One important component for successful air quality modelling is the utilisation of a reliable meteorological simulator. Evaluating the model with respect to its overall performance in predicting natural processes is no easy task. The problem is twofold, firstly there is the availability and suitability of field data with which to compare a model with and secondly there is the method of evaluation. The Air Pollution Model (TAPM), developed by the CSIRO was used to simulate the winds in Southeast Queensland (SEQ). The complex nature of the airshed makes it difficult to compare modelled data with observational data as the observational data may be influenced by local phenomena. Evaluation of the model through the use of standard statistics and monthly and seasonal statistics illustrated that overall the model predicted the annual average wind speeds and temperatures well. Through the use of synoptic clustering, more detail on model performance was gained and it was found that TAPM predicted sea breezes that occurred on high pollution days. The sensitivity of the model to the selection of input parameters such as soil type, land use, vegetation, and rain processes was also investigated.