Precision improvements in the geo-fit retrieval of pressure and temperature from MIPAS limb observations by modeling CO2 line-mixing

In this paper, we present an improvement of the retrieval of pressure and temperature from the observations of the Michelson Interferometer for Passive Atmospheric Sounding (MIPAS) on board of the environmental satellite (ENVISAT). The improvement has been obtained by integrating a state-of-the-art CO2 line-mixing model in a two-dimensional inversion system (Geo-fit). We describe the implementation of this model in the Geo-fit system and we show its capability to reproduce the CO2 spectral features affected by line-mixing. The upgraded analysis algorithm provides a better fit of the set of MIPAS observations analyzed by the European Space Agency (ESA) ground segment although these observations have been selected with criteria that should avoid line-mixing effects. Moreover, we show that this set of observations can be extended improving the precision of the retrieved pressure and temperature fields without increasing the computing demands. Since the CO2Q branches are very sensitive to pressure and temperature, the capability to model accurately the line-mixing effects opens the possibility to exploit at best these spectral regions to infer pressure and temperature distributions. According to this idea, we propose a new set of spectral intervals, including the most intense CO2Q branches. It is shown that the analysis of these intervals provides a significant improvement (up to 70%) in the precision of the retrieved pressure and temperature profiles, while using a smaller number of observations with respect to the ESA ground segment analysis. Since the knowledge of pressure and temperature is necessary for the retrieval of the altitude distribution of all the atmospheric constituents, the benefits of more precise pressure and temperature fields obtained in this work propagate into the quality of all the MIPAS products.