A blind test retrieval experiment for infrared limb emission spectrometry

• Published in: Journal of Geophysical Research. D. Atmospheres 2003-12, Vol.108 (D23), p.ACH12.1-n/a
• Main Authors: von Clarmann, T., Ceccherini, S., Doicu, A., Dudhia, A., Funke, B., Grabowski, U., Hilgers, S., Jay, V., Linden, A., López-Puertas, M., Martín-Torres, F.-J., Payne, V., Reburn, J., Ridolfi, M., Schreier, F., Schwarz, G., Siddans, R., Steck, T.
• Format: Article
• Language: English
• Subjects: Earth, ocean, space; Exact sciences and technology; MIPAS; remote sensing; retrieval
• ISSN: 0148-0227; 2169-8996; 2169-897X; 2156-2202
• DOI: 10.1029/2003JD003835

The functionality and characteristics of six different data processors (i.e., retrieval codes in their actual software and hardware environment) for analysis of high‐resolution limb emission infrared spectra recorded by the space‐borne Michelson Interferometer for Passive Atmospheric Sounding (MIPAS) have been validated by means of a blind test retrieval experiment based on synthetic spectra. For this purpose a self‐consistent set of atmospheric state parameters, including pressure, temperature, vibrational temperatures, and abundances of trace gases and aerosols, has been generated and used as input for radiative transfer calculations for MIPAS measurement geometry and configuration. These spectra were convolved with the MIPAS field of view, spectrally degraded by the MIPAS instrument line shape, and, finally, superimposed with synthetic measurement noise. These synthetic MIPAS measurements were distributed among the participants of the project “Advanced MIPAS level‐2 data analysis” (AMIL2DA), who performed temperature and species abundance profile retrievals by inverse radiative transfer calculations. While the retrieved profiles of atmospheric state parameters reflect some characteristics of the individual data processors, it was shown that all the data processors under investigation are capable of producing reliable results in the sense that deviations of retrieved results from the reference profiles are within the margin that is consistent with analytical error estimation.