Study On Spatial-temporal Characteristics And Influencing Factors Of Aerosol Optical Depth In Taiyuan City

Aerosol Optical Depth(AOD)is the integral of the extinction coefficient of the medium on the vertical path.It is one of the characterization indicators of air quality and can reflect the state of air pollution.Taiyuan is one of the old industrial base cities in the country.With the rapid development of industrialization and urbanization,the fog and haze weather has become more and more serious,which has had a great impact on local economic development and social life.Therefore,monitoring and evaluating air pollution through satellite remote sensing technology is of great significance for improving the quality of local ecological environment.Based on the MODIS data from 2009 to 2018,this paper takes Taiyuan as the research area,and carries out the AOD inversion work by combining and improving the dark pixel algorithm(Dark Target,DT)and the deep blue algorithm(Deep Blue,DB).The main contents are as follows :(1)The traditional MODIS data preprocessing method is not efficient,which cannot meet the actual needs of batch inversion.Therefore,this paper selects the MOD35＿L2 cloud mask product,combines MRTSwath,MRT data processing software and IDL programming,and proposes an improved MODIS data preprocessing process.The results show that compared with the original preprocessing method,the preprocessing time of this method is greatly shortened,the problem of pixel matching between different data is solved,and the available time length of the inversion data is guaranteed at the same time.(2)The DT algorithm has higher accuracy for the inversion of the dense vegetation surface,but it is more difficult for the inversion of the urban high-brightness surface.At present,there are still some challenges in the application of the combined DT and DB inversion AOD method in long time series,Aiming at this problem,this paper uses the improved DT algorithm(V5.2algorithm)and the method of constructing a priori surface reflectance library based on MOD09A1 data to obtain the corresponding surface reflectance by dividing bright and dark surfaces,and uses the 6S radiation transfer model Create a lookup table for AOD inversion.The results show that the AOD time series inverted by the method proposed in this paper has better stability and is more conducive to long-term series analysis.(3)The AOD retrieval results proposed in this paper are compared with the MOD04＿L210km spatial resolution aerosol products in many aspects,The results show that the spatial distribution trend of AOD retrieved by the method proposed in this paper is consistent with that of MOD04＿L2 AOD,and the correlation coefficient reaches 0.872;At the same time,the spatial resolution of the inversion results in this paper is higher,the transition between areas with different vegetation coverage is more detailed,the overall spatial continuity is better,and there are fewer AOD missing values in the urban center area,which has good applicability.(4)Based on the AOD time series data inverted in this paper,using Empirical Orthogonal Function(EOF),Complete EEMD with Adaptive Noise(CEEMDAN),HYSPLIT model(Hybrid Single Particle Lagrangian Integrated Trajectory Model),geographic detector and other methods to analyze the temporal and spatial variation characteristics of AOD and its influencing factors in Taiyuan from 2009 to 2018,and the results showed that: From the perspective of spatial distribution,the AOD in Taiyuan presents a distribution characteristic of high in the southeast and low in the northwest,and The high value area of AOD shows a banded distribution,mainly showing two spatial distribution patterns: global decline type and the reverse phase change type;From the perspective of time change characteristics,AOD shows a fluctuating downward trend,and the air quality will continue to improve in the future,and the seasonal fluctuation cycle dominates the time series change of AOD in Taiyuan;From the perspective of the regional transmission of aerosol pollutants,the spatial distribution of the backward airflow trajectory is greatly affected by the monsoon climate;The potential source areas are mainly distributed on the west side of the Taihang Mountain,the influence of aerosol pollutant transport process along the Hexi Corridor to the east is more significant;In the past ten years,the overall explanatory power of social and economic factors is greater than that of natural environment factors in Taiyuan,and the primary influencing factor is population density,indicating that the spatial differentiation of AOD in Taiyuan is mainly caused by frequent human activities.At the same time,the explanatory power of NDVI is on the rise,and its influence on AOD changes is increasing.