Characteristics of multi-year-long regional CO2 fluxes estimated from GOSAT XCO2 retrievals

Abstract:

Since its successful launch in early 2009, the Greenhouse gases Observing SATellite (GOSAT) has been making global measurement of reflected light spectra from which column-averaged CO₂ concentrations (X_CO2) are retrieved. The X_CO2 values, retrieved under clear-sky conditions, have been utilized together with the surface-based GLOBALVIEW-CO2 (GV) data in the inference of monthly CO₂ fluxes on sub-continental and ocean-basin scales (42 terrestrial and 22 oceanic regions). The results of the flux inference calculation have been made public as the GOSAT Level 4 data product by the National Institute for Environmental Studies GOSAT Project. We herein present the overview of the GOSAT Level 4 inversion approach and the characteristics of multi-year time series of the obtained regional-scale CO₂ flux estimates.

The GOSAT-GV combined inversion estimates for many of the land regions were found to exhibit larger seasonal-cycle amplitudes compared to those of the GV-only estimates. The annual global-total terrestrial uptakes inferred by the GOSAT-GV inversion turned out to be smaller than those by the GV-only inversion, but the opposite trends were found in terrestrial uptakes for some latitudinal-bands, particularly in the Southern Hemisphere.

Touched also in this report is the evaluation of balancing GOSAT data amount that vary with time and locations over the globe. The data amount balancing was considered necessary since successful GOSAT retrievals are obtained mostly over sunny periods and areas and thus their number distributions over the globe can be uneven. In our current inversion approach, individual GOSAT X_CO2 retrievals are gridded to 5°×5° cells and averaged on a monthly basis (GV data are also monthly averaged). Inversion results based on “single-shot” concentration data were obtained for comparisons. We will discuss the differences observed and their implications for better satellite-based flux estimation.