Impact of local and non-local eddy-diffusivity schemes on calculating the concentration of pollutants in environmental models

Abstract

In this paper, we studied a local and a non-local scheme for vertical diffusion in the atmospheric boundary layer and their impact on the concentration of pollutants calculated with an environmental model. In the local diffusion scheme, the eddy-diffusivity is determined independently at each point in the vertical based on local vertical gradients of wind and potential temperature. The non-local scheme determines an eddy-diffusivity profile based on a diagnosed boundary layer height and a turbulent vertical scale. To compare these two approaches, we used a local (O'Brien [1970]) and a modified non-local (Alapaty [2003]) diffusivity scheme. Compared to the original, the modification introduced in the non-local scheme included different values of the parameterin the equation for eddy-diffusivity depending on the stability regime. For the unstable conditions, the parameterwas set to 2, a value commonly used in parameterizations in regional climate modeling, while for stable conditions we used a value of 1.5 based on numerical experiments. To examine the performance of the schemes, simulated and measured concentrations of the pollutant believed to be one of the most affected ones by the processes in the atmospheric boundary layer () were compared for the years 1999, 2001 and 2002. The comparison was made for the whole domain used in simulations performed by the chemical EMEP Unified model (version UNI-ACID, rv2.0). pp2NO.