seismologists-tapped-into-the-fiber-optic-cable-network-to-study-offshore-faults

“`html

An aerial image captured from above the Homer Spit, a 4.5 mile stretch that projects into the water. On the left, fiber optic cables are submerged.

This overhead image depicts the Homer Spit in Alaska, located just south of the Cook Inlet. Although not visible in the image, fiber optic cables situated beneath the water on the upper left side supplied data for these studies.Qibin Shi

The Pacific Northwest features a comprehensive grid of over 600 seismic monitoring stations that aid scientists in observing tectonic and volcanic activities, inclusive of earthquakes. This information grants essential insights into local faults and feeds into early warning systems, allowing communities critical moments to brace for natural calamities. A major danger to this area, however, resides several miles offshore, where the Juan de Fuca plate is being pushed beneath the North American plate, creating the Cascadia Subduction Zone.

Tracking activity at ocean floor faults poses difficulties, and existing techniques frequently lack sufficient data for thorough investigations. To address this challenge, scholars are testing a method known as Distributed Acoustic Sensing, or DAS, which encompasses measuring vibrations on the ocean floor with fiber optic cables used for global telecommunications. Recent developments allow researchers to gather information from active cables and apply artificial intelligence to detect distant earthquakes that would typically go unnoticed.

In a new investigation, University of Washington scientists utilized the Ocean Observatory Initiative’s Regional Cabled Array, which stretches across the offshore plate boundary and sends data through fiber optic cables. Unlike past studies that depended on offline or “dark fibers” for data gathering, this latest research illustrates that DAS technology can function without disrupting the OOI network.

The researchers released their findings on February 28 in Seismological Research Letters.

“What we have developed is the foundation for any earthquake analysis,” stated co-author Marine Denolle, an associate professor in the Earth and space sciences department at UW. “Once our AI algorithm enhances the data, we can effectively employ the fluctuations for scientific inquiry.”

The fiber optic cable grid came to the attention of researchers over the last ten years when its potential for recording solid Earth data became evident. The cables convey bits of information across vast distances in the form of photons, or light particles. An apparatus — known as an interrogator — sends a pulse of light down the cable, but irregularities in the core can sometimes cause light to reflect back to its source.

Disruptions near the cable can divert the reflected particles off their intended path, and when they return to the source, scientists can trace their trajectory to identify the disturbance.

“When the earthquake is minor or distant, the energy on the cable is relatively low compared to the ocean, resulting in the signal being obscured by background noise,” noted co-author Qibin Shi, a former postdoctoral researcher at UW’s Earth and space sciences department who is currently a seismologist at Rice University.

Looking out towards the Cook Inlet and the Kenai Mountains beyond it

The Kenai Mountains rise above the Cook Inlet, under which two fiber optic cables gathered seismic data from the bustling seafloor for researchers to enhance using AI.Qibin Shi

In a prior study, UW scientists developed an algorithm that isolates signals and amplifies them by up to 2.5 times over the surrounding noise. All they need to do is let the algorithm sift through the data, and it will learn to recognize the signal — in this instance, an earthquake. The researchers utilized data from 285 earthquakes that occurred in Alaska’s Cook Inlet in 2023 as the training set.

“A well-trained model will discern earthquakes that are invisible to the human eye,” Shi remarked. “This represents the initial stride towards a broad-spectrum foundational model for earthquakes”

To ensure its efficacy in filtering data from additional locations, the researchers validated their model at the testing site in Oregon, employing a live cable. Previous studies, including the tests in Alaska, have obtained data from inactive cables, or dark fibers.

In Oregon, the researchers showcased their ability to obtain high-quality data while the cables were transmitting information. They accessed the Ocean Observatory Initiative’s Regional Cabled Array, which consists of fiber optic cables, and adjusted the algorithm to the frequency of seismic waves generated by small- to medium-sized distant earthquakes. The researchers then traced the signal back to distinct areas of the subduction zone and accurately pinpointed the location of an earthquake.

“It’s the closest we can get to where the action is,” Denolle said. “For addressing scientific inquiries, monitoring, and providing early tsunami and earthquake alerts, it’s our best opportunity.”

The system is also mobile, requiring only a modest amount of computational power for operation.

The recent trial in Oregon lasted just three days and yielded vast quantities of high-quality data, arguably more than the team knows how to handle, Denolle noted. Their current challenge lies in figuring out how to manage this data. Both datasets have been made freely accessible, and the one from Alaska is the largest single-site dataset of its type. The team is now working on arranging long-term placements for their monitoring setup and pursuing collaborative ventures.

“This is the future,” Denolle stated. “We will gain an understanding of plate movements by examining small earthquakes and this system provides us unprecedented access to such data.”

Additional co-authors of this paper include Ethan F. Williams, a postdoctoral researcher at the Earth and space sciences department; Bradley P. Lipovsky, an assistant professor at the Earth and space sciences department; William S. D. Wilcock, a professor in the oceanography department; Deborah S. Kelley, a professor in the oceanography department and director of the Ocean Observatory Initiative’s Regional Cabled Array and Katelyn Schoedl, a research coordinator in Earth and space sciences.

This investigation was financed by the National Science Foundation, U.S. Geological Survey, David and Lucile Packard Foundation, UW Geohazard Initiative, and Jerome M. Paros Endowed Chair in Sensor Networks.

For more details, contact Shi at [email protected] and Denolle at [email protected].







“`


Leave a Reply

Your email address will not be published. Required fields are marked *

Share This