The development of a hybrid meteorological field to improve ozone simulation results in the CCOS domain

The Central California Ozone Study (CCOS) modeling domain is geographically complex. There are ocean/land contrasts in a west-to-east direction while the major terrain features run primarily north-to-south. The land/sea breezes, mountain/valley flows, and regions of bifurication are a challenge for any meteorological model.

In the evaluation of ozone episodes using air quality models, the development of data fields that adequately represent meteorological conditions occurring during the episodes is a major challenge. For the July/August, 2000 CCOS episode, a number of researchers have attempted to develop suitable meteorological fields using the MM5 prognostic model. Although different configurations of MM5 have been applied to this episode, the resulting meteorological fields have been disappointing when compared to observed meteorological parameters, and when used in subsequent air quality model simulations.

Objective/diagnostic meteorological models such as CALMET are an alternative to the prognostic models for developing the meteorological fields needed for air quality modeling. There are numerous examples of their use in air quality model simulations. Reservations about the use of these models for developing inputs to air quality models have sometimes been expressed; however, the greatest limitation to their use is that they can only be applied where observational data are available. In large portions of the CCOS modeling domain, such as over the ocean, such data were not available.

To improve the comparisons between observed and simulated parameter values the CALMET model was used to develop hybrid meteorological fields. In this case, the hybrid field were developed as a combination of the objective analysis and the MM5 prognostic model results. The extensive archive of observational data collected during the CCOS field study provided the basis for the development of improved wind and temperature fields in portions of the domain where such observations were available, and the MM5 prognostic results were utilized where they were not. The resulting hybrid meteorological fields were used to develop better representations of the emissions inventory and to simulate improved ozone concentration fields using an air quality model.