Estimating the evolution of CMS-Flux fossil fuel emissions from air quality measurements

Abstract:

Rapid regional changes in anthropogenic emissions in response to the COVID-19 pandemic have underscored the contribution of fossil fuel (FF) emission uncertainty to regional carbon budgets, which typically assumes no anthropogenic uncertainty. Typical methods for spatially-explicit FF emissions are dependent on national reporting, which can incur substantial latencies. However, the concomitant changes in short-lived pollutants from common emission sources point to opportunities to develop independent low-latency estimates of fossil fuel emissions and to better understand anthropogenic processes. Here we combine state-of-the-art Multiple Model Multi Constituent chemical data assimilation system (MOMO-Chem) with bottom-up FF emissions to repartition the net carbon fluxes from the NASA Carbon Monitoring System Flux (CMS-Flux) project. To that end, we implement a novel Kalman filtering algorithm that predicts emission ratio co-evolution of air quality (AQ) and carbon species. Based upon top-down estimates of AQ emissions, FF CO2 emissions and uncertainties can be rapidly determined. We show overall good agreement between predicted FF fluxes and the latest bottom-up inventories. These data are in turn used to interpret the decadal evolution of CMS-Flux net carbon exchange. This approach is an important step in quantifying both regional fossil fuel and natural carbon fluxes contributions to the atmospheric CO2 growth rate.