Developing, coupling, and applying a gas, aerosol, transport, and radiation model to study urban and regional air pollution

This thesis discusses the development of a gas, aerosol, transport, and radiation air quality model (GATOR), the coupling of GATOR to a mesoscale meteorological and tracer dispersion model (MMTD), and the application of the resulting GATOR/MMTD air pollution modeling system (APMS) to studies of gas and aerosol pollution buildup in the Los Angeles Basin. GATOR consists of computer algorithms that simulate four groups of atmospheric processes: gas, aerosol, transport, and radiation processes. Gas processes include chemistry, emissions, gas-to-particle conversion, optical depth attenuation, and deposition. Aerosol processes include size-resolved emissions, nucleation, coagulation, condensational growth, dissolutional growth, evaporation, chemical equilibrium, aqueous chemistry, optical depth attenuation, deposition, and sedimentation. Transport processes include horizontal advection and diffusion and vertical diffusion of all gases and particles, and they require meteorological data as inputs. To drive the transport processes, the MMTD, developed by R. Lu and R. P. Turco, was coupled to GATOR. The MMTD predicts wind speed, wind direction, temperature, humidity, and pressure, among other variables. Finally, radiation processes in GATOR include spectrally-resolved scattering and absorption by gases, aerosols, fogs, and clouds, and calculation of mean intensities and heating rates. I used the GATOR/MMTD modeling system to predict pollution buildup in Los Angeles during the Southern California Air Quality Study (SCAQS) period of August 26-28, 1987. Among the model inputs were emissions, soil moisture, albedo, topographical, landuse, and chemical rate data. To validate the model, surface observations were compared to model predictions of gas-phase ozone, nitric oxide, nitrogen dioxide, carbon monoxide, sulfur dioxide, methane, total non-methane hydrocarbons, formaldehyde, peroxyacetylnitrate, hydrogen peroxide, nitric acid, nitrous acid, and ammonia concentrations. Observations were also compared to predictions of aerosol-phase ammonium, nitrate, sodium, chloride, sulfate, and total particulate concentrations. Finally, observations were compared to predictions of solar radiation and scattering coefficients. In sum, the GATOR/MMTD system predicted ozone to within a normalized gross error of 20-35% during the study period. Additional statistics and time-series plots are shown.