Développement et évaluation d'un modèle explicite de formation d'aérosols organiques secondaires : sensibilité aux paramètres physico-chimiques

Abstract : Fine aerosols have an important impact on health, visibility and climate. Secondary Organic Aerosols (SOA) represent an important fraction of fine aerosol composition. SOA are formed by nucleation or condensation onto pre-existing particles of gaseous species formed during the oxidation of emitted volatile organic compounds (VOC). VOC oxidation implies a huge number of secondary intermediates which are potentially involved in SOA formation. In order to study SOA formation, it is necessary to develop chemical schemes describing explicitly the formation and condensation of the gaseous secondary intermediates. The LISA has thus developed in collaboration with NCAR (National Center of Atmospheric Research) a generator of explicit chemical schemes : GECKO-A (Generator for Explicit Chemistry and Kinetics of Organics in the Atmosphere). This work aims at testing (i) the reliability of GECKO-A to simulate observed SOA concentrations in Atmospheric Simulation Chamber (ASC) and (ii) exploring the SOA sensitivity to physico-chemical parameters such as saturation vapour pressures, chamber walls effects or kinetics rate constants. In order to assess GECKO-A's chemical schemes, the model has been confronted to chamber experiments performed to study SOA. Saturation vapour pressure (Pvap) is the key parameter controlling the gas/particles partitioning of organic compounds The three Pvap estimation methods considered as the more reliable in the literature have been implemented in GECKO-A. Pvap estimated by the three methods differs highly, up to several orders of magnitude. Despite of these discrepancies, simulated SOA concentration and speciation show a low sensitivity to the method used to estimate the Pvap. Moreover, none of the methods were able to make the model fit the observations. SOA concentration is systematically overestimated of a factor 2. Semi volatile organic compounds deposition on a chamber walls has been investigated. The implementation of this process in the model leads to a significant decrease of the simulated SOA concentrations, up to factor of 2. Simulated SOA yields are in good agreement with measured SOA yields. The hypothesis of a misrepresentation of some gaseous processes has then been investigated through sensitivity tests. SOA formation sensitivity to COV+ OH reactions rate constants has been explored. Results exhibited a high sensitivity to the rate constants estimations (regarding the rate constants values estimation, as well as the determination of the OH attack sites). The estimated alkoxy radicals decomposition rate constants have also been tested. This test showed however no significant impact on the simulated SOA yields
Document type :
Sciences de la Terre. Université Paris-Est, 2011. Français. <NNT : 2011PEST1126>

Contributor : Abes Star <>
Submitted on : Monday, February 27, 2012 - 1:52:40 PM
Last modification on : Tuesday, October 11, 2016 - 2:04:35 PM
Document(s) archivé(s) le : Monday, May 28, 2012 - 2:45:57 AM


Version validated by the jury (STAR)


  • HAL Id : tel-00674473, version 1



Richard Valorso. Développement et évaluation d'un modèle explicite de formation d'aérosols organiques secondaires : sensibilité aux paramètres physico-chimiques. Sciences de la Terre. Université Paris-Est, 2011. Français. <NNT : 2011PEST1126>. <tel-00674473>




Record views


Document downloads