TY - JOUR
T1 - Near-surface seismic imaging of tunnels using 3D elastic full waveform inversion
AU - Smith, James
AU - Borisov, Dmitry
AU - Modrak, Ryan
AU - Tromp, Jeroen
AU - Cudney, Harley
AU - Moran, Mark
AU - Sloan, Steven D.
AU - Miller, Richard D.
AU - Peterie, Shelby L.
N1 - Publisher Copyright:
© 2017 SEG.
PY - 2017/8/17
Y1 - 2017/8/17
N2 - Detection and imaging of subterranean tunnels using seismic methods has historically been a difficult problem. Major challenges include the lack of separation (time and frequency) of body waves from surface waves, poor signal-to-noise ratio, and rapid attenuation of higher-frequency seismic energy in the shallow surface. In this proof-of-concept research experiment, we apply 3D elastic full waveform inversion to a field data-set collected by the Kansas Geological Society (KGS) at Yuma Proving Grounds (YPG) over an existing hand-dug, research tunnel with no distinguishing surface expressions. To image the tunnel, we used ten lines of twelve vertical sources spaced two meters apart. The void is observed as an elongated low S-wave velocity anomaly. Additional inversion experiments using fewer sources, further offset from the tunnel, and using horizontal sensors instead of vertical sensors, were also able to evaluate optimal approaches to locate the anomaly successfully.
AB - Detection and imaging of subterranean tunnels using seismic methods has historically been a difficult problem. Major challenges include the lack of separation (time and frequency) of body waves from surface waves, poor signal-to-noise ratio, and rapid attenuation of higher-frequency seismic energy in the shallow surface. In this proof-of-concept research experiment, we apply 3D elastic full waveform inversion to a field data-set collected by the Kansas Geological Society (KGS) at Yuma Proving Grounds (YPG) over an existing hand-dug, research tunnel with no distinguishing surface expressions. To image the tunnel, we used ten lines of twelve vertical sources spaced two meters apart. The void is observed as an elongated low S-wave velocity anomaly. Additional inversion experiments using fewer sources, further offset from the tunnel, and using horizontal sensors instead of vertical sensors, were also able to evaluate optimal approaches to locate the anomaly successfully.
UR - http://www.scopus.com/inward/record.url?scp=85044036597&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85044036597&partnerID=8YFLogxK
U2 - 10.1190/segam2017-17641505.1
DO - 10.1190/segam2017-17641505.1
M3 - Conference article
AN - SCOPUS:85044036597
SN - 1052-3812
SP - 2637
EP - 2641
JO - SEG Technical Program Expanded Abstracts
JF - SEG Technical Program Expanded Abstracts
T2 - Society of Exploration Geophysicists International Exposition and 87th Annual Meeting, SEG 2017
Y2 - 24 September 2017 through 29 September 2017
ER -