Abstract:Air pollution in California's San Joaquin Valley (SJV) rivals that of the Los Angeles area and ranks among the worst in the United States for particulate matter (PM) and ozone. The application and validation of an atmospheric chemical transport model to the SJV will aid in the design of emissions control programs to improve air quality. The extensive data required for meaningful Eulerian modeling of airborne PM in the SJV region were collected as a part of the 1995 Integrated Monitoring Study (IMS95). In the current study, the CIT-UCD source-oriented air quality model is applied to the IMS95 data set to verify model performance in the SJV. This article represents the first published application of a full-scale photochemical grid model with diagnostic meteorological data to simulate PM concentrations in the SJV and is the first study outside of the Los Angeles area to include complete PM model performance statistics.

The CIT-UCD model results show excellent agreement with most measurements collected during the IMS95 4–6 January 1996 modeling episode. The fractional bias (FBIAS) for SJV ozone and PM10 are approximately 0.16 and −0.19, respectively, for all SJV stations excluding Bakersfield. Most modeled criteria gases, PM precursor gases, and chemically speciated PM concentrations show strong agreement with their corresponding measurements at the Chowchilla, Fresno, Kern Wildlife Refuge, and Bakersfield sampling stations. Furthermore, the modeled PM size distribution of nitrate, ammonium ion, and sulfate agree well with cascade impactor measurements made at Bakersfield on 5 January 1996. Given the robust model agreement with both gas and condensed phase measurements, it appears that the CIT-UCD model adequately captures the fundamental transport and chemical reactivity of air pollutants in the IMS95 domain during a typical severe pollution episode. These results suggest that the CIT-UCD model can be used to explore control scenarios designed to improve air quality in the SJV.