Assessment of TOMS UV bias due to absorbing aerosols
Loading...
Author/Creator ORCID
Date
2005-12-10
Type of Work
Department
Program
Citation of Original Publication
Arola, A., Kazadzis, S., Krotkov, N., Bais, A., Gröbner, J., and Herman, J. R. (2005), Assessment of TOMS UV bias due to absorbing aerosols, J. Geophys. Res., 110, D23211, doi:10.1029/2005JD005913.
Rights
This work was written as part of one of the author's official duties as an Employee of the United States Government and is therefore a work of the United States Government. In accordance with 17 U.S.C. 105, no copyright protection is available for such works under U.S. Law.
Public Domain Mark 1.0
Public Domain Mark 1.0
Subjects
Abstract
We compared NASA Total Ozone Mapping Spectrometer (TOMS) overpass irradiances against ground-based Brewer measurements at Ispra, Italy, and Thessaloniki, Greece, with the main objective of evaluating the effect of absorbing aerosols on TOMS UV bias using direct measurements of aerosol optical properties. Dependence of the TOMS/Brewer bias on aerosol absorption optical depth is significant. Our study demonstrates that the bias between TOMS and measured Brewer 324 nm irradiances increases with increasing aerosol absorption optical depth τₐbₛ. The correlation coefficients between the ratio TOMS/Brewer and τₐbₛ at Ispra and Thessaloniki are ∼0.7 or more and ∼0.8 or more, respectively, depending on the range of solar zenith angle values selected for the analysis. We found that the correlation with single-scattering albedo ω or aerosol optical depth is significantly smaller than with τabs. If measurements or climatology of τₐbₛ are available in the UVB range, the TOMS UV product can be postcorrected for absorbing aerosols using similar site-dependent regressions as established in our study. If no correction is applied, the mean positive biases between TOMS and Brewer 324 nm irradiances in Thessaloniki and Ispra are 19.2% and 29.7%, respectively, while the standard deviations are 8.9% and 16.5%. Depending on the level of correction, the mean positive bias can be essentially removed, and the standard deviations can be reduced down to ∼6% and ∼10%, respectively.