Dispersion Modeling of Agricultural Low Level Point Sources

Cotton gins, feed mills, and grain elevators are examples of low level point sources that are permitted based upon the predicted concentrations from dispersion modeling. ISCST3, the current EPA approved dispersion model for low level point sources, uses an emission rate developed from the emission factors listed in AP42 to predict a 10-minute downwind concentration. This paper addresses the problems associated with the errors in the emission factors listed in AP42 as well as those associated with the assumption by the developers of ISCST3 that a 10-minute concentration is equal to a 60-minute concentration. The six Pasquill-Gifford stability classes, used by ISCST3, were developed from 10-minute average field data and can only be used to calculate 10-minute concentrations. In addition, the meteorological data (wind speed and direction) used in ISCST3 is a 10-minute average that is assumed to be a 60-minute average. These assumptions cause ISCST3 to over-predict downwind concentrations by approximately 2.5 times. Monte Carlo simulations and the power law model were used to develop P values (dependent upon stability class and down wind distance) that were used to give a more accurate prediction of downwind concentrations.