Optimized forward model and retrieval scheme for MIPAS near-real-time dataprocessing

An optimized code to perform the near-real-time retrieval of profiles of pr essure, temperature, and volume mixing ratio (VMR) of five key species (O-3 , H2O, HNO3, CH4, and N2O) from infrared limb spectra recorded by the Miche lson Interferometer for Passive Atmospheric Sounding (MIPAS) experiment on board the European Space Agency (ESA) Environmental Satellite ENVISAT-1 was developed as part of a ESA-supported study. The implementation uses the gl obal fit approach on selected narrow spectral intervals (microwindows) to r etrieve each profile in sequence. The trade-off between run time and accura cy of the retrieval was optimized from both the physical and the mathematic al points of view, with optimizations in the program structure, in the radi ative transfer model, and in the computation of the retrieval Jacobian. The attained performances of the retrieval code are noise error on temperature <2 K at all the altitudes covered by the typical MIPAS scan (8-53 km with 3-km resolution), noise error on tangent pressure <3%, and noise error on V MR of the target species <5% at most of the altitudes covered by the standa rd MIPAS scan, with a total run time of less than 1 min on a modern worksta tion. (C) 2000 Optical Society of America.