Geographically Weighted Regression Model Research Articles

Construction of A Geographically Weighted Nonparametric Regression Model Fit Test

One of the approach of Geographically Weighted Regression (GWR) models is the Geographically Weighted Nonparametric Regression (GWNR) has more parameters than the GWR model. Models with more parameters usually have better match values, which is an advantage, while models with fewer parameters have the advantage of being easier to use and interpret. However, a model with more parameters should be used if it is proven to be significantly superior. Therefore, the purpose of this study was to develop a hypothesis test of goodness of fit test for GWNR model. The goodness of fit test was performed for the real data. We found that the GWNR model was more suitable than the mixed nonparametric regression model. Some highlights of the proposed method are:• A new model for GWR to overcome the unknown regression function by using mixed estimator spline truncated and fourier series at nonparametric regression• Goodness of fit for GWNR to testing the model fit between the mixed nonparametric regression model and GWNR• Applied goodness of fit test to poverty data in Sulawesi Island and infant mortality in East Java

Generalized Geographically Weighted Regression Model within a Modularized Bayesian Framework.

Geographically weighted regression (GWR) models handle geographical dependence through a spatially varying coefficient model and have been widely used in applied science, but its general Bayesian extension is unclear because it involves a weighted log-likelihood which does not imply a probability distribution on data. We present a Bayesian GWR model and show that its essence is dealing with partial misspecification of the model. Current modularized Bayesian inference models accommodate partial misspecification from a single component of the model. We extend these models to handle partial misspecification in more than one component of the model, as required for our Bayesian GWR model. Information from the various spatial locations is manipulated via a geographically weighted kernel and the optimal manipulation is chosen according to a Kullback-Leibler (KL) divergence. We justify the model via an information risk minimization approach and show the consistency of the proposed estimator in terms of a geographically weighted KL divergence.

The spatial variation and drivers of soil δ13C in an alpine-gorge region on the eastern Qinghai-Tibetan Plateau

Soil stable carbon isotope (soil δ13C) can reflect soil carbon metabolic processes and record environmental and vegetation change information, so mapping the spatial distribution of soil δ13C (isoscapes of soil δ13C) can help us better understand the spatial variability of ecosystem carbon cycles. However, efficient approaches for obtaining the isoscapes of soil δ13C remain challenging, especially in mountainous areas with complex terrain. A total of 150 soil samples were collected and their soil δ13C was measured to investigate the spatial variation of soil δ13C in an alpine-gorge region on the eastern Qinghai-Tibetan Plateau which is characterized by complex terrain. Four prediction methods included ordinary kriging interpolation, multiple linear regression, random forest regression, and geographically weighted regression (GWR) were selected and compared to find the best prediction model for mapping the isoscapes of soil δ13C. The pathways of vegetation, topography, soil, and spatial factors influencing the spatial variability of soil δ13C were explored using variance partitioning analysis and structural equation model. The soil δ13C was significantly different among vegetation types, and ranged from −27.155‰ to −9.647‰. The spatial heterogeneity of soil δ13C showed moderate variation, and was dominated by spatial structural factors. The GWR model had higher prediction accuracy in the modeling soil δ13C in comparison to other models. Soil carbon to nitrogen ratio and normalized difference vegetation index were the main factors determining the spatial variability of soil δ13C, and other topographic and soil factors indirectly regulated the spatial distribution of soil δ13C by influencing these two factors. Our results provided a useful tool (GWR model) for mapping the isoscapes of soil δ13C and explored the factors controlling the spatial variability of soil δ13C in the alpine-gorge region on the eastern Qinghai-Tibetan Plateau, which are important for an in-depth understanding of soil carbon cycling in mountain ecosystems.

Distribution Pattern of Species Richness of Endemic Genera in Mountainous Areas of Southwest China and Its Influencing Factors

China’s southwest mountainous region is one of the hotspots of biodiversity in the world. However, the study on the pattern of species richness distribution of endemic plants in this region and its influencing factors is still in the exploration stage. A clear regional spatial distribution pattern of plants and its influencing factors are very important for the protection and management of plant diversity in southwest China. In this study, 511 county-level regions in the southwest mountainous region were taken as examples to establish a plant species resource bank in the southwest mountainous region, and the distribution pattern of the proportion and abundance of endemic plant families and genera was analyzed based on 19 climate and environmental factors and 8 human disturbance factors in the southwest mountainous region. The geographical weighted regression model (GWR) was used to reveal the main factors affecting the spatial distribution of species. The results showed that: (1) A total of 4688 species investigated in the southwest mountain belong to 19 families and 32 genera. (2) The endemic species richness in the southwest mountainous region showed a significant uneven spatial pattern, mainly concentrated in the Heng Duan Mountain area, southwest Sichuan and northern Yunnan area, and the Ruoergai-Hongyuan area of West Sichuan Plateau. (3) The results of the GWR model showed that the diurnal mean temperature range, precipitation seasonality and distance from the main road had significant effects on species richness. The study on the distribution of endemic genera and their influencing factors in the southwest mountainous region has important implications for the exploration of the evolutionary history and species conservation of the mountain flora.

The Heterogeneous Effects of Microscale-Built Environments on Land Surface Temperature Based on Machine Learning and Street View Images

Global climate change has exacerbated alterations in urban thermal environments, significantly impacting the daily lives and health of city residents. Measuring and understanding urban land surface temperatures (LST) and their influencing factors is important in addressing global climate change and enhancing the well-being of residents. However, due to limitations in data precision and analytical methods, existing studies often overlook the microscale examination closely related to residents’ daily lives, and lack a deep exploration of the spatial heterogeneity of the influencing factors. This leads to these results being ineffective in guiding the planning and construction of cities. Taking Shenzhen as a case study, our study investigates the effects of various microscale build environment characteristics of LST using street view images and machine learning. A convolutional neural network model adopting the SegNet architecture is used to perform semantic segmentation on street view images, extracting features of the microscale urban-built environment. The LST is inverted through the Google Earth Engine (GEE) platform. By using Multiscale Geographically Weighted Regression (MGWR) models, our study reveals the comprehensive impact of the urban-built environment on LST and its significant spatial heterogeneity. The findings indicate that the proportions of sky, roads, and buildings are positively correlated with LST, while trees have a significant cooling effect. Although earth and water can reduce LST, their overall contribution is minimal due to limitations in their area and distribution patterns. This study not only reveals the key factors affecting urban LST at the microscale but also emphasizes the necessity of considering the spatial heterogeneity of these factors’ impacts. This suggests the need for targeted strategies for different areas to effectively improve the urban thermal environment and achieve sustainable urban development.

Exploring impacts of COVID-19 on spatial and temporal patterns of visitors to Canadian Rocky Mountain National Parks from social media big data

COVID-19 posed challenges for global tourism management. Changes in visitor temporal and spatial patterns and their associated determinants pre- and peri-pandemic in Canadian Rocky Mountain National Parks are analyzed. Data was collected through social media programming and analyzed using spatiotemporal analysis and a geographically weighted regression (GWR) model. Results highlight that COVID-19 significantly changed park visitation patterns. Visitors tended to explore more remote areas peri-pandemic. The GWR model also indicated distance to nearby trails was a significant influence on visitor density. Our results indicate that the pandemic influenced tourism temporal and spatial imbalance. This research presents a novel approach using combined social media big data which can be extended to the field of tourism management, and has important implications to manage visitor patterns and to allocate resources efficiently to satisfy multiple objectives of park management.

Understanding the Spatiotemporal Dynamics and Influencing Factors of the Rice–Crayfish Field in Jianghan Plain, China

The rice–crayfish co-culture system, a representative of Agri-aqua food systems, has emerged worldwide as an effective strategy for enhancing agricultural land use efficiency and boosting sustainability, particularly in China and Southeast Asia. Despite its widespread adoption in China’s Jianghan Plain, the exact spatiotemporal dynamics and factors influencing this practice in this region are yet to be clarified. Therefore, understanding the spatiotemporal dynamics and influencing factors of the rice–crayfish fields (RCFs) is crucial for promoting the rice–crayfish co-culture system, and optimizing land use policies. In this study, we identified the spatial distribution of RCF using Sentinel-2 images and land use spatiotemporal data to analyze its spatiotemporal dynamics during the period of 2016–2020. Additionally, we used the Multiscale Geographically Weighted Regression model to explore the key factors influencing RCF’s spatiotemporal changes. Our findings reveal that (1). the RCF area in Jianghan Plain expanded from 1216.04 km2 to 2429.76 km2 between 2016 and 2020, marking a 99.81% increase. (2). RCF in Jianghan Plain evolved toward a more contiguous and clustered spatial pattern, suggesting a clear industrial agglomeration in this area. (3). The expansion of the RCFs was majorly influenced by its landscape and local agricultural conditions. Significantly, the Aggregation and Landscape Shape Indexes positively impacted this expansion, whereas proximity to rural areas and towns had a negative impact. This study provides a solid foundation for promoting the rice–crayfish co-culture system and sustainably developing related industries. To ensure the sustainable development of rice–crayfish co-culture industries in Jianghan Plain, we recommend that local governments optimize the spatial layout of rural settlements, improve transportation infrastructure, and enhance regional agricultural water sources and irrigation system construction, all in line with the national strategy of rural revitalization and village planning. Additionally, promoting the concentration and contiguity of RCF through land consolidation can achieve efficient development of these industries.

Analysis of the situations and influencing factors of public anxiety in China: based on Baidu index data.

Anxiety disorders have emerged as one of the most prevalent mental health problems and health concerns. However, previous research has paid limited attention to measuring public anxiety from a broader perspective. Furthermore, while we know many factors that influence anxiety disorders, we still have an incomplete understanding of how these factors affect public anxiety. We aimed to quantify public anxiety from the perspective of Internet searches, and to analyze its spatiotemporal changing characteristics and influencing factors. This study collected Baidu Index from 2014 to 2022 in 31 provinces in mainland China to measure the degree of public anxiety based on the Baidu Index from 2014 to 2022. The spatial autocorrelation analysis method was used to study the changing trends and spatial distribution characteristics of public anxiety. The influencing factors of public anxiety were studied using spatial statistical modeling methods. Empirical analysis shows that the level of public anxiety in my country has continued to rise in recent years, with significant spatial clustering characteristics, especially in the eastern and central-southern regions. In addition, we constructed ordinary least squares (OLS) and geographically weighted regression (GWR) spatial statistical models to examine the relationship between social, economic, and environmental factors and public anxiety levels. We found that the GWR model that considers spatial correlation and dependence is significantly better than the OLS model in terms of fitting accuracy. Factors such as the number of college graduates, Internet traffic, and urbanization rate are significantly positively correlated with the level of public anxiety. Our research results draw attention to public anxiety among policymakers, highlighting the necessity for a more extensive examination of anxiety issues, especially among university graduates, by the public and relevant authorities.

Digitalization era: Investigating the spatial interplay between cyber human activity and economy with a hierarchical framework

The development of Information and Communication Technology has shifted human activity from offline to online, and promoted digital economy. These changes challenge traditional methodologies relying on physical human activity as a micro-level reflection of the macro-level economy. To address this, a hierarchical framework is proposed to characterize cyber human activity, incorporating activity diversity, size, and preference. Then, Ordinary Least Squares and Geographically Weighted Regression models are used to examine the spatial interplay between cyber human activity and township-level economy. K-Medoids clustering is further applied to coefficients in cyber GWR model to reveal mechanisms influencing local economies. Taking diverse geospatial data in Jilin Province, China as an example, the result indicates that the significant extreme values of cyber human activity have divided Jilin Province into four subregions, closely aligning with local economic segmentation. Moreover, the local economy can be better reflected by cyber human activity rather than physical ones, especially in highly digitized regions. Furthermore, the regions with similar influence mechanisms cluster geographically. These clusters are categorized into three types, namely, dynamics, robust, and balanced, representing different mechanisms influencing the local economy. Practical and theoretical implications are discussed, including using cyber human activity in assessing the economy and implementing adaptable economic policies at township level.

SGWR: similarity and geographically weighted regression

Geographically weighted regression (GWR) offers a local approach to modeling spatial data, considering geographical location and spatial relationships between observations. A salient feature of GWR is the emphasis on geographical proximity, in accordance with Tobler’s First Law of Geography, which assumes that closer entities have a greater influence on the target location. Traditional GWR models have been augmented to consider various forms of physical distances aimed at enhancing model performance, and they often disregarded the potential influence of other data attributes, a shortcoming that extends to most GWR extensions. In this study, we introduce a novel weight matrix construction, which integrates data attribute similarity alongside the conventional geographically weighted matrix. The two weights are integrated in a manner that results in improved model performance. The proposed model, called Similarity and Geographically Weighted Regression or SGWR, was applied to five distinct datasets: housing prices, crime rates, and three health outcomes including mental health, depression, and HIV. Results show that SGWR significantly improved model performance based on several statistical measures, outperforming the global regression model and the traditional GWR.

Spatial Relationship of Inter-City Population Movement and Socio-Economic Determinants: A Case Study in China Using Multiscale Geographically Weighted Regression

In the current field of regional studies, there is a growing focus on regional spatial relationships from the perspective of functional linkages between cities. Inter-city population movement serves as an embodiment of the integrated functionality of cities within a region, and this is closely tied to the socio-economic development of urban areas. This study utilized Location-Based Services (LBSs) to collect the scale of inter-city population movement across 355 cities in China. Additionally, socio-economic data published by local governments were incorporated. By establishing a Multiscale Geographically Weighted Regression (MGWR) model, this research explores the spatial relationships between inter-city population movement and socio-economic influencing factors in China. This study aims to elucidate the spatial scales of the relationships between various variables. Our research findings indicate that the relationship between inter-city population movement and potential socio-economic determinants exhibits spatial non-stationarity. It is better to explore this spatial relationship through the MGWR model as there are different determinants operating on inter-city population movement at different spatial scales. The spatial distribution of the coefficient estimates shows significant regional differences and numerical variations. In China’s economically developed coastal regions, there is relatively balanced development among cities, with advanced manufacturing and producer service industries acting as significant drivers of mobility. In inland regions of China, city size is the most influential variable, directing a substantial flow of human and economic resources towards regional socio-economic hubs such as provincial capitals. The main contribution of this study is the re-examination of the relationship between inter-city population movement and socio-economic factors from the perspective of spatial scales. This approach will help China to consider the heterogeneity of different regions more extensively when formulating regional development policies, thereby facilitating the targeted promotion of regional element flow.

Unveiling the Patterns and Drivers of Ecological Efficiency in Chinese Cities: A Comprehensive Study Using Super-Efficiency Slacks-Based Measure and Geographically Weighted Regression Approaches

Urban ecological efficiency stands as a pivotal indicator that mirrors the level of sustainable development within cities. To unravel the sustainable development status of Chinese cities and illuminate the factors impacting the diverse developments among them, this study leveraged the super-efficiency SBM (slacks-based measure) model to assess the ecological efficiency of 284 prefectural-level and above cities across China in 2019, divulging their spatial distribution. Furthermore, a GWR (geographically weighted regression) model was also employed to scrutinize the factors influencing the ecological efficiency of these cities. Key findings include: (1) The mean ecological efficiency of Chinese cities in 2019 stood at 0.555, signaling moderate urban sustainability, with southern cities outperforming their northern counterparts. (2) A pronounced spatial clustering of ecological efficiency was evident, featuring positive spillover effects around high-efficiency cities and conversely, negative spillover effects around low-efficiency cities. (3) Economic development and population density positively influenced urban ecological efficiency, while urbanization levels exhibited a negative impact. The influences of industrial structure, technological level, and opening-up level varied, showcasing both positive and negative impacts contingent upon the spatial disposition of the cities. Hence, policymakers are advised to recognize the spatial nuances in the impacts of distinct factors on urban ecological efficiency and tailor measures accordingly to fortify urban sustainability.

Determinants of COVID-19 Prevalence Rate in Asia: A study using Spatial Analysis

This study aims at finding out the important determinants of prevalence rate of COVID-19 in the Asian continent using spatial analysis. The impact of climatic, socioeconomic, demographic, and health status variables on the prevalence rate of COVID-19 is seen through various spatial models such as Spatial lag, Spatial error, Geographically Weighted regression model, and Multiscale Geographically Weighted regression model. The performance of the models is compared under different comparison criteria. It is found that among all, Multiscale Geographically Weighted regression model outperformed other competitive models. Findings also indicate that cardiovascular health, prevalence of smoking habit, human development index, and net migration rate played significant role in defining the prevalence rate of COVID-19 in Asia.

A Spatiotemporal Comparative Analysis of Docked and Dockless Shared Micromobility Services

Sustainable urban mobility is an imperative concern in contemporary cities, and shared micromobility systems, such as docked bike-sharing, dockless bike-sharing, and dockless e-scooter-sharing, are recognized as essential contributors to sustainable behaviors in cities, both complementing and enhancing public transport options. Most of the literature on this subject predominantly focuses on individual assessments of these systems, overlooking the comparative analysis necessary for a comprehensive understanding. This study aims to bridge this gap by conducting a spatiotemporal analysis of two different shared micromobility modes of transportation, docked bike-sharing systems and dockless e-scooter-sharing systems operating in the municipality of Lisbon. The analysis is further segmented into arrivals and departures on weekdays and weekends. Additionally, this study explores the impact of sociodemographic factors, the population’s commuting modes, and points of interest (POIs) on the demand for both docked bike-sharing and dockless e-scooter-sharing. Multiscale Geographically Weighted Regression (MGWR) models are employed to estimate the influence of these factors on system usage in different parishes in Lisbon. Comparative analysis reveals that the temporal distribution of trips is similar for both docked bike-sharing and dockless e-scooter-sharing systems on weekdays and weekends. However, differences in spatial distribution between the two systems were observed. The MGWR results indicate that the number of individuals commuting by bike in each parish has a positive effect on docked bike-sharing, while it exerts a negative influence on dockless e-scooter-sharing. Also, the number of commercial points of interest (POIs) for weekday arrivals positively affects the usage of both systems. This study contributes to a deeper understanding of shared micromobility patterns in urban environments and can aid cities in developing effective strategies that not only promote and increase the utilization of these shared micromobility systems but also contribute to sustainable urban mobility.

Spatio-Temporal Variation Analysis of Soil Salinization in the Ougan-Kuqa River Oasis of China

In order to investigate the mechanism of environmental factors in soil salinization, this study focused on analyzing the temporal-spatial variation of soil salinity in the Ogan-Kuqa River Oasis in Xinjiang, China. The research aimed to predict soil salinity using a combination of satellite data, environmental covariates, and advanced modeling techniques. Firstly, Boruta and ReliefF algorithms were employed to select variables that significantly affect soil salinity from the Sentinel-2 satellite data and environmental covariates. Subsequently, a soil salinity inversion model was established using three advanced strategies: comprehensive variable analysis, a Boruta-based variable selection algorithm, and a ReliefF-based variable selection algorithm. Each strategy was modeled using a Light Gradient Boosting Machine (LightGBM), an Extreme Learning Machine (ELM), and a Support Vector Machine (SVM). Finally, the Boruta-LightGBM strategy was proven to be the most effective in predicting soil electrical conductivity (EC), with a coefficient of determination (R2) of 0.72 and a Root Mean Square Error (RMSE) of 12.49 ds/m. The experimental results show that the red-edge band index is the foremost variable in predicting soil salinity, succeeded by the salinity index and soil attribute data, while the topographic index has the least influence, which further demonstrates that proper variable selection could significantly improve model functionality and predictive precision. Furthermore, the Multiscale Geographically Weighted Regression (MGWR) model was utilized to reveal the influence and temporal-temporal-spatial heterogeneity of environmental factors such as soil organic carbon (SOC), precipitation (PRE), pH value, and temperature (TEM) on soil EC. This research offers not just a viable methodological framework for monitoring soil salinization but also new perspectives on the environmental drivers of soil salinity changes, which have implications for sustainable land management and provide valuable information for decision-making in soil salinity control and mitigation efforts.

Exploring the relationships between 3D urban landscape patterns and PM2.5 pollution using the multiscale geographic weighted regression model

Fine particulate matter (PM2.5) is a major source of air pollution and exerts serious impacts on human health. The 3D urban landscape patterns can significantly affect the diffusion and emissions of PM2.5. However, studies on the relationships between 3D urban landscape patterns and PM2.5 pollution across different seasons remain understudied. With the ground-level air pollutants estimated by the remote sensing and fine-scale building information, this study applied the multiscale geographically weighted regression model to explore such relationships. Wuhan, the largest metropolis in Central China, was selected as the study area for the application of our methodology. The results showed that the direction, degree, and scale of the effect of 3D urban landscape patterns on PM2.5 pollution varied across seasons. For building height, the standard deviation of building height had a significant positive correlation with PM2.5 all year round. For building density, the building count density showed a significant positive correlation with PM2.5 in general, with the bandwidth in winter and autumn smaller than in spring and summer. The building plan area fraction exerted both positive and negative influences on PM2.5, dependent on season and location. The bandwidth of it gradually increased from spring to winter, with the effect changing from local to regional scale. For building volume, the floor area ratio showed a significant negative correlation with PM2.5 in winter and autumn, and a localized effect was found, especially in winter. The findings of this study provide practical implications for urban planning and policy making to mitigate PM2.5 pollution in the rapidly urbanizing regions.

Modelling and dynamic water analysis for the ecosystem service in the Central Citarum watershed, Indonesia

Exploring the drivers of changes in ecosystem services is crucial to maintain ecosystem functionality, especially in the diverse Central Citarum watershed. This study utilises the integrated valuation of ecosystem service and trade-offs (InVEST) model and multiscale geographically weighted regression (MGWR) model to examine ecosystem services patterns from 2006 to 2018. The InVEST is a hydrological model to calculate water availability and evaluate benefits provided by nature through simulating alterations in the amount of water yields driven by land use/cover changes. Economic, topographic, climate, and vegetation factors are considered, with an emphasis on their essential components. The presence of a geographical link between dependent and explanatory variables was investigated using a multiscale geographic weighted regression model. The MGWR model is employed to analyse spatial impacts. The integration of both models simplified the process and enhanced its understanding. The findings reveal the following patterns: 1) decreasing land cover and increasing ecosystem services demand in the watershed, along with a decline in water yield, e.g. certain sub-districts encounter water scarcity, while others have abundant water resources; 2) the impact of natural factors on water yield shifts along vegetation > climate > topography (2006) changes to climate > vegetation > topography (2018).

Influencing factors and spatial differentiation of rental housing in a smart city: A GWR model analysis

Smart cities leverage technology, data, and digital infrastructure to enhance residents' quality of life, fostering sustainability and operational efficiency. To this end, the price data of rental housing in Hefei in 2022 were hereby collected to build a geographically weighted regression (GWR) model, and the mechanism of various factors on the price of rental housing in Hefei was discussed. Spatial correlation analysis revealed a notable autocorrelation of rental housing prices in Hefei, exhibiting a distribution pattern of decreasing values from the city center outward. Besides, the impact of 13 factors on house rent was examined. Overall, factors such as building area, proximity to wet markets, and distance from the top three hospitals were demonstrated to exert the most significant influence on rental housing prices in Hefei. The decision-making of smart cities needs to be scientifically analyzed to find out the factors affecting the price of rental housing in the region, and the research conclusions can provide certain references for urban decision-makers in the overall management of urban resources and targeted solutions to regional housing rental and other issues.

Hedonic Approach to Vertical Residential Rentals in the Brazilian Amazon: The Case of Belém, Pará

This study set out to identify and analyze the factors that influence the formation of rental prices for residential apartments in the city of Belém. The approach adopted was based on the hedonic price theory, which considers that the rental price of an apartment reflects the implicit prices of its attributes—structural, locational, and neighborhood quality. The sample used consisted of 259 observations, corresponding to the rental advertisements of the representative apartments in each building. The ordinary least squares (OLS), spatial lag model (SLM), and geographically weighted regression (GWR) techniques were used in the statistical analysis in this study. The results of the OLS model showed statistical significance between the attributes analyzed and the rental price of the apartments. In turn, the SLM indicated that the structural attributes have an impact on the rental prices of neighboring apartments, configuring a contagion effect in the real estate market. The GWR model showed that there was no spatial heterogeneity in the effects of the determinants on apartment rental prices throughout the sample.

Multi-scale Driving Mechanism of Urbanization on PM2.5 Concentration in Urban Agglomeration in the Middle Reaches of the Yangtze River

Exploring the nonlinear effect of urbanization on PM2.5 concentration and its driving mechanism is crucial for controlling urban air pollution. Based on remote sensing data and statistical data from 2002 to 2020, spatial autocorrelation, systematic dynamic panel regression, and spatio-temporal geographical weighted regression models were used to analyze the spatio-temporal evolution of PM2.5 concentration in the urban agglomeration of the middle reaches of the Yangtze River and explore the driving mechanism of urbanization on PM2.5 concentration at different spatial scales. The results showed that:① PM2.5 concentration in the middle reaches of the Yangtze River showed an overall decreasing trend from 2002 to 2020, with a spatial distribution pattern of "high in the north and low in the south." ② Hot spot cities expanded towards the western part of the urban agglomeration, whereas cold spot cities showed enhanced spatial correlation. ③ The relationship between PM2.5 concentration and economic, land, and population urbanization followed N-shaped, U-shaped, and U-shaped curves, respectively. Secondary industry and energy consumption significantly promoted the change in PM2.5 concentration, and precipitation and vegetation helped mitigate PM2.5 pollution. ④ The overall driving effects of all urbanization factors in local areas were transformed, and the main areas of influence were concentrated in the southeast, northwest, and southwest of the study area. Considering the current urban development status and regional characteristics of the urban agglomeration in the middle reaches of the Yangtze River, promoting green industrial transformation, rational planning of urban spatial distribution and population distribution, and enhancing infrastructure construction will facilitate the coordinated development between urbanization and environmental protection.