Search

A growing body of research has incorporated the Social Vulnerability Index (SVI) into an expanded understanding of the social determinants of health. Although each component of SVI and its association with individual-level mental health conditions have been well discussed, variation in mentally unhealthy days (MUDs) at a county level is still unexplored. To systematically examine the geographically varying relationships between SVI and MUDs across the US counties, our study adopted two different methods: 1) aspatial regression modeling (ordinary least square [OLS]); and 2) locally calibrated spatial regression (geographically weighted regression [GWR]). This study used a cross-sectional statistical design and geospatial data manipulation/analysis techniques. Analytical unit is each of the 3109 counties in the continental USA. We tested the model performance of two different methods and suggest using both methods to reduce potential issues (e.g., Simpson's paradox) when researchers apply aspatial analysis to spatially coded data sets. We applied GWR after checking the spatial dependence of residuals and non-stationary issues in OLS. GWR split a single OLS equation into 3109 equations for each county. Among 15 SVI variables, a combination of eight variables showed the best model performance. Notably, unemployment, person with a disability, and single-parent households with children aged under 18 years especially impacted the variation of MUDs in OLS. GWR showed better model performance than OLS and specified each county's varying relationships between subcomponents of SVI and MUDs. For example, GWR specified that 69.3% (2157 of 3109) of counties showed positive relationships between single-parent households and MUDs across the USA. Higher positive relationships were concentrated in Michigan, Kansas, Texas, and Louisiana. Our findings could contribute to the literature regarding social determinants of community mental health by specifying spatially varying relationships between SVI and MUDs across US counties. Regarding policy implementation, in counties containing more social and physical minorities (e.g., single-parent households and disabled population), policymakers should attend to these groups of people and increase intervention programs to reduce potential or current mental health illness. The results of GWR could help policymakers determine the specific counties that need more support to reduce regional mental health disparities. [ABSTRACT FROM AUTHOR]

Poverty is one of the world's main problems, which has been very difficult to overcome. The poverty rate has influences on economic growth in an area. The greater the poverty rate in an area, the more difficult it is for an area to grow economically. Many factors affect the level of poverty. Geographical location, level of education, population growth in each region, significantly affect the poverty level. The many factors that affect the poverty level make it difficult for us to predict which factors will influence the level of poverty in an area, making it difficult for us to solve the problem. This research was made to predict what factors influence the level of poverty in Papua, Indonesia. This study uses a Geographically Weighted Regression (GWR) method due to the existence of spatial heterogeneity. It produces model estimates that are local for each location and different from other locations to get a model with the most influential factors in each region. The results of this study are as follows. First, agricultural employees affected the poverty rate in each location in Papua. Second, house area and households with access to safe water are significant variables in Yapen, Biak Numfor, and Supiori. [ABSTRACT FROM AUTHOR]

The purpose of this study was to examine the relationship between test site availability and testing rate within the context of social determinants of health. A retrospective ecological investigation was conducted using statewide COVID-19 testing data between March 2020 and December 2021. Ordinary least squares and geographically weighted regression were used to estimate state and ZIP code level associations between testing rate and testing sites per capita, adjusting for neighbourhood-level confounders. The findings indicate that site availability is positively associated with the ZIP code level testing rate and that this association is amplified in communities of greater economic deprivation. In addition, economic deprivation is a key factor for consideration when examining ethnic differences in testing in medically underserved states. The study findings could be used to guide the delivery of testing facilities in resource-constrained states. [ABSTRACT FROM AUTHOR]

Air quality degradation presents a significant public health challenge, particularly in rapidly urbanizing regions where changes in land use/land cover (LULC) can dramatically influence pollution levels. This study investigates the association between LULC changes and air pollution (AP) in the five fastest-growing cities of Bangladesh from 1998 to 2021. Leveraging satellite data from Landsat and Sentinel-5P, the analysis reveals a substantial increase in urban areas and sparse vegetation, with declines in dense vegetation and water bodies over this period. Urban expansion was most pronounced in Sylhet (22–254%), while Khulna experienced the largest increase in sparse vegetation (2–124%). Dense vegetation loss was highest in Dhaka (20–77%) and water bodies (9–59%) over this period. Concentrations of six major air pollutants (APTs) - aerosol index, CO, HCHO, NO 2 , O 3 , and SO 2 - were quantified, showing alarmingly high levels in densely populated industrial and commercial zones. Pearson's correlation indicates strong positive associations between APTs and urban land indices (R > 0.8), while negative correlations exist with vegetation indices. Geographically weighted regression modeling identifies city centers with dense urban built-up as pollution hotspots, where APTs exhibited stronger impacts on land cover changes (R2 > 0.8) compared to other land classes. The highest daily emissions were observed for O 3 (1031 tons) and CO (356 tons) at Chittagong in 2021. In contrast, areas with substantial green cover displayed weaker pollutant-land cover associations. These findings underscore how unplanned urbanization drives AP by replacing natural land cover with emission sources, providing crucial insights to guide sustainable urban planning strategies integrating pollution mitigation and environmental resilience. [Display omitted] • Study reveals urban-air quality nexus in 5 fastest-growing Bangladesh cities. • Urban expansion (22–254%) linked to air quality degradation (1998–2021). • GWR reveals strong pollutant-land cover association in city centers (R2 > 0.8). • Highest daily emissions: O 3 (1031.36 tons), CO (355.78 tons) in Chittagong. • Urban indices strongly correlate with air pollutants (R > 0.8) across cities. [ABSTRACT FROM AUTHOR]

Randomised designs are often preferred over systematic designs by agronomists and biometricians. For on-farm trials, however, the choice may depend on the objective of the experiment. If the purpose is to create a prescription map of a continuous input for each plot in a grid covering a large strip trial, a systematic design may be a better choice, although it often attracts less discussion and attention. This study aims to evaluate the performance of systematic designs with geographically weighted regression (GWR) models in addressing spatial variation and estimating continuous treatment effects in large strip trials through numeric simulations. A hierarchical model with spatially correlated random parameters is utilised to generate simulated data for various scenarios of large strip on-farm trials. The study employs GWR models to analyse the simulated data for two assumptions: a linear response and a quadratic response of yield to the treatment effects. With the assumption of a quadratic response, a systematic design is superior to a randomised design in terms of achieving lower mean squared errors (MSEs) with GWR. With the assumption of a linear response, the difference of MSE between a systematic design and a randomised design is not significant, regardless of the presence of spatial variation. The findings highlight the superiority of systematic designs in producing smooth spatial maps of optimal input levels for quadratic response models in large strip trials, even when impacted by significant spatial variation. Additionally, we recommend selecting fixed bandwidths in GWR analysis based on the plot configurations used in experimental designs. For a large strip trial, to produce estimates of spatially-varying treatment effects across strips, a systemic design should be used as it allows us to obtain better estimates than those obtained from a randomised design in post-experiment statistical modelling. The findings offer practical recommendations for designing large strip trials. By drawing attention to the experiment's main inferential purpose, this research contributes valuable insights for improving the efficacy and planning of large strip trials. • Simulation of diverse scenarios generated by Bayesian hierarchical model. • Systematic designs are preferred for producing continuous input maps. • The presence of spatial variation affects the performance of GWR. • A fixed bandwidth based on the experiment is recommended for GWR. [ABSTRACT FROM AUTHOR]

The x-minute city/neighborhood concept has been introduced in numerous cities globally and is considered a new management paradigm to support low-carbon and sustainable non-motorized travel mode. This paper provides the first empirical evidence of the economic benefits of the 10-min neighborhood by investigating the impact of functional diversity, functional density, and street connectivity within a 10-min walking distance on rental prices. Based on approximately 74,000 rental transaction data points in the Munich Metropolitan Area, Bavaria, Germany, spanning from 2015 to 2022, we have identified significant price premiums associated with functional diversity. And the economic significance diminishes when distances extend beyond 10 min. Additionally, in regions characterized by relatively lower levels of accessibility and income, residents pay more for functional diversity in their rents. Our results offer intuitive insights for urban planning on the economic value of the x-minute city/neighborhood, which can also serve as a reference for effectively allocating resources. [Display omitted] • Quantify the WTP for four dimensions of the x-min neighborhood: proximity, diversity, density, and connectivity. • The price premium primarily stems from the functional diversity of urban service amenities close to properties. • Beyond a 10-min walking distance, residents' willingness to pay for access to urban services notably diminishes. • WTP for functional diversity varies spatially, with greater impact in regions of lower accessibility and resident income. [ABSTRACT FROM AUTHOR]

• This paper measures the CUEE with the Super-SBM with undesired outputs. • This paper explores the evolution of CUEE from a micro perspective with the CNA. • With the link prediction, the development trend of the CUEE is analyzed. • The Geodetector and GWR are used to excavate the influencing factors of CUEE. • This paper innovatively considers the impact of network attributes on the CUEE. Urban eco-efficiency (EE) refers to the ability of a city to balance economic growth with environmental protection through the optimal use of resources and environmental management, reflecting the city's performance in sustainable development. Quantitative EE analysis is a crucial evaluation criterion for measuring the process of sustainable development (SD). To assess the current state of sustainable development in China, this study employs the slacks-based super efficiency model (Super-SBM) to measure the the urban eco-efficiency in China (CUEE) of 284 cities. The study utilizes Kernel Density Estimation (KDE) and Spatial Autocorrelation Analysis (SAA) to investigate the fundamental patterns of CUEE in terms of inter-city differences and relationships. Complex Network Analysis (CNA) and Link Prediction methods are applied to examine the characteristics and future trends of the CUEE network. Additionally, Geodetector and Geographically Weighted Regression (GWR) are used to analyze the influencing factors of CUEE. The findings are as follows: Overall CUEE is relatively low but shows a steady increase. The spatial association network of CUEE is becoming more complex, with increasing association strengths. The Chengdu-Chongqing Economic Circle (CCEC) is expected to further enhance its spatial association strength and may emerge as a new core area. The effectiveness of the interactions among selected factors is greater than that of individual factors. Degree centrality (X1) is identified as a core explanatory factor influencing CUEE, showing a positive correlation with CUEE. Scientific expenditure (X5) is also positively correlated with CUEE. Local fiscal budget expenditure (X3), industrial output value of foreign-invested enterprises (X7), and total cargo transportation volume (X9) are negatively correlated with CUEE. The analytical framework and research conclusions of this study offer valuable insights into the development of urban ecological efficiency in China and provide constructive recommendations for promoting sustainable urban development. [ABSTRACT FROM AUTHOR]

The forestland expansion is the joint efforts by various behavioral actors, and among them governments and farmers play a more important role in it. However, their afforestation behaviors haven't be together investigated. Examining Beijing metropolitan area, this research distinguishes between government-led (GA) and farmer-led (FA) afforestation on arable land, and analyzes their spatial characteristics and the influencing factors using Ordinary Least Squares and Geographically Weighted Regression models. Our findings show: (1) Spatial distributions of GA and FA are non-random and systematically organized. GA dominates both in the areas close to the central city and at the outermost edge of the metropolitan area. FA leads in intermediate zones between GA dominant areas. (2) Both GA and FA are rational strategies for land allocation. GA in areas close to the central city is regarded as a trade-off between public benefits of afforestation and potential loss of fiscal revenue, while GA at the metropolitan area's outermost edge prioritizes ecological gains and lower land costs. FA, however, is driven by maximizing household welfare amidst the challenges brought by rural exodus and a rapid aging society. The findings deepen the existing investigations and provide a guidance for spatially organizing GA and synthetically regulating FA. • Government-led (GA) and farmer-led (FA) afforestation show organized, non-random spatial patterns. • Areas close to the central city are dominated by GA due to afforestation's social and ecological value. • GA also predominates at the outermost edge of the metropolitan area, driven by ecological needs and attraction of lower land prices. • FA occurs in intermediate zones, influenced by off-farm work and land transfer limitations. [ABSTRACT FROM AUTHOR]

Distance holds significant importance in the decision-making processes of migration flows. Previous studies have predominantly adopted a global perspective to understand distance constraints. However, same spatial heterogeneity exists in distance constraints, and this heterogeneity contributes to understanding of the effect of regional characteristics in human migration. This study answers three questions: Does distance work for all counties in the same way? Why distance works for some counties but does not for others? And what makes distance decay differently among counties? We adopt geographically weighted regression, binary logistics regression, and random forest regression to analyze the migration from counties all over the China to Shenzhen to explore the spatial heterogeneity of distance decay. The results show that distance does not always work, and that demographics and transportation facilities are important determinants of whether distance works. For counties where distance works, distance decay is nonlinearly related to regional development. This non-linear relationship is due to the game between the appeal of intervention opportunities in the nearest new first-tier city, the constraints represented by the socio-economic development of the source county, and the attractiveness of Shenzhen. This further reflects the dynamic interplay between costs and opportunities, give and gain in the decision process of migration. • Proposing that distance doesn't always work: It does not work in the western part and some well-developed counties. • The demographics and transportation facilities of counties play an important role in whether distance works or not. • Revealing migration does not decay linearly with distance, but rather is characterized by a non-linear. • The nonlinear effect of distance decay as a game of intervening opportunity, county development, and Shenzhen attraction. [ABSTRACT FROM AUTHOR]

• This study investigates the factors contributing to bicycle accidents. • Impacts of cycling lanes and exposure and built-environment factors are examined. • GWPR model is used to capture spatial heterogeneity across accident locations. • Impacts of bicycle lanes on accidents vary depending on their type. This study investigates the factors contributing to bicycle accidents, focusing on four types of bicycle lanes and other exposure and built environment characteristics of census blocks. Using Seoul as a case study, three years of bicycle accident spot data from 2018 to 2020 was collected, resulting in 1,330 bicycle accident spots and a total of 2,072 accidents. The geographically weighted Poisson regression (GWPR) model was used as a methodological approach to investigate the spatially varying relationships between the accident frequency and explanatory variables across the space, as opposed to the Poisson regression model. The results indicated that the GWPR model outperforms the global Poisson regression model in capturing unobserved spatial heterogeneity. For example, the value of deviance that determines the goodness of fit for a model was 0.244 for the Poisson regression model and 0.500 for the far better-fitting GWPR model. Further findings revealed that the factors affecting bicycle accidents have varying impacts depending on the location and distribution of accidents. For example, despite the presence of bicycle lanes, some census blocks, particularly in the northeast part of the city, still pose a risk for bicycle accidents. These findings can provide valuable insights for urban planners and policymakers in developing bicycle safety measures and regulations. [ABSTRACT FROM AUTHOR]

[Display omitted] • YREB land structure is stable, landscape is fragmented and connectivity is reduced. • HQ distribution varies significantly, with high HQ widely distributed at best. • High HQ gradually converts to low, with some risk of HQ degradation. • Global view, there are different correlations between different LSPI and HQ. • Local view, LSPI promote or inhibit HQ, with clear spatiotemporal heterogeneity. Recently, rapid economic development of the Yangtze River Economic Belt region in China and the growing problem of regional habitat quality have seriously threatened the process of biodiversity conservation in the Yangtze River Basin. Studying the impact of changes in landscape pattern indexes on habitat quality in the Yangtze River Economic Belt can provide policy insights for biodiversity conservation in the Yangtze River Basin. However, existing studies are still lacking in exploring the effects of changing landscape pattern indices on habitat quality under future scenarios. Based on the land use data of the Yangtze River Economic Belt from 2000 to 2020, this study simulates the land use situation in 2030, and analyzes the landscape pattern indexes and habitat quality during the 30 years, to investigate the impact of landscape pattern indexes on habitat quality at global and local scales through various methods. The results show that landscape patterns were significantly affected by human activities and land use types, with the Yangtze River Economic Belt experiencing increased levels of land use and structural stabilization during the three decades (Shannon Diversity Index increased from 0.90 in 2000 to 0.94 in 2030), fragmentation of the landscape (Landscape fragmentation index increase from 0.094 in 2000 to 0.102 in 2030), and weakened connectivity (contagion index decrease from 48.213 in 2000 to 45.437 in 2030). The Habitat Quality has a large variation in spatial distribution at the county scale, with an average change in habitat quality of 1.04 % from 2000 to 2030, and an overall downward trend. There is a strong localized spatial autocorrelation between landscape pattern indices and habitat quality, with significant spatial heterogeneity and some form of variability at time scales. The results of this study help to further understand the eco-environmental problems of the Yangtze River Economic Belt, and provide theoretical references for the formulation of ecological environmental protection policies and ecological functional area planning. [ABSTRACT FROM AUTHOR]

• The agglomeration scale of SMMs was characterized by kernel density analysis. • GWR was used to explore the spatial association between SMMs cluster scale and LER. • The different effects of SMMs with different cluster scales on LER were observed. • The aggregated distribution of SMMs caused a "1 + 1 > 2" diffusion effect on LER. Mining is one of the most pervasive of the various anthropogenic disturbances to landscape ecology. Currently, more attention is paid to the landscape ecological impacts caused by large-scale opencast mining, while small- to medium-scale mines (SMMs) are often neglected due to their small individual size. However, for SMMs, which account for up to 80 % of total mines in China, the impacts of their different agglomeration scales on regional landscapes, especially in ecologically fragile regions, are not yet clear. Therefore, a typical watershed with SMMs aggregated in the semi-arid ecologically fragile area of northwestern Liaoning Province was selected to explore the spatiotemporal dynamics of landscape patterns and landscape ecological risk (LER) under the influence of long-time mining by SMMs. Based on the geographically weighted regression model, the differences in the impact of SMMs on LER at different agglomeration scales were identified. The results revealed that (1) from 1989 to 2022, cropland as the dominant landscape decreased by 1504.82 hm2, while the area of mining land increased the largest, by 668.51 hm2. (2) SMMs were an important factor affecting the landscape pattern, and their multi-point dispersed mining destroyed the original large and intact landscape patches (e.g., cropland), leading to an increase in landscape fragmentation and heterogeneity, and a decrease in connectivity. (3) From 1989 to 2022, the total proportion of very high-risk and high-risk areas increased from 8.18 % to 38.29 %, which are primarily concentrated in the northeast of SHCRB where SMMs were aggregated. (4) The impact on LER was negligible when SMMs did not form an agglomeration scale, while the impact had a diffusion effect when SMMs formed a certain or larger agglomeration scale (the total number of mines was more than 100 and the total area was more than 500 hm2 in a kernel density region with a bandwidth of 1 km), i.e., the overall impact of aggregated SMMs on LER was much greater than the sum of the individual impacts of these mines. Additionally, LER was positively correlated with mine kernel density. The findings can provide a reference for ecological protection and systematic restoration of many similar watersheds with SMMs aggregated, as well as for SMMs around the world in future mining and planning. [ABSTRACT FROM AUTHOR]

• LER improved from 1990 to 2030, but the risk intensity did not converge. • Mainly land use changes over the past 30 years: more impervious surfaces, less grassland. • Human activities have intensified LER, while natural environmental factors have mitigated LER. • The high-risk driving factors vary across different regions. Hills and mountains cover approximately 25 % of Earth's landmass; therefore, investigating landscape ecological risk (LER) in these regions is imperative for ecosystem management. Hilly and mountainous regions susceptible to influences such as desertification and erosion include the Southern Hill and Mountainous Belt (SHMB) in southern China. Over the past 30 years, the SHMB has undergone significant changes, offering an excellent research case. A comprehensive assessment of the ecological status of this region is crucial for the protection and management of the ecosystems in the SHMB. This study assessed the LER of the SHMB, identified influencing factors over the past 30 years, and utilized a patch-generating land-use simulation model (PLUS) to project future land use and LER. Accounting for more than 90 % of the total area, cropland and forestland were the two most common land use types in the SHMB. From 1990 to 2020, impervious surface and water area saw obvious increases, rising by 154 % (from 1209.37 km2 to 3073.59 km2) and 21 % (from 1584.78 km2 to 1924.8 km2), respectively. Grasslands and shrublands decreased by 54 % and 34 %. It is anticipated that there will be little change in land-use types by 2030 compared to 2020. Additionally, the overall LER in the SHMB was low, with low- and medium–low-risk representing around 70 % of the overall risk. The general state of LER improved during the research period but did not converge. Specifically, the LER intensity did not decrease although the total LER showed a declining trend. In the western portion of the research region, the karst area displayed an increasing LER, and the severity of the high-risk area increased. Furthermore, natural factors promote the restoration of LER. For example, the abundance of hydrothermal conditions (i.e., increased temperature and precipitation) promotes the recovery of vegetation growth and reduces the LER. Human activities, such as the construction of stations, highways, and an increasing population, exacerbate LER, contributing to an increase in LER. This study offers helpful guidance for formulating ecological conservation strategies to maintain the ecological balance and sustainability of hilly and mountainous regions. [ABSTRACT FROM AUTHOR]

The Affordable Connectivity Program (ACP) represents a historic investment by the federal government to address the digital divide. Its goals are both to increase the rate of broadband adoption and make it easier for low-income households to sustain service. But what have been the drivers of ACP enrollment? We know a great deal about trends in adoption and the geography of enrollment among eligible households. But those metrics – especially a focus on enrollment rates among eligible households – are only so useful in explaining ACP sign-ups. This invites analysis that investigates the drivers of ACP enrollment in a comprehensive way, which this paper proposes to do. The paper hypothesizes that ACP enrollment decisions are not solely individual (i.e., that a household's enrollment decision is structured only by income level), but also influenced by community-wide considerations, such as housing costs, share of occupied houses, presence of anchor institutions such as public libraries, and population density (i.e., whether a place is urban or rural). The paper develops a regression model that predicts ACP enrollment rates among eligible households at the 5-digit zip code geography as a function of the variables discussed above, as well as others such as the racial and ethnic make-up of a zip code area. The analysis also controls for existing levels of broadband subscriptions and computer ownership in a given area. The difference between predicted and actual ACP enrollment rates at the 5-digit zip code level is proposed as a metric of ACP performance. These findings are supplemented with a geographically weighted regression (GWR) model that allows the relationship between each explanatory variable and ACP enrollment to vary by locality. This allows for analyzing and visualizing how these relationships vary across space. The empirical findings show that high levels of economic distress are a strong determinant of ACP enrollment in a given area, but social and community-wide indicators are important as well. The GWR results demonstrate that important differences in these relationships exist across geographies. Understanding spatial variations in ACP performance can help policymakers and other stakeholders better target resources to address the digital divide. • We provide background on the ACP, a recent program that subsidizes broadband service for low-income U.S. households. • We estimate ACP enrollment using both ordinary least squares (OLS) and geographically weighted regression (GWR). • OLS results show that economic distress and local library presence are strongly positively associated with ACP enrollment. • Higher rates of foreign-born or elderly residents are negatively associated with ACP enrollment. • GWR results show that relationships between ACP enrollment and local characteristics vary widely and are context-specific. [ABSTRACT FROM AUTHOR]

This paper aims to analyze how the associations of built environment and usage of urban space changed due to the social restrictions during COVID-19 pandemic in the United States. A total of 57,185 census block groups (CBGs) in 459 cities were analyzed. Built environment elements including road network, green space, and land use associated with usage of urban space (measured by over two billion mobile phone data) before/during COVID-19 were evaluated at both CBG and city level. Our results showed drastic decreases of urban space usage, ranging from 0.25 % to 76.35 %, among cities during COVID-19 social restriction. Specifically, CBG-level road length-node ratio, road intersection index and land use densities contributed less to usage of urban space during COVID-19. Meanwhile, the heterogeneity of city-level road network, commercial, public places, as well as hospitals and health cares exhibited stronger relationship with usage of urban space. The heterogeneity of green space at both CBG and city level showed less associations with usage of urban space during COVID-19. The discrepancies of the built environment effects on usage of urban space between CBG and city level suggest the importance of varying-scale analysis. Future interventions should consider space usages and health inequality from a space-varying perspective for post-pandemic period. • The changing associations of built environment with usage of urban space during COVID-19 were investigated • The changing associations were evaluated based on the census-block-group (CBG) level and city level in 459 U.S. cities • CBG-level road complexity and urban facility density contributed less on the usage of urban space during social restriction • City-level built environment exhibited stronger relationship with usage of urban space during social restriction [ABSTRACT FROM AUTHOR]

Proper knowledge regarding the spatial dependence of ecosystem service values (ESVs) on the changes in land use dynamics is essential for environmental management as well as sustainable urban development. To investigate this complex association concerning a fast-growing secondary city, this study estimated the integrated land use dynamic degree (ILUDD), uniform annual land use transition intensity (W tn), and entropy index of land use mix (LUM), and the ESVs for Raiganj urban agglomeration (UA). The overall association between the three indices and the ESVs is explored using the spatial lag and spatial error model. For the local level association, the geographically weighted regression (GWR) is applied. The output showed that changes in ILUDD and LUM indicate over −20 unit and −2 unit changes in the ESVs of built-up with 1 unit increase in ILUDD and LUM, respectively. However, the ILUDD coefficient score concerning vegetation ESV was observed around +25. The W tn also represented a significant loss of natural vegetation from every administrative unit of Raiganj UA. The relation between urban growth type and landscape patches is analyzed using Getis-Ord-Gi. It is found that the promotion of compact growth can be the best possible solution to preserve the environmental quality in this region. • The spatial dependence between changes in land use dynamics and ESVs is examined. • Four different indices, such as SLUDD, ILUDD, Wtn, and LUM were incorporated. • Changes in ESVs were strongly associated with the changes in land use dynamics. • The association between urban growth types and LULC characteristics were explored. • Infilling expansion type was the most suitable for urban growth in Raiganj UA. [ABSTRACT FROM AUTHOR]

Exploring the relationships between variables is a crucial component in comprehending geographical phenomena. Most existing methods ignore the vagueness hidden in spatial data when quantifying this relation, which may lead to a partial or even wrong understanding of geographical phenomena as vagueness is an intrinsic property of them. This paper uses fuzzy rough sets for quantifying local and overall variable relationships to address this limitation, relying on the consistent degree between variables. This approach uses a sliding window to scan the entire study area and build a local region for each object. The local variable relation is quantified using the local average membership degree to the positive region for each object during the scan. The overall variable relation in the whole study area is quantified using the median value of the local consistent degree between variables in every local region, and the entropy of the normalized local consistent degree is used to measure the corresponding spatial heterogeneity. The proposed method can detect and compare local and overall variable relations. Comparison experiments on five publicly accessible datasets demonstrate the effectiveness of the proposed method and show that it can reveal patterns missed by geographically weighted regression and geographical detectors, as it models rather than ignores vagueness uncertainty. • Effective in modelling vagueness relations between variables in spatial data. • Be adaptable to both nominal and continuous-valued variables. • Data-driven and not bounded to any classifiers. [ABSTRACT FROM AUTHOR]

• Ridesplitting adoption rate (RAR) presents spatial heterogeneity across regions. • Factors affecting the RAR during different time periods are examined. • Subway station density is positive with the RAR in the southern area. • Potential substitutive and complementary effects between ridesplitting and transit. • Percent zero-vehicle households is negative with the RAR for most regions. Ridesplitting, a form of ride-hailing service where passengers with similar travel routes are matched to the same driver, can reduce the negative effects of solo ride-hailing trips and bring various environmental and social benefits. However, limited efforts were made to examine the spatial variation of ridesplitting trips, which was not conducive to the formulation of ridesplitting policies. To fill the gap, this work investigates the spatial variation of ridesplitting adoption rate (the proportion of ride-hailing trips with shared trip authorized, RAR) and its association with built environment and socio-economic factors at the census tract level, using the ride-hailing trip data in Chicago. To addressing the spatial heterogeneity, geographically weighted regression models are established to detect the factors influencing the RAR during different time periods, such as weekday, weekend, weekday morning peak and evening peak. Modeling results show that GWR models outperform the traditional global models in terms of model fit. The census tract level factors including subway station density, frequency of transit, land use mix, homicide density, percent female, the share of nonwhite, and percent zero-vehicle households have impacts on RAR, and the coefficient estimates of each explanatory variable vary across regions. The research results can help urban planners and transportation network companies develop refined policies to promote shared ride-hailing trips. [ABSTRACT FROM AUTHOR]

• This model addresses spatial-temporal challenges for predicting urban electricity consumption in diverse and mature city. • Agent-based approach evaluates shifts in electricity usage, accounting for cluster transitions and socio-economic factors. • Integrating GWR and DT mitigates heterogeneity in predicting citywide energy consumption patterns. • Findings reveal decision probabilities of cluster shifts and impacts on the occupants' behavior and urban energy dynamics. This study introduces a model to predict regional electricity consumption in urban areas with mixed land use patterns and high energy intensity. It employs an agent-based approach to tackle the temporal-spatial and heterogeneity distribution challenges brought about by rapid urbanization. The model evaluates potential shifts in electricity usage driven by occupants' behavior, considering the urban environment, demographics, and energy policies. Using the GAMA platform, it forecasts electricity consumption for 907 agents in Taipei City, Taiwan. Integrating 5 clusters of 5 independent variables, the model utilizes Geographically Weighted Regression (GWR) and Decision Trees (DT) for analysis. Findings indicate a higher shift likelihood among residential occupants (Cluster 1 and Cluster 4) than commercial occupants (Cluster 2 and Cluster 3), especially in areas with diverse clusters. Consequently, the projected electricity consumption in 2030 is only 2 kWh/household lower than in 2020, underscoring the challenge of achieving carbon neutrality through changes in occupants' electricity usage behavior in a mature city. Future research aims to enhance accuracy by delving into cluster shift mechanisms and socio-economic backgrounds. Despite the hurdles, the model contributes to understanding urban energy dynamics and informs policy formulation for energy-efficient practices. [ABSTRACT FROM AUTHOR]

Urban landscapes, characterized by intricate interactions within a three-dimensional (3D) framework, give rise to complex mechanisms that underlie the Surface Urban Heat Island phenomenon. While global case studies have emphasized the impact of 3D urban morphological components on urban thermal environments, conducting quantitative assessments to identify spatial and seasonal variations in these relationships poses significant challenges. Based on Lidar and land cover dataset, this study first investigated the 3D urban morphology characteristics across nine components, and then assessed their spatial and seasonal impact on the urban thermal environment through Geographically Weighted Regression (GWR). We found that Impervious Surface Fraction (ISF) has the greatest warming effect, whereas the Vegetation Coverage Fraction (VCF) exerts the strongest cooling influence on the thermal environment across all seasons. Building Height (BH) and Street Aspect Ratio (SAR) exhibit spatially varying cooling effects, while Building Number (BN) and Standard Deviation of Building Height (BH_SD) have minimal warming impacts. Additionally, the influence of BH on the thermal environment is significant primarily in areas characterized by higher building coverage. VCF has a stronger cooling effect in central regions with limited vegetation during spring and winter, but this effect diminishes compared to suburban ones in summer and autumn. These findings provide valuable insights into developing effective heat mitigation strategies at both city and community levels, ultimately fostering sustainable urban development. • The paper analyzed the impact of urban morphology on LST from a 3D perspective. • The impact of urban morphology characteristics on LST was analyzed in seasons. • The GWR model was adopted to reveal the spatial heterogeneity of these impacts. [ABSTRACT FROM AUTHOR]

Urban tourists experience neighborhoods in a Peer-to-Peer context. However, literature overlooked the neighborhood concept in guests' accommodation location decisions, emphasizing the centrality (for a given quality) as the primary driver of urban touristification. This paper assesses how guests' preferences for neighborhoods could disrupt tourists' preferences for centrality. We discuss the exogeneity of the neighborhood tourism concept, verify its significance, and explore the relationship between neighborhoods and centrality in the urban touristification process. By extracting latent dimensions of 2,186 neighborhoods in 31 French cities and using a two-step hedonic approach (GW/2SLS regressions) on 45,600 Airbnb listings, we identify significant guests' preferences for neighborhood dimensions independently of embedded accommodation characteristics. Findings indicate that guests' preference for touristy neighborhoods reinforces touristification but non-linearly (concavity), suggesting a touristification maximum threshold. With threshold effects, the other neighborhood dimensions (youthful-hub, popular, well-off) shape the touristification overflow, suggesting new regulations of the accommodations gathering. [ABSTRACT FROM AUTHOR]

[Display omitted] • A global–local collaborative architecture that accounts for spatial non-stationarity in three-dimensional mineral prospectivity modeling. • An interpretable mineral prospectivity model that indicates the relative contributions of predictor variables and reflects spatial nonstationarity. • Expectation-Maximization algorithm for integration of machine learning results into the non-stationary spatial association modeled by geographically weighted regression. The three-dimensional (3D) modeling of mineral prospectivity plays a pivotal role in the mineral exploration process, enabling explorers to identify, evaluate, and prioritize deep-seated targets more effectively. Accurately predicting the 3D distribution of mineral prospectivity relies on effectively quantifying correlations to mineralization. The ore formation process is extremely intricate, marked by spatial non-stationary correlations. While state-of-the-art 3D mineral prospectivity modeling methods use machine learning (including deep learning) to capture correlation between geological evidence and mineralization hidden in the geosciences data, limited attention has been given to quantifying the local non-stationary nature in ore formation. This paper proposes a novel method that adapts to local spatial non-stationarity in 3D mineral prospectivity modeling. The approach integrates a prospectively model with a local regression model, creating a global–local collaborative model to quantify spatial non-stationary associations with mineralization. Random Forest (RF) and Reset are employed to construct mineral prospectivity models that quantify the global associations with mineralization. Simultaneously, the geographically weighted regression (GWR) model serves as a local model to capture the local non-stationarity within mineralization association. An Expectation Maximization algorithm is tailored to bridge the gap between global and local models, resulting in an iterative framework that progressively fuses global spatial associations and local non-stationarity related to mineralization. The iterative process generates a mineral perspectivity model that adapts to local non-stationarity in predictive areas. A case study on the Dingjiashan Pb-Zn deposit validates the effectiveness of the proposed method. The results demonstrate that the global–local collaborative mineral perspectivity model significantly boosts prediction accuracy in predictive areas. Moreover, the regression parameters of GWR derived from the model construction process effectively reveal the spatial non-stationary impacts of different ore-controlling factors, thereby improving the interpretability and reliability of predicted mineral prospectivity. [ABSTRACT FROM AUTHOR]

In recent years, mobility-on-demand services have experienced rapid growth and become an integral part of the urban transport system. As one of the prevalent mobility-on-demand services, online designated driving allows car-owners booking professional drivers via mobile apps to deliver private cars (frequently with the car-owners inside) to a designated location for a fee. Online designated driving can play a crucial role in promoting road safety by providing a responsible alternative for those who cannot drive themselves due to drinking or fatigue, but there is still a lack of empirical study on its basic travel patterns. Using massive trip records of online designated driving, this study investigates the collective and individual spatio-temporal travel patterns, and reveals the spatio-temporal impact mechanism of various land-use types on online designated driving demand. Take Beijing as a case study, the results show that unlike ride-hailing trips that primarily occur during the daytime, 93.07 % of online designated driving trips occurs between 18:00 and 7:00 the next day, with the longest peak observed on Friday nights. Meanwhile, the average distance of online designated driving trips is approximately twice as long as that of ride-hailing trips (16.58 km vs. 8.48 km). The collective travel patterns of online designated driving exhibit a shrinking trend in the form of "surface-line-point" over time. Regarding individual patterns, we rigorously demonstrate that 81.55 % of high-frequency users exhibit a "fixed-destination" phenomenon, indicating a clear centripetal tendency. Furthermore, online designated driving is primarily associated with limited land-use types, revealing that it may serve specific target audiences, like travelers returning from entertainment and catering activities after consuming alcohol, or those in need of transportation for business and automotive services. These findings can contribute to refining operational practices and enhancing regulatory measures for online designated driving. [ABSTRACT FROM AUTHOR]

Analyzing the carbon sink of cropland and its relationship with influencing factors is of great significance to improve carbon sink and develop sustainable agriculture. However, fewer studies evaluated the spatial relationship at multiple scales. In this study, we first quantified the carbon sink of cropland in China in 2020 using empirical model. Then, we used the Geographical Weighted Regression model to quantify the relationship between carbon sink of cropland and influencing factors at the national, agricultural zone and provincial scales. Finally, we discussed potential ways to enhance carbon sink of cropland. The results found that the total carbon sink of cropland in 2020 was in surplus, with 2.56 billion tons, showing a spatial distribution of "high north and low south". There was significant spatial heterogeneity in the relationship between carbon sink of cropland and influencing factors. Labor size and agricultural inputs were most closely related to carbon sink of cropland. The areas with significant correlation were 116.99 and 108.79 million ha, respectively, or 93.17% and 86.64% of the total cropland. Carbon sink of cropland will have strong economic value, with 1.17 trillion yuan in China in 2020, which can increase farmers' per capita income by 1.34%. In order to enhance the carbon sink of cropland, we suggest that China's agricultural sector should reduce farmers' production costs, further improve the carbon trading platform for realizing the economic value of carbon sink of cropland, and then promote the sustainable development of agriculture. [ABSTRACT FROM AUTHOR]

In response to changes in climatic patterns, a profound comprehension of air pollutants (AP) variability is vital for enhancing climate models and facilitating informed decision-making in nations susceptible to climate change. Earlier research primarily depended on limited models, potentially neglecting intricate relationships and not fully encapsulating associations. This study, in contrast, probed the spatiotemporal variability of airborne particles (CO, CH 4 , SO 2 , and NO 2) under varying climatic conditions within a climate-sensitive nation, utilizing multiple regression models. Spatial and seasonal AP data were acquired via the Google Earth Engine platform, which indicated elevated AP concentrations in primarily urban areas. Remarkably, the average airborne particle levels were lower in 2020 than in 2019, though they escalated during winter. The study employed linear regression, Pearson's correlation (PC), Spearman rank correlation models, and Geographically Weighted Regression (GWR) models to probe the relationship between pollutant variability and climatic elements such as rainfall, temperature, and humidity. Across all seasons, APs showed a negative correlation with rainfall while displaying positive correlations with temperature and humidity. The GWR and PC models produced the most reliable results from all the models employed, with the GWR model superseding the rest. Moreover, heightened aerosol levels were detected within a rainfall range of 600 mm/season, a temperature range of 25–30 °C, and humidity levels of 75 %–85 %. Overall, this study emphasizes the growing levels of APs in correlation with meteorological changes. By adopting a comprehensive approach and considering multiple factors, this research provides a more sophisticated understanding of the relationship between AP variability and climatic shifts. [Display omitted] • Air pollutants peaked at 600 mm rainfall, 25–30 °C temp, 75–85 % humidity. • Temperature and Humidity positively correlated with CH 4 , NO 2 , and SO 2 levels. • Rainfall showed a consistent negative correlation with GHGs across all seasons. • Rapid urbanization and climate change may worsen Bangladesh's air quality. • Mitigation strategies needed to curb air pollution amid climate change. [ABSTRACT FROM AUTHOR]

The rapid growth of heatwaves' severity have increasingly endangered citizens' health in the last decade. Evidence points to the environmental injustice of heatwaves: inequal heatwave exposure among socioeconomic groups. Failing to use an adequate indicator of thermal comfort at a large scale, the previous studies have not adequately scrutinized the environmental justice of heatwaves and their variations across a large-scale territory. This study is novel in an unprecedented analysis of psychological equivalent temperature (PET), a comprehensive measure of thermal comfort, across socioeconomic groups and the urban-rural gradient of the Netherlands, as a proxy for factors affecting heatwave vulnerability. The results show that heatwave inequality (measured by the Gini coefficient) is higher in less urbanized areas. It shows that the population aged 25–44, immigrants, tenants, and females are the most heat-exposed groups across all levels of urbanization. However, the population aged 25–44 is more likely to be overexposed in urbanized areas, and immigrants are more likely to be overexposed in rural areas. The results open discussion on the necessity of location-specific policies protecting the most heat-exposed groups in different areas. It also paves the way for future studies using broader PET simulations and expanding their scope to include citizens' daily movements. • The study is novel in using PET to study environmental justice of heatwaves. • Heatwave inequality is higher in the less urbanized areas. • All social groups across urban-rural gradient experience heat stress. • The population aged 25–44, immigrants, tenants, females are the most heat-exposed. • Heat exposure of population aged 25–44 and immigrants differ by urbanization level. [ABSTRACT FROM AUTHOR]

• Italian waste taxation and its components are analysed at the provincial level. • The regressivity (progressivity) of the Italian taxation is investigated. • Geographically Weighted Regression (GWR) is applied on Italian provincial data. • Pay-As-You-Throw scheme allows to achieve a higher equality. The Italian financing system for waste collection and disposal services, still far from a price-based system at the national level, is similar to a real property tax system. Among other concerns, this raises equity issues. The aim of this paper is threefold. First, the structure of the waste tax in its components (i.e., fixed and variable) is analysed at the provincial level. This allows to investigate the magnitude of the two tax components in Italian provinces, while showing the differences among macro-areas in the financing system for waste collection and disposal services. Second, the regressivity (progressivity) of the tax system adopted is investigated, so as to highlight the differences among geographical areas. Third, an alternative base for the waste tax is proposed, in order to avoid regressivity and achieve higher equality. For these purposes, official data referring to 2017 are used and a Geographically Weighted Regression is implemented. The results highlight the importance of an alternative base of the waste tax to design an equal and non-regressive tax, more similar to a Pay-As-You-Throw scheme. [ABSTRACT FROM AUTHOR]

Idle rural residential land (IRRL) is an important manifestation of changes in the human-land relationship during rural development. Studies on this topic form an important field in the study of sustainable land use, but quantitative analyses of IRRL in the academic community are still lacking. In this paper, we take Pinggu District, a metropolitan suburb that has experienced rapid urbanization, to analyze the spatial differentiation of IRRL and explore the spatial differentiation in the impact of different factors on IRRL. Our findings showed that IRRL was a common phenomenon in Pinggu. There was a spatial pattern of "one belt, three cores" in Pinggu District, and its scale decreased from the southeast to the northwest. Industrial areas, semimountainous ecotourism areas and urban fringe areas were the high-incidence areas of IRRL, while the idle rate of rural residential land in mountainous areas and agricultural areas on the plain was relatively low. The IRRL was the result of a combination of different factors, and there were differences among the factors and regions. The transfer of rural labor, nonagriculturalization of the industrial structure and the mode of production and lifestyle caused by urbanization and industrialization were the major driving forces, and the lagging village planning and imperfect land use system increased the risk of IRRL at the management level. Our study contributes to filling the gap in quantitative research on IRRL to enrich the research regarding the land use system by exploring the interaction between humans and land in rural areas and thus has significance for the rural restructuring and sustainable use of land in China. • The transfer of rural labour, non-agriculturalization of the industrial structure and mode of production and lifestyle caused by urbanization and industrialization were the major driving forces. • Influence of different driving factors on idle rural residential land varies in different regions, and the leading factors in different regions are also different. • The government should enhance the positive interaction of industry, population and land to promote the coordinated development of urban and rural areas. [ABSTRACT FROM AUTHOR]

• A computationally efficient geographically weighted regression (GWR) approach, RNN-GWR, which utilizes reverse nearest neighbor (RNN) strategy, is proposed to handle frequently updated data. • The performance of the RNN-GWR approach is compared with the performances of a Naive GWR and FastGWR approaches. • Algorithms are experimentally evaluated using real and synthetic meteorological datasets. Geographically weighted regression (GWR) is a local spatial regression technique to model varying relationships in many application domains, such as ecology, environmental management, public health, meteorology, and tourism. In the literature, most of the studies dealing with GWR do not take into account if the dataset is frequently updated and so these techniques are not efficient to handle such datasets. In this study, to handle frequently updated data on given locations, a computationally efficient GWR approach, RNN-GWR, which utilizes reverse nearest neighbor (RNN) strategy, is proposed. The performance of the proposed RNN-GWR approach is compared with the performances of a Naïve-GWR and FastGWR approaches. Experimental evaluations show that the proposed approach is computationally efficient than the other approaches on handling frequently updated data. [ABSTRACT FROM AUTHOR]

Economic performance is one of the important things that have to be evaluated by the government. There are many factors that affect economic performance. This paper aims to evaluate the effect of fiscal decentralization indicators on regional economic performance in Sumatra Island, Indonesia. Based on the results of spatial dependency testing, can be concluded that there are spatial dependencies on economic performance and spatial heterogeneity between locations in Sumatra Island. Therefore, Geographically Weighted Regression (GWR) which is a spatial modelling can be used in this case. There are six independent variables that be used in this research, there are original local revenue, general allocation fund, special allocation fund, and profit share fund, poverty rate and human development index. The result showed that Geographically Weighted Regression model performed better than multiple linear regression of Ordinary Least Square (OLS) model, providing different effect information at each location. Based on model evaluation using RMSE, MAPE and AIC showed that GWR model is better than OLS model. [ABSTRACT FROM AUTHOR]

In this paper the impact of High Speed Rail (HSR) on the tourism market is analysed. The original and added value of this contribution is in the proposed methodology, which considers the Geographically Weighted Regression technique, incorporated within a Poisson model. This approach allows measuring the relationship between independent and dependent variables with respect to space. The case study comprises 99 Italian provinces, analysed in the time period 2006–2016. The main outcome of the analysis is that HSR affects tourists 'choices of a given destination. [ABSTRACT FROM AUTHOR]

In this paper, we use a newly constructed dataset to study the geographic distribution of fuel price across the US at a very high resolution. We study the influence of socio-economic variables through different and complementary statistical methods. We highlight an optimal spatial range roughly corresponding to stationarity scale, and significant influence of variables such as median income, wage with a non-simple spatial behavior that confirms the importance of geographical particularities. On the other hand, multi-level modeling reveals a strong influence of the state in the level of price but also of some local characteristics including population density. Through the combination of such methods, we unveil the superposition of a governance process with a local socio-economical spatial process. The influence of population density on prices is furthermore consistent with a minimal theoretical model of competition between gas stations, that we introduce and solve numerically. We discuss developments and applications, including the elaboration of locally parametrized car-regulation policies. [ABSTRACT FROM AUTHOR]

Immediately after the initial COVID-19 outbreak in January 2020 in China, an intriguing trend of distrust and anti-protection among the older generation appeared. Based on 1926 samples collected across the Chinese mainland within a short timeframe, this paper adopted a series of global and spatial statistical techniques to explore the socio-cultural reasons underlying the older generation's risk response and their spatial heterogeneity in China. First, the global Ordinary Least Square (OLS) model validated the significant effects of trust in government information, cultural traditions, and social conformity, but the spatial autocorrelation (Moran's I) test of the residuals suggested spatial nonstationarity of the model. Second, the results from the Local Indicators of Spatial Association (LISA) and Geographically Weighted Regression (GWR) confirmed the spatial heterogeneity of the older generation's protective patterns. An 'inverted T-shaped' high-protection belt centering on Wuhan City was identified, indicating that physical proximity to and social connectedness with the pandemic epicenter is crucial in motivating the older generation's protection. In areas distant from and less connected with the risk center, strong collectivistic values may play a vital role in stimulating trust in government information and fostering positive protective intentions of the older generation. However, in most areas, such strong government trust is absent, and the inhibitory cultural and social factors often prevail to result in an anti-protection norm of the older generation. Finally, suggestions were provided to develop locally effective risk management measures from the socio-cultural perspective, and more geographic investigations on public risk response were proposed. • The older generation's protective actions are shaped by trust in government, cultural traditions, and social conformity, but spatial heterogeneity exists. • In areas close to the risk epicenter, positive social norms form quickly and override government distrust and cultural barriers. • In less connected regions, collectivist values enhance government trust and strengthen protection. • Limited government trust and cultural-social impediments lead to widespread anti-protection norms in most regions. [ABSTRACT FROM AUTHOR]

The escalation of thermal risks is attributed to accelerating pace of urbanization. However, assessment and response to green infrastructure with respect to heat risk under different climate and function have been inadequate. This study intends to address these gaps by focusing on local climate zones (LCZs). Firstly, spatial characteristics of heat risk indexes (HRIs) constructed based on heat hazard-exposure-vulnerability for three large cities in 2010, 2015 and 2020 were explored. Secondly, whether HRI cross LCZs have significant differences was examined. Third, proportion and heat contribution of different HRI classes under different LCZs were quantified. Finally, effects of green infrastructure under different LCZs on HRI were analyzed. The results revealed consistent upward trends in the prevalence of sub-high and high HRIs from 2010 to 2015. HRIs exhibited significant spatial aggregation characteristics. Importantly, more than 95.83% of HRIs cross LCZs exhibited significant variations. The HRI for open building type was lower when compared to the compact LCZ types. Additionally, Normalized Difference Vegetation Index (NDVI) had a more pronounced mitigating effect on HRIs in compact high-rise (LCZ 1), compact mid-rise (LCZ 2), open high-rise (LCZ 4) and open mid-rise (LCZ 5). Updating the compact LCZ types to open LCZ types, avoiding configuration of LCZ 1 and 2, and prioritizing the configuration of NDVI enhancement in the existing LCZ 1, 2, 4 and 5, and increasing the amount of greenery by upgrading mono-structures to composite structures consisting of trees, shrubs and grasses and by implementing greening of façade are suggested to alleviate heat risk. • Elevated levels of heat risk demonstrated discernible expansion from 2010 to 2020. • Heat risk was distinguished by a notable degree of spatial aggregation. • There were significant differences in heat risk between local climate zones (LCZs). • Green infrastructure exhibited varied abilities in mitigating heat risk across LCZs. [ABSTRACT FROM AUTHOR]

• Spatially examines the relationships between crash frequency and various factors, including demographics, traffic exposure, and roadway characteristics. • MGWR outperforms OLS and GWR in identifying spatial relationships between contributing factors and crash frequencies. • Regions with more low-wage workers and higher employment and household entropy tend to have increased crash frequencies. Understanding the relationship between social disparities and traffic crash frequency is essential for long-term transportation planning and policymaking. Few studies have systemically examined the influence of socioeconomic and infrastructure-related disparities in macro-level traffic crash frequency. This study provides a framework to spatially examine the relationships between crash rates and demographic and socioeconomic characteristics, as well as roadway infrastructure and traffic characteristics at the Census Block Groups (CBGs) level. Spatial autocorrelation analysis was first performed on the residual of the Ordinary Least Squares (OLS) model to identify whether non-stationarity exists. Then, the Geographically Weighted Regression (GWR) model and the Multiscale Geographically Weighted Regression (MGWR) model were applied to assess the impacts of these factors on crash rates spatially and statistically. Our findings indicate that MGWR outperforms both OLS and GWR in uncovering the spatial relationships between contributing factors and both fatal and injury (FI) crashes as well as property damage only (PDO) crashes. A thorough examination of local coefficient maps highlighted six pivotal variables that significantly influenced a majority of CBGs. Improving infrastructure, including pedestrian pathways and public transit facilities, in low-income areas can offer significant benefits. These findings and recommendations can inform the development of effective strategies for reducing crashes and guide the appropriate selection of modeling techniques for macro-level crash analysis. [ABSTRACT FROM AUTHOR]

In recent years, car-sharing services have emerged as viable alternatives to private individual mobility, promising more sustainable and resource-efficient, but still comfortable transportation. Research on short-term prediction and optimization methods has improved operations and fleet control of car-sharing services; however, long-term projections and spatial analysis are sparse in the literature. We propose to analyze the average monthly demand in a station-based car-sharing service with spatially-aware learning algorithms that offer high predictive performance as well as interpretability. Our study utilizes a rich set of socio-demographic, location-based (e.g., POIs), and car-sharing-specific features as input, extracted from a large proprietary car-sharing dataset and publicly available datasets. We first compare the performance of different modeling approaches and find that a global Random Forest with geo-coordinates as part of input features achieves the highest predictive performance with an R-squared score of 0.87 on test data. While a local linear model, Geographically Weighted Regression, performs almost on par in terms of out-of-sample prediction accuracy. We further leverage the models to identify spatial and socio-demographic drivers of car-sharing demand. An analysis of the Random Forest via SHAP values, as well as the coefficients of GWR and MGWR models, reveals that besides population density and the car-sharing supply, other spatial features such as surrounding POIs play a major role. In addition, MGWR yields exciting insights into the multiscale heterogeneous spatial distributions of factors influencing car-sharing behaviour. Together, our study offers insights for selecting effective and interpretable methods for diagnosing and planning the placement of car-sharing stations. • Geographical analysis of long-term car sharing demand with spatially-explicit regression models • Local and global models achieve high predictive performance when considering spatial features and sociodemographics • Geographically Weighted Regression analysis and SHAP value analysis foster spatial interpretability • Case study on utilizing the methods as a decision support system for car-sharing station planning [ABSTRACT FROM AUTHOR]

[Display omitted] • Between 1980 and 2018, greenspaces decreased, yet the river system still developed. • The greenspaces exhibit a trend towards fragmentation and complexity. • Changes in surface runoff and groundwater flow varied greatly in space. • Water yield is affected by both relief ratio and forest land morphology. • The Geographically Weighted Regression has high accuracy in water yield simulation. The rapid shifts in landuse and the frequency of flooding events brought by urbanization significantly impact the urban basins, which means that optimizing landscape patterns is crucial for improving urban hydrology. Therefore, the Liuxi River Basin (LRB), an important river in the Pearl River Delta region, was utilized as an example in the study to examine how the landscape configuration of greenspaces and water bodies influences the hydrological components. However, there are few studies on landscape hydrology in the Pearl River Delta (PRD) and the spatial linkages between landscape metrics and hydrological components are not clear. Based on this, this paper aims to explore the mathematical and spatial associations between landscape configuration and hydrological variables in the PRD region. The SWAT model was constructed to simulate hydrological components based on landuse changes from 1980 to 2018, and regressions were established for landscape pattern indices, water morphology parameters, and hydrological components using stepwise multiple linear regression and geographically weighted regression (GWR). According to the results, the LRB continued to develop, although the greenspaces spread out and changed shape considerably. The influence of water bodies on the hydrological component is mainly in evapotranspiration, and some three-dimensional parameters such as relief ratio favor surface runoff. In the case of greenspace, the runoff is less in connected and wide farmland areas. Irregular forest land will promote water yield but suppress evapotranspiration, and the effect of forest's shape on hydrological components is more obvious in upstream areas because the regression coefficients in this range are higher, such as 90.27. Methodologically, we found that for water yield, the accuracy of GWR was significantly higher (R2 = 0.814) than that of multiple linear regression (R2 = 0.243). In addition, understanding land use changes before conducting landscape hydrology studies can help make the results more interpretable. We supplemented the relevant research gaps in the PRD region and revealed the spatial correlation between landscape metrics and the hydrological cycle. These findings can aid urban planners in improving their comprehension of hydrological processes and landscape design, leading to more sustainable management of the urban basins. [ABSTRACT FROM AUTHOR]

In recent years, the biogeochemical behavior and environmental impact of Selenium (Se) on soil-plant systems have received widespread attention, and traditional statistical methods reveal generally positive correlations between rice Se and soil Se. However, that initial positive relationship may have been obscured by local external factors. Using local scale data from the geochemical evaluation of land quality project, this work employed geographically weighted regression (GWR) to examine the spatial variation of rice Se (as the dependent variable) and soil Se (as the independent variable) in Guangxi. Strong and weak correlation coefficients occur between rice Se and soil Se, thereby indicating that their relationships are spatially varying. Guangxi is characterized by significantly positive correlations in most areas, with weak correlations mostly found in the south-western and central-eastern regions. Areas with weak correlation can be divided into two patterns: high soil Se with low rice Se and high rice Se with low soil Se. The unique patterns are correlated with distinct natural factors, particularly the abundance of Fe-rich soils in the carbonate area; by contrast, sandstone areas in central Guangxi may have been affected by anthropogenic activities. To reveal the spatially varying relationships at the local scale, we employed GWR, an effective tool that allowed us to identify the association between environmental variables and influencing factors and explore spatially varying relationships between them. This study breaks through the existing understanding that soil Se is completely positively correlated with rice Se for the first time, and concludes that their correlation is spatially variable, providing an effective approach for the study of complex relationships. [Display omitted] • Spatially varying relationships between rice Se and soil Se are revealed by GWR. • Weak correlations are observed in north-western and central-eastern Guangxi. • Fe-enriched soil is the main influencing factor of the weak relationship. • The utility of Se-enriched land is proposed based on the spatial relationships. [ABSTRACT FROM AUTHOR]

Poisson and Negative Binomial Regression Models are often used to describe the relationship between a count dependent variable and a set of independent variables. However, these models fail to analyze data with an excess of zeros, being Zero-Inflated Poisson (ZIP) and Zero-Inflated Negative Binomial (ZINB) models the most appropriate to fit this kind of data. To Incorporate the spatial dimension into the count data models, Geographically Weighted Poisson Regression (GWPR), Geographically Weighted Negative Binomial Regression (GWNBR) and Geographically Weighted Zero-Inflated Poisson Regression (GWZIPR) have been developed, but the zero-inflation part of the negative binomial distribution is undeveloped in order to incorporate the overdispersion and the excess of zeros, as was at the beginning of the COVID-19 pandemic, whereas some places were having an outbreak of cases and in others places, there were no cases yet. Therefore, we propose a Geographically Weighted Zero-Inflated Negative Binomial Regression (GWZINBR) model which can be considered a general case for count data, since locally it can become a GWZIPR, GWNBR or a GWPR model. We applied this model to simulated data and to the cases of COVID-19 in South Korea at the beginning of the pandemic in 2020 and the results showed a better understanding of the phenomenon compared to the GWNBR model. [ABSTRACT FROM AUTHOR]

• This study considers time of the day in self-report seat belt use. • This study considers heterogeneity in the analysis. • local coefficients displayed relatively constant variation across space. • GWR model outperformed random parameter and and binary logit model. Time of day and heterogeneity are two common factors that received less attention in self-reported studies concerning seat belt use. Although nighttime seat belt use rate is relatively lower than daytime, previous studies have often not separated the analysis of seat belt use between daytime and nighttime. To incorporate heterogeneity in our analysis, we used random parameters and geographically weighted regression models to explore the factors influencing seat belt use. This study consists of a self-reported seat belt use survey conducted in a sample of 814 respondents aged 18–50 years in six counties (50 zip codes area) in East Tennessee. Comparison of the models indicated that the geographically weighted regression models outperformed other models. Considering the non-stationary test, we learned that the local coefficients displayed relatively constant variation across space in the study area, which indicates behaviors, at least across a large metropolitan area, does not vary spatially. For the random parameter models, age and income had random parameter effects. Perception of receiving a ticket for not wearing a seat belt, uncomfortable seat belt design, driving for a short distance, and driving exposure also had significant negative associations with self-reported seat belt use in both models. Moreover, exposure to educational programs had a significant correlation with seat belt use only in nighttime, whereas this correlation was insignificant for daytime. Findings provide new insight for design and convoy new messages to promote seat belt use by targeting factors predicting seat belt use. Results are discussed in line with road safety analysis. [ABSTRACT FROM AUTHOR]

• Dynamics of net primary productivity in Inner Mongolia show spatial heterogeneity. • Spatial statistics and ADMA are combined to study NPP clustering patterns. • NPP correlates negatively with temperature, but positively with precipitation. • Negative spatial association in NPP is likely associated with human activities. Vegetation Net Primary Productivity (NPP) is an important indicator for agriculture production. Understanding spatio-temporal dynamics of NPP and their driving factors have attracted much attention from both academic field and practical applications. In this paper, we coupled spatial statistics and a new approach called accumulated density map analysis (ADMA) to explore spatio-temporal variations in NPP distribution and possible contributing factors to the variations in Inner Mongolia Autonomous Region (IMAR), China. Density of Spatio-temporally Aggregated Clustering (D-STAC) index of NPP distribution, as output from ADMA, was proposed to indicate the impact of local factors on NPP. The study showed that spatially averaged NPP did not exhibit significant changes over time, but inter-annual variability of NPP presented critical spatial heterogeneity. Local spatial association analysis, as a preliminary step for ADMA analysis, detected two positive autocorrelation patterns, namely H-H (high NPP enclosed by high NPP) and L-L (low NPP enclosed by low NPP), and two negative autocorrelation patterns, including H-L (high NPP surrounded by low NPP) and L-H (low NPP surrounded by high NPP), for localized places in the study area. While positive autocorrelation patterns were found to dominate most parts of the study area, D-STAC for negative autocorrelation patterns was closely associated with Neighborhood to Cities (NC), an index for urbanization level. To evaluate the relationship between the NPP variation and possible driving factors, local regression analysis using geographically weighted regression (GWR) revealed that NPP largely had positive correlation with precipitation, but showed substantial spatial variations in the relationships with temperature and NC. We concluded that, through LISA, ADMA and GWR, the associations between the spatio-temporal NPP variations and their driving factors could be examined under localized context. [ABSTRACT FROM AUTHOR]

• HBA attribute crash to the home-address of traffic victim. • We explored comprehensive cost of traffic crashes at the census tract level (CCCPC). • GWPR model was used to evaluate factors influencing CCCPC. • Increase in population density and non-motorized transportation reduce CCCPC. Global road safety records demonstrate spatial variation of comprehensive cost of traffic crashes across countries. To the best of our knowledge, no study has explored the variation of this matter at a local geographical level. This study proposes a method to estimate the comprehensive crash cost at the zonal level by using person-injury cost. The current metric of road safety attributes safety to the location of the crash, which makes it challenging to assign the crash cost to home-location of the individuals who were involved in traffic crashes. To overcome this limitation, we defined Home-Based Approach crash frequency as the expected number of crashes by severity that road users who live in a certain geographic area have during a specified period. Using crash data from Tennessee, we assign those involved in traffic crashes to the census tract corresponding to their home address. The average Comprehensive Crash Cost at the Zonal Level (CCCAZ) for the period of the study was $18.2 million (2018 dollars). Poisson and Geographically Weighted Poisson Regression (GWPR) models were used to analyzing the data. The GWPR model was more suitable compared to the global model to address spatial heterogeneity. Findings indicate population of people over 60-years-old, the proportion of residents that use non-motorized transportation, household income, population density, household size, and metropolitan indicator have a negative association with CCCAZ. Alternatively, VMT, vehicle per capita, percent educated over 25-year-old, population under 16-year-old, and proportion of non-white races and individuals who use a motorcycle as their commute mode have a positive association with CCCAZ. Findings are discussed in line with road safety literature. [ABSTRACT FROM AUTHOR]

Cities experience the urban heat island (UHI), which continue to pose challenges for humanity's increasingly urban population. Past research has revealed that land cover composition and configuration, along with other geographical phenomena (i.e., albedo), can explain much of the spatial pattern of UHI, yet advances await. In response, this research was made to: (i) assess the spatial pattern of mean ambient night temperature across 34 streetscapes in New York City (NYC); (ii) create and differentiate global and local regression models between-natural and built streetscape characteristics- and mean ambient night temperature; and (iii) use geographically weighted regression (GWR) to assess local patterns of correlated associations. Urban canopy layer (UCL) temperatures were recorded across 34 weather stations, and landscape metrics calculated from 0.914 m land cover data with 96% accuracy. Local Getis-Ord Gi* statistic exhibited significant spatial cold and hot spots of UHI in NYC. Global inferential tests revealed that sky-view factor, photosynthesis activity, elevation, and road configuration were the strongest predictors of mean ambient night temperature. Six multiple regression models were ultimately made with GWR fitting the UHI aptly (R 2 = 65–74%). Important explanatory covariates were illustrated using local pseudo- t statistics and linked to mean ambient night temperature, supporting the importance of GWR for understanding local UHI interactions. Results also confirm that landscape configuration metrics are stronger predictors of UHI than composition measures. Streetscape design, particularly road patterns and process, requires more consideration when attempting to mitigate UHI during future sustainability planning, urban renewal projects, and research. • New York City's urban heat island (UHI) was illustrated using Getis-Ord Gi* test. • Over 90 streetscape variables were calculated to examine their link to NYC's UHI. • Sky-view factor was the single best predictor of NYC's mean night temperature. • Land cover configuration was better than composition at modeling NYC's UHI. • Road configuration plays a larger role in the UHI than previously understood. [ABSTRACT FROM AUTHOR]

Global warming and climate change have become a serious environmental problem and China's carbon emissions are currently the highest in the world. Cities are the main sources of carbon emissions and the key to solving these problems. Therefore, research on reducing carbon dioxide emissions is a matter of concern. In this study, a spatial autocorrelation analysis was performed to understand the spatial characteristics of carbon dioxide emissions in 171 Chinese cities. Then, stepwise and geographically weighted regressions were used to explore the processes that drive carbon dioxide emissions in Chinese cities. A two-step cluster was used to classify Chinese cities into different categories based on the degree of impact of each driver. The results showed that there is a spatial aggregation relationship between urban carbon dioxide emissions. High-high clusters mainly occur in the Beijing-Tianjin-Hebei and Yangtze River Delta urban agglomerations, while low-low clusters occur in the central, western, and southwestern cities. Among all variables, freight volume, per capita gross domestic product, population density, and the proportion of secondary industries correlate positively with carbon dioxide emissions, whereas the number of buses per 10,000 people correlates negatively with carbon dioxide emissions. The geographically weighted regression model provided more detailed results and revealed the spatial heterogeneity of the effects of the different drivers. The impact of population, economic factors, and industrial factors in the eastern region is significantly greater than that in the central and western regions. Freight volume and public transport have the most significant impact in the northeast region. The clustering results showed that cities can be divided into four types. These findings provide a reference and policy suggestions for how cities in different regions should reduce carbon dioxide emissions. Four types of cities obtained by clustering of the GWR coefficients. (a) The distribution of clustering results and (b) the clustering results for the GWR coefficients. Unlabelled Image • Chinese cities showed significant spatial correlations in carbon dioxide emissions. • Geographically weighted regression indicated a regional variation in driving factors. • Cities were divided into four categories based on the coefficients of regression. • Targeted policy suggestions were presented for four types of cities. [ABSTRACT FROM AUTHOR]

Bathymetric mapping is a prerequisite procedure to conduct assessments of water quality, habitat and environmental flow for riverine ecosystems using hydraulic modelling. This study evaluates the capability of a geographically weighted regression (GWR) model, which can capture a spatially heterogeneous relationship between inputs and an output, to retrieve bathymetry of a shallow stream, of which water depth is less than about 1 m from simple RGB imagery. A field experiment was performed for measuring water depth and simultaneously for acquiring remotely-sensed data with RGB digital numbers (DN) using a digital camera mounted on an unmanned aerial vehicle (UAV). A 2D shallow water model, which was validated by comparison with the field-surveyed data, was used to simulate the water depth of unmeasured regions. Band ratios of ln(DN G / DN R) was selected as an optimal spectral input of bathymetric inversion models through the principal component analysis (PCA). Results showed that global inversion models based on multiple linear regression (MLR) and artificial neural network (ANN) resulted in large discrepancy between estimation and observation due to the spatially varying response of the PCA-selected band ratio to water depth over the experimental channel. In contrast, the GWR model successfully alleviated the biases of the conventional models as R2 increased to 0.85 from 0.60 by accurately modelling the effect of spatial heterogeneity, which arose from variable bottom types attributed to submerged vegetation, on the remote-sensing radiance-water depth relationship. • Shallow stream bathymetry was derived from UAV-aided RGB imagery using a GWR model. • Response of water depth to a band ratio was heterogeneous due to bottom reflection. • The GWR model enhanced the estimation accuracy, resolving the heterogeneity effect. [ABSTRACT FROM AUTHOR]