Construction Of Regional Air Quality Assessment And Early Warning System Based On Remote Sensing Information

The traditional method of forecasting the air quality is the air quality model. Limited by the preparation of emission inventories, the air quality model from the 1st to 3rd generation air quality models had not been used widely. With the development of satellite remote sensing technology,we could get the air quality estimation by aerosol optical depth retrieved by remote sensing image retrieval in the very short period of time. IDEAI Trajectory model adopts remote sensing data for air quality forecast, avoiding the difficulty of counting emissions inventory. The experiment achieved good prediction effect.The studying area of this paper are mainly in north of China. IDEA-I Trajectory model uses MODIS’s AOD product as sources of pollution. Remote sensing images still exist vast areas with no value affected by the clouds. Benefiting from the ground monitoring site all over the country, through the empirical formula, making air quality corresponds to aerosol optical depth, using ground data interpolate into remote sensing data, then adopting Inverse distance weighting spatial interpolation smooth full figure acquiring AOD product with high data coverage, inputting to IDEA-I Trajectory model. The main research results of this paper are as follows:(1) Complete localization application of IDEA-I trajectory model through making it forecast air quality of any domestic area during 48 hours with meteorological data and remote sensing AOD data. Take north China area as the research object, analysis trajectory of aerosol within 48 hours, compare running results and monitoring results. The results show that IDEA-I can acquire good experimental results and accurately estimate the change trend of air quality when the coverage of AOD data is relatively complete.(2) Limited by weather conditions as there are much clouds across research area, so adopt interpolation of the ground site data into remote sensing image as to make up lack of large data. Set up air quality automatic data acquisition system based on the JAVA platform. By using statistics of seven air quality monitoring sites’ data of Chengdu third quarter and fourth quarter, set up during PM2.5 and PM10 normalized experience model, obtain another value with known value. At the same time reference to related research results, set up normalized model experience between AOD and AQI to achieve the corresponding monitoring station data and AOD data. At the same time using Inverse distance weighting spatial interpolation smooth full figure to obtain better experiment results.(3) Building an automated early warning system based on JAVA platform, which could download of remote sensing data and configure IDEA-I related operating parameters automatically. This system also is able to be fill the lose value of the remote image with the data from the ground monitoring station and spatially interpolate. After running the IDEA-I system, the program could calculate the point of the aerosol trajectory and make quantitative prediction of air quality.The result show that the system can accurately judge the air quality changes of pollutants and air mass trajectories, migration direction, diffusion range. There is a good experimental result for the regional air quality prediction.