doi:10.1016/j.cpc.2004.08.006
Copyright © 2004 Elsevier B.V. All rights reserved.
Multi-parameter regularization method for atmospheric remote sensing
References and further reading may be available for this article. To view references and further reading you must
purchase this article.
Adrian Doicu
,
, Franz Schreier, Siegfried Hilgers and Michael Hess
DLR—Remote Sensing Technology Institute, Oberpfaffenhofen, 82234 Wessling, Germany
Received 16 June 2004.
Available online 17 November 2004.
Abstract
The iteratively regularized Gauss–Newton algorithm with simple bounds on the variables is extended to the multi-parameter case. The global regularization matrix is computed by using the L-curve method for a sequence of one-parameter regularization problems. Numerical results concerning the joint retrieval of O3 and NO2 profiles using the scattered light from the limb of the atmosphere are presented.
Keywords: Inverse problems; Nonlinear least squares; Multi-parameter regularization; Atmospheric spectroscopy; Remote sensing
PACS: 02.60.-x; 07.05.Tp; 95.75.Fg; 95.75.Rs
Fig. 1. Geometry of limb scatter measurements.
Fig. 2. L-curve and GCV function for the O3 profile.
Fig. 3. Retrieved profiles using the regularization matrix selection algorithm.
Fig. 4. Retrieved profiles using a constant sequence of regularization parameters.
Fig. 5. Relative errors of the O3 profile.
Fig. 6. Relative errors of the NO2 profile.
Fig. 7. Retrieved NO2 profiles with and without auxiliary parameters regularization.

Corresponding author.