Source‐receptor relationships of column‐average CO2 and implications for the impact of observations on flux inversions

Abstract

AbstractSource‐receptor relationships are the fundamental quantities used in atmosphere CO2 flux inversions. In this study, we systematically investigate the global source‐receptor relationships of column CO2 (XCO2) in 12 continental‐scale receptors in terms of transport and local versus nonlocal flux contributions using the GEOS‐Chem adjoint model. Using simulated Greenhouse gases Observing Satellite (GOSAT) XCO2, we quantify the impact of inclusion (add‐on) or exclusion of observations (data‐denial) within a receptor region on flux inversion. We discuss the connections between the observation impact and the underlying source‐receptor relationships. The strong sensitivity of XCO2 to nonlocal fluxes makes the XCO2 observations have strong impact on nonlocal flux estimation. On an annual mean, the impact of GOSAT XCO2 over Europe on North America flux estimation is 10%–39% of the full observation impact. Because the Southern Hemisphere midlatitude XCO2 are most sensitive to local and tropical fluxes, the mean impact of GOSAT XCO2 over Southern South America on Northern South America (N‐S‐America) flux estimation is 30%–59% of the full observation impact. Because of the strong sensitivity of the XCO2 over N‐S‐America to Central Africa (C‐Africa) fluxes, the mean impact on C‐Africa flux estimation is between 14% and 31% of the full observation impact in spite of the sparse observation coverage. The results also show that XCO2 have similar sensitivity to local and nonlocal fluxes occurring 3 months before, which indicates that XCO2 observations cannot differentiate any fluxes occurring beyond 3 months.