Impacts of prescribed fossil fuel CO2 emissions on subnational level inverse flux estimates

Atmospheric-based approaches have been recognized as promising tools for QA/QC and verification of greenhouse gas (GHG) emission inventories reported by countries. Atmospheric-based approaches also help provide GHG estimates for countries and regions with less robust inventory building capacities, and direct estimates for key subnational levels, which are not often covered by national inventories. Conventional CO₂ flux inversion approaches, unlike urban inversion applications, often prescribe fossil fuel CO₂ emissions (FFCO2) and mostly optimized natural fluxes. Thus, errors in prescribed FFCO2 impact the final flux estimates.

We implemented two sets of inversions with two different emission inventories in order to examine the impact of the prescribed FFCO2 on the inverse flux estimates. The emission inventory difference was used to approximate potential errors in the prescribed FFCO2. Our inversion result demonstrated how the FFCO2 errors, particularly due to sub-annual seasonal emission pattern differences, can have an impact on posterior emission estimates via flux optimization. Our result also demonstrated that FFCO2 errors from large emitting countries could significantly bias sub-national flux estimates by mis-attributing their flux corrections to natural fluxes to compensate for the FFCO2 errors. The magnitude of the potential errors might be small compared to that of large-scale fluxes. However, the errors could be significant in relation to small sub-national scale fluxes or emissions from lesser emitting countries. We also examined the impact of two observation systems, such as global in-situ network and a satellite, to FFCO2 errors.

We discuss the existing challenges that need to be addressed to further enhance the use of atmospheric inversions with country level inventories. In addition to improvements in inversions, improving inventories (prescribed FFCO2) should have a direct benefit to improved accuracy of inverse flux estimates. For example, extended data collection at sub-national scales should greatly mitigate potential errors in the prescribed FFCO2. This study highlights the importance of developing GHG emission information in a hybrid fashion to support science and the emission reporting and monitoring.