Atmospheric Measurement Techniques
The differential optical absorption spectroscopy (DOAS) method is a well-known remote sensing technique that is nowadays widely used for measurements of atmospheric trace gases, creating the need for harmonization and characterization efforts. In this study, an intercomparison exercise of DOAS retrieval codes from 17 international groups is presented, focusing on NO2 slant columns. The study is based on data collected by one instrument during the Multi-Axis DOAS Comparison campaign for Aerosols and Trace gases (MAD-CAT) in Mainz, Germany, in summer 2013. As data from the same instrument are used by all groups, the results are free of biases due to instrumental differences, which is in contrast to previous intercomparison exercises.While in general an excellent correlation of NO2 slant columns between groups of > 99.98% (noon reference fits) and > 99.2% (sequential reference fits) for all elevation angles is found, differences between individual retrievals are as large as 8% for NO2 slant columns and 100% for rms residuals in small elevation angles above the horizon.Comprehensive sensitivity studies revealed that absolute slant column differences result predominantly from the choice of the reference spectrum while relative differences originate from the numerical approach for solving the DOAS equation as well as the treatment of the slit function. Furthermore, differences in the implementation of the intensity offset correction were found to produce disagreements for measurements close to sunrise (8-10% for NO2, 80% for rms residual). The largest effect of [approximate] 8% difference in NO2 was found to arise from the reference treatment; in particular for fits using a sequential reference. In terms of rms fit residual, the reference treatment has only a minor impact. In contrast, the wavelength calibration as well as the intensity offset correction were found to have the largest impact (up to 80%) on rms residual while having only a minor impact on retrieved NO2 slant columns.
Koenig, Theodore; Ortega, Ivan; Volkamer, Rainer; and for full list of authors., See bottom of the page, "Investigating differences in DOAS retrieval codes using MAD-CAT campaign data" (2017). Chemistry & Biochemistry Faculty Contributions. 67.