Applying the CO2M candidate algorithm FOCAL to OCO-2

Maximilian Reuter,  University of Bremen, Institute of Environmental Physics,  reuterm@loz.de (Presenter)
Michael Hilker,  University of Bremen, Institute of Environmental Physics,  mhilker@iup.physik.uni-bremen.de
Michael Buchwitz,  University of Bremen, Institute of Environmental Physics,  michael.buchwitz@iup.physik.uni-bremen.de
Stefan Noël,  University of Bremen, Institute of Environmental Physics,  stefan.noel@iup.physik.uni-bremen.de
Heinrich Bovensmann,  University of Bremen, Institute of Environmental Physics,  heinrich.bovensmann@iup.physik.uni-bremen.de
John P. Burrows,  University of Bremen, Institute of Environmental Physics,  burrows@iup.physik.uni-bremen.de

CO2 (carbon dioxide) and CH4 (methane) are the most important anthropogenic greenhouse gases and a reduction of their emissions is necessary to stop or slow down man-made climate change. The European Copernicus anthropogenic CO2 monitoring mission (CO2M) will be part of a global monitoring system aiming to verify mitigation strategies and to improve our understanding of anthropogenic and natural carbon fluxes. EUMETSAT will operate CO2M and is preparing for analyzing the data in its ground segment with three independently developed retrieval algorithms. One of these algorithms is FOCAL (fast atmospheric trace gas retrieval), which has been developed at the University of Bremen. FOCAL has successfully been applied to observations from OCO-2, GOSAT, and GOSAT-2. Retrieval studies with simulated data are essential when preparing for a new satellite mission. However, it turned out, that the retrieval algorithms usually need significant adaptations when applying them to the observations from instruments in orbit. Amongst other issues, this usually requires pre-processing steps, adaptations of the measurement error covariance, handling of systematic residuals, outlier detection, and bias correction. The CO2M measurements are similar to but more extensive than those of NASA’s OCO instruments. Consequently, a key goal for the CO2M FOCAL algorithm development is to apply it to OCO-2 data and develop universal strategies to optimize the algorithm for an existing instrument; an activity which is co-funded by EUMETSAT’s CO2M GHG L2 scientific support services. In this context, we will present the most recent FOCAL developments and optimizations for OCO-2. These advances comprise, e.g., a new pre-processor which includes cloud detection, a climatology-based CO2 a priori, and a random forest-based post processor.

Presentation: Talk_Reuter_36_25.pdf 

Presentation Type: Talk

Session: 2.1 Retrieval algorithms and methods for inter-instrument and product Cal/Val

Session Date: Tuesday (6/15) 9:00 AM