← Back to Sessions

Enhancing Seismic Monitoring in Cook Inlet, Alaska: Integration of Distributed Acoustic Sensing with the Existing Seismic Network for Advanced Earthquake Denoising, Detection and Location

Description: 

Offshore Distributed Acoustic Sensing (DAS) has emerged as a powerful technology for seismic monitoring, complementing the existing seismic networks in coastal regions. However, the offshore DAS data often combine signals that are new to the eyes of seismologists, including new types of instrumental noise, fiber coupling issues, ocean waves, and microseism that overprint earthquake signals. The overlap of these signals hinders accurate detection and location of earthquakes. In this study, we propose an advanced architecture for earthquake monitoring in Cook Inlet, Alaska, by integrating the offshore DAS with the Alaskan regional seismic network. The proposed architecture incorporates machine-learning-based denoising, phase picking, and event relocation, leveraging an ensemble approach with DAS and broadband observations. A denoising neural network, employing self-supervised learning, is trained on randomly masked real DAS recordings for 2000 earthquakes. The trained denoiser effectively removes DAS signals lacking spatial-temporal coherence. Subsequently, the denoised DAS data enables the detection of lower-magnitude earthquakes and provides significantly more phase picks than the original noisy data, particularly for the smallest events. We then apply the trained denoiser to the M0.6+ earthquakes in the Cook Inlet region. The ensemble-learning approach is applied to pick P and S arrival times on both the denoised DAS and broadband recordings. In the final step, we relocate the detected earthquakes using their P and S phase picks of the integrated DAS and broadband seismic network. We quantify the improvement of adding DAS to a regional network by quantifying uncertainties earthquake source parameters This comprehensive approach results in more efficient and robust earthquake detection and location compared to traditional seismic networks in the offshore region. The proposed architecture holds significant promise for earthquake early warning and hazard mitigation, demonstrating the potential for enhanced seismic monitoring networks in offshore environments.

Session: Advancing Seismology with Distributed Fiber Optic Sensing - III

Type: Oral

Date: 5/3/2024

Presentation Time: 02:00 PM (local time)

Presenting Author: Qibin

Student Presenter: No

Invited Presentation: 

Authors