Retrieval of aerosol mass load (PM₁₀) from MERIS/Envisat top of atmosphere spectral reflectance measurements

Abstract

Abstract. A retrieval of particulate matter concentration (PM10) from satellite data is presented as well as improvements of the aerosol optical depth retrieval by the Bremen aerosol algorithm. The add-on retrieval of particulate matter uses the derivation of the effective radii from the Ångstroem exponents and an assumed log-normal size distribution function. The Ångstroem exponent is derived through the multi-channel approach using MERIS/Envisat data, and benefits from the fitting of a smooth spectral slope of aerosol optical depth and the surface reflectance. The advantage of the retrieval is that this retrieval of the aerosol mass, i.e., in particular the effective radius, is exclusively based on spectral information from satellite measurements, global aerosol models, and meteorological parameters. ECMWF Boundary layer height, humidity, temperature, and pressure data are used in the retrieval; a retrieval of PM inferred without meteorological information and a proper BRDF is shown to be not very promising. Over the city of Hamburg, the aerosol optical depths agree within a standard deviation of 0.03 and 0.068 for all thirteen wavelengths between 412 and 885 nm, compared with AERONET and ground based air quality measurements; the particulate matter concentrations show agreement with a correlation factor of 0.64. In addition to the urban site of Hamburg, comparisons of PM10 measurements over rural sites in Germany exhibit a correlation coefficient of 0.75.