Sensitivity of the regional ozone modeling results to different physical processes within a meteorological model

Ambient ozone concentrations during summertime, continues to exceed the 1-hour standard in many regions of the Northeastern United States. Emission control strategies to mitigate the problem of ozone nonattainment are usually based on simulations of historic ozone events. It is well known that photochemical models used for these simulations are known to be sensitive to meteorological inputs.

The ozone process is nonlinear because of the complex interplay between chemistry and meteorology. Model sensitivity runs are needed to evaluate the contribution of different meteorological parameters to the variability in the meteorological model predictions. Biswas and Rao(1999) examined the variability in the predictions of the Urban Airshed Model (UAM-V), when the meteorological input for the UAM-V are derived from two different meteorological models, namely, RAMS and MM5. This study extends the previous work by investigating the sensitivity of the UAM -V predictions to differing assumtions used in the MM5 meteorological model.

Meteorological fields for the UAM-V are obtained from two versions of MM5 (versions 1 and 2). The features prescribed differently in the two meteorological model simulations include the upper boundary, the planetary boundary layer specification, cloud microphysics, vertical diffusion, convective schemes and radiative cooling. The modeling results for the July 12 to July 15, 1995 ozone episode are analyzed with the objective of quantifying the variability in the ozone concentrations predicted by the UAM-V model. In this paper, we present the results of 1-hr daily maximum ozone and 8-hr daily maximum ozone and discuss the implications of these results to policy-making.