Special session of California 2000 field study - Air Pollution Modeling and Its Application - page 549

8.11 Evaluating CMAQ Particulate Matter Simulations
in Central Valley California with Ground
and Airborne LIDAR Observations

Dazhong Yin 1 , Bruce Jackson 1 , and Ajith Kaduwela 1,2

1

Planning and Technical Support Division, Air Resources Board, California Environmental
Protection Agency, Sacramento, CA, USA
2

Department of Land, Air, and Water Resources, University of California at Davis, CA, USA

Abstract Models-3/Community Multi-scale Air Quality (CMAQ) model particulate
matter (PM) simulations for February 11-18 in the Central Valley California were
evaluated against ground-based PM2.5 observations and the United Stated (US)
Environmental Protection Agency (EPA) Advanced Monitoring Initiative (AMI)
field campaign aerosol optical depth (AOD) data. The comparisons showed that
the modeled AOD peaks corresponding to the observed peaks well, although the
modeled AOD values were generally smaller than the observations. The modeled
surface PM2.5 concentrations showed different temporal variation trend than that
of the AOD sometimes. The modeled hourly dry PM2.5 (with which water com-
ponents excluded) compared better with the Beta Attenuation Monitors observations
than modeled PM2.5 including water contents.

Keywords Particulate matter, aerosol optical depth, air quality modeling, model
evaluation

California's Central Valley is one of the largest PM2.5 and PM10 non-attainment
areas in the United States (Chow et al., 2006). In this study, we used EPA's Models-3
Community Multi-scale Air Quality (CMAQ) 4.6 (Byun and Ching, 1999) to simu-
late particulate matter (PM) in Central California from February 10 to February 18,
2007. The results were compared with surface PM2.5 and aerosol optical depth
(AOD) observations from a High Spectral Resolution LIDAR (HSRL) onboard a
NASA B200 King Air aircraft. In the simulations, two chemical mechanisms,
Carbon Bond (CB) 05 (Yarwood et al., 2005) and the Statewide Air Pollution
Research Center (SAPRC) 99 (Carter, 2000) were used. We used aerosol module
AERO4 (Binkowski and Roselle, 2003) in CMAQ. Emissions were for the year
2005. We assume there are no significant emission changes from 2005 to 2007.
Meteorological conditions were simulated with MM5 3.7.4 (Grell et al., 1994).
Two level nesting domains, one 36 km and the other 12 km, were employed in a
one-way nesting model run. There were 30 vertical model layers from the ground