PM₂.₅ analog forecast and Kalman filter post-processing for the Community Multiscale Air Quality (CMAQ) model

AMS Citation:
Djalalova, I., L. Delle Monache, and J. Wilczak, 2015: PM₂.₅ analog forecast and Kalman filter post-processing for the Community Multiscale Air Quality (CMAQ) model. Atmospheric Environment, 108, 76-87, doi:10.1016/j.atmosenv.2015.02.021.
Resource Type:article
Title:PM₂.₅ analog forecast and Kalman filter post-processing for the Community Multiscale Air Quality (CMAQ) model
Abstract: A new post-processing method for surface particulate matter (PM₂.₅) predictions from the National Oceanic and Atmospheric Administration (NOAA) developmental air quality forecasting system using the Community Multiscale Air Quality (CMAQ) model is described. It includes three main components: • A real-time quality control procedure for surface PM₂.₅ observations; • Model post-processing at each observational site using historical forecast analogs and Kalman filtering; • Spreading the forecast corrections from the observation locations to the entire gridded domain. The methodology is tested using 12 months of CMAQ forecasts of hourly PM2₂.₅, from December 01, 2009 through November 30, 2010. The model domain covers the contiguous USA, and model data are verified against U.S. Environmental Prediction Agency AIRNow PM₂.₅ observations measured at 716 stations over the CMAQ domain. The model bias is found to have a strong seasonal dependency, with a large positive bias in winter and a small bias in the summer months, and also to have a strong diurnal cycle. Five different post-processing techniques are compared, including a seven-day running mean subtraction, Kalman-filtering, analogs, and combinations of analogs and Kalman filtering. The most accurate PM₂.₅ forecasts have been found to be produced when using historical analogs of the hourly Kalman-filtered forecasts, referred to as KFAN. The choice of meteorological variables used in the hourly analog search is also found to have a significant effect. A monthly error analysis is computed, in each case using the remaining 11 months of the data set for the analog searches. The improvement of KFAN errors over the raw CMAQ model errors ranges from 50 to 75% for MAE and from 40 to 60% for the correlation coefficient. Since the post-processing analysis is only done at the locations where observations are available, the spreading of post-processing correction information over nearby model grid points is necessary to make forecast contour maps. This spreading of information is accomplished with an eight-pass Barnes-type iterative objective analysis scheme. The final corrected CMAQ forecast over the entire domain is composed of the sum of the original CMAQ forecasts and the KFAN bias information interpolated over the entire domain, and is applied on an hourly basis
Peer Review:Refereed
Copyright Information:Copyright 2015 Elsevier.
OpenSky citable URL: ark:/85065/d7rf5w6s
Publisher's Version: 10.1016/j.atmosenv.2015.02.021
  • Irina Djalalova
  • Luca Delle Monache - NCAR/UCAR
  • James Wilczak
  • Random Profile


    Recent & Upcoming Visitors