Abstract for presentation at 14th IUAPPA World Congress

Air pollution dispersion models require input fields of meteorological data including wind conditions near the surface and plume height, and data for the estimation of boundary layer mixing height. In the absence of measurements, these data can be modelled with the use of prognostic meteorological models (such as MM5), or the meteorological component of dispersion models (such as TAPM).
Meteorological modelling was performed for the Collie region in Western Australia using MM5 and TAPM. Terrain elevation, vegetation and soil data were revised from the default data in an attempt to improve the accuracy of the model simulations.
Model performance was evaluated by comparing observations from meteorological stations not used for nudging purposes, with those predicted by MM5 and TAPM. The methodology employed for model validation included standard statistical tests.
Sensitivity studies on improved land-use and soil data on TAPM predictions show that improving the land-use did not significantly improve the model performance with respect to wind speed prediction, but did improve the model performance with respect to temperature and humidity prediction.
Comparison of outputs of MM5 and TAPM with observed data shows that with the exception of cloud amount, TAPM predicted meteorological parameters such as wind speed, direction, temperature, rainfall and relative humidity with more accuracy than MM5.