1Laboratory for Atmospheric Research, Dept. of Civil and Environmental Engineering, Washington State University, Pullman, WA, USA
2Molina Center for Energy and the Environment (MCE2), La Jolla, CA, USA
3Dept. of Earth, Atmospheric and Planetary Sciences, Massachusetts Institute of Technology, Cambridge, MA, USA
4Secretaría del Medio Ambiente del Gobierno del Distrito Federal, México D. F., México
*now at: Dept. of Geography, Faculty of Arts and Social Sciences, National University of Singapore, Singapore
**now at: William J. Clinton Foundation, Clinton Climate Initiative, México D. F., México
Abstract. Eddy covariance (EC) flux measurements of the atmosphere/surface exchange of gases over an urban area are a direct way to improve and evaluate emissions inventories, and, in turn, to better understand urban atmospheric chemistry and the role that cities play in regional and global chemical cycles. As part of the MCMA-2003 study, we demonstrated the feasibility of using eddy covariance techniques to measure fluxes of selected volatile organic compounds (VOCs) and CO2 from a residential district of Mexico City (Velasco et al., 2005a, b). During the MILAGRO/MCMA-2006 field campaign, a second flux measurement study was conducted in a different district of Mexico City to corroborate the 2003 flux measurements, to expand the number of species measured, and to obtain additional data for evaluation of the local emissions inventory. Fluxes of CO2 and olefins were measured by the conventional EC technique using an open path CO2 sensor and a Fast Isoprene Sensor calibrated with a propylene standard. In addition, fluxes of toluene, benzene, methanol and C2-benzenes were measured using a virtual disjunct EC method with a Proton Transfer Reaction Mass Spectrometer. The flux measurements were analyzed in terms of diurnal patterns and vehicular activity and were compared with the most recent gridded emissions inventory. In both studies, the results showed that the urban surface of Mexico City is a net source of CO2 and VOCs with significant contributions from vehicular traffic. Evaporative emissions from commercial and other anthropogenic activities were significant sources of toluene and methanol. The data show that the emissions inventory is in reasonable agreement with measured olefin and CO2 fluxes, while C2-benzenes and toluene emissions from evaporative sources are overestimated in the inventory. It appears that methanol emissions from mobile sources occur, but are not present in the mobile emissions inventory.