Thesis
Evaluation of the ClearSky smoke dispersion ensemble forecast system for agricultural field burning in Eastern Washington and Northern Idaho
Washington State University
Master of Science (MS), Washington State University
2006
Handle:
https://hdl.handle.net/2376/515
Abstract
An ensemble air quality forecast system (ensemble ClearSky) was developed to predict hourly surface level particulate matter concentrations downwind of agricultural field burns in eastern Washington and northern Idaho. A suite of 17 meteorological forecasts is used as input to a Lagrangian dispersion model (CALPUFF) along with a single PM2.5 emissions scenario (representing agricultural field burns) to yield an ensemble of PM2.5 dispersion forecasts. Thus, for a selected day, the ensemble ClearSky system produces a range of possible PM2.5 forecasts by using 17 different, similarly possible, meteorological forecasts. The ensemble average (un-weighted average of ensemble members), individual ensemble members, and original ClearSky PM2.5 forecasts were evaluated against observed PM2.5 concentrations for two burn days during the 2004 burn season. The evaluation used observations from five monitoring sites on the August 17, 2004 burn day and seven monitoring sites on the September 8, 2004 burn day. Normalized mean error (NME) and unpaired peak prediction error (UPPE) were calculated for the PM2.5 concentrations, and the mean absolute error (MAE) in wind direction was calculated for the forecast meteorology. For PM2.5 prediction, the ensemble average had lower NME values at 11 of the monitoring sites and lower UPPE values at five of the sites compared to original ClearSky. With the exception of one monitoring site (Plummer, ID) where significant PM2.5 over-prediction occurred, the lowest UPPE values ranged from 1%-68% for the two burn days. The UPPE values for the ensemble average PM2.5 forecasts ranged from 11%-78%; and the UPPE values for the original ClearSky PM2.5 forecasts ranged from 8%-88%. The lowest NME values ranged from 27%-98% for the two burn days. The NME values for the ensemble average PM2.5 forecasts ranged from 36%-117%; and the NME values for the original ClearSky PM2.5 forecasts ranged from 41%-131%. The lowest MAE in wind direction values ranged from 13-73 degrees. Using the ensemble meteorology, the dispersion model was capable of predicting the maximum PM2.5 concentrations within the range of 1%-68% based upon the UPPE statistics.
Metrics
2 File views/ downloads
14 Record Views
Details
- Title
- Evaluation of the ClearSky smoke dispersion ensemble forecast system for agricultural field burning in Eastern Washington and Northern Idaho
- Creators
- Kyle Matthew Heitkamp
- Contributors
- Brian Lamb (Degree Supervisor)
- Awarding Institution
- Washington State University
- Academic Unit
- Civil and Environmental Engineering, Department of
- Theses and Dissertations
- Master of Science (MS), Washington State University
- Publisher
- Washington State University; [Pullman, Washington] :
- Identifiers
- 99900525397301842
- Language
- English
- Resource Type
- Thesis