ABSTRACT

Cotton gins are required to have an air quality permit in order to operate. Some states permit gins are based upon predicted PM₁₀ concentration levels from dispersion modeling. The use of dispersion modeling allows regulators to estimate downwind concentrations that potentially could impact the public off-property. Industrial Source Complex - Short Term version 3 (ISCST3) is the EPA approved dispersion modeling program used by most states to regulate industrial sources. The product of ISCST3 is a 24-hour average downwind concentration of particulate matter given an emission rate. Concentrations pre-dicted by ISCST3 are based upon the Gaussian dispersion model, using Pasquill-Gifford dispersion parameters (σ_y, σ_z)

It is generally known that ISCST3 over-predicts downwind concentration from low-level point sources (LLPS) such as cotton gins, grain elevators and feed mills. ISCST3 does not account for horizontal wind direction variations within an hour. Ac-counting for these variations results in a significantly lower predicted PM₁₀ concentration at the receptor. This paper describes a unique approach for estimating 24-hour concentrations using USEPA (2000) data on horizontal wind direction variations, Monte-Carlo simulations, and the Gaussian model. The results include power law coefficients by stability class for estimating 60-minute concentrations from 10-minute concentrations (Gaussian equation) and the ratios of C₆₀/C₁₀ for 10 days of mete-orological data. It is concluded that the current downwind concentrations estimated by regulatory agencies using ISCST3 over-predicts PM₁₀ by a factor of 2.5.