Global CO2 flux estimation using GOSAT: An inter-comparison of inversion results
Abstract
A unique global data archive is under construction of total column CO2 measurements retrieved from the Greenhouse gas Observing SATellite, currently spanning more than three years of data. Several groups are investigating the application of these data to global atmospheric inverse modelling for studying the global carbon cycle. It is known from inverse modeling using surface measurements that the robustness of the inversion-estimated fluxes is best analyzed using a multi-model approach. So far, this has not been demonstrated for inversions using satellite data, but but some of the known sources of uncertainty are difficult to account for in a single inversion, such as transport model uncertainties and differences between retrieval methods. We have organized an inversion inter-comparison experiment to investigate whether, despite these uncertainties, robust signals of sources and sinks can be inferred from the GOSAT data. The current experiment allows full freedom in inversion set-up in order to avoid limiting the range of possible outcomes. Each participating group is free to use their preferred inversion set-up, transport model, and measurements, but is asked to report in a common format and for a common time period of one year to allow one-to-one comparison. We will present an overview of the status of the experiment, including a preliminary synthesis of large-scale CO2 fluxes from a statistical analysis of the ensemble of inversion results and verification of the performance of the inversions using independent measurements.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2012
- Bibcode:
- 2012AGUFM.A42F..01H
- Keywords:
-
- 0315 ATMOSPHERIC COMPOSITION AND STRUCTURE / Biosphere/atmosphere interactions;
- 0322 ATMOSPHERIC COMPOSITION AND STRUCTURE / Constituent sources and sinks;
- 0368 ATMOSPHERIC COMPOSITION AND STRUCTURE / Troposphere: constituent transport and chemistry;
- 0545 COMPUTATIONAL GEOPHYSICS / Modeling