S41A-2707
Ambient Noise Surface Wave Tomography for Geotechnical Monitoring Using “Large N” Distributed Acoustic Sensing

Thursday, 17 December 2015
Poster Hall (Moscone South)
Jonathan Blair Ajo Franklin1, Nate Lindsey1, Eileen Rose Martin2, Anna M Wagner3, Michelle Robertson1, Kevin Bjella4, Arthur Gelvin4, Craig Ulrich1, Yuxin Wu5, Barry M Freifeld1 and Thomas M Daley1, (1)Lawrence Berkeley National Laboratory, Berkeley, CA, United States, (2)Stanford University, Stanford, CA, United States, (3)Cold Regions Research and Engineering Laboratory Alaska, Fort Wainwright, AK, United States, (4)U.S. Army Cold Regions Research and Engineering Laboratory Alaska, Fairbanks, AK, United States, (5)Lawrence Berkeley National Lab, Berkeley, CA, United States
Abstract:
Surface wave tomography using ambient noise sources has found broad application at the regional scale but has not been adopted fully for geotechnical applications despite the abundance of noise sources in this context. The recent development of Distributed Acoustic Sensing (DAS) provides a clear path for inexpensively recording high spatial resolution (< 1m sampling) surface wave data in the context of infrastructure monitoring over significant spatial domains (10s of km). Infrastructure monitoring is particularly crucial in the context of high-latitude installations where a changing global climate can trigger reductions in soil strength due to permafrost thaw. DAS surface wave monitoring systems, particularly those installed in/near transport corridors and coupled to ambient noise inversion algorithms, could be a critical “early warning” system to detect zones of decreased shear strength before failure.

We present preliminary ambient noise tomography results from a 1.3 km continuously recording subsurface DAS array used to record traffic noise next to an active road in Fairbanks, AK. The array, depolyed at the Farmer's Loop Permafrost Test Station, was designed as a narrow 2D array and installed via trenching at ~30 cm. We develop a pre-processing and QC approach to analyze the large resulting volume of data, equivalent to a 1300 geophone array sampled at 1 khz. We utilize automated dispersion analysis and a quasi-2D MC inversion to generate a shear wave velocity profile underneath the road in a region of discontinuous permafrost. The results are validated against a high-resolution ERT survey as well as direct-push data on ice content. We also compare vintages of ambient noise DAS data to evaluate the short-term repeatability of the technique in the face of changing noise environments. The resulting dataset demonstrates the utility of using DAS for real-time shear-modulus monitoring in support of critical infrastructure.

<< Previous Abstract | Next Abstract >>