The final project of my PhD was the development of a minimalist infrasonic signal detection and characterization technique requiring just one microphone and one three-component seismometer. This technique could be advantageous in situations where resources are limited. The development of this signal detection and characterization technique for ground-coupled airwaves using a nearly collocated seismometer and microphone was feasible because of groundwork laid out in Ichihara et al. [2012] and Matoza and Fee [2014]. This work was a poster presentation at the AGU Fall Meeting 2016 (Poster) and was presented at the IAVCEI 2017 Scientific Assembly in August; abstract below. Details of this method have been published [McKee et al., 2018].


Infrasound signal detection and characterization using ground-coupled airwaves on a single seismo-acoustic sensor pair

Kathleen McKee1, David Fee1, Matthew Haney2, John Lyons2, Robin Matoza3 

1Alaska Volcano Observatory, Geophysical Institute, University of Alaska Fairbanks, Fairbanks, Alaska

2Alaska Volcano Observatory, U. S. Geological Survey Volcano Science Center, Anchorage, Alaska

3Department of Earth Science, University of California, Santa Barbara, California

Here we aim to develop a minimalist infrasound signal detection and characterization technique requiring just one microphone and one three-component seismometer. Ground-coupled airwaves (GCA) are commonly sourced from volcanic eruptions, bolides, meteors and explosions and detected 100s of kilometers away across seismic networks and infrasound arrays. GCAs occur when an incident atmospheric pressure wave encounters the Earth’s surface and part of the energy of the wave is transferred to the ground (i.e. coupled to the ground) as a seismic wave. This seismic wave typically propagates as a Rayleigh wave evidenced by the retrograde particle motion detected on a three-component seismometer. When these acoustic waves propagate along the surface exciting the assumed dominant Rayleigh waves in the subsurface and are recorded on a collocated microphone and seismometer, they can be coherent and have a 90° phase. If the sensors are separated, usually 10s to 100s of meters, then recorded wind noise becomes incoherent relative to wind speed and frequency and an additional phase shift is present due to the separation distance.


Determining a source azimuth should be possible using a single seismo-acoustic sensor pair by utilizing the phase difference, coherence, and exploiting the characteristic particle motion. The phase difference from 90° depends on the direction the pressure wave arrives from, as each back-azimuth will have a different apparent distance between the sensors. However, the apparent sensor separation determined from the additional phase alone does not provide a unique source azimuth. In turn, we incorporate the arrival times and particle motion to determine a unique solution. Here we use synthetic seismo-acoustic data generated by a coupled Earth-atmosphere 3D finite difference code to test and tune the detection and characterization method. These simulations have the expected high coherence, 90° phase and elliptical retrograde particle motion. The method is then further tested using various well-constrained sources (e.g. Antares Rocket, Chelyabinsk meteor, Pagan and Cleveland Volcanoes) and existing high signal-noise data (e.g. EarthScope Transportable Array, IMS infrasound network, USGS volcano monitoring networks). Our proposed technique would provide a new method to detect infrasound signals and determining the back-azimuth, and would be particularly useful in situations where resources are limited and large sensor networks or arrays are not feasible.