sensor-rich, high-noise conditions. Real-map simulations and hourly risk heatmaps validate the effectiveness of our approach, highlighting its potential to inform scalable, data-driven patrol strategies in next-generation smart cities. Full article

Refining the land use structure can boost land utilization efficiency and curtail regional carbon emissions. Nevertheless, prior research has predominantly concentrated on static linear planning analysis. It has failed to account for how future dynamic alterations in driving factors (such as GDP and population) affect simulation outcomes and how the land use spatial configuration impacts the attainment of the carbon-neutrality goal. In this research, 1 km spatial resolution LULC products were employed to meticulously simulate multiple land use scenarios across China at the national level from 2030 to 2060. This was performed by taking into account the dynamic changes in driving factors. Subsequently, an analysis was carried out on the low-carbon land use spatial structure required to reach the carbon-neutrality target. The findings are as follows: (1) When employing the PLUS (Patch—based Land Use Simulation) model to conduct simulations of various land use scenarios in China by taking into account the dynamic alterations in driving factors, a high degree of precision was attained across diverse scenarios. The sustainable development scenario demonstrated the best performance, with kappa, OA, and FoM values of 0.9101, 93.15%, and 0.3895, respectively. This implies that the simulation approach based on dynamic factors is highly suitable for national-scale applications. (2) The simulation accuracy of the PLUS and GeoSOS-FLUS (Systems for Geographical Modeling and Optimization, Simulation of Future Land Utilization) models was validated for six scenarios by extrapolating the trends of influencing factors. Moreover, a set of scenarios was added to each model as a control group without extrapolation. The present research demonstrated that projecting the trends of factors having an impact notably improved the simulation precision of both the PLUS and GeoSOS-FLUS models. When contrasted with the GeoSOS-FLUS model, the PLUS model attained superior simulation accuracy across all six scenarios. The highest precision indicators were observed in the sustainable development scenario, with kappa, OA, and FoM values reaching 0.9101, 93.15%, and 0.3895, respectively. The precise simulation method of the PLUS model, which considers the dynamic changes in influencing factors, is highly applicable at the national scale. (3) Under the sustainable development scenario, it is anticipated that China’s land use carbon emissions will reach their peak in 2030 and achieve the carbon-neutrality target by 2060. Net carbon emissions are expected to decline by 14.36% compared to the 2020 levels. From the perspective of dynamic changes in influencing factors, the PLUS model was used to accurately simulate China’s future land use. Based on these simulations, multi-scenario predictions of future carbon emissions were made, and the results uncover the spatiotemporal evolution characteristics of China’s carbon emissions. This study aims to offer a solid scientific basis for policy-making related to China’s low-carbon economy and high-quality development. It also intends to present Chinese solutions and key paths for achieving carbon peak and carbon neutrality. Full article

Background: Non-communicable diseases (NCDs) are a leading cause of mortality globally, contributing significantly to the burden on healthcare systems. Understanding the spatiotemporal patterns of NCD mortality is crucial for identifying vulnerable populations and regions at high risk. Objectives: Here, we evaluated the spatiotemporal patterns of NCD mortality in the Metropolitan Area of the Valley of Mexico (MAVM) from 2000 to 2019 for five International Classification of Diseases chapters (4, 5, 6, 9, and 10) at two spatial scales: the municipal level and metropolitan region. Methods: Mortality rates were calculated for the total population and stratified by sex and age groups at both spatial scales. In addition, the relative risk (RR) of mortality was estimated to identify vulnerable population groups and regions with a high risk of mortality, using women and the 25–34 age group as reference categories for population-level analysis, and the overall MAVM mortality rate as the reference for municipal-level analysis. Results: Mortality trends showed that circulatory-system diseases (Chapter 9) are emerging as a concerning health issue, with 45 municipalities showing increasing mortality trends, especially among older adults. Respiratory-system diseases (Chapter 10), mental and behavioral disorders (Chapter 5) and nervous-system diseases (Chapter 6) predominantly did not exhibit a consistent general mortality trend. However, upon disaggregating by sex and age groups, specific negative or positive trends emerged at the municipal level for some of these chapters or subgroups. Endocrine, nutritional, and metabolic diseases (Chapter 4) showed a complex pattern, with some age groups presenting increasing mortality trends, and 52 municipalities showing increasing trends overall. The RR showed men and older age groups (≥35 years) exhibiting higher mortality risks. The temporal trend of RR allowed us to identify spatial mortality hotspots mainly in chapters related to circulatory, endocrine, and respiratory diseases, forming four geographical clusters in Mexico City that show persistent high risk of mortality. Conclusions: The spatiotemporal analysis highlights municipalities and vulnerable populations with a consistently elevated mortality risk. These findings emphasize the need for monitoring NCD mortality patterns at both the municipal and metropolitan levels to address disparities and guide the implementation of health policies aimed at reducing mortality risk in vulnerable populations. Full article

Understanding the carbon emissions of multimodal travel—comprising walking, metro, bus, cycling, and ride-hailing—is essential for promoting sustainable urban mobility. However, most existing studies focus on single-mode travel, while underlying spatiotemporal and behavioral determinants remain insufficiently explored due to the lack of fine-grained data and interpretable analytical frameworks. This study proposes a novel integration of high-frequency, real-world mobility trajectory data with interpretable machine learning to systematically identify the key drivers of carbon emissions at the individual trip level. Firstly, multimodal travel chains are reconstructed using continuous GPS trajectory data collected in Beijing. Secondly, a model based on Calculate Emissions from Road Transport (COPERT) is developed to quantify trip-level CO₂ emissions. Thirdly, four interpretable machine learning models based on gradient boosting—XGBoost, GBDT, LightGBM, and CatBoost—are trained using transportation and built environment features to model the relationship between CO₂ emissions and a set of explanatory variables; finally, Shapley Additive exPlanations (SHAP) and partial dependence plots (PDPs) are used to interpret the model outputs, revealing key determinants and their non-linear interaction effects. The results show that transportation-related features account for 75.1% of the explained variance in emissions, with bus usage being the most influential single factor (contributing 22.6%). Built environment features explain the remaining 24.9%. The PDP analysis reveals that substantial emission reductions occur only when the shares of bus, metro, and cycling surpass threshold levels of approximately 40%, 40%, and 30%, respectively. Additionally, travel carbon emissions are minimized when trip origins and destinations are located within a 10 to 11 km radius of the central business district (CBD). This study advances the field by establishing a scalable, interpretable, and behaviorally grounded framework to assess carbon emissions from multimodal travel, providing actionable insights for low-carbon transport planning and policy design. Full article

Petrochemical complexes are spatially expansive and host diverse emission sources, making accurate monitoring of volatile organic compounds (VOCs) challenging using conventional two-dimensional methods. This study introduces Mobile-extraction Differential Optical Absorption Spectroscopy (Me-DOAS), a real-time, three-dimensional remote sensing technique for assessing benzene emissions in the Ulsan petrochemical complex, South Korea. A vehicle-mounted Me-DOAS system conducted monthly measurements throughout 2024, capturing data during four daily intervals to evaluate diurnal variation. Routes included perimeter loops and grid-based transects within core industrial zones. The highest benzene concentrations were observed in February (mean: 64.28 ± 194.69 µg/m³; geometric mean: 5.13 µg/m³), with exceedances of the national annual standard (5 µg/m³) in several months. Notably, nighttime and early morning sessions showed elevated levels, suggesting contributions from nocturnal operations and meteorological conditions such as atmospheric inversion. A total of 179 exceedances (≥30 µg/m³) were identified, predominantly in zones with benzene-handling activities. Correlation analysis revealed a significant relationship between high concentrations and specific emission sources. These results demonstrate the utility of Me-DOAS in capturing spatiotemporal emission dynamics and support its application in exposure risk assessment and industrial emission control. The findings provide a robust framework for targeted management strategies and call for integration with source apportionment and dispersion modeling tools. Full article

Gully erosion refers to the landform formed by soil and water loss through gully development, which is a critical manifestation of soil degradation. However, research on the spatio-temporal variations in erosion gullies at the county scale remains insufficient, particularly regarding changes in gully aggregation and their driving factors. This study utilized high-resolution remote sensing imagery, gully interpretation information, topographic data, meteorological records, vegetation coverage, soil texture, and land use datasets to analyze the spatio-temporal patterns and influencing factors of erosion gully evolution in Bin County, Heilongjiang Province of China, from 2012 to 2022. Kernel density evaluation (KDE) analysis was also employed to explore these dynamics. The results indicate that the gully number in Bin County has significantly increased over the past decade. Gully development involves not only headward erosion of gully heads but also lateral expansion of gully channels. Gully evolution is most pronounced in slope intervals. While gentle slopes and slope intervals host the highest density of gullies, the aspect does not significantly influence gully development. Vegetation coverage exhibits a clear threshold effect of 0.6 in inhibiting erosion gully formation. Additionally, cultivated areas contain the largest number of gullies and experience the most intense changes; gully aggregation in forested and grassland regions shows an upward trend; the central part of the black soil region has witnessed a marked decrease in gully aggregation; and meadow soil areas exhibit relatively stable spatio-temporal variations in gully distribution. These findings provide valuable data and decision-making support for soil erosion control and transformation efforts. Full article

Accurate assessment of green productivity is essential for advancing sustainable agriculture in ecologically fragile regions. This study examined the evolution of agricultural green total factor productivity (AGTFP) in Tibet over the period 2002–2021 by applying a super-efficiency SBM-GML model that accounts for undesirable outputs. We decompose AGTFP into technical change and efficiency change, conduct redundancy analysis to identify sources of inefficiency and explore its spatiotemporal dynamics through kernel density estimation and convergence analysis. Results show that (1) AGTFP in Tibet grew at an average annual rate of 0.78%, slower than the national average of 1.6%; (2) labor input, livestock scale, and agricultural carbon emissions are major sources of redundancy, especially in pastoral regions; (3) technological progress is the main driver of AGTFP growth, while efficiency gains have a limited impact, reflecting a technology-led growth pattern; (4) AGTFP follows a “convergence-divergence-reconvergence” trend, with signs of conditional β convergence after controlling for regional heterogeneity. These findings highlight the need for region-specific green agricultural policies. Priority should be given to improving green technology diffusion and input allocation in high-altitude pastoral areas, alongside strengthening ecological compensation and interregional coordination to enhance green efficiency and promote high-quality development across Tibet. Full article

The Xiong’an New Area, a newly established national-level zone in China, faces the threat of land subsidence and ground fissure due to groundwater overexploitation and geothermal extraction, threatening urban safety. This study integrates time-series InSAR and GNSS monitoring to analyze spatiotemporal deformation patterns from 2017/05 to 2025/03. The key results show: (1) Three subsidence hotspots, namely northern Xiongxian (max. cumulative subsidence: 591 mm; 70 mm/yr), Luzhuang, and Liulizhuang, strongly correlate with geothermal wells and F4/F5 fault zones; (2) GNSS baseline analysis (e.g., XA01-XA02) reveals fissure-induced differential deformation (max. horizontal/vertical rates: 40.04 mm/yr and 19.8 mm/yr); and (3) InSAR–GNSS cross-validation confirms the high consistency of the results (Pearson’s correlation coefficient = 0.86). Subsidence in Xiongxian is driven by geothermal/industrial groundwater use, without any seasonal variations, while Anxin exhibits agricultural pumping-linked seasonal fluctuations. The use of rooftop GNSS stations reduces multipath effects and improves urban monitoring accuracy. The spatiotemporal heterogeneity stems from coupled resource exploitation and tectonic activity. We propose prioritizing rooftop GNSS deployments to enhance east–west deformation monitoring. This framework balances regional and local-scale precision, offering a replicable solution for geological risk assessments in emerging cities. Full article

In the context of intelligent coal mine safety monitoring, an integrated prediction and early-warning method named MTGNN-Bayesian-IF-DBSCAN (Multi-Task Graph Neural Network–Bayesian Optimization–Isolation Forest–Density-Based Spatial Clustering of Applications with Noise) is proposed to address the challenges of gas concentration prediction and anomaly detection in coal mining faces. The MTGNN (Multi-Task Graph Neural Network) is first employed to model the spatiotemporal coupling characteristics of gas concentration and wind speed data. By constructing a graph structure based on sensor spatial dependencies and utilizing temporal convolutional layers to capture long short-term time-series features, the high-precision dynamic prediction of gas concentrations is achieved via the MTGNN. Experimental results indicate that the MTGNN outperforms comparative algorithms, such as CrossGNN and FourierGNN, in prediction accuracy, with the mean absolute error (MAE) being as low as 0.00237 and the root mean square error (RMSE) maintained below 0.0203 across different sensor locations (T0, T1, T2). For anomaly detection, a Bayesian optimization framework is introduced to adaptively optimize the fusion weights of IF (Isolation Forest) and DBSCAN (Density-Based Spatial Clustering of Applications with Noise). Through defining the objective function as the F1 score and employing Gaussian process surrogate models, the optimal weight combination (w_if = 0.43, w_dbscan = 0.52) is determined, achieving an F1 score of 1.0. By integrating original concentration data and residual features, gas anomalies are effectively identified by the proposed method, with the detection rate reaching a range of 93–96% and the false alarm rate controlled below 5%. Multidimensional analysis diagrams (e.g., residual distribution, 45° diagonal error plot, and boxplots) further validate the model’s robustness in different spatial locations, particularly in capturing abrupt changes and low-concentration anomalies. This study provides a new technical pathway for intelligent gas warning in coal mines, integrating spatiotemporal modeling, multi-algorithm fusion, and statistical optimization. The proposed framework not only enhances the accuracy and reliability of gas prediction and anomaly detection but also demonstrates potential for generalization to other industrial sensor networks. Full article

Urban–rural integration realises the coordinated development and prosperity of urban and rural areas as a whole by optimising the allocation of resources and the flow of factors, and its connotations have been extended from a single dimension to multiple dimensions such as people, land and industry. The Beijing–Tianjin–Hebei Region has a typical “Core–Periphery Structure”, and this paper took the 187 county units within the region as the research object, taking into account indicators of development and coordination to construct an evaluation index system of urban–rural integration of the Beijing–Tianjin–Hebei region counties in the dimensions of “people–land–industry”. Global principal component analysis was used to measure the evolutionary pattern of the urban–rural integration level between 2005 and 2020, and its spatiotemporal drivers were analysed by using the Geographical and Temporal Weighted Regression model (GTWR). The results of the study show that (1) the level of urban–rural integration in the Beijing–Tianjin–Hebei region showed an increasing trend during the 15-year study period, the high-value areas of urban–rural integration were mainly distributed in Beijing and the Bohai Rim region in the eastern part of the Tianjin–Hebei region, and the level of urban–rural integration of the peri-urban county units of the city was better than that of the remote counties and cities as a whole. (2) In terms of spatial agglomeration, all dimensions were characterised by significant spatial agglomeration. The degree of agglomeration was categorised as urban–rural comprehensive integration (U-RCI) > urban–rural industry integration (U-RII) > urban–rural land integration (U-RLI) > urban–rural people integration (U-RPI). (3) In terms of spatial and temporal driving factors for urban–rural integration, the driving role of U-RPI, U-RLI and U-RII for U-RCI has gradually weakened during the past 15 years, and urban–rural integration in the counties shifted from a single role to a more central coordinated and multidimensional driving role. Full article