Spatiotemporal Variation Of PM_2.5 In Guanzhong Basin Based On Satellite Remote Sensing

Air pollution is closely related to human health,it is a worldwide environmental problem.The particulate matter with aerodynamic diameter less than 2.5μm(PM_2.5)is an important component of air pollutants.PM_2.5 dataset with high precision and coverage is very significant for studying the temporal and spatial variation of PM_2.5 concentration,which can provide an important science base for air pollution control policies.However,the observation data of PM_2.5obtained by traditional ground monitoring methods are scarce,and not enough to be used in-depth research.Based on the extinction characteristics of PM_2.5 particles,the aerosol optical depth（AOD）data derived from satellite remote sensing can be used to retrieve PM_2.5concentration,which can effectively make up for the shortage of ground monitoring stations and plays an important role in the study of PM_2.5 concentration in recent years.As a result of the rapid urbanization and industrialization,air pollution has become a severe problem in China.Guanzhong Basin,located in the northwest of China,has been one of the most serious air pollution areas in China because of its rapid economic development,dense population and special terrain conditions.In this study,a high-precision PM_2.5 retrieve model for Guanzhong Basin is developed based on satellite remote sensing AOD data,and the temporal and spatial variation characteristics of PM_2.5concentration in Guanzhong Basin are analyzed.The main research contents and results are as follows:1.Based on fitting regression analysis,the VIIRS AOD products with different quality flags（QF）are validated by using AOD data from Sun-sky radiometer Observation NETwork（SONET）ground monitoring network.The results in Guanzhong Basin show that there is a significant linear correlation between VIIRS AOD QF=3 and SONET AOD（R²=0.72）,and more than 80%of VIIRS AOD fall within the range of expected error.The VIIRS AOD data is accurate and reliable in Guanzhong Basin,and can be used as the basic data for developing PM_2.5 concentration retrieve model.2.We develop a multiscale geographically and temporally weighted regression（GTWR-Ms）model for interpreting the spatiotemporal variation of PM_2.5-AOD in Guanzhong Basin,considering monthly and seasonal variations of PM_2.5 concentration.In this GTWR-Ms model,multiple variables are involved,including VIIRS AOD QF=3 data,meteorological parameters（temperature,planetary boundary layer height,relative humidity）,normalized difference vegetation index（NDVI）,population density,meridional wind speed.The performance and the over fitting problem of the model are evaluated by using cross validation（CV）.The results show that the PM_2.5 concentration estimated by GTWR-Ms model has higher accuracy compared with other common models which can explain the temporal and spatial variation of PM_2.5-AOD.The fitting R² of GTWR-Ms is 0.85,and the R² after CV is 0.76.GTWR-Ms model is more accurate of estimating PM_2.5 concentration in summer and autumn than in spring and winter.3.By the GTWR-Ms model is developed in this study,PM_2.5 concentration in Guanzhong Basin from 2013 to 2019 are estimated.We analysis the temporal and spatial variation of PM_2.5on three time scales（month,season and year）,and discuss the effect of social and economic factors on PM_2.5.The results show that the ground PM_2.5 concentration estimated by GTWR-Ms model from 2013 to 2019 could be used in this research,and R² ranges from 0.80 to 0.87.The average annual PM_2.5concentration in Guanzhong Basin show a downward trend from2013 to 2019.The decrease in 2015 is the largest,reaching 12%compared with 2014.The PM_2.5concentration in Xi’an,Xianyang and Weinan is generally higher than that in Baoji and Tongchuan.In addition to the local emission,the terrain and weather are also the key factors for the high concentration of PM_2.5.PM_2.5 concentration are always higher in the areas with high GDP,strong night light and large proportion of urban land.4.In view of the Corona Virus Disease-2019（COVID-19）special event,we study the effects of epidemic control on PM_2.5 concentration and air pollution in Guanzhong Basin.We retrieve the surface PM_2.5 concentrations with spatial resolution of 6 km×6 km by GTWR-Ms model during the epidemic period in Guanzhong Basin,and combining analysis the air pollution(PM₁₀,SO₂,NO₂,CO and O₃)concentrations from the ground monitoring station.The results show that the air quality in Guanzhong Basin has been improved obviously.After the control measures carried out,the PM_2.5 concentration retrieved from satellite remote sensing data and observed by ground monitoring station decreased by 27%and 37%respectively.With the epidemic under control in Guanzhong Basin,air pollutants began to rebound because of the control measures gradually relaxed.