Direct-fitting: a new approach for trace gases retrieval in the UV-visible spectral range
Michel Van Roozendael(1), Robert Spurr(2), Christophe Lerot(3), Jean-Christopher Lambert(3), Jos van Geffen(3), Jeroen van Gent(3) and Claus Zehner(4)
(1) BIRA-IASB, Ringlaan, 3, 1180 Brussels, Belgium
(2) RT-Solutions, Inc., 9, Channing Street, MA 02138, United States
(3) BIRA-IASB, 3, Ringlaan, 1180 Brussels, Belgium
(4) ESA-ESRIN, Via Galileo Galilei, 00044 Frascati, Italy
We present a new direct fitting algorithm developed for the retrieval of trace gas column amounts from remote sensing UV/visible spectrometers. In contrast to the commonly used Differential Optical Absorption Spectroscopy (DOAS) method, which relies on a two-step approach involving a slant column fitting followed by an air mass factor calculation, the algorithm makes a direct fit for the vertical column obtained through a direct comparison between simulated and measured spectral radiances. The inversion is based on a non-linear least-squares minimization where simulated radiances and weighting functions are calculated “on-the-fly” using, as forward model, the linearized radiative transfer (RT) model LIDORT. Ring effect resulting from rotational Raman scattering by molecular nitrogen and oxygen is treated using a parameterization of the Raman scattering source term that accounts for atmospheric absorption or using look-up-tables of correction spectra pre-calculated using LIDORT-RRS, an off-line version of the RT code where Raman scattering is included. In the retrieval process, ozone profiles are iteratively adjusted to the retrieved columns according to the TOMS version 8 ozone profile climatology. The code also includes handles to alternative atmospheric input data sets such as the Bremen DOC or the recently developed CHEOPS climatology. Temperature profiles are taken from the TOMS climatology and further adjusted as part of the retrieval.
The direct-fitting code is primarily designed for accurate total ozone retrieval from space nadir-looking sensors like GOME, SCIAMACHY, OMI and GOME-2. However it can also be applied to the retrieval of minor absorbers in the UV and visible range, like SO2, BrO, HCHO and NO2. Unlike DOAS, direct-fitting is not restricted to optically thin conditions and can handle strong non-linear absorption features such as those of ozone below 325 nm. In the present work, we concentrate on total ozone retrieval from GOME and SCIAMACHY nadir observations. First a characterization of the global error budget is presented according to Rodgers formulation. Sources of uncertainties at low sun in polar regions are highlighted. The study also includes comparisons with alternative retrievals using DOAS approaches. A particular attention is paid to the sensitivity of the inversion to the wavelengths selected for retrieval. Intervals in the range from 317 to 335 nm are investigated with the aim to establish the link between DOAS-like and TOMS-like algorithms and therefore contribute to a better assessment of the consistency of the long-term multi-platform total ozone data record.
Keywords: ESA European
Space Agency - Agence spatiale europeenne,
observation de la terre, earth observation,
satellite remote sensing,
teledetection, geophysique, altimetrie, radar,
chimique atmospherique, geophysics, altimetry, radar,