Single-Channel Infrasound Detection Using Machine Learning
Description:
Infrasound, low frequency sound less than 20 Hz, is generated by both natural and anthropogenic sources. A source exerts a force on the atmosphere, generating infrasonic waves. These waves have the potential to travel thousands of kilometers due to their low frequency nature, making infrasound particularly useful for explosion monitoring. Regional and global networks of infrasound arrays are sparse, but many single-sensor infrasound stations exist. However, current processing methods rely on the presence of an infrasound array for signal detection and event association, location, and characterization efforts. Here we present a method using machine learning to detect infrasound arrivals at single-channel infrasound stations. We show that single-channel infrasound detection is possible and reliable as well as discuss efforts by the University of Utah to create a standardized and automated infrasound event catalog. This event catalog will be used as ground truth to verify single-channel infrasound detections, serving as a test for model generalization.
This research was funded by the National Nuclear Security Administration, Defense Nuclear Nonproliferation Research and Development (NNSA DNN R&D). The authors acknowledge important interdisciplinary collaboration with scientists and engineers from LANL, LLNL, MSTS, PNNL, and SNL. SNL is managed and operated by NTESS under DOE NNSA contract DE-NA0003525.
Session: Detecting, Locating, Characterizing and Monitoring Non-earthquake Seismoacoustic Sources
Type: Oral
Date: 4/19/2023
Presentation Time: 11:30 AM (local time)
Presenting Author: Sarah A. Albert
Student Presenter: No
Invited Presentation:
Authors
Sarah Albert Presenting Author Corresponding Author salber@sandia.gov Sandia National Laboratories |
J. Mark Hale hale.jmark@utah.edu University of Utah |
Kristine Pankow kris.pankow@utah.edu University of Utah |
|
|
|
|
|
|
Single-Channel Infrasound Detection Using Machine Learning
Category
Detecting, Locating, Characterizing and Monitoring Non-earthquake Seismoacoustic Sources