Spatiotemporal Pattern And Influencing Factors Of PM_2.5 In Central Plains Urban Agglomeration Based On Multi-Source Data

PM_2.5 pollution is a typical atmospheric pollution problem in our country,which not only restricts the sustainable development of cities but also interferes with various aspects of human production,life,and physical health,and has now presented regional and complex new characteristics.As the emerging growth pole in the central region of our country,the Central Plains Urban Agglomeration bears vital significance for regional coordination and the promotion of the central region’s rise strategy.The"Central Plains Urban Agglomeration Development Plan"sets the goal of creating an ecologically livable urban agglomeration.However,at present,PM_2.5 pollution in the Central Plains Urban Agglomeration is still severe.Therefore,it is particularly important to detect the pollution status,identify key influencing factors,and clarify the internal mechanisms.This study focuses on the Central Plains Urban Agglomeration as the research area.It comprehensively analyzes various data sources,including PM_2.5 remote sensing inversion data,meteorology,terrain,NDVI,population density,nighttime lights,road network,and socioeconomic statistics.Firstly,it reveals the spatiotemporal evolution characteristics of PM_2.5 concentration in the Central Plains Urban Agglomeration from 2012 to 2020 using methods such as spatial autocorrelation,Theil-Sen Median method,Mann-Kendall test,and rescale range analysis.Secondly,the study utilizes the geodetector to deeply investigate the impact and interaction differences of terrain,meteorology,vegetation,and socioeconomic factors on PM_2.5 concentration from two temporal scales of season and year and the regional scale of spatial clustering division.The multi-scale geographic weighted regression model is also incorporated into the analysis of influencing factors,which initially analyzes the spatial scale and effect of the influencing factors,and then combines the geodetector to analyze the spatial heterogeneity of interaction between the factors.Finally,the study uses a systematic dynamic panel data model to explore the non-linear response relationship between socioeconomic factors and PM_2.5 concentration,and based on the research results and practical situation,proposes suggestions for the prevention and control of PM_2.5 pollution in the Central Plains Urban Agglomeration.The research findings are as follows:（1）In terms of time,the concentration of PM_2.5 in the Central Plains Urban Agglomeration shows a clear pattern of change:the annual average concentration shows a trend of first rising and then continuously declining,and the overall pollution situation has been alleviated.The overall trend of the seasonal average concentration is downward,with the most significant decrease in the winter.The monthly average concentration shows a U-shaped trend of change,with the highest average concentration and the most severe pollution occurring in January and February,while the average concentration in July and August is relatively low.In terms of space,the distribution pattern of PM_2.5concentration shows a decreasing trend from northeast to southwest,with high-pollution areas distributed in the north,middle,and east of the urban agglomeration,while low-pollution areas have a distribution pattern on the mountainous edges of the Qin Ling and Tai Hang Mountains.The concentration of PM_2.5 in the Central Plains Urban Agglomeration shows significant spatial correlation and obvious spatial clustering characteristics,which only exhibit positive spatial correlation,and there are differences in the spatial evolution of different clusters.In terms of the trend of change,the overall trend of PM_2.5 concentration is downward,but there are significant regional differences,and the descending trend among regions ranks from north to south:the north>the middle>the two wings.The future trend of change shows a spatial pattern of"reverse in the middle and maintaining in the surroundings",with the mountainous and hilly areas in the west of the urban agglomeration and the Ta-pieh Mountainous region of Xinyang City exhibiting more obvious persistent characteristics.（2）On an annual scale,the terrain relief has the strongest explanatory power for the PM_2.5 concentration in the Central Plains Urban Agglomeration,followed by nighttime lights and DEM.Per capita GDP and road density have weaker explanatory power.After averaging the explanatory power by factor type,the ranking is terrain>socio-economic>meteorological>vegetation.The maximum explanatory power factors for spring and autumn are still terrain relief,nighttime lights dominate in winter,and NDVI is the maximum explanatory power factor in summer,with a seasonal pattern of high in autumn and summer and low in spring and winter.Different key influencing factors exist under different spatial clusters,with temperature and NDVI being the key factors for high-high clusters and low-low clusters,respectively.There are only two types of interactive types:nonlinear enhancement and bivariate enhancement.DEM∩precipitation is the combination with the largest interaction value in the year,spring and winter,while the strongest combination for summer and autumn are DEM∩NDVI and temperature∩terrain relief,indicating that the synergy between terrain and meteorological factors is more effective than with other factors.Terrain conditions are an important factor leading to significant differences in the explanatory power of factor interactions under different spatial clusters.（3）The multi-scale geographic weighted regression model has exhibited superior performance in quantifying the influencing factors of PM_2.5 concentration.The effects of terrain fluctuation,NDVI,precipitation,humidity,and wind speed were found to have relatively small spatial scales and significant spatial heterogeneity.Among them,terrain fluctuation,NDVI,and precipitation were all negatively correlated with PM_2.5concentration throughout the study area,suggesting that they can inhibit the growth of PM_2.5 concentration to a certain extent,while temperature was mainly positively correlated,especially in the northern part of the urban agglomeration where Changzhi,Jincheng,Handan,Liaocheng,Anyang,and Jiaozuo are located,with the strongest positive correlation effects.In the eastern part of the Central Plains urban agglomeration,such as Suzhou and Bengbu,temperature was negatively correlated with PM_2.5concentration.Humidity was generally positively correlated with PM_2.5 concentration in most areas of the urban agglomeration,while wind speed was mainly negatively correlated.There was also significant spatial heterogeneity in the interactions between factors,and the explanatory power of their interaction combinations was significantly influenced by the direction of action of the internal factors.（4）The per capita GDP,population density,foreign investment proportion,nighttime lighting,and the proportion of the secondary industry all have non-linear effects on PM_2.5concentration.Among them,population density and foreign investment proportion have a"U"-shaped relationship with PM_2.5 concentration.When the population density is greater than 411 people per square kilometer or the foreign investment proportion is greater than 3.78%,they both have a suppressing effect on PM_2.5 concentration,indicating that controlling the regional population density within a reasonable range and adjusting the threshold for foreign investment are important measures to control PM_2.5 pollution in the Central Plains Urban Agglomeration.There is an inverted"U"-shaped relationship between nighttime lighting and PM_2.5 concentration,indicating that improving the existing level of urbanization and building a new type of urbanization can help reduce PM_2.5 concentration.Per capita GDP and the proportion of the secondary industry both have an"N"-shaped relationship with PM_2.5 concentration.They both show a significant inhibitory effect on PM_2.5 concentration when per capita GDP is between 25,835 and 95,798 yuan or the proportion of the secondary industry is between 38.46%and 61.48%.Based on the research results,policy recommendations for the prevention and control of PM_2.5 pollution in the Central Plains Urban Agglomeration have been proposed from the perspectives of laws and regulations,urban planning,industrial structure,environmental regulation,and energy structure.