Coupling and evaluating gas/particle mass transfer treatments for aerosol simulation and forecast

Simulating gas/particle mass transfer in three-dimensional (3-D) air quality models (AQMs) represents one of the major challenges for both hindcasting and forecasting. The lack of an efficient yet accurate gas/particle mass transfer treatment for aerosol simulation and forecast in 3-D AQMs warrants its development, improvement, and evaluation. In this paper, several condensation/evaporation schemes (e. g., the Bott, Trajectory-Grid (T-G), Walcek, and analytical predictor of condensation (APC)) are first tested in a condensation-only case. The APC and Walcek schemes are shown to be more accurate than the Bott and T-G schemes. The Walcek and the APC schemes are then incorporated into the Model of Aerosol Dynamics, Reaction, Ionization, and Dissolution (MADRID) to solve the gas/particle mass transfer process explicitly. The test simulations with MADRID are initialized with measurements available from three sites with representative meteorological and emission characteristics. The results are evaluated using benchmark based on the kinetic approach with 500-section for all cases and available measurements from two sites. The box MADRID tests have shown that the bulk equilibrium approach fails to predict the distribution of semivolatile species (e. g., ammonium, chloride, and nitrate) because of the equilibrium and internal mixture assumptions. The hybrid approach exhibits the same problem for some cases as the bulk equilibrium approach since it assumes bulk equilibrium for fine particles. The kinetic approaches (including the APC and Walcek schemes for the condensation/evaporation equations) predict the most accurate solutions. Among all approaches tested, the bulk equilibrium approach is the most computationally efficient, and the kinetic/Walcek scheme provides an accurate solution but is the slowest due to its requirement for a small time step. Overall, the kinetic/APC and hybrid/APC schemes are attractive for 3-D applications in terms of both accuracy and computational efficiency.