Previous Issues

 ABSTRACT

 The modelling of the results of accidental releases of hazardous chemicals is important for the making of contingency plans, such as the evacuation of the public from affected areas. Meteoro logical conditions such as wind speed and the degree of cloudiness, and the quantity of such chemicals released, are the main input parameters to mathematical models for the prediction of directions about an accident center that can be affected by the resulting puff or plume and the general rates at which pollutants are dispersed into the atmosphere.

Gaussian plume models have proved to be reasonably successful in accounting for observed concentration patterns. The Sutton (1953) and Pasquill (1962) approaches are the most com mon examples of this basic model and they have been used in this study. Sutton’s model has been applied frequently in reactor hazard analysis and air pollution studies and there has been considerable experience with Sutton’s diffusion coefficients and good verification of diffusion predictions for distances up to several kilometres from a release site. Pasquill’s approach is based on estimating diffusion in terms of the angular spread of a diffusing plume using more commonly observed weather parameters.

In this paper we report our comparative study using both Sutton and Pasquill models to simulate the consequences of accidental release of two hazardous gases (chlorine under pressure and refrigerated ammonia). The results are analyzed and assessed in the light of the specific features of each model.