Study On The Retrieval Of Aerosol Optical Depth Base On MODIS Data

Atmospheric aerosol is the micro suspended solids which liquid or solid particles forms in the air, it plays a very important role in the earth atmosphere radiation balance and global climate change, one of the most important parameters is the optical thickness, accurate detection aerosol optical thickness makes great significance for research of climate change, the satellite remote sensing retrievals of aerosol optical thickness is better than the conventional ground-based detection.Comparing with several current algorithm of remote sensing inverting of the aerosol optical depth, the Dark Pixel algorithm, which is currently of mature technology and wide application, is chosen to do the inversion. The angle data of MOD021 KM and MOD03 from MODIS remote sensing data is applied. A lookup table is edited with the 6S radiation transfer models. And the MODIS remote-sensing images mainly of sunny days in Nanchang city from March to September in the year of 2014 are taken as the primary data. Then the radiation and geometric correction as well as data synthesis for the same will be carried out under ENVI. Data of the processed Band 1( the red band 0.66um), Band 3(the blue band 0.47um) and Band 7(near-infrared band 2.1um)will be extracted to inverting the aerosol optical depth(AOD).According to the lookup table defined in this paper, it is assumed that the inversion result of aerosol is negatively coorelated with the PM10 concentration. After checking the AOD value in the province site, and combining it with the average value of PM10 concentration abtained from nine ground monitoring stations in Nanchang city, it is found that the relation between AOD value and the PM10 concentration is not a simple negative coorelation but analogous to a normal distribution. When the AOD value is 1.2, the PM10 concentration will reach the maximum. The AOD value between 0-1.2 will increase with the increase of PM10 concentration, and this is not correspondent with practice; while the AOD value between 1.2-2 will decrease with the increase of PM10 concentration, and this relatively conforms to the fact.It can be concluded that the AOD value abtained from inversion of MODIS remote sensing images cannot be directly applied to forecast the level of air pollution. However it can indicate the distribution of air pollution at macro level, and provide reference to the forecast and abatement of air pollution.