Quantifying low-frequency acoustic fields in urban environments

Abstract

Infrasound data contain contributions from incoherent noise, coherent noise and signals of interest. The design of an infrasound array to target sources of interest requires a quantification of array response, individual sensor response, propagation effects (topography and meteorological conditions), signal spectrum and the noise environment. The Comprehensive Nuclear-Test-Ban Treaty community has spent significant effort in quantifying the acoustic field in rural environments for frequencies up to 7 Hz. Given that the nuclear monitoring and tactical infrasound community have growing interests in monitoring sources in or near populated regions, there is an emergent need to measure and understand acoustic fields in these environments as well. This paper focuses on quantification of the acoustic field in three different urban environments: (1) arrays installed within Dallas, TX, a metropolitan area, (2) an array installed at the rural-suburban interface near San Diego, CA and (3) an array installed in Vicksburg, MS, a small city with multiple major transportation corridors. A minimum of five months of data was recorded and used for each site characterization. The analysis focuses on frequencies from 0.1 to 45 Hz. The quantification of the data from these three sites is accomplished with statistical noise models that capture the total ambient acoustic field, separation of the coherent portion of the field and trend analysis to link temporal and seasonal variations to wind speed and anthropogenic activities. The resulting physical interpretation of the data demonstrates that the total acoustic field in urban regions is overall higher and inversely related to the number of coherent detections observed by an array. Furthermore, this study presents an analysis framework for characterizing additional urban arrays and provides a basis for future array site selection and installation.