Preparation for assimilation of Meteosat Third Generation infrared sounder radiances into an atmospheric composition model for ozone and carbon monoxide forecasts

Abstract In 2025, a new generation of infrared sounders was launched aboard a geostationary platform from the European Meteosat Third Generation (MTG) series. The MTG IRS is capable of providing an accurate four‐dimensional image of the state of the atmosphere for many meteorological and chemical variables. The objective of this study is to prepare for the assimilation of IRS radiances into the Modèle de Chimie Atmosphérique de Grande Échelle (MOCAGE) regional atmospheric composition model at Météo‐France and to assess the impact of these observations on ozone and carbon monoxide forecasts. At the time of the study, no real IRS data wwere available, which is why an Observing system simulation experiment framework was set up to construct accurate synthetic observations based on a realistic state of the atmosphere. A sensitivity analysis of these observations was conducted using the RTTOV radiative transfer model, which enabled the selection of 76 and 35 channels sensitive to ozone and carbon monoxide respectively. An experiment to assimilate these channels was carried out using a three‐dimensional variational system in the MOCAGE regional model, under clear skies, taking into account the respective observation and background errors. The results show significantly positive impacts for ozone and carbon monoxide forecasts, with reductions in forecast error of up to 60% and 40% respectively over a 24‐hr range.