Modeling concepts in DEGADIS to characterize important behavior of ground-level, denser-than-air contaminant clouds

Many commercially important materials in the chemical process industries form denser-than-air clouds when released to the atmosphere. The dispersion of episodic releases of these materials has been studied extensively because of the potential impact of such releases to the public. Experimental programs aimed at quantifying uncertainties related to denser-than-air clouds were conducted focusing on the material released (such as ammonia, nitrogen tetroxide, hydrogen fluoride, liquefied natural gas, and liquefied petroleum gas, for example) as well as stimulant clouds such as the Thorney Island test series and carbon dioxide over unobstructed terrain and billboard arrays (Kit Fox tests). The DEGADIS model was originally developed for the US Coast Guard and the Gas Research Institute to model the dispersion of liquefied natural gas. In the model development, the objective was to take the best understanding of denser-than-air cloud behavior and incorporate these principles into a framework which would be asymptotically consistent with current passive dispersion models as the density difference approaches zero between the dispersing cloud and surrounding atmosphere. Furthermore, the modeling principles developed in DEGADIS were based largely on laboratory scale experiments with field-scale test comparisons reserved for validation. This paper summarizes the modeling concepts included in DEGADIS to model the behavior of ground-level, denser-than-air contaminant clouds including: air entrainment into a stably stratified layer, a model of the maximum flux of (denser-than-air) material that can be taken up directly into the atmosphere (important in modeling cloud persistence near the source), a model based on experimental data to describe the spreading and air entrainment of a denser-than-air cloud, and experimental evidence indicating how denser-than-air modifies concentration profiles from standard Gaussian profiles.