Modeling the Dispersion of Vapor and Aerosol Particulates in the Atmospheric Boundary Layer

Abstract

Abstract : When dealing with dispersion modeling in topographically complex surroundings there are several aspects to be considered. Does the topography give rise to preferred flow patterns, sea and land breeze cycles, mountain and valley winds, canalizations, etc.? When developing operational models there is also the question of available computer capacity. The selection of measurement locations in complex terrain, frequency of sampling, and selection of synoptic situations always place limits on the completeness of information about local flows. Observational studies provide useful information about the structure of parameters such as wind speed and direction, temperature, and humidity. However, one of the most critical issues is the structure of turbulence, which varies greatly over space and time in complex terrain and requires a great number of routine measurements to characterize. Determination of the turbulent transfer is essential to understanding and predicting the dispersion of atmospheric pollutants. Mountain flows are generally nonlinear and fully three-dimensional phenomena, and analytical models are applicable only to simple cases. In the present project, comparison of different dispersion models - a semi-Gaussian trajectory model and a higher-order closure dispersion model - will be performed. The observed or numerically simulated, by different dynamic models, wind, temperature, and turbulence fields will be used as inputs to these dispersion models. When choosing a modeling system, or combination of dynamic model and dispersion model, it is important to take into consideration the purpose of the simulations. If it is important to get good agreement between measurements and simulations for short-term sampling periods, a more complex modeling system needs to be used than if one is only interested in monthly, seasonally, or yearly mean values.