Sensitivity Analysis of the Spatial and Altitude Distributions of Pollutants Using the Weather Research and Forecasting Model With Chemistry (WRF/CHEM)

Abstract

The Weather Research and Forecasting (W RF) model is a new “state -of-the-ar t” meteorological model which offers the users many different options for physical parameterizations. The recent introduction of a Chemical module (WRF/Chem) allows to per fo rm an “ on -line” description of the chemi cal evolution of trace pollut ant s by coupling a time-dependent chemical mechanism to the primitive meteorological equations. The advantage of WRF/Chem model over traditional dispersion models (CALPUFF, ISC3, etc.) is that a 3-D+temporal description of the pollutants distribution can be obtained. In this work, the WRF/Chem model has been used to study the dependence of the spatial and temporal distribution of point sources of pollutants with the altitude at which they are emitted. The study considers a detailed 24 hours winter period over a 200 km × 200 km regional domain centered at Buenos Aires, including only the pollutants emissions from typical power-plant stack with variable height. A sensitivity analysis of the model was performed considering different scenarios where the altitude and emission rate of the stack were modified, keeping the total emission constant. The response of WRF/Chem model to the proposed cases was statistically analyzed by computing several types of difference measures, as mean bias error, mean absolute error and index of agreement. Larger sensitivity to emissions schemes was found for the scenarios emitting near surface or at elevated levels. For those emitting at intermediate levels, the height of the time-variant Planetary Boundary Layer (PBL) is a relevant parameter.