Atmospheric Chemistry and Physics Discussions www.atmos-chem-phys-discuss.net 1680-7367 1680-7375 6 2 2006 10.5194/acpd-6-2765-2006 http://www.atmos-chem-phys-discuss.net/6/2765/2006/ http://www.atmos-chem-phys-discuss.net/6/2765/2006/acpd-6-2765-2006.html http://www.atmos-chem-phys-discuss.net/6/2765/2006/acpd-6-2765-2006.pdf 2765 2807 2006-04-10 Measuring atmospheric CO₂ from space using Full Spectral Initiation (FSI) WFM-DOAS M. P. Barkley U. Frieß P. S. Monks EOS, Space Research Centre, Department of Physics & Astronomy, University of Leicester, Leicester, LE1 7RH, UK Department of Chemistry, University of Leicester, Leicester, LE1 7RH, UK now at: Institute of Environmental Physics, Heidelberg, Germany Satellite measurements of atmospheric CO₂ concentrations are a rapidly evolving area of scientific research which can help reduce the uncertainties in the global carbon cycle fluxes and provide insight into surface sources and sinks. One of the emerging CO₂ measurement techniques is a relatively new retrieval algorithm called Weighting Function Modified Differential Optical Absorption Spectroscopy (WFM-DOAS) that has been developed by Buchwitz et al. (2000). This algorithm is designed to measure the total columns of CO₂ (and other greenhouse gases) through the application to spectral measurements in the near infrared (NIR), made by the SCIAMACHY instrument on-board ENVISAT. The algorithm itself is based on fitting the logarithm of a model reference spectrum and its derivatives to the logarithm of the ratio of a measured nadir radiance and solar irradiance spectrum. In this work, a detailed error assessment of this technique has been conducted and it has been found necessary to include suitable a priori information within the retrieval in order to minimize the errors on the retrieved CO₂ columns. Hence, a new CO₂ retrieval algorithm called Full Spectral Initiation (FSI) WFM-DOAS has been developed which generates a reference spectrum for each individual SCIAMACHY observation using the known properties of the atmosphere and surface at the time of the measurement. Initial retrievals over Siberia during the summer of 2003 show that the measured CO₂ columns are not biased from the input a priori data and that whilst the monthly averaged CO₂ distributions contain a high degree of variability, they also contain significant spatial features.