Wind noise and the spectrum of atmospheric turbulence pressure fluctuations

Abstract

Previous research [S. Morgan and R. Raspet, J. Acoust. Soc. Am. 92, 1180–1183 (1992)] has shown that wind noise is predominantly caused by pressure fluctuations intrinsic to the turbulent atmospheric flow. Therefore it should be possible to predict wind noise from models for turbulent pressure spectra in the atmosphere. Based on simple dimensional analysis and an application of Taylor’s hypothesis, the inertial-subrange power spectrum for turbulent pressure fluctuations should be proportional to f−7/3, where f is frequency. But more recent atmospheric observations and theoretical arguments [J. D. Albertson, G. G. Katul, M. B. Parlange, and W. E. Eichinger, Phys. Fluids 10, 1725–1732 (1998)] suggest that the power spectrum actually goes as f−3/2. In this paper, it is shown that the f−3/2 dependence predicts a much slower decay in wind noise with increasing acoustic frequency than is typically observed. Possible reasons for this discrepancy are discussed.