HR: 09:30h
AN: A21F-07    [Abstracts]
TI: The Evaluation of PM2.5 Forecasts from Several Regional Air Quality Models Using Data from the ICARTT/NEAQS-2K4 Field Study
AU: * McKeen, S A
EM: Stuart.A.McKeen@noaa.gov
AF: NOAA Aeronomy Lab/CIRES, NOAA R/AL4 325 Broadway, Boulder, CO 80305-3328 United States
AU: Hsie, E
EM: EirhYu.Hsie@noaa.gov
AF: NOAA Aeronomy Lab/CIRES, NOAA R/AL4 325 Broadway, Boulder, CO 80305-3328 United States
AU: Grell, G
EM: georg.a.grell@noaa.gov
AF: NOAA FSL/CIRES, NOAA R/FSL 325 Broadway, Boulder, CO 80305-3328 United States
AU: Peckham, S
EM: steven.peckham@noaa.gov
AF: NOAA FSL/CIRES, NOAA R/FSL 325 Broadway, Boulder, CO 80305-3328 United States
AU: Mathur, R
EM: Mathur.Rohit@epamail.epa.gov
AF: NOAA Air Resources Laboratory, 109 T.W. Alexander Drive U.S. EPA - Mail Drop E243-03, RTP, NC 27711 United States
AU: Yu, S
EM: yu.shaocai@epa.gov
AF: NOAA Air Resources Laboratory, 109 T.W. Alexander Drive U.S. EPA - Mail Drop E243-03, RTP, NC 27711 United States
AU: Gong, W
EM: wanmin.gong@ec.gc.ca
AF: Meteorological Services of Canada, 4905 Dufferin Street, Downsview, ON M3H-5T4 Canada
AU: Bouchet, V
EM: Veronique.Bouchet@ec.gc.ca
AF: Meteorological Services of Canada, 2121 Trans-Canada N., Dorval, QU H9P-1J3 Canada
AU: Menard, S
EM: sylvain.menard@ec.gc.ca
AF: Meteorological Services of Canada, 2121 Trans-Canada N., Dorval, QU H9P-1J3 Canada
AU: Tang, Y
EM: ytang@cgrer.uiowa.edu
AF: University of Iowa, 402 IATL, Iowa City, IA 52242 United States
AU: Carmichael, G
EM: gregory-carmichael@uiowa.edu
AF: University of Iowa, 402 IATL, Iowa City, IA 52242 United States
AB: Real-time air quality forecasts of aerosol PM2.5 for the Eastern U.S. and Southern Canada are currently available through a number forecast offices and research centers. This study presents results from an evaluation of six regional air quality models compared to surface data from the U.S. EPA AIRNow PM2.5 network and Speciation Trends Network (STN), as well as data collected during the ICARTT/NEAQS-2K4 field campaign from the NOAA WP-3 aircraft and Ronald H. Brown research vessel. The forecast models include two versions of the NOAA/FSL WRF/Chem model, a developmental version of the NWS/NCEP CMAQ/ETA model, the Canadian Meteorological Services CHRONOS and AURAMS models, and the University of Iowa STEM-2K3 model. Statistical evaluations of each model with the AIRNow PM2.5 network characterize the models' ability to forecast surface PM2.5 over a large region of Eastern North America for a seven-week period during the summer of 2004. The composition of aerosol PM2.5 is statistically evaluated by comparing model predicted PM2.5 sulfate, nitrate, ammonium, organic carbon and elemental carbon with aerosol compositional data from the aircraft, ship, and STN surface network. Several important conclusions have emerged from the analysis completed to date. Comparisons of model forecasts with surface AIRNow PM2.5 network data show that all models possess some skill in predicting daytime average PM2.5 levels when compared to simple persistence forecasts. An ensemble PM2.5 forecast, constructed by taking the geometric mean of the six forecasts, shows significant statistical improvement over any individual forecast. Comparisons of model forecasts with aerosol composition measurements from the W-P3 aircraft show that all models under-predict the organic carbon fraction of aerosol. SO₂ oxidation rates downwind of urban centers are examined by comparing SO₂/SO₄ ratios between the models and WP-3 observations, and those models that include SO₂ conversion to SO₄ by cloud oxidation are found to over-predict aerosol sulfate. Comparisons with Ronald H. Brown data provide additional diagnostic tests of various model components, particularly the emission of organic and elemental carbon relative to SO₂, CO, and sulfate aerosol from urban and forested up-wind sources. The evaluation of multiple forecast models with multiple observing platforms leads directly to recommendations for future research and improvements in PM2.5 forecasting which are summarized in this presentation.
DE: 0345 Pollution: urban and regional (0305, 0478, 4251)
DE: 0550 Model verification and validation
DE: 3238 Prediction (3245, 4263)
DE: 3355 Regional modeling
SC: Atmospheric Sciences [A]
MN: Fall Meeting 2005