Loading...
Near-surface seismic characterization and monitoring: a dense seismic acquisition perspective
Yang, Jihyun
Yang, Jihyun
Citations
Altmetric:
Advisor
Editor
Date
Date Issued
2023
Date Submitted
Collections
Research Projects
Organizational Units
Journal Issue
Embargo Expires
Abstract
Geophysical analyses made available by sensors such as geophones that are commonly used to measure signals from within the Earth are frequently constrained by intrinsic aliasing mainly due to sensor spacing. While in many cases it is possible and meaningful to deploy dense geophone arrays, the number of sensors installed has been constrained by the cost and effort required to install those devices. In contrast to conventional geophones, recently distributed acoustic sensing (DAS) using fiber offers both rapid deployment and dense spatial and temporal sampling. DAS has been widely used in many applications; however, horizontal (dark-)fiber DAS presents challenges including variable fiber-ground coupling, amplitude fluctuations in cross-correlation gathers, and (crooked) deployment directionality. To realize the potential of horizontal DAS, the work examines such deployments for near-surface characterization and monitoring. I present research work in a case study that uses low-frequency ambient DAS data. I demonstrate that applying interferometric analysis can construct virtual shot gathers (VSGs) that are used in a multi-channel analysis of surface waves (MASW) to constrain the shear-wave velocity (VS) up to 0.5 km depth. However, the amplitude variations by factors such as source types, gauge length, and orientation hinder the application of advanced inversion that require greater amplitude fidelity. I investigate the usability of filtered deformation-rate DAS data under improved fiber-ground coupling. To achieve near-elastic coupling, I freeze a fiber in a trench and compare observations to those measured on fiber laid on the ground. The trenched-freezing method improves the data quality and shows enhanced signal quality and reduced time-varying effects from variable coupling. I also compare different filtering techniques and show that the 2D filtering approach leads to improved data quality. I demonstrate the potential of DAS for long-term monitoring using a pre-existing fiber deployment. I acquire short-duration data every hour for ten months and use cross-coherence to convert observations into weekly sliding-window VSGs. I observe surface-wave travel-time variations of up to 6% for a range of fixed source-receiver offsets. I estimate time-lapse averaged S-wave velocity for the top 30 m (VS30) using the computed VSGs and MASW. The surface-wave inversions reveal coherent 10% maximum fluctuations of VS30, with the maximum slowdowns occurring after periods of heavy precipitation suggesting a negative empirical correlation between rainfall and VS30. The final study uses seismic data from a dense geophone array to assess refraction tomography and elastic full-waveform inversion (E-FWI) to comprehend the subsurface in Majes, Peru, near a suspected landslide. I use P-wave refraction travel-time tomography to generate a starting velocity model. I then estimate a VS model using low-frequency surface-wave data that exhibits a layered structure with a 450 m/s average velocity and a velocity reversal from 5 m to 15 m away from the cliff. However, the VS model closer to the cliff has VS values between 250 m/s and 400 m/s to 20 m depth. The significant surface-wave backscattering observed at 90 m along the line is likely to be caused by a strong sub-vertical discontinuity, while the corresponding S-wave velocity slowdown may delineate a former landslide complex. Further geotechnical research is needed to determine whether these geophysical insights are related to suspected landslide events.
Associated Publications
Rights
Copyright of the original work is retained by the author.