A comparative study of nucleation parameterizations: 2. Three‐dimensional model application and evaluation

摘要: [1] Following the examination and evaluation of 12 nucleation parameterizations presented in part 1, 11 them representing binary, ternary, kinetic, cluster-activated theories are evaluated U.S. Environmental Protection Agency Community Multiscale Air Quality (CMAQ) modeling system version 4.4. The 12–28 June 1999 Southern Oxidants Study episode is selected as a testbed to evaluate simulated particulate matter (PM) number size predictions CMAQ with different parameterizations. shows that domain-wide maximum PM2.5 concentrations can vary by 3 orders magnitude. All overpredict (by factor 1.4 1.7) total accumulation-mode PM significantly underpredict factors 1.3 65.7) those Aitken-mode PM, resulting net underprediction 13.7) under polluted urban environment at downtown station Atlanta. predicted for this site up magnitude, 3.2, best power law Sihto et al. (2006) (NMB −31.7%) worst ternary parameterization Merikanto (2007) −93.1%). Napari (2002) gives relatively good agreement observations but wrong reason. Kuang (2008) binary Harrington Kreidenweis (1998) give better than remaining fail reproduce observed temporal variations number, volume, surface area concentrations. significant variation performance these caused their theoretical bases, formulations, dependence on temperature, relative humidity, ambient levels H2SO4 NH3. controlling processes mass, areas. At urban/rural locations, some (e.g., homogeneous nucleation) and/or vertical transport may dominate production emissions, or processes, combinations mass area. Dry deposition such coagulation loss, horizontal transport, cloud scavenging wet deposition) loss Sensitivity simulations show distribution most sensitive prescribed emission fractions Aitken assumed initial distribution, addition