Meteorological and air quality modeling present many challenges in regards to accuracy and verification. This dissertation focuses on three major challenges and provides a discussion of the modeling results. The Fifth-Generation Penn State University–National Center for Atmospheric Research (NCAR) Mesoscale Model (MM5) was used to prepare meteorological data for the Models-3/Community Multiscale Air Quality (CMAQ) modeling system to simulate air quality in the context of atmospheric chemistry. The three topics addressed were (1) the assimilation of a newly archived satellite (WINDSAT) data from National Polar-orbiting Operational Environmental Satellite System (NPOESS) into MM5 via 3D-Var (Three-Dimensional Variational data assimilation system) to investigate potential increase in accuracy of the modeling forecast; (2) quantitative estimation of surface ozone (O₃) in California and Nevada from grid-averaged wildfire emissions from NEI02 (National Emissions Inventory 2002); and (3) investigation of two prescribed fire events to the contribution of O3 peaks to background concentration levels.

The 3D-Var code was first modified to use WINDSAT wind vectors as MM5 model inputs; and the effect of WINDSAT data via 3D-Var was later investigated by use of MM5 model outputs with and without (free forecast) WINDSAT data. MM5 results via 3D-Var of a 14-20 July 2005 assimilation period were compared with the corresponding free run weather forecast results and observations. The primary results show that the root mean square error (RMSE) of temperatures were improved from 4% in chaotic coastal stations to more than 40% in the buoy sites offshore of California. Similarly, RMSE of wind speeds were improved from 3% at the chaotic coastal buoy stations to 43% at the offshore buoy sites.

The focus of the other two experiments was to understand the contribution of O₃ concentrations and exceedances from wildfire and prescribed fire emissions by incorporating the emissions with the NEI02 to simulate CMAQ over the Southern Sierra Nevada and the San Joaquin Valley (SJV) of central California. The results from this study show that wildfire and prescribed fires contribute increased O₃ concentration downwind of the burn area during high actinic flux episodes. O

3

concentration formed from the NEI02 averaged wildfires showed an increase of 16 ppb for the representative day of maximum 8-hour average O₃. The prescribed fire events also contributed a maximum O₃ of 31 ppb per hour downwind of the burn area. These results suggest that both wild and prescribed fires can potentially contribute to violations in both state and federal air quality standards (non-attainment limits), though for the examined cases exceedances did not occur. Air quality standard violation areas are classified as non-attainment zones by federal regulators, which can lead to punitive actions by federal agencies.