Dispersion modeling of ground-level area sources of particulate

Abstract

The use of dispersion modeling by State Air Pollution hics. Regulatory Agencies (SAPRAS) is increasing. Dispersion modeling provides a quick and efficient means of determining the downwind impact of pollutant release from a source. The SAPRAS are charged with the task of insuring that public exposure levels of these pollutants are less than the standards set by the United States Environmental Protection Agency (US EPA). Estimating the concentration of pollutant at some distance downwind, in most cases the property line, allows that SAPRAS to determine whether or not a source needs to install additional means of control in order to decrease the rate of pollutant release. One set of models approved for regulatory use by the US EPA is Industrial Source Complex (ISC). ISC includes SCREEN: which is a simple screening model with imbedded meteorological data, and ST3 which is a more refined model requiring meteorological data to be provided. ISC models are based on the concept of Gaussian dispersion. The concentration determined using the ISC models are a result of determining concentrations based on a single wind speed and direction for a one hour time period. The first step in the formulation of a new model was to incorporate smaller time periods into the concentration predictions, in order to account for variation or wind speed and direction within an hour period. For ground-level sources, the vertical distribution involves mathematically dispersing the pollutant underground, then reflecting it back up. The next step taken in the new model development was the incorporation of a triangular distribution in the Medical plane. The triangular distribution is entirely above ground. Once the new model was formulated, a sample modeling procedure was performed in order to compare the behavior of the new model as compared to ISC ST3. Examination of the characteristics of the two models, the meteorological data, and the output from the modeling procedures allows the comparison and contrast of the behavioral characteristics of the two models.