Kearney, CS, Watson, IM (2009). Correcting satellite-based infrared sulfur dioxide retrievals for the presence of silicate ash. JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES, 114, D22208.
Previous studies have suggested that the presence of silicate ash may cause the 8.6-mu m sulfur dioxide (SO2) retrieval to overestimate total column abundances and contribute to variations between ultraviolet and 8.6-mu m SO2 estimates. Tested here is a volcanic aerosol forward model which allows the determination of transmission spectra. The forward model is embedded into a three-layer atmospheric radiative transfer model, based upon the moderate spectral resolution atmospheric transmittance algorithm (MODTRAN) radiative transfer code. This model is used to investigate the effects of silicate ash on the 8.6-mu m SO2 retrieval from satellite data. Forward modeling of transmission through volcanic clouds enables us to evaluate the assumptions and errors of current retrievals. Four free variables, (1) effective radius, (2) variance, (3) number of particles, and (4) spectral refractive index, are used to determine the transmission and single scattering albedo at 7-13 mu m. Here we present simplistic characterizations of three eruptions, containing sulfates, ice, or andesitic ash using this forward model. The ash correction model based on the aerosol forward model is tested on the ash-rich lobe of the 19 February 2001 eruption plume of Mount Cleveland volcano using Moderate Resolution Imaging Spectroradiometer (MODIS) data. Uncorrected MODIS radiance values estimate 22 kt of SO2, while ash-corrected radiances estimate 17 kt. The minimal reduction in total SO2 indicates that previous variations in ultraviolet and 8.6-mu m SO2 retrievals are unlikely to be due solely to the presence of ash.