The relative importance of these processes depends on the source location, the emission properties, the building geometry and orientation, and the meteorological conditions. Consequently it is important to test any building effects model in a wide variety of circumstances so as to obtain a true overall view of its performance. At the same time a full appreciation of the sensitivity of predictions to changes in input conditions must be established. This activity, together with feedback from practical applications, feeds a development cycle as deficiencies are brought to light when the model is applied in more and more extreme circumstances.
The nature of the ADMS building effects model is first summarised, paying particular attention to the origins and basis of the algorithms used. The performance and applicability of the current model is then assessed against a wide range of experimental studies. Some matters that arose in model development and application are then discussed. Finally, some outstanding modelling and evaluation issues are discussed, together with the inherent limitations of building effects modelling in ‘regulatory’ dispersion models.