Air quality models have traditionally been applied to few high ozone episodes. Emission reduction simulations are carried out for these episodes to determine which control strategies are most effective in reducing maximum ozone concentrations for this episode in a given region, and this information has been used as a basis for policy-making (Tesche et al., 1996). In contrast, analyses of observations have shown that the ozone is the result of processes acting on a number of different time scales (Rao et al., 1997), many of which cannot be simulated by photochemical models if they are applied only to a few days, as is the case in episodic modeling.
In this paper, we present results from a modeling study in which the UAM-V photochemical model was applied for the entire summer of 1995 (i.e. from June 1, 1995 to August 31, 1995) for
the Eastern United States. In addition to a base case simulation, various emission reduction scenarios were also simulated. In addition to conventional model evaluation approaches, which have used metrics such as mean bias, RMSE, or an overall "index of agreement" (e.g. Tesche et al., 1996), we compare the model predictions for ozone to observations on different scales in space and time. The rationale for this approach is that different processes (e.g. chemical production vs. synoptic-scale transport) occur on different time scales and, thus, through the analysis of spectrally decomposed signals one can make evaluate the strengths or shortcomings of the model. To this end, we separate each time series into four components, reflecting forcings on the intra-day, diurnal, synoptic and longer-term time scales, respectively.
The results show that correlations between the modeled and observed data are high on longer time scales and poor on the intraday time scale. It is also shown that emission reductions are more effective in reducing longer-term components rather than the short-term component. These findings are discussed in the policy-making context, since they suggest that photochemical models should be used to evaluate the effectiveness of a certain control strategy on a longer-term basis rather than on an episodic basis.
References:
Rao, S. T., et al., 1997. "Space and Time Scales in Ambient Ozone Data", Bull. Amer. Meteor. Soc., 78, 2153 - 2166
Tesche, T.W., et al., 1996. "Initial Estimates of Emission Reductions needed for Ozone Attainment in the Pittsburgh - Beaver Valley Ozone Nonattainment Area", interim report to the Southwestern Pennsylvania Clean Air Stakeholder Groups, prepared by Alpine Geophysics, Golden, CO
Symposium on Interdisciplinary Issues in Atmospheric Chemistry