Constraints on the oxidative capacity of the atmosphere: Implications for top-down CO emission estimates

Atmospheric carbon monoxide (CO) emissions estimated from inverse modeling analyses exhibit large uncertainties, due, in part, to discrepancies in the tropospheric chemistry in atmospheric models. In particular, the inversion analyses are sensitive to discrepancies in tropospheric OH, the main sink of CO. We attempted to reduce the uncertainties in CO emission estimates by constraining the modeled abundance of ozone (O₃), nitrogen dioxide (NO₂), nitric acid (HNO₃), and formaldehyde (HCHO), which are constituents that play a key role in tropospheric chemistry. Using the GEOS-Chem four-dimensional Variational (4D-Var) data assimilation system, we estimated CO emissions by assimilating observations of CO from the Measurement of Pollution In the Troposphere (MOPITT) instrument, together with observations of O₃from the Optical Spectrograph and InfraRed Imager System (OSIRIS) and the Infrared Atmospheric Sounding Interferometer (IASI), NO₂and HCHO from the Ozone Monitoring Instrument (OMI), and HNO₃from the Microwave Limb Sounder (MLS). By integrating these data in the 4D-Var scheme, we obtained an estimate of global tropospheric mean OH was that was 8% lower than the a priori, with the reductions mainly to the tropics and subtropics. The a posteriori OH estimate was 1.2 x 10⁶molec cm^-1compared to the a priori of 1.3 x 10⁶molec cm^-1. As a result of the changes in OH, the assimilation suggested anthropogenic CO emissions for the United States, Europe, and Asia of 47, 56, and 264 Tg CO, respectively. The estimated global anthropogenic source was 473 Tg CO, compared to the a priori of 401 Tg CO. In contrast, we obtained a total anthropogenic source of 562 Tg CO when we assimilated only MOPITT data and did not correct for model discrepancies in OH.