A MODEL FOR TROPOSPHERIC MULTIPHASE CHEMISTRY: APPLICATION TO ONE CLOUDY EVENT DURING THE CIME EXPERIMENT

摘要

In this study, we have used a multiphase box model which takes into account an explicit chemistry mechanism for both gas and aqueous phase for a rural environment and the kinetic of mass transfer between phases (Schwartz, 1986). The model is, then, initialized with multiphase measurements performed by Voisin et al. (2000) during the CIME experiment. The 13th December 1997, the cloud chemistry is mainly governed by high NO_x and high formaldehyde levels and by an acidic pH in the droplets. A comparison of the model results is performed versus recent theoretical results from Herrmann et al. (1999), who proposed a slightly different chemical scheme, including C2 chemistry and transition metal chemistry but neglecting some reaction pathways, such as the one involving OHCH_2O_2 radical and using unrealistic microphysical cloud conditions (cloud duration, constant liquid water content, small droplet radius). On the basis of this confrontation with theoretical results from Herrmann et al. (1999), a detailed analysis of radical chemistry and of the relative importance of those radicals in the oxidation of volatile organic compounds is performed. Depending on the stage of evolution of the cloud, either growing or dissipating, a different partitioning of organic volatile compounds is observed between the gas phase and the aqueous phase. These differences lead to different oxidation pathways that should be accounted for in simulating multiphase chemistry. Then, evidence for the possible role of aqueous peroxonitric acid in the conversion of S(Ⅳ) to S(Ⅵ), already underlined by Amels et al. (1996) has been found in case of low hydrogen peroxide concentration levels associated to the high NO_x regime observed during the experiment. These results show the capability of such a chemical box model, including exhaustive and explicit multiphase chemistry of highlighting wild variability in reaction and oxidation pathways, that cannot be directly put in evidence through actual measurements. They also indicate the need for lacking observations of gases such as formaldehyde that would greatly help this kind of approach.