Numerical predictions of atmospheric sound‐pressure levels in shadow zones

Abstract

This letter is a study of some practical considerations that must be addressed when using modern numerical methods to predict atmospheric sound‐pressure levels in upward refracting environments. Conventional formulations of the shadow zone problem usually assume that the sound speed decreases indefinitely with height above the ground. A version of the fast field program (FFP) was used, which was formulated for a linear sound‐speed profile to illustrate the effect of ‘‘capping’’ the linear sound‐speed decrease with a homogeneous layer. This is found to give significantly increased sound‐pressure levels in the ‘‘shadow zone’’ when the height of the homogeneous layer becomes comparable to the thickness of the creeping wave region defined in the theory for an indefinitely decreasing profile. Also considered was the accuracy of the homogeneous layer approximation when used to represent the sound‐speed profile for noise propagation into a shadow zone. This approximation was used in a recent formulation of the fast field program (FFP) for the atmospheric propagation problem. The approximation is studied for some realistic examples of long‐distance propagation of low‐frequency (<500 Hz) acoustic signals. To assess the accuracy of the homogeneous layer approximation, a comparison of numerical calculations of sound‐pressure levels computed using the FFP introduced by Lee et al. [J. Acoust. Soc. Am. 79, 628–634 (1986)] with the FFP formulated for the linear speed profile must be made. A realistic ground impedance model (Delaney–Bazley–Chessel) M. E. Delaney and E. N. Bazley [J. Sound Vib. 16, 315–322 (1971); C. I. Chessel, J. Acoust. Soc. Am. 62, 825–834 (1977)] is used so that this study will apply to the case of sound propagation over a typical grass covered surface.