← Back to Sessions

Landslide Monitoring with a Local Infrasound Array in Barry Arm, Alaska

Description: 

Large surficial mass movements generate energetic, low-frequency sound waves in the atmosphere, or infrasound, that can propagate and be detected tens to hundreds of kilometers from source areas. The recent proliferation of infrasound sensors in regional networks provides event databases with a growing variety of sources and drives improvements in signal detection and classification algorithms. Yet challenges remain in automated, real-time detection and characterization of infrasound from small to moderate mass movements due to the often emergent, low-amplitude signals from these events.

Here we present results from the first 5 months of local infrasound data from a 6-element array installed in summer 2022 just 2 kilometers from a recently identified, large (500 M m3) landslide in Barry Arm, Prince William Sound, Alaska. The array recorded signals from a wide variety of sources in the acoustically rich environment such as rockfalls, calving signals from multiple local glaciers, and snow avalanches. We process the array data in real-time using a least-squares beamforming algorithm to distinguish coherent signals from noise and to constrain source back-azimuths. Investigation of infrasound detections in 2022 indicate an increase in small mass movements from a portion of the landslide during September and October, coincident with heavy rains and high rates of landslide motion detected by terrestrial synthetic aperture radar. We investigate individual events from the dominant infrasound sources (landslide, glacier calving, snow avalanches) that were also captured in radar data and imagery to better understand how they generate infrasound. Identification of signal characteristics unique to the various infrasound sources active in Barry Arm is a necessary next step toward an automated detection and classification workflow to better monitor this potentially hazardous landslide. While the current focus is on Barry Arm, the techniques and lessons learned here should be applicable to other locations and will also provide insights into how best to monitor similar mass movement processes at regional scales.

Session: Detecting, Locating, Characterizing and Monitoring Non-earthquake Seismoacoustic Sources

Type: Oral

Date: 4/19/2023

Presentation Time: 10:30 AM (local time)

Presenting Author: John J. Lyons

Student Presenter: No

Invited Presentation: 

Authors