 |
| Titel |
A self-adapting and altitude-dependent regularization method for atmospheric profile retrievals |
| VerfasserIn |
M. Ridolfi, L. Sgheri |
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| ISSN |
1680-7316
|
| Digitales Dokument |
URL |
| Erschienen |
In: Atmospheric Chemistry and Physics ; 9, no. 6 ; Nr. 9, no. 6 (2009-03-17), S.1883-1897 |
| Datensatznummer |
250007082
|
| Publikation (Nr.) |
 copernicus.org/acp-9-1883-2009.pdf |
|
|
|
|
|
| Zusammenfassung |
MIPAS is a Fourier transform spectrometer, operating onboard of the ENVISAT
satellite since July 2002. The online retrieval algorithm produces geolocated
profiles of temperature and of volume mixing ratios of six key atmospheric
constituents: H2O, O3, HNO3, CH4, N2O
and NO2. In the validation phase, oscillations beyond the error bars
were observed in several profiles, particularly in CH4 and
N2O.
To tackle this problem, a Tikhonov regularization scheme has been implemented
in the retrieval algorithm. The applied regularization is however rather weak
in order to preserve the vertical resolution of the profiles.
In this paper we present a self-adapting and altitude-dependent
regularization approach that detects whether the analyzed observations
contain information about small-scale profile features, and determines the
strength of the regularization accordingly. The objective of the method is to
smooth out artificial oscillations as much as possible, while preserving the
fine detail features of the profile when related information is detected in
the observations.
The proposed method is checked for self consistency, its performance is
tested on MIPAS observations and compared with that of some other
regularization schemes available in the literature. In all the considered
cases the proposed scheme achieves a good performance, thanks to its altitude
dependence and to the constraints employed, which are specific of the
inversion problem under consideration. The proposed method is generally
applicable to iterative Gauss-Newton algorithms for the retrieval of vertical
distribution profiles from atmospheric remote sounding measurements. |
| |
|
| Teil von |
|
|
|
|
|
|
|
|