Improved definition of prior uncertainties in CO2 and CO fossil fuel fluxes and its impact on multi-species inversion with GEOS-Chem (v12.5)

Monitoring, reporting, and verification frameworks for greenhouse gas emissions are being developed by countries across the world to keep track of progress towards national emission reduction targets. Data assimilation plays an important role in monitoring frameworks, combining different sources of information to achieve the best possible estimate of fossil fuel emissions and, as a consequence, better estimates for fluxes from the natural biosphere. Robust estimates for fossil fuel emissions rely on accurate estimates of uncertainties corresponding to different pieces of information. We describe prior uncertainties in CO2 and CO fossil fuel fluxes, paying special attention to spatial error correlations and the covariance structure between CO2 and CO. This represents the first time that prior uncertainties in CO2 and the important co-emitted trace gas CO are defined consistently, with error correlations included, which allows us to make use of the synergy between the two trace gases to better constrain CO2 fossil fuel fluxes. CO:CO2 error correlations differ by sector, depending on the diversity of sub-processes occurring within a sector, and also show a large range of values between pixels within the same sector. For example, for other stationary combustion, pixel correlation values range from 0.1 to 1.0, whereas for road transport, the correlation is mostly larger than 0.6. We illustrate the added value of our definition of prior uncertainties using closed-loop numerical experiments over mainland Europe and the UK, which isolate the influence of using error correlations between CO2 and CO and the influence of prescribing more detailed information about prior emission uncertainties. For the experiments, synthetic in situ observations are used, allowing us to validate the results against a “truth”. The “true” emissions are made by perturbing the prior emissions (from an emission inventory) according to the prescribed prior uncertainties. We find that using our realistic definition of prior uncertainties helps our data assimilation system to differentiate more easily between CO2 fluxes from biogenic and fossil fuel sources. Using improved prior emission uncertainties, we find fewer geographic regions with significant deviations from the prior compared to when using default prior uncertainties (32 vs. 80 grid cells of 0.25°×0.3125°, with an absolute difference of more than 1 kg s-1 between the prior and posterior), but these deviations from the prior almost c