A Modified Power Law Representation of the Pasquill-Gifford Dispersion Coefficients

Abstract

In Gaussian plume modeling, the rate of plume dilution is described by empirical dispersion coefficients, which specify the horizontal extent {sigma}{sub y} and vertical extent {sigma}{sub z} of the plume versus downwind distance x from the source under different weather conditions, as characterized by six atmospheric stability catagories A through F. Several sigma schemes have been derived from measurements but probably the most common schemes currently in use are those due to Pasquill and Gifford and those due to Briggs. For machine computations, Briggs sigme curves are convenient to use since they are specified in an explicit functional form; {sigma}{sub y}, for example is given by {sigma}{sub y} = ax(1 + bx){sup {minus}1/2} where a and b are constants, each of which takes on a particular value for each atmospheric stability class. A similar relation describes {sigma}{sub z}. Thus, by inputting the above functional form and relatively few constants into a calculator, computer program or spreadsheet, {sigma}{sub y} and {sigma}{sub z} can be easily calculated at any x location of interest. The Pasquill-Gifford sigma curves, on the other hand, were originally specified graphically. While graphs as presented by Turner{sup 3}, for example, are convenient for manual abstraction of sigma values,more » a functional representation comparable to Equation 1 is clearly more suitable for machine computations. Accordingly, various curve fit approximations to the Pasquill-Gifford sigma graphs have been presented.« less