Estimation of urban PM10 concentration, based on MODIS and MERIS/AATSR synergistic observations

Abstract

Multiple linear and higher order equations were developed for the estimation of the PM10 spatial distribution over the broader area of Athens, Greece, based on satellite data. In situ PM10 measurements for the period 2007–2010 were used to calibrate the satellite observations. The latter included Aerosol Optical Thickness (AOT) derived from MODIS (Moderate Resolution Imaging Spectroradiometer) and MERIS (MEdium Resolution Imaging Spectrometer) and AATSR (Advanced Along-Track Scanning Radiometer) sensors, as well as surface relative humidity, surface temperature and K-Index derived from MODIS. The performance of linear and non-linear relationships and the contribution of each satellite-derived parameter were analyzed based on a stepwise validation. Results found that non-linear relations perform better than simple linear, while the K-Index, an estimator of atmospheric static stability, can be used as a proxy of the vertical aerosol mixing. Coefficients of determination around 0.7 were achieved. Seasonal mean PM10 distributions were also computed, revealing an intra-annual variation, which obtained higher values during the summer, along with spatial differences in PM10 concentrations which were attributed to air pollution at the city center and at industrial areas.