Estimates of European uptake of CO2 inferred from GOSAT XCO2 retrievals: sensitivity to measurement bias inside and outside Europe

摘要

Estimates of the natural CO flux over Europe inferred from in situmeasurements of atmospheric CO mole fraction have been used previouslyto check top-down flux estimates inferred from space-borne dry-air COcolumn (X) retrievals. Several recent studies have shown thatCO fluxes inferred from X data from the JapaneseGreenhouse gases Observing SATellite (GOSAT) and the Scanning ImagingAbsorption Spectrometer for Atmospheric CHartographY (SCIAMACHY) have largerseasonal amplitudes and a more negative annual net CO balance thanthose inferred from the in situ data. The cause of this elevated Europeanuptake of CO is still unclear, but some recent studies have suggestedthat this is a genuine scientific phenomenon. Here, we put forward analternative hypothesis and show that realistic levels of bias in GOSAT datacan result in an erroneous estimate of elevated uptake over Europe. We use aglobal flux inversion system to examine the relationship between measurementbiases and estimates of CO uptake from Europe. We establish a referencein situ inversion that uses an Ensemble Kalman Filter (EnKF) to assimilateconventional surface mole fraction observations and Xretrievals from the surface-based Total Carbon Column Observing Network(TCCON). We use the same EnKF system to assimilate two independent versionsof GOSAT X data. We find that the GOSAT-inferred Europeanterrestrial biosphere uptake peaks during the summer, similar to thereference inversion, but the net annual flux is1.40 ± 0.19 GtC a compared to a value of0.58 ± 0.14 GtC a for our control inversion that uses only insitu data. To reconcile these two estimates, we perform a series of numericalexperiments that assimilate observations with added biases or assimilatesynthetic observations for which part or all of the GOSAT Xdata are replaced with model data. We find that for our global fluxinversions, a large portion (60–90 %) of the elevated European uptakeinferred from GOSAT data in 2010 is due to retrievals outside the immediateEuropean region, while the remainder can largely be explained by a sub-ppmretrieval bias over Europe. We use a data assimilation approach to estimatemonthly GOSAT X biases from the joint assimilation of in situobservations and GOSAT X retrievals. The inferred biasesrepresent an estimate of systematic differences between GOSATX retrievals and the inversion system at regional orsub-regional scales. We find that a monthly varying bias of up to 0.5 ppmcan explain an overestimate of the annual sink of up to 0.20 GtC a.Our results highlight the sensitivity of CO flux estimates to regionalobservation biases, which have not been fully characterized by the currentobservation network. Without further dedicated measurements we cannot proveor disprove that European ecosystems are taking up a larger-than-expectedamount of CO. More robust inversion systems are also needed to inferconsistent fluxes from multiple observation types.