Study On The Relationship Model Between Near-surface AOD Based On MODIS Remote Sensing Inversion And PM2.5

With the development of economic,the process of industrialization and urbanization is also accelerating,and the living environment of human beings is getting worse and worse.The increasingly frequent haze weather has seriously affected the health of people and the safety of travel.This is due to fine particles(PM2.5)with aerodynamic equivalent diameter less than or equal to 2.5 microns in ambient air.these fine particles are only a small part of the earth’s atmospheric composition,but it can have a very important impact on air quality and visibility distance.As air pollution problems continue to intensify,The smog weather continues to emerge,The impact on human respiratory diseases is getting worse,The government and some public organizations have established atmospheric environmental quality monitoring projects and established ground monitoring stations to monitor atmospheric air quality and pollutant content.However,due to the limited number of ground monitoring stations,the sparse distribution of monitoring stations is uneven,and the cost of establishing monitoring stations is high and the maintenance is time-consuming and labor-intensive.It is impossible to fully and effectively reflect the spatial and temporal distribution of PM2.5 in a wide range of space.In recent years,with the emergence and development of quantitative remote sensing,this problem has been effectively and effectively solved.Due to the wide coverage area,strong timeliness and low cost,satellite remote sensing observations have greatly compensated for the high cost of establishing and maintaining ground monitoring stations and the small coverage area.Therefore,the use of satellite quantitative remote sensing to monitor particulate matter and the concentration of pollutant gases has been widely used.The use of satellite remote sensing images to achieve monitoring of PM2.5 concentration values has also become a hot area.This paper selects Shuangliu County of Chengdu city District as the study area.which ranks in the top three in Chengdu’s air quality index ranking in 2017,and is a heavily polluted area in various districts and counties of Chengdu.Choosing the optimal inversion algorithm-extended dark pixel algorithm,The aerosol optical thickness(referred to as AOD for different regions in different periods)was calculated by Terra-MODIS satellite remote sensing data.According to the vertical distribution characteristics of aerosol and the geographical environment of the double-flow zone,Use weather station visibility data,Calculate the corresponding near-surface AOD and construct a relationship model between PM2.5 concentration and near-surface AOD value.Analysis of the influence of meteorological factors(temperature,pressure,humidity)on PM2.5 concentration distribution and even diffusion,And add these influence factors to the relational model with near-surface AOD and PM2.5 as variables to improve the accuracy of their relationship model.PM2.5 concentration value of air quality monitoring site measured by Cross-validation method Verify the optimized model accuracy,This study is used to realize the feasibility of in-depth real-time dynamic monitoring of PM2.5 concentration by satellite quantitative remote sensing.Throughout the research work,the research results include the following five aspects.(1)In this paper,the extended dark pixel algorithm V5.2 algorithm is used to invert the AOD in the dual-flow region,and then the vertical polarization of the aerosol is used to vertically correct the inversion AOD value,and then PM2.5 and AOD,near-surface AOD are analyzed separately.The correlation between PM2.5 and the near-surface AOD value is stronger,indicating that the vertical correction method is feasible and can improve the correlation between PM2.5 and aerosol optical thickness.(2)Interpolation analysis was carried out on the temporal and spatial variation characteristics of PM2.5 concentration values in the dual-flow monitoring station.In the time dimension,PM2.5 concentration was the lowest in summer,followed by autumn and highest in winter.In spatial dimension,PM2.5 concentration was from It gradually rises from south to north,and has the highest concentration in the center of Chengdu.(3)The concentration of PM2.5 near Shuangliu Airport is lower than other areas,which may be caused by the cyclone generated by the takeoff and landing of the aircraft in the airport to accelerate the diffusion of PM2.5 in this area.(4)The near-surface AOD-PM2.5 relationship model is established,and the fitting degree of the one-dimensional equation is higher.The correlation between PM2.5 and meteorological factors including temperature,pressure and humidity is analyzed respectively,and PM2 is obtained.The concentration of 5 is negative,negative and positively correlated with temperature,humidity and pressure,and the relationship between the near-surface AOD-PM2.5 relationship model is optimized to obtain an optimized multiple regression equation.The cross-validation method is used to verify the concentration.Optimize the feasibility of the model,and the result is that the accuracy is higher and feasible.