Resilience research is a critical component in achieving Sustainable Development Goals, and its importance has become increasingly evident against the backdrop of complex and volatile external shocks. The existing literature predominantly focuses on the resilience of specific regions and industries, while market resilience, which connects the production and consumption sides, receives relatively insufficient attention. The “Wholesale and Retail Industries”, encompassing both corporate manufacturing and individual consumption, constitute a critical window for observing market resilience. Under the impact of the COVID-19 pandemic, individual offline visitation can effectively reflect market resilience in specific spaces. Therefore, this study develops an analytical framework based on Mobile Signaling Data to observe market resilience at a fine-grained scale through individual activity. Focusing on Guangzhou, we decode the spatiotemporal patterns of market resilience in the “Wholesale and Retail Industries” under the impact of the COVID-19 shock. Key findings include: (1) The “retail industry” outperforms the “wholesale industry” in both market resistance (− 0.37 vs. −0.41) and market recovery (66
People with disabilities (PWDs) may exhibit distinctive spatiotemporal patterns of bus travel because of their specific physical conditions and travel preferences, which may in turn lead to different associations with the built environment compared with people without disabilities. However, these differences have received limited scholarly attention. This study addresses this gap by comparing the bus travel behaviors of people with and without disabilities in Wuhan, China. The results show that PWDs have significantly higher bus travel frequencies than people without disabilities on both weekdays (3.58 vs. 2.86 trips) and weekends (3.32 vs. 2.63 trips). In terms of spatial distribution, bus trips made by PWDs are more concentrated in central urban areas, whereas those made by people without disabilities are more clustered around transport hubs and employment centers. Further analysis using XGBoost and SHAP reveals significant heterogeneity in the effects of the built environment on bus ridership between the two groups. For PWDs, the three most important factors influencing bus ridership on weekday are road density (14.71
This research looks into the pattern and mechanism of occupational structure change of the Pearl River Delta, China, from 2000 to 2020, testing the polarization and professionalization hypothesis of western cities in a developing country on a regional scale. It is found that the region had a polarizing trend and its cities followed a trajectory from industrialization, proletarianization to polarization. Centers of the region tended to polarize with faster growth in the high-skilled, while the sub-centers polarized with faster growth in the low-skilled occupations, and the peripheries displayed industrialization and proletarianization in early years and shifted to polarization in the recent decade. Industrial policies promoting advanced manufacturing and services enhanced growth of the high-skilled occupations. Large-scale domestic migration to the region fueled the growth of low-skilled occupations, but also acted as important sources of talents. Functional division and difference in development stage between the centers, sub-centers and peripheries led to the heterogeneity in occupational restructuring patterns in the region. This research complements to the debate on occupational structural changes by providing evidence from the “Global South”, proposing differences between central and peripheral cities and a trajectory of occupational change covering different stages of development, and further interpreting how localized factors shaped the occupational restructuring process.
Exploring the spatial variation of emotion has attracted increasing attention in recent years. However, few scholarly attempts have focused on the spatiotemporal process of emotional dynamics. This study aimed to depict continuous and fine-grained human emotions and investigate the patterns of emotional changes from a spatiotemporal perspective. Electrodermal activity (EDA), audio and GPS data were collected from 256 volunteers in the Fanta Oriental Heritage theme park. We developed an integrated framework that combines multimodal emotion recognition, emotional trajectory construction, and pattern mining to examine affective dynamics in a theme park environment. The emotional trajectories were constructed by matching emotional status and GPS. With the support of the hierarchical clustering, the similarities among the emotional trajectories were investigated. Compared with relying solely on single-modal data, our recognition method exhibited significant improvement by fusing EDA and audio, achieving a four-class classification accuracy of more than 0.68. Our results demonstrated that the trajectory model can represent the continuous spatiotemporal changes of minute-by-minute emotional statuses effectively. Further, we have discovered three patterns from emotional trajectories within the theme park. The “Peak-decrease” pattern is sensitive to the exciting experiences and includes most young individuals and females. All tourism projects can stimulate the emotion of individuals belong to the “High-low” pattern significantly. The “Stable-fluctuation” pattern showed the stable emotional dynamic and contained half of the individuals older than 40. This work provides new insights into the fine-grained human emotional changes to enrich the spatiotemporal theories of human emotions, and highlights the potential applications of our findings in urban planning and experience monitoring.
Artificial intelligence is reshaping urban energy systems at an unprecedented pace, yet its net contribution to carbon-constrained energy efficiency remains an unsettled empirical question. The mechanisms of impact, the role of digital infrastructure, and the spatial patterns of energy savings constitute a critical but underexamined research frontier. Drawing on panel data from 297 Chinese prefecture-level cities over 2011–2024, we employ a parallel multiple mediation model, a panel threshold model, and a spatial Durbin model within a unified framework. We present our main findings as follows. First, AI development exerts a positive effect on carbon-constrained energy efficiency. Raising AI_level by one standard deviation improves carbon-constrained energy efficiency by roughly 10.3
The health impacts of urban greenspace are increasingly acknowledged, yet their relationship with subjective wellbeing may be nonlinear and conditioned by individual capacity. This study examines how capability of physical activity moderates the non-linear association between residential greenspace and health-related quality of life among 331 clinically diagnosed COPD patients in Beijing. Using objective geospatial indicators—tree canopy, total vegetation coverage, and park proximity—and validated wellbeing scores (EQ-5D), we estimated nonlinear and moderated quadratic models. Results reveal a robust U-shaped association between tree canopy and wellbeing, with the association reversing in sign beyond moderate coverage levels. In contrast, vegetation coverage showed no consistent pattern, and park distance displayed a weaker nonlinear relationship. Crucially, physical activity capacity, measured via the 1-minute sit-to-stand and 6-minute walk tests, significantly moderated both the shape and direction of the canopy–wellbeing curve. Higher activity levels flattened or inverted the curve, suggesting that wellbeing gains depend on an individual’s ability to engage with their environment. These findings highlight the importance of integrating ecological design with patient mobility needs, offering implications for precision greening in dense urban settings. Urban health strategies for vulnerable populations may benefit from tailoring greenspace interventions to match physical capacity thresholds.
Urban planning and public health policies are increasingly recognizing urban running as a key activity that reflects and shapes the livability and health of cities. Urban running is not merely an individual behavior but a spatially embedded practice influenced by both built environment characteristics and social context. To inform effective spatial planning, it is essential to move beyond single-level explanations and account for the nested social-spatial processes shaping activity patterns. This study introduces a spatial multilevel regression approach that integrates multi-source data to examine urban running across multiple spatial scales. Using running trajectory big data from the Hangzhou central districts, I construct grid level indicators to capture local activity contexts and district level measures to reflect broader structural conditions. The multilevel model explicitly tests cross-scale interactions between built environment features and sociodemographic factors, revealing that urban running patterns emerge from the interplay of local environments and broader social structures. These findings provide actionable insights for planners and policymakers, highlighting areas where interventions can promote equitable, healthy, and sustainable urban environments. More broadly, the study offers a generalizable analytical framework for understanding health-related mobility in rapidly transforming cities, demonstrating the value of moving beyond single-level approaches in spatial analysis and policy design.
Conventional assessments of urban green space usually rely on planar indicators, which cannot fully capture changes in vegetation height, canopy volume, and layered structure during urbanization. To address this limitation, this study estimated three-dimensional green volume (3DGV) and compared its spatiotemporal evolution with two-dimensional green space (2DGS) in the main city of Nanjing from 1990 to 2020. Field-based reference samples, Landsat-derived vegetation indices, and machine-learning regression were integrated to reconstruct decadal-scale 3DGV, while 2DGS was extracted from high-resolution remote-sensing images using support vector machine classification. Spatial autocorrelation analysis and spatial regression models were then used to examine temporal changes, clustering patterns, and associations with socioeconomic, land-use, and natural-environment variables. The results show that both 3DGV and 2DGS experienced a decline–recovery trajectory and exhibited significant positive spatial autocorrelation, but their recovery magnitudes and local clustering patterns differed across phases. The two indicators shared several associated variables, yet differed in coefficient direction, statistical significance, and phase-specific patterns. Land-use variables were more closely associated with changes in 2DGS, whereas socioeconomic and natural-environment variables showed more frequent or indicator-specific associations with 3DGV in several phases. These findings indicate that 3DGV adds a structural dimension to planar green-space assessment by revealing spatially uneven changes in vegetation volume, canopy development, and layered planting structure. The combined 2DGS–3DGV framework provides an additional and complementary basis for diagnosing urban green-space change and supporting structure-oriented greening strategies in high-density cities.
This article asks what event-level spatial analysis adds after the concentration of urban violence is already known. Using 2,800 geolocated and timestamped armed-violence events recorded by Crossfire (Fogo Cruzado) app across 21 municipalities in metropolitan Rio de Janeiro between December 2024 and March 2026, we separate four decision-relevant objects, persistent place burden, short-window event co-occurrence, excess space–time dependence, and model-based self-excitation. The design combines hotspot persistence screening, a 5 km/72 h event graph, shuffled-time and location-permutation benchmarks, and Hawkes point-process models. Linked pairs exceed the shuffled-time null by 27.5 z=17.57 , empirical p=0.0033 ), but visually strong graph fragmentation is not unusual under the spatially grounded null and attenuates sharply as thresholds widen. Persistent hotspots and temporary event communities therefore represent different policy layers rather than interchangeable maps. The former support sustained place-based prioritization; the latter can support short-run coordination only when their threshold sensitivity and provisional status are displayed. Crossfire app records are produced through a verified multi-source civic-monitoring workflow rather than passive app-location traces, although geographically uneven ascertainment remains possible and low counts in peripheral municipalities must be treated as lower bounds. The study does not identify the causal effect of any policy implemented during the observation period. Its contribution is a transferable framework for matching spatial evidence, uncertainty, and policy time horizons without reifying temporary clusters as fixed territories.
The paper presents the results from a study, devoted to modelling internal migrations in Poland at the municipality level. Thus, data concerns some 2,500 units across 20 years from the first two decades of the 21st century. The basic model studied assumes the dependence of migration flows on the unemployment rate. Then, the results of this model (identified for each of 20 years, i.e. actually 20 models) are compared with several of its extensions. The main conclusion is that the basic hypothesis of migration dependence upon unemployment is strongly confirmed. What is, however, even more important, is that the analysis of model errors, both of the basic model and its extensions, provides essential and quite specific information on the spatial character of the phenomena and processes analysed. These can be directly used for further, more detailed studies, but also for development policy purposes. While the results do not categorically state that this relationship must be positive for all units considered, they provide strong evidence that this may be the case to a much greater extent than we previously thought. On the top of this, the very clear spatial pattern constitutes an important value added.
Provincial average scores are widely used to monitor educational performance, but they may obscure how inequality is distributed within and across administrative units. Using annual data for all 63 provinces in Vietnam’s national upper secondary graduation examination from 2017 to 2024, this paper develops a sequential spatial diagnostic framework for three compulsory subjects: Mathematics, Literature, and Foreign Language. The framework first evaluates the adequacy of provincial averages and then examines the spatial structure, persistence, and neighborhood-conditioned mobility of the between-province signal that remains visible at the administrative scale. Theil decomposition indicates that provincial mean scores capture only 6–18
Tourism expansion and urban development in the Yucatan Peninsula have intensified pressures on ecosystems, altering the natural contributions to people (NCP) that sustain regional wellbeing. To better understand emerging socioecological conflicts, we applied a multi-criteria analysis to integrate and spatially represent the perspectives of key institutional stakeholders regarding: (1) the prioritization of NCP groups per ecosystem, (2) the NCP they consider most important, (3) the main ecosystem threats, and (4) the ecosystems most affected by tourism and urban development. To achieve these objectives, the following ecosystems were assessed: Coastal lagoon, Cultivated grasslands, Popal, Savanna, Tropical Forest, Tular, Coastal dunes, Halophytic-hydrophytic vegetation, Peten, Mangrove, Palm Forest, Cenotes. Regulatory contributions—such as water and air quality regulation and protection against extreme weather—were consistently identified as the most valuable across coastal ecosystems, whereas cultivated grasslands were primarily valued for their material contributions, particularly food production. Perceptions of threats and ecosystem vulnerability varied across states, reflecting distinct tourism and urbanization trajectories. Incorporating stakeholder perspectives allowed us to identify major socioecological trade-offs and land-use transformation patterns by state. Overall, current development dynamics appear to prioritize short-term socioeconomic gains at the expense of ecological functions, particularly regulatory NCP, underscoring growing concerns about the long-term integrity of the peninsula’s socioecological systems.
Using six waves of the Chinese General Social Survey (2010–2021) and cross-classified random effects modeling, this study investigates regional and cohort variations in intergenerational mobility and transmission processes in China. Results show that regional patterns in the levels of intergenerational mobility and in the effects of determinants of status attainment are relatively stable across birth cohorts from the 1940s to the 1990s. First-tier cities and eastern regions consistently exhibit higher absolute mobility and greater influence of socioeconomic achieved factors, while the influence of socioeconomic ascribed factors and institutional factors such as party membership and hukou is weaker. Fixed-effects panel analyses demonstrate that regional economic development is associated with greater absolute mobility and weaker institutional effects, while coinciding with strengthened roles of socioeconomic ascribed and achieved factors. Educational inequality aligns with heavier reliance on institutional factors and correlates positively with relative and absolute mobility. Higher public education expenditure is linked to greater absolute mobility and diminished ascriptive influences. The findings highlight that China’s intergenerational mobility is shaped by intertwined economic, institutional, and policy dynamics, with enduring regional disparities despite overall progress towards a merit-based system.
Can promoting the transfer and transformation of scientific and technological achievements open up new pathways for air pollution control? Based on data from Chinese cities and listed companies, this study employs a difference-in-differences (DID) model to examine the impacts and underlying mechanisms of National Demonstration Zones for the Transfer and Transformation of Scientific and Technological Achievements (NDZTTSTA) on urban air pollution. The findings show that NDZTTSTA can reduce air pollution. Mechanism analysis indicates that the demonstration zones curb air pollution through the scale effects of introducing emerging production factors, the technological effects of knowledge spillovers, and the structural effects of achievement transformation. Additional analysis reveals that the air pollution control effects of NDZTTSTA vary significantly across locational conditions and levels of resource dependence. Moreover, NDZTTSTA generate spillover effects on air pollution control both in regions with strong technological linkages and in neighboring cities.
Rural residence and employment spaces are key components of rural regional systems and are closely related to the coordination of rural living, production, and ecological functions. Taking the Chengdu-Chongqing Metropolitan Area (CCMA) as the study area and using county-level data for 2011, 2016, and 2020, this study examines the spatio-temporal differentiation and explanatory mechanisms of urban-rural integrated development from the perspective of rural residence-employment coordination. The results show that: (1) Rural residence-employment coordination (RREC) in the CCMA increased steadily from 2011 to 2020 and exhibited clear spatial heterogeneity and clustering characteristics. High-value areas were mainly distributed around Chengdu and central Chongqing, while low-value areas were concentrated in peripheral mountainous counties. (2) The Pearson correlation results indicate that RREC is not dominated by a single public-service variable, but is associated with rural employment structure, transportation accessibility, fiscal-economic conditions, and infrastructure-related factors. (3) The Geodetector results further show that the proportion of rural workers has the strongest and statistically significant explanatory power across all three years, while economic non-agriculturalization becomes a significant explanatory factor in 2020. Interaction detection shows that paired interactions involving rural employment structure consistently generate high explanatory power, suggesting that RREC differentiation is shaped by the combined mechanisms of local employment absorption, residential stability, and spatial connectivity. These findings suggest that rural residence-employment coordination in mountainous metropolitan regions should be promoted by strengthening local employment opportunities, improving spatial accessibility, and enhancing county-level public service and infrastructure capacity.
This study examines the factors affecting the ecological deficit levels of countries in the Black Sea basin within a spatial panel data framework. Using data covering the period 1996–2021, the General Nested Spatial (GNS) model is employed to simultaneously evaluate cross-country interactions and local determinants. The results show that trade in low-carbon technology products reduces the ecological deficit through channels such as clean technology transfer and structural transformation in production. Similarly, foreign direct investment improves environmental performance through technology diffusion and production efficiency, while political globalization limits environmental pressure through international environmental norms and institutional cooperation. In contrast, consumption expenditures and financial globalization expand the ecological deficit by increasing the use of natural resources, whereas information globalization increases environmental pressure through digitalization, information flows, and the diffusion of consumption patterns. The analysis of spatial effects indicates that increases in political globalization and foreign direct investment in neighboring countries generate positive spatial externalities that reduce the ecological deficit through regional policy coordination and production networks. Supported by robustness and sensitivity analyses, these findings suggest that environmental outcomes in the Black Sea basin are shaped not only by national dynamics but also by regional interactions, and they provide important implications for the design of sustainable development policies.
Child malnutrition remains a major public health challenge in Sub-Saharan Africa, with persistent disparities in stunting and underweight among children under five. This study examined subnational spatial variation and structural determinants of child growth failure across the region using Demographic and Health Survey data from 24 Sub-Saharan African countries collected between 2015 and 2023 (n = 198,443). Generalized additive models with spatial smoothing were used to assess geospatial patterns of stunting and underweight. Stunting prevalence ranged from 16
Under the “neighborhood enrollment” policy, the spatial rationality of school district delineation is important for evaluating educational accessibility. Taking the main urban area of Hangzhou as a case study, this research introduces an excess commuting framework to measure the distance discrepancy between officially assigned public primary schools and geographically nearest public primary schools at the residential community scale. Based on GIS-based spatial analysis, XGBoost regression, and SHAP explainability methods, this study examines the spatial distribution of excess school commuting and its associated institutional, locational, and built-environment factors. The results show that: (1) around 21.1
Digital access, as an alternative resource for human capital growth, exerts a profound influence on return migrant entrepreneurship in developing regions globally, empowering entrepreneurs to adapt and thrive digitally. Although digital human capital is recognized as pivotal for return migrant entrepreneurship, how return migrants leverage accumulated skills during entrepreneurship remains a pressing concern. Using China Family Panel Studies (CFPS) data (2018–2020), we examine the effect of digital access on entrepreneurship, incorporating mediating effects of entrepreneurial opportunity mechanisms. We find that digital access significantly enhances entrepreneurial behaviors in developing areas. Accumulated digital human capital significantly strengthens both the capability of entrepreneurial information acquisition and opportunity identification, which in turn independently boosts return migrants’ entrepreneurial behaviors. Concurrently, digital access exerts an influence on return entrepreneurial behaviors through a sequential process-digital access first enhances information acquisition capability, then improves opportunity identification capability, and ultimately increases return entrepreneurial behaviors-that remains robust after conducting sensitivity tests. Furthermore, digital access demonstrate varying levels and manners of influence on necessity and opportunity return migrant entrepreneurship. While digital access significantly facilitates both, the former hinges predominantly on opportunity identification, and the latter relies on information acquisition. Our analysis reveals how return migrants harness digital democratization-particularly information access and opportunity recognition-to navigate evolving entrepreneurial landscapes. This study enriches theoretical understanding of digital access’ impacts and provides policy-calibrated pathways for rural revitalization in the digital era.
Dockless bike sharing (DBS) is an important component of urban shared micromobility that helps bridge the last-mile gap to public transport, supporting short-distance daily travel, and advancing low-carbon urban mobility. Using four representative districts in Shenzhen as the study area, this paper defines 500 m × 500 m grid cells as spatial analysis units and integrates DBS orders with Baidu heatmap data, road networks, public transport facilities, buildings, and land-use data. A residual correction framework combining Tabular Prior-Data Fitted Network (TabPFN) and geographically weighted regression (GWR), hereafter TabPFN-GWR, is applied to examine how built-environment conditions are associated with weekday and weekend DBS usage and to diagnose local spatial deviations that may require planning responses. The findings show that TabPFN-GWR improves predictive performance over standalone TabPFN, with R2 values rising to 0.698 for weekdays and 0.716 for weekends. Local surrogate results indicate that Baidu heatmap density is the dominant local predictor in most grid cells, while road density, building count, metro station presence, industrial land-use share, and land-use mix further explain spatial differences in DBS usage. PDP results show that Baidu heatmap density, road density, building count, and land-use mix are positively associated with predicted DBS usage, whereas industrial land-use share exerts a persistent negative effect and metro station presence provides clear feeder benefits. The findings suggest that DBS planning should move beyond uniform fleet deployment toward area-based governance: improving non-motorized transport connectivity in poorly connected areas, organizing parking and transfers around transit nodes, adopting differentiated dispatching for commercial, residential, and industrial districts, and designating residual clusters as priority areas for field inspection and fine-grained spatial management.
