Simulation of sound propagation through high-resolution atmospheric boundary layer turbulence fields

We describe our development of a finite-difference, time-domain (FDTD) technique for sound propagation through the atmosphere. The approach includes the effect of turbulent wind and temperature fields in the propagation medium. When executed on the current generation of parallel processing computers using spatial domain decomposition, the new FDTD code provides calculations of wave propagation through the atmospheric boundary layer at unprecedented fidelity and detail. We drive the codes with dynamic wind and temperature fields from atmospheric large-eddy simulations, which themselves are run at very high resolution on parallel computers. The results vividly illustrate the dynamic behavior of sound waves propagating through the near-ground atmosphere and the differing signal characteristics of various atmospheric stability regimes. We also use the FDTD, with statistically controlled kinematic turbulence fields synthesized by the quasi-wavelet method, to validate widely used scattering theories for wave propagation through turbulence. With the FDTD, controlled numerical experiments can be devised that are not contaminated by factors such as refraction and ground reflections, which are inevitably present during experiments performed in the atmosphere. Comparisons between FDTD calculations of wavefield extinction and coherence and the theory of line-of-sight wave propagation (based on the parabolic and Markov approximations) show favorable agreement.