Near real-time seismic data processing helps scientists understand aftershocks
A new method of automatically detecting and locating earthquakes uses artificial intelligence. The system has successfully located recent earthquakes in Taiwan and could help scientists monitor ongoing events.
By Hao Kuo-Chen, Ph.D., Sun Wei-Fang, Chun-Ming Huangand Pan Sheng YanNational Taiwan University
Quote: Kuo-Chen, H., Sun, W., Huang, C., Pan, S., 2022, Near real-time seismic data processing helps scientists understand aftershocks, Temblor, http://doi.org /10.32858/temblor. 276
This article is also available in Traditional Chinese.
A magnitude 6.5 earthquake hit eastern Taiwan on Saturday, September 17. Just 17 hours later, a 6.9 magnitude main shock struck, rupturing the surface. The main shock damaged buildings, roads and bridges along the southern longitudinal valley, the boundary between the Eurasian and Philippine tectonic plates.
Many aftershocks followed. Although the larger Main Shock and Precursor Shock are important for understanding seismic hazards in the region, these smaller earthquakes, which are common after a large event, provide valuable information about the rupture process.
When an earthquake strikes, scientists analyze seismograms – plots of seismic waves recorded at stations scattered across the landscape. Accurately determining when primary (P) and secondary (S) seismic waves arrive at a station is critical to locating an earthquake. However, conventional methods for spotting the arrival of P and S waves in a sea of seismic data are time-consuming, especially for aftershock sequences, when a considerable number of small earthquakes occur in a relatively short time. .
Being able to quickly assess aftershock distribution can help scientists track seismic hazard following large earthquakes and into the future, which is crucial for seismic risk management.
We recently developed artificial intelligence (AI) technology for wave detection, and since 2021 we have deployed five seismic stations in eastern Taiwan equipped with such capabilities. Fortunately, this network of stations was well placed to record the latest earthquake sequence that hit eastern Taiwan. We were able to successfully monitor the entire event, from seismicity to aftershocks, in near real time, as our artificial intelligence system, called SeisBlue, automatically detected the incoming waves. SeisBlue is a deep learning data processing platform for creating earthquake catalogs offline or in near real time.
A catalog of earthquakes in near real time
In the 39 hours following the magnitude 6.5 shock, SeisBlue detected 6,104 earthquakes and located 1,223. Most of the aftershocks were triggered along a 20 kilometer (12 mile) stretch of the southern longitudinal valley. After a week, aftershocks extended 90 kilometers (55 miles). According to residents’ recollections, the damage to the construction was mainly caused by the main shock, which is consistent with our detection results.
Within two days of the 6.5 magnitude shock, a catalog of earthquakes from the event, detected by SeisBlue, was presented to the public. Such a catalog of earthquakes assembled by conventional methods would have taken more than six months to create. The catalog will be used by seismologists and community planners for risk assessment and disaster relief and by the research community to identify potential locations of surface failures.
Mousavi, SM, Ellsworth, WL, Zhu, W., Chuang, LY, and Beroza, GC (2020). Seismic Transformer—an attentive deep learning model for simultaneous earthquake detection and phase selection. Natural Communications, 11(1), 1-12.