HR: 12:05h
AN: A41F-08    [PDF]
TI: Regional Air Quality Forecasts during NEAQS-2K2: Comparisons with Observations
AU: * McKeen, S A
EM: stu@al.noaa.gov
AF: NOAA Aeronomy Laboratory/CIRES, R/AL4 325 Broadway, Boulder, CO 80305-3328
AU: Grell, G A
EM: grell@fsl.noaa.gov
AF: NOAA Forecast Systems Lab/CIRES, R/FS1 325 Broadway, Boulder, CO 80305-3328
AU: McHenry, J
EM: mchenry@emc.mcnc.org
AF: Environmental Modeling Center, Baron Advanced Meteorology Systems, Research Triangle Pk, NC 27709
AU: Peckham, S E
EM: steven.peckham@noaa.gov
AF: NOAA Forecast Systems Lab/CIRES, R/FS1 325 Broadway, Boulder, CO 80305-3328
AU: Williams, E
EM: ewilliams@al.noaa.gov
AF: NOAA Aeronomy Laboratory/CIRES, R/AL4 325 Broadway, Boulder, CO 80305-3328
AU: Lerner, B
EM: B.Lerner@noaa.gov
AF: NOAA Aeronomy Laboratory/CIRES, R/AL4 325 Broadway, Boulder, CO 80305-3328
AU: Munger, J W
EM: jwm@io.harvard.edu
AF: Harvard University, Dept. of Earth and Planetary Sciences, 20 Oxford Street, Cambridge, MA 02420
AU: Angevine, W M
EM: wangevine@al.noaa.gov
AF: NOAA Aeronomy Laboratory/CIRES, R/AL4 325 Broadway, Boulder, CO 80305-3328
AB: The results of two Eulerian-based air quality forecast models (AQFMs) are compared to observations collected during the summer of 2002 North East Air Quality Study (NEAQS-2K2) field campaign. The focus of the comparison is on ozone, and several other key photo-oxidants measured during July and August of 2002 at four AIRMAP surface sites in New Hampshire, the Harvard Forest site in central Massachusetts, and aboard the NOAA Research Vessel Ron Brown that was deployed along the northeast U.S. coastline. The air quality forecast models are NOAA Forecast System Laboratory's MM5-CHEM, and MCNC's Multiscale Air Quality Simulation Platform (MAQSIP). Each model provides forecasts of meteorology and gas-phase oxidants at three horizontal grid resolutions; the coarsest (45 km and 27 km) covering much of the United States, and the finest (5 km and 3 km) covering the Northeast U.S. The modeling systems are completely independent, with different grid structures, precursor emissions estimates, physical parameterizations, and photochemical mechanisms. Two types of model-measurement comparisons are presented; standard statistical measures of variance and model bias, and species-species relationships between key photochemical species. The standard statistical measures allow forecast skill for the six independent AQFMs to be analyzed in terms of model system, horizontal resolution, and lead time of the forecast. In terms of forecast skill, neither model platform is significantly better than the other, and nearly no improvement is seen with increased horizontal resolution. Model biases, however, are noticeably different between model platforms, model horizontal resolution, and distance from coastline. The comparisons between model and measured species-species relationships provide limited information on the reliability of specific model processes such as emissions and O$_{3}$ production efficiencies. The models generally reproduce species-species relationships between O$_{3}$, nitrogen oxides, and CO with some dependence on model horizontal resolution. Deposition of HNO$_{3}$ is shown to be a strong influence on relationships involving nitrogen oxides. These results allow recommendations to be made concerning AQFM formulation, further observations needed to assess individual model components, and provide a statistical base-line for further AQFM development in the Northeast U.S.
DE: 0305 Aerosols and particles (0345, 4801)
DE: 0317 Chemical kinetic and photochemical properties
DE: 0345 Pollution--urban and regional (0305)
DE: 0365 Troposphere--composition and chemistry
DE: 0394 Instruments and techniques
SC: Atmospheric Sciences [A]
MN: 2003 Fall Meeting