On the development of approximate models for outdoor sound propagation

The suitable prediction model for outdoor sound propagation depends on the situation and the application. Computationally intensive method such as Parabolic Equation methods, FFP methods and Boundary Element Methods all have advantages in certain situations. None of these approaches are at present able to handle all the relevant aspects of propagation such as terrain features, atmospheric wind and temperature gradients and turbulence.How much wind can be tolerated when a model based upon a still homogeneous atmosphereis used? How does a terrain feature affect the sound propagation? An attempt is made to answer these questions and to find the limits between which various approaches are applicable.This should help identifying the more promising approach for various applications.