Development and application of a new air pollution modeling system-part I: Gas-phase simulations

A new air pollution modeling system is discussed and applied. The system consists of GATOR, a gas, aerosol, transport, and radiation air quality model and MMTD, a mesoscale meteorological and tracer dispersion model. The gas-phase processes treated by GATOR include photochemistry, deposition, emissions, and gas-to-particle conversion. To solve stiff chemical rate equations, a sparse-matrix, vectorized Gear-type code (SMVGEAR) was used. The aerosol processes in GATOR include coagulation, aqueous chemistry, chemical equilibrium, condensational growth, dissolutional growth, evaporation, nucleation, emissions, deposition, and sedimentation. The transport processes include horizontal advection and diffusion and vertical convection and diffusion. Finally, the radiation algorithm calculates ultraviolet, visible, and infrared optical depths, mean intensities for photodissociation rates, and radiative heat fluxes for temperature calculations. The MMTD predicts winds, diffusion, temperature, pressure, humidity, soil moisture, and rainfall. These variables are fed to GATOR and radiative heating rates from GATOR are fed back to the MMTD. With the GATOR/MMTD system, gas-phase pollution was simulated for the Southern California Air Quality Study (SCAQ) days of 26–28 August 1987. Results were compared to surface measurements for many parameters. The model predicted normalized gross errors for ozone of 17.6% and 23.4% at 2:30 p.m. on the first and second days of simulation, respectively. Also, the normalized gross error during the first 12 h of simulation was 22%. Correct emissions and initial mixing ratios appear to be necessary for obtaining good results. Initial conditions outside the basin seem to affect results by the second and third days. Time-series plots, statistics, and a sensitivity test are discussed. Aerosol simulation results will be shown in a later work.