Skip to Main content Skip to Navigation

Modeling of particle mixing in the atmosphere

Abstract : This thesis present a newly developed size-composition resolved aerosol model (SCRAM), which is able to simulate the dynamics of externally-mixed particles in the atmosphere, and it evaluates its performance in three-dimensional air-quality simulations. The main work is split into four parts. First, the research context of external mixing and aerosol modelling is introduced. Secondly, the development of the SCRAM box model is presented along with validation tests. Each particle composition is defined by the combination of mass-fraction sections of its chemical components or aggregates of components. The three main processes involved in aerosol dynamic (nucleation, coagulation, condensation/evaporation) are included in SCRAM. The model is first validated by comparisons with published reference solutions for coagulation and condensation/evaporation of internally-mixed particles. The particle mixing state is investigated in a 0-D simulation using data representative of air pollution at a traffic site in Paris. The relative influence on the mixing state of the different aerosol processes and of the algorithm used to model condensation/evaporation (dynamic evolution or bulk equilibrium between particles and gas) is studied. Then, SCRAM is integrated into the Polyphemus air quality platform and used to conduct simulations over Greater Paris during the summer period of 2009. This evaluation showed that SCRAM gives satisfactory results for both PM2.5/PM10 concentrations and aerosol optical depths, as assessed from comparisons to observations. Besides, the model allows us to analyze the particle mixing state, as well as the impact of the mixing state assumption made in the modelling on particle formation, aerosols optical properties, and cloud condensation nuclei activation. Finally, two simulations are conducted during the winter campaign of MEGAPOLI (Megacities : Emissions, urban, regional and Global Atmospheric POLlution and climate effects, and Integrated tools for assessment and mitigation) in January 2010 where the composition of individual particles was measured. One simulation assumes that particles are internally mixed, while the other explicitly models the mixing state with SCRAM. The simulation results of both bulk concentrations of chemical species and concentrations of individual particle classes are compared with the measurements. Then, the single particle diversity and the mixing-state index are computed using a quantification approach based on information-theoretic entropy, and they are compared to those derived from the measurements at a urban site in Paris: the simulated mixing-state index is equal to 69% against 59% from the measurements, indicating that particles are not internally mixed over Paris
Document type :
Complete list of metadata
Contributor : ABES STAR :  Contact
Submitted on : Thursday, October 20, 2016 - 11:19:32 AM
Last modification on : Thursday, October 17, 2019 - 9:40:04 PM


Version validated by the jury (STAR)


  • HAL Id : tel-01384607, version 1



Shupeng Zhu. Modeling of particle mixing in the atmosphere. Environmental Engineering. Université Paris-Est, 2015. English. ⟨NNT : 2015PESC1156⟩. ⟨tel-01384607⟩



Record views


Files downloads