Geographically Weighted Research Articles

Detecting Spatially Non-Stationary between Vegetation and Related Factors in the Yellow River Basin from 1986 to 2021 Using Multiscale Geographically Weighted Regression Based on Landsat

As an important ecological barrier in northern China, the ecological environment of the Yellow River Basin (YRB) has been greatly improved in recent decades. However, due to spatially non-stationarity, the contribution of human activities and natural factors to vegetation restoration may exhibit different coupling effects in various areas. In this paper, the Normalized Difference Vegetation Index (NDVI) of the YRB from 1986 to 2021 was used as the dependent variable, and terrain, meteorological, and socioeconomic factors were used as independent variables. With the help of Multiscale Geographically Weighted Regression (MGWR), which could handle the scale difference well, combined with Ordinary Least Squares (OLS) and traditional Geographically Weighted Regression (GWR), the spatial non-stationary relationship between vegetation and related factors was discussed. The results showed that: (1) The vegetation was subject to fluctuating changes from 1986 to 2021, mainly improving, with a growth rate of 0.0018/year; the spatial distribution pattern of vegetation in the basin was high in the southeast and low in the northwest. (2) Compared with the OLS and GWR, the MGWR could better explain the relationship between vegetation and various factors. (3) The response scale of vegetation and related factors was significantly variant, and this scale changed with time. The effect scale of terrain factor is lower than climate and social factors. (4) There was obvious spatial heterogeneity in the effects of various influencing factors on vegetation. The vegetation of the upstream was mainly positively affected by mean annual temperature (coefficients ∈ [1.507, 1.784]); while potential evapotranspiration was the dominant factor of vegetation in the middle and lower reaches of the basin (coefficients ∈ [−1.724, −1.704]); it was worth noting that the influence of social factors on vegetation was relatively small. This study deeply explores the spatial non-stationarity of vegetation and various related factors, thereby revealing the evolution law of vegetation pattern and providing scientific support for monitoring and improving the ecological environment quality of the YRB.

Analysis of the energy justice in natural gas distribution with Multiscale Geographically Weighted Regression (MGWR)

Energy justice is violated when particular customers and locations are excluded from a variety of urban energy service distribution. This study explores energy justice in terms of natural gas distribution by providing empirical evidence from 356 neighborhoods of Izmir Metropolitan Area (IMA). The aim is to reveal driving factors of natural gas investment and the spatial reflections of the relationships between investment, and socio-economic and physical characteristics. A global regression model, OLS, and two local spatial regression models, GWR and MGWR, are conducted. Population, income, employment and disadvantaged areas are the significant determinants of natural gas investments. Due to the presence of spatial autocorrelation in OLS residuals, GWR and MGWR are utilized to account for spatial variation in the response variable. Local models are superior to the global model according to AICc values, which are 607.3, 586.7, and 558.5, in OLS, GWR, and MGWR, respectively. MGWR further improves the overall fit with higher R-sq and lower AICc values. The local R-sq values indicate at least 70% variability is explained in 85% of the study area in MGWR, and in 73% of IMA in GWR. Parameters are slightly overestimated in GWR at the mean level. None of the local models are subject to multicollinearity according to local condition numbers, but local variance decomposition proportions indicate the effects of multicollinearity in some observations. Spatial modeling of investments helps to demonstrate local variations of energy injustice and to develop site-specific policies. Population and employment are related to potential customers, which lead to higher investments. Income depends on purchasing power, and there are economic barriers that need to be regulated through subsidies and incentives for the low-income households. Awareness raising policies can also be developed to better inform households about energy alternatives. Disadvantaged areas, either declared as urban transformation areas or currently under urban transformation, lack the natural gas service in general, while the investments can be considered in those areas along with the new development plans.

Spatio-temporal analysis of the COVID-19 pandemic in Iran

Globally, the COVID-19 pandemic is a top-level public health concern. This paper is an attempt to identify and COVID-19 pandemic in Iran using spatial analysis approaches. This study was based on secondary data of confirmed cases, deaths, recoveries, number of hospitals, hospital beds and population from March 2, 2019 to the end of November 2021 in 31 provinces of Iran from hospitals and the website of the National Institute of Health. In this paper, three geographical models in ArcGIS10.3 were utilized to analyze and evaluate COVID-19, including Geographic Weight Regression (GWR), Getis-OrdGi* (G-i-star) statistics (hot and cold spot), and Moran autocorrelation spatial analysis. Moran statistics, based on the GWR model, demonstrated that deaths and recoveries followed a clustering pattern for the confirmed cases index during the study period. The Moran Z-score for all three indicators confirmed cases, deaths, and recoveries, which was greater than 2.5 (95% confidence level). The Getis-OrdGi* (G-I-Star) (hot and cold spot) data revealed a wide range of levels for six variables (confirmed cases, deaths, recoveries, population, hospital beds, and hospital) across Iran's provinces. The overall number of deaths exceeded the population and the number of hospitals in the central and southern regions, including the provinces of Qom, Alborz, Tehran, Markazi, Isfahan, Razavi Khorasan, East Azerbaijan, Fars, and Yazd, which had the largest number and The Z-score for the deaths Index is greater than 14.314. The results of this research can pave the way for future studies.

Effects of landscape pattern change on ecosystem services and its interactions in karst cities: A case study of Guiyang City in China

Exploring the effects of landscape pattern changes on ecosystem services (ESs) and its interactions could provide a new method to reduce the trade-offs between ESs from the perspective of landscape pattern optimization, thus contributing to the sustainable management of ESs. In this study, the Guiyang City which is a typical karst mountainous city in China was used as the study region, using InVEST model, Spearman’s correlation and Multiscale Geographically Weighted Regression (MGWR) model, the study assessed the changes and trade-offs of four key types of ESs (i.e., carbon storage (CS), habitat quality (HQ), soil conservation (SC), water yield (WY)) from 1995 to 2019, and analyzed the impacts of landscape pattern on the synergies of ESs. The results showed that ESs in Guiyang City steadily increased during 1995–2019, with CS, HQ, SC, WY, and total ecosystem services (TES) increased by 6.67%, 7.66%, 27.16%, 13.08%, and 11.81%, respectively. The relationship of CS-SC and SC-WY were mainly trade-offs, while the other ESs types were mainly synergies. Moreover, the interaction of static ESs was reversed with that of dynamic spatiotemporal ESs when the lower interaction relationship between static ESs. The effects of landscape pattern changes on ESs and ESs interactions have obvious spatial and scale heterogeneity. Landscape composition had a greater impact on ESs than landscape configuration. Proportion of woodland (Wood_per) and Shannon’s Diversity Index (SHDI) were the main factors influencing the synergies effect of TES. The results could provide practical references for ESs improvement-oriented landscape pattern planning and ecosystem service management in the Guiyang City and in similar fragile ecological regions, which is helpful for the improvement of ESs and high-quality sustainable development.

A Rejoinder to the Commentaries on “A Route Map for Successful Applications of Geographically Weighted Regression” by Comber et al. (2022)

A Rejoinder to the Commentaries on “A Route Map for Successful Applications of Geographically Weighted Regression” by Comber et al. (2022)

Urban vulnerability to floods investigation in jakarta, Indonesia: A hybrid optimized fuzzy spatial clustering and news media analysis approach

Jakarta, known as the capital city of Indonesia and the national economic activity center, is also known as the market for floods. Moreover, the regions of Jakarta become vulnerable, especially the slum areas and this COVID-19 pandemic. This study proposes implementing Fuzzy Geographically Weighted Clustering (FGWC) with Chaotic Flower Pollination Algorithm (CFPA) in urban vulnerability analysis of Jakarta at the subdistrict level. Moreover, this study also implements the named entity recognition (NER) from local news media to extract the history of floods in Jakarta during 2019–2020 to support the urban vulnerability. This study found that the CFPA can improve the performance of FGWC. Subsequently, the clustering results showed that each cluster has its urban vulnerability problem. Most regions outside central Jakarta have more vulnerability problems, particularly in population dynamics, floods history, and slum areas. On the other hand, the news analysis showed that the most frequent flood happened between October and February. Therefore, the government and society must collaborates to overcome the urban vulnerability problems and mitigates the floods to increase the regional resilience of Jakarta.

Multi-Dimensional Threshold Effects of the Digital Economy on Green Economic Growth?—New Evidence from China

China’s economy has shifted from a high-speed growth stage to a high-quality development stage. Achieving green and sustainable growth driven by the digital economy is its most important purpose. Based on provincial panel data from 2013 to 2019, this paper examines the spatial heterogeneity of the digital economy (Dig) and its nonlinear impact on green economic growth (Geg) from multiple dimensions, using threshold models and spatiotemporal geographic weighting models. The study found that: (1) Dig can significantly promote Geg, but it presents a typical nonlinear characteristic: double thresholds with a trend of “weak negative → strong positive → weak positive”; (2) in this process, the three critical nonlinear factors are financial agglomeration (Fina) with a single threshold of “0.186” and the trend of “weak negative → strong positive”, the marketization level (Market) with double thresholds of “4.448 and 9.06” and the trend of “weak negative → strong positive → positive”, and green technology innovation (Grti) with double thresholds of “5.236 and 9.152” and the trend of “strong negative → strong positive → positive”; (3) the multi-dimensional composite threshold effect indicates that when 0.061 < Dig < 0.539, 5.236 < Grti < 9.152, 4.448 < Market < 9.06, and 0.186 < Fina, the digital economy has the greatest effect on the promotion of green economic growth; and (4) the spatial distribution regression results show that there is obvious regional heterogeneity. The paper has reference value in maximizing the promotion effect of the digital economy on green economic growth.

Spatial impact of the built environment on street vitality: A case study of the Tianhe District, Guangzhou

Taking the Tianhe District in Guangzhou, China, as a case and the urban street (road) as the basic research unit, this study analyzes the relationship between the built environment and street vitality to analyze the influencing factors of street vitality. The Tencent location service data are used to characterize street vitality, and the OLS and GWR models are used to construct the statistical relationship between the built environment and street vitality after establishing the urban built environment index. The results show that spatial heterogeneity is considered in the GWR model based on local geographic weighting, and its fitting effect is better than that of the OLS model, which can reveal the micro-local characteristics of the built environment’s effect on street vitality. Increasing the land-use mixing degree, new and old building mixing degrees, and land-use intensity (building density and volume ratio) can significantly increase street vitality, which proves to a certain extent that Jacobs’s relevant discussion is still highly practical for the Tianhe District.

Drivers of spatiotemporal disparities in the supply-demand budget of ecosystem services: A case study in the Beijing-Tianjin-Hebei urban agglomeration, China

Assessing the spatiotemporal patterns of ecosystem services (ESs) supply and demand, as well as the drivers thereof during specific time periods, is critical for regional policy making and sustainable management. Taking the Beijing-Tianjin-Hebei (BTH) urban agglomeration of China as an example, we studied four ES supply-demand budgets: carbon storage, water provision, food provision and soil conservation from 2000 to 2015. Through the geodetector model, canonical analysis and Multiscale Geographically Weighted Regression (MGWR) model, the drivers of the ES supply-demand budget were explored. The results showed that the areas supplying high amounts of ESs in the northern region usually did not overlap those areas consuming intensive ESs, which were mainly distributed in metropolitan areas. The anthropological factors, including per capita gross domestic product (Per.GDP) and population density (POP.Den), were the dominant influencing factors for the imbalance between the supply and demand of carbon storage, water provision and food provision, which were mainly distributed in the central and southern regions of the study area. Geomorphological factors (ELE and SLO) were the key driving factors of soil conservation, which was mainly distributed in the eastern regions. In all, our findings could provide comprehensive information for decision-making and ES management.

Transit Ridership Modeling at the Bus Stop Level: Comparison of Approaches Focusing on Count and Spatially Dependent Data

Boarding and alighting modeling at the bus stop level is an important tool for operational planning of public transport systems, in addition to contributing to transit-oriented development. The interest variables, in this case, present two particularities that strongly influence the performance of proposed estimates: they demonstrate spatial dependence and are count data. Moreover, in most cases, these data are not easy to collect. Thus, the present study proposes a comparison of approaches for transit ridership modeling at the bus stop level, applying linear, Poisson, Geographically Weighted and Geographically Weighted Poisson (GWPR) regressions, as well as Universal Kriging (UK), to the boarding and alighting data along a bus line in the city of São Paulo, Brazil. The results from goodness-of-fit measures confirmed the assumption that adding asymmetry and spatial autocorrelation, isolated and together, to the transportation demand modeling, contributes to a gradual improvement in the estimates, highlighting the GWPR and UK spatial estimation techniques. Moreover, the spatially varying relationships between the variables of interest (boardings and alightings) and their predictors (land use and transport system features around the bus stops), shown in the present study, may support land use policies toward transit-oriented development. In addition, by using an approach with little information, the good results achieved proved that satisfactory boarding and alighting modeling can be done in regions where there is a lack of travel demand data, as in the case of emerging countries.

Investigating Accessibility of Hospitals in Cold Regions: A Case Study of Harbin in China.

The objective of this study is to explore healthcare resource accessibility in Harbin, a typical city in a cold region in China. Recently, investments in the construction of medical resources have been increasing annually in China, and consequently, the allocation of these resources has improved. Snow and ice on surfaces in China's cold regions have certain effects on the traffic capacity of urban roads, leading to a great difference in the accessibility of medical resources in winter and summer. The basic spatial data, including spatial road data, medical facility data, and population distribution data, are analyzed using geographic information system. Then, a spatial barrier model is used to measure healthcare accessibility based on geographic and population weighting; we explore the accessibility of hospitals under the influence of weather by defining a novel distance attenuation function. Finally, the accessibility of medical institutions in the study area is explored by analyzing data about the related separation factors. It was found that the spatial distribution of medical resources was not equal, and the dominant resources were concentrated in the city center. Some regions are always in an advantageous position regardless of traffic conditions. In contrast, in areas far from the city center, the accessibility of medical resources significantly decreases in winter. These results will help optimize the layout of medical institutions and improve medical equality and propose strategies for the optimization of the accessibility of urban medical institutions in cold regions of China.

Monitoring spatiotemporal characteristics of land-use carbon emissions and their driving mechanisms in the Yellow River Delta: A grid-scale analysis

Comprehensive and accurate grasp of land-use carbon emissions (LCE) level and its driving mechanism is key to success in China's pursuit of low-carbon development, and it is also the scientific basis for the formulation and implementation of regional carbon emissions strategies. Based on fossil fuel carbon emissions raster data (published by the Open-Data Inventory for Anthropogenic Carbon dioxide (ODIAC) platform) and land use data, this manuscript selects the Yellow River Delta as the study area and uses an improved LCE measurement model, exploratory spatial data analysis, multiscale geographical weighting regression (MGWR), and other models to explore the spatiotemporal heterogeneity and driving mechanisms of LCE at the grid level. The results showed the following: ① The total amount of LCE in the study area continued to increase from 2000 to 2019, the growth rate decreased, but the peak of LCE had not yet been reached. ② The LCE of the study area showed a significant positive global autocorrelation. The H–H aggregation region showed a relatively stable spatial distribution range; the L-L aggregation region showed wide distribution characteristics that covered the entire study area; and the aggregation regions of H-L and L-H, which have not yet reached the scale. ③ At the global dimension, the mean correlation coefficients between LCE and driving factors (net primary productivity (NPP), nighttime light (NTL), and population density (PD)) from 2000 to 2019 were −0.11, 0.28, and 0.12; at the local dimension, the strength (from strong to weak) of the effect of each factor on LCE was PD, NTL, NPP (2000) and NTL, PD, NPP (2019). The research results provide a scientific basis and basic guarantee for the development, and implementation of regional carbon emission strategies.

Spatial interdependence and spillovers of fiscal grants in Benin: Static and dynamic diffusions

This paper investigates the spatial diffusion of an intergovernmental grant in Benin. Using static and dynamic spatial models, we estimate the spillover effects of the Fonds d’Appui au Développement des Communes (FADeC) on per capita local government expenditure in the 77 municipalities from 2008 to 2015. Neighborliness – a measure of interdependence – is captured through geographic and distance-based spatial weighting schemes. In addition, we constructed a measurement of ethnic affinity as an alternative spatial weighting scheme to test for the existence of an ethno-spatial interdependence in local public finance in Benin. The empirical results suggest that a statistically significant share of the total effects of the FADeC stems from indirect elasticities or the diffusion process of grants received by neighboring jurisdictions, regardless of how we measure neighborliness. The results also confirm the existence of a robust ethno-spatial interdependence and complementarity in local government expenditure in Benin. The spillovers across ethnic neighbors are estimated to be 13.9% of the total effects in the short-run and 15.5% in the long-run. Put differently, the effects of the FADeC in a given municipality are influenced by the transfers received by its ethnic (and linguistic) neighbors. The findings point to the appeal of inter-governmental transfers for the decentralized financing of public services, especially in low-income countries where local bureaucratic capacity in raising own-source revenues might be limited. Supporting local governments with well-structured grants can not only be a channel to foster local public provision but also contribute to pushing geographic or ethnic neighboring localities to increase their own spending and generate positive spillovers that are crucial for wholistic regional development.

Soil Erosion Process Simulation and Factor Analysis of Jihe Basin

Soil erosion is a notable contributor to carbon emissions. Distributed erosion model can be used to study erosion distribution in different land cover, identify the main influential factors, and hence guide soil and water conservation. In this study, Regional Soil Erosion model (RSEM) was used to simulate the soil erosion processes of Jihe Basin in 2015, and the Multiscale Geographically Weighted Regression (MGWR) implementation was applied to compare the erosion regression of Geographically Weighted Regression (GWR) and MGWR as well as study the impact of influential factors on sediment modulus in hillslopes. The results are as follows: (1) MGWR results indicated slope was the dominant factors affecting soil erosion at the catchment scales, where the average coefficients of slope, forest coverage, and grass coverage descended in the value of 0.90, −0.11, and −0.19, and the influences of factors operate over scales; (2) MGWR with the adoptive bandwidths performed well in the goodness of fit, t-test of variables, scales that variables operate, and interactive interpretation of soil erosion; (3) the coupling effects and scales of vegetation and topography factors are an important approach to study soil erosion at a larger scale.

Spatial change and scale dependence of built-up land expansion and landscape pattern evolution—Case study of affected area of the lower Yellow River

Land use change are essential for sustaining human wellbeing, but empirical studies have shown that the urban development requires built-up land, yet, the disorderly expansion of built-up land poses a threat to sustaining the stability of landscape patterns. Reconciling these seemingly conflicting results requires better understanding how the built-up land expansion and the landscape patterns change relationship varies with space and scales. Here we systematically analyzed the landscape pattern and expansion mode of built-up land in the affected area of the Lower Yellow River on three spatial scales (i.e., 5 km, 10 km, 15 km), using landscape pattern analysis, the landscape expansion index and the Multiscale Geographically Weighted Regression (MGWR) model. Our results indicated that: (1) infilling has been the dominant mode of built-up land expansion, and the spreading tendency become more obvious with a decrease in scale; (2) Rapid expansion of built-up land had led to the significant spatial differences in landscape fragmentation and agglomeration at different scales and the complexity of the landscape shape gradually increased; (3) the MGWR model can explain the relationship between the built-up land expansion and landscape pattern more than the Ordinary Least Squares model (OLS), and shows that the dependence of the two is increasing with the increase of scale, Further, the effect of control variables on the landscape pattern evolution differed. This study provides a valuable reference for sustainable development in other large river basins experiencing similar ecological landscape pattern changes (landscape elements, spatial locations and connections that are critical to the health and security of ecological processes).

DEVELOPING GEOGRAPHICALLY WEIGHTED PANEL REGRESSION MODEL FOR SPATIO-TEMPORAL ANALYSIS OF COVID-19 POSITIVE CASES IN KALIMANTAN, INDONESIA

This article describes spatial-temporal analysis using the innovation of developing a Geographically Weighted Panel Regression model with a distance weighting function that includes the interaction between spatial and time aspects (GWPR-st). The method is a local regression technique that provides a parameter model that varies in each location through cross-sectional and time-series data observation units. This study develops a new model in spatial statistics and offers new methodologies in Geographic Models and Geographic Information Systems (GIS). This study aims to determine the factors that influence the increase in positive cases and map the spread of COVID-19 on the Kalimantan Regency/City Scale. The model applied in this study involves geographic weighting functions, including the Gaussian kernel, Bisquare kernel, and tricube kernel, which spatial interactions and time series have modified. This study uses national COVID-19 data from 56 regencies/cities until August 2021. According to the research results, the developed model with the geographic weighting of the Bisquare kernel function was considered the most acceptable method. The developed model, which is deemed capable of information the most substantial influence on the number of COVID-19 cases in Kalimantan, is health services, such as a shortage of doctors, number of hospitals, number of community health centers, and number of tuberculosis cases. The study results provide the local governments with decision-making recommendations for overcoming the COVID-19 problems in their regions.

Gwverse: A Template for a New Generic Geographically Weighted R Package

GWR is a popular approach for investigating the spatial variation in relationships between response and predictor variables, and critically for investigating and understanding process spatial heterogeneity. The geographically weighted (GW) framework is increasingly used to accommodate different types of models and analyses, reflecting a wider desire to explore spatial variation in model parameters and outputs. However, the growth in the use of GWR and different GW models has only been partially supported by package development in both R and Python, the major coding environments for spatial analysis. The result is that refinements have been inconsistently included within GWR and GW functions in any given package. This paper outlines the structure of a new gwverse package, that may over time replace GWmodel, that takes advantage of recent developments in the composition of complex, integrated packages. It conceptualizes gwverse as having a modular structure, that separates core GW functionality and applications such as GWR. It adopts a function factory approach, in which bespoke functions are created and returned to the user based on user‐defined parameters. The paper introduces two demonstrator modules that can be used to undertake GWR and identifies a number of key considerations and next steps.

A Geographically Weighted Random Forest Approach to Predict Corn Yield in the US Corn Belt

Crop yield prediction before the harvest is crucial for food security, grain trade, and policy making. Previously, several machine learning methods have been applied to predict crop yield using different types of variables. In this study, we propose using the Geographically Weighted Random Forest Regression (GWRFR) approach to improve crop yield prediction at the county level in the US Corn Belt. We trained the GWRFR and five other popular machine learning algorithms (Multiple Linear Regression (MLR), Partial Least Square Regression (PLSR), Support Vector Regression (SVR), Decision Tree Regression (DTR), and Random Forest Regression (RFR)) with the following different sets of features: (1) full length features; (2) vegetation indices; (3) gross primary production (GPP); (4) climate data; and (5) soil data. We compared the results of the GWRFR with those of the other five models. The results show that the GWRFR with full length features (R2 = 0.90 and RMSE = 0.764 MT/ha) outperforms other machine learning algorithms. For individual categories of features such as GPP, vegetation indices, climate, and soil features, the GWRFR also outperforms other models. The Moran’s I value of the residuals generated by GWRFR is smaller than that of other models, which shows that GWRFR can better address the spatial non-stationarity issue. The proposed method in this article can also be potentially used to improve yield prediction for other types of crops in other regions.

Spatial interpolation of PM2.5 concentrations during holidays in south-central China considering multiple factors

Investigations on PM2.5 regional interpolation models during specific periods with abnormal PM2.5 variation patterns (e.g., during holidays) remain relatively rare. In this paper, we first explored PM2.5 concentration patterns in major provincial capitals in south-central China during the Spring Festival and National Day periods. Then, we analysed the correlation between atmospheric pollutants, meteorological conditions, Zenith Tropospheric Delay (ZTD), elevation and Normalized Difference Vegetation Index (NDVI) with PM2.5 in south-central China during the Spring Festival and National Day periods from 2016 to 2018, followed by GeoDetector application to explore the spatial stratified heterogeneity. In addition, we developed Geographically Weighted Regression (GWR) model, Multiscale Geographically Weighted Regression (MGWR) model, and Geographically and Temporally Weighted Regression (GTWR) model. Then, we applied three interpolation methods (Kriging, Empirical Bayesian Kriging, and Tensor Spline Function) to further optimize the residuals of each regression class model. The results indicated that the model created by the Tensor Spline Function (TSF) interpolation method correcting for the residuals of the GWR model (GWR-TSF) attained the best fit, the GTWR model achieved the worst fit during the Spring Festival period, and the MGWR model attained the worst fit during the National Day period. In terms of the interpolation accuracy, the GWR-TSF model also achieved the highest interpolation accuracy, while the GTWR model attained the lowest interpolation accuracy.

Do underserved and socially vulnerable communities observe more crashes? A spatial examination of social vulnerability and crash risks in Texas

Understanding the relationship between social vulnerability and traffic crashes is a cornerstone for promoting social justice in transportation planning and policymaking. However, few studies have examined the disparities in traffic crashes by systemically considering the influence of social vulnerability via spatial analysis approaches. This study puts forward a new approach to assess the inequity in transportation safety by spatially examining the relationships between crash risks and the social vulnerability index (SVI) established by the Centers for Disease Control and Prevention (CDC). We performed spatial autocorrelation analyses to identify the clusters of high-risk and high-vulnerable census tracts in Texas. Meanwhile, we innovatively applied the Multiscale Geographically Weighted Regression model (MGWR) to assess the impacts of CDC SVI on crash risks spatially and statistically. The results demonstrate that the crash rate and the social vulnerability are significantly correlated in the highly urbanized regions as well as the southern border along the Rio Grande in Texas. The MGWR results indicate the minority status of census tracts is strongly correlated with overall crashes in north-central and northeastern Texas, and the socioeconomic status is tightly correlated with fatal crashes across Texas. The outcomes from this study have significant implications for transportation planning and policymaking.