Development of a Merged CO Climate Data Record from IASI and MOPITT Observations
Résumé
Carbon monoxide (CO) is a key atmospheric compound that can be remotely sensed by satellite on a global scale. Continuous observations have been available since 2000 from the MOPITT/Terra instrument. Since 2007, the IASI/Metop instrument series has provided another homogeneous CO data record, thanks to the recent reprocessing of Metop-A and Metop-B data by EUMETSAT, resulting in the IASI CO Climate Data Record (IASI CO-CDR). Measuring the variability and trends of CO on a global scale is crucial as it serves as a precursor for ozone and carbon dioxide and regulates the troposphere's oxidizing capacity through its destruction cycle involving the hydroxyl radical (OH). As part of the ESA CCI+ Ozone Precursors project, we have been developing a merged CO Climate Data Record dataset combining IASI and MOPITT data to analyze long-term variability and trends. Monthly averaged gridded CO total columns (Level 3, 1°x1° resolution) are used as input. For IASI, we first apply an additional cloud mask to the Level 2 official data available on the Aeris French Database (https://iasi.aeris-data.fr/). We then compute monthly averages using IASI CO data from all Metop satellites, resulting in an intermediate (non-public) IASI CO monthly Level 3 product. For MOPITT, we use the official monthly Level 3 (version 9T) data available on the NASA Earth Data Portal (https://www.earthdata.nasa.gov/). We tested various methodologies for merging IASI and MOPITT CO Level 3 monthly grids. We performed averages with weighting schemes based on MOPITT priors and/or IASI/MOPITT uncertainties. In this poster, we will present the final version of the CO CCI merged product, which uses MOPITT CO total column/MOPITT prior ratios as weights for averaging. Among the different algorithm versions tested, this approach showed the best performance when validated against ground-based FTIR NDACC measurements, achieving a mean absolute bias below 5%, low standard deviation, and excellent correlation.