Inversions of CO2 Emissions from the Paris Area Using Yearlong Measurement Series

Abstract:

Estimates of fossil fuel CO2 emission from urban areas rely on inventories that are based upon energy-use consumption for different fuel types, various socio-economic activity data and emission factors. There is growing interest in improving those estimates by using an atmospheric inversion approach based on transport modeling and CO2 measurements.

Bréon et al. (ACPD, 2014) have recently developed a Bayesian inversion framework to control the daily CO2 fluxes of the Paris urban area during a 2 months period. The inversion framework relies on the transport model CHIMERE with a 2 km spatial resolution and uses atmospheric CO2 concentrations measurements obtained from three monitoring stations at the edge of the Paris urban area, installed as part of the CO2-MEGAPARIS and ICOS-France projects. The method relies on the measured daytime CO2 gradients between up- and downwind stations to correct the prior 6-hour mean emissions of the Paris area, given by the AIRPARIF regional air quality monitoring agency. The system, however, relies on the spatial distribution of the AIRPARIF inventory and does not attempt at correcting it.

Here, we apply the inversion framework to one year of atmospheric CO2 observations (August 2010 - July 2011). We show that the results for the monthly budgets exhibit a reasonable seasonal cycle. We check that the sensitivity to the prior estimates of the monthly budgets and to the meteorological forcing to CHIMERE is low, which demonstrates that the system is strongly controlled by observations.