Air pollution mapping: relationship between satellite- made observations and air quality parameters

The mapping of air quality over the city of Strasbourg and its vicinity requires an increase of the number of measurements aiming at spatializing information about air pollutants concentrations. A methodology based on a multisource approach for mapping air pollutants concentrations over the city was presented in previous communications. One step of the methodology is the estimation of pollutants concentrations by remote sensing techniques. This paper investigates the strength of linear relationship between satellite-made observations and air quality parameters. Multidate correlations were calculated between reflectances measured by Landsat satellite and measurements of the various pollutants. Those measurements are concentrations of pollutants on a daily average basis, or for the time of image acquisition, or on a basis of average of several hours before the image acquisition. The results add evidence on linear correlation between air quality parameters and satellite measurements not only in thermal infrared range but also in the infrared range. Keys-words: remote sensing, air pollution, radiometric calibration, reflectance, correlation.