Research by a scientist from the University of Alaska Fairbanks shows that the public could know days to months before a major earthquake based on the identification of prior widespread low-level tectonic activity. This analysis focuses on two major earthquakes in Alaska and California. New research suggests that it may be possible to predict large earthquakes months in advance by using machine learning to detect early signs of seismic activity. However, the effectiveness and ethical implications of this predictive technology remain a matter of debate.
The work was led by Társilo Girona, a research professor at the University of Alaska Geophysical Institute.
Girona is a physicist and data scientist who studies the precursor activity of volcanic eruptions and earthquakes. Kyriaki Drymoni, a geologist at Ludwig-Maximilians University in Munich, Germany, is a co-author of the study.
This machine learning-based detection method was published in Nature Communications on August 28.
“Our paper shows that advanced statistical techniques, particularly machine learning, have the potential to identify precursors of large-scale earthquakes by examining datasets from earthquake catalogs,” said Girona.
The authors wrote a computer algorithm to scour the data, looking for unusual seismic activity. An algorithm is a set of computer instructions that guides a program to interpret data, learn from it, and make predictions or informed decisions. Case Study: Anchoring and the Ridgecrest Earthquake
They focused on two major earthquakes: the 2018 magnitude 7.1 Anchorage earthquake and the 2019 Ridgecrest, California, magnitude 6.4 to 7.1 earthquake series. They found that about 15% to 25% of south-central Alaska and Southern California were affected by unusual low-magnitude regional earthquakes for about three months before the two earthquakes they studied. study of earthquakes caused by seismic activity of magnitude 1.5.
The Anchorage earthquake occurred at 8:29 a.m. on November 30, 2018, with the epicenter about 10.5 miles north of the city. The earthquake severely damaged some roads and highways and damaged some buildings.
Using their data training program, Girona and Drymoni found that in the Anchorage earthquake, the chance of a major earthquake occurring in 30 days or less suddenly increased to about 80 per about three months before the November 30 earthquake. Just days before the earthquake, the chance rose to about 85%. They also made similar findings for a period beginning about 40 days before the Ridgecrest earthquake series.
Girona and Drymoni suggested a geologic reason for low-volume foreland activity: a significant increase in pore fluid pressure within the fault. Pore fluid pressure refers to the pressure of the fluid within the rock. High pore fluid pressure has the potential to cause fault slippage if the pore fluid pressure is sufficient to overcome the frictional resistance between the rock masses on either side of the fault.
An increase in pore fluid pressure in faults that cause large earthquakes can change the mechanical properties of the fault, leading to non-uniform changes in the regional stress field. Research suggests that these uneven changes control unusual and premature earthquakes.
Machine learning is having a major impact on earthquake research, Girona said: “Today’s seismic networks generate large data sets that, if analyzed properly, can provide valuable insights into what has onset of seismic events. in high-performance computing it can play a transformative role, allowing researchers to identify meaningful patterns that could indicate impending earthquakes.”
The authors note that their algorithm will be tested in near-real-time situations to identify and address potential challenges in earthquake forecasting. This approach should not be adopted in a new area without training the algorithm on the area’s historical earthquakes. Producing reliable earthquake forecasts has a “very important and often controversial aspect”. Accurate forecasts have the potential to save lives and reduce economic losses by providing early warnings for timely evacuation and preparation. However, the inherent uncertainty in earthquake prediction also raises important ethical and practical questions. False alarms can lead to unnecessary panic, economic disruption and loss of public trust, and forecasting errors can have dire consequences.
