Regression Analysis On Air Pollution Index And MODIS Aerosol Optical Thickness In Nanjing

In today's world, industrialization has greatly accelerated the economic development, which results in vast consumption of resources. Due to this, the atmospheric conditions have been severely deteriorating, threatening the sustainable development of both our society and economy. Thus the government and the public are paying more attention to aggravating air pollution. The most direct influence people's life from the air is its everyday quality, hence the API (Air Pollution Index ) is of considerable importance. That is why a timely and effective forecast of air pollution quality needs immediate solution. Besides, more and more countries are attaching increasing importance to air pollution supervision of Remote Sensing, because it is a quick and economic means to provide an objective and accurate dynamic scan of air pollution for decision-making.What this thesis mainly discusses are as follows:1 .Discussing the scientific basis of the correlation analysis of AOT(Aerosol Optics Thickness)and API;2. Using a new method to predict air quality from MODIS AOT product and according to the grade standards of API, cutting the API to many groups with certain steps; then establishing correlation models of mean values for each group. Discussing their precisions in the end;3. Dividing the data into 4 seasons by their characteristics and discussing the models of each. The results of the research show that:1 .MODIS AOT data can be used to analyze the distributing and changing trend of aerosol opticsthickness, and have more advantage in time and space ;2. The relative of MODIS AOT data and API can reflect the pollution status in land to a certainextent Especially in Summer Model, their relativity and the prediction precision are better. Withthe model to forecast the value of API, the rate of the predicted value and the exact value reaches86%.