Earthquake monitoring at the speed of light

Earthquakes and tsunamis caused by them Nearly a million victims in the past thirty years. Several warning systems have been developed to reduce the human and material costs of these natural disasters. However, these systems have difficulties in quickly and accurately estimating the magnitude of very large earthquakes.

Now, a study was published in temper nature Describe a machine learning model that recognizes patterns in seismic data to better estimate the magnitude and location of a large earthquake.

Using 350,000 earthquake modeling scenarios starting at 1,400 potential earthquake sites in Japan, Andrea LisiardiA geophysicist at the University of Côte d’Azur in France and his colleagues have succeeded in instantaneous estimation of the magnitude of large earthquakes based on rapid elastic gravity (PEGS) signals.

PEGS are gravitational disturbances caused by the movement of large masses of rock during an earthquake. They propagate at the speed of light, and carry earthquake information much faster than the seismic waves traditionally used in early warning systems.

“The innovative thing about this paper is the use of machine learning techniques, which makes it possible to improve the detection of these very small signals.”

Scientists have realized that although PEGS can in principle help speed up earthquake warnings, its too weak amplitude has prevented its use in warning systems. The researchers in the new study overcome this limitation thanks to an artificial intelligence algorithm that relies on global navigation system data via satellite. Using the algorithm, they showed that the magnitude of large earthquakes can be accurately estimated based on PEGS seconds after the earthquake began and tracked as the earthquake grows.

“I think this paper is interesting. In fact, PEGS has already been discovered in my paper for 2016 Then confirm in the paper Valley et al in 2017,” He said Jean Paul Montagner, a geologist at the Institute of International Physics in Paris was not involved in the new research. “So the innovative thing about this paper is the use of machine learning techniques, which makes it possible to improve the detection of these very small signals.”

Licciardi agreed. “The main advantage of our model is based on the underlying data, the elastic gravitational signals,” he explained. “Once an earthquake occurs, these signals travel faster than seismic waves and are highly sensitive to earthquake magnitude. Because of this, our model can estimate earthquake magnitude faster and more accurately than conventional early warning systems based on s waves at least for large earthquakes (magnitude greater than 8.3 / 8.4). “

Improving tsunami early warning

Licciardi noted that the model’s response time of about a minute could significantly improve tsunami early warning predictions. In a real-time scenario, he said, the volume recovered by the model could be used to rapidly estimate the magnitude of the induced tsunami wave and thus mitigate its impact.

Classic early warning systems rely on s “The waves cannot distinguish between a magnitude 8 and 9 earthquake, whereas our model does not suffer from this limitation,” Lisiardi said. “It provides the most accurate estimate of the amount as a function of time.”

Lisiardi explained that the new model’s strength in predicting large earthquakes “is due to the fact that the gravitational-elastic signal is very sensitive to the magnitude of such large earthquakes.” “In fact, the application of our model is limited to such large earthquakes (magnitude above 8.3/8.4) because the signal amplitude of relatively smaller earthquakes is very small and buried in background noise. This is why other tools and data are still needed in the context of early warning.”

“This is important for earthquake early warning systems because for the largest earthquakes, there is an extended time (up to minutes) in which magnitude and strength combine,” he explained. Andreas Bleich, a senior Earth scientist at Harvard University and was not involved in the new work. “The authors correctly point out that over this extended time the method, especially if combined with other methods, has operational potential to track the growth of such an earthquake earlier and more accurately.”

Bleich further noted that with the new model, tsunami alerts can be issued not only earlier (with tens of seconds or perhaps even minutes) but also with greater confidence and with better estimates of wave height derived from the improved magnitude estimates.

– Mohamed El-Sayed (@MOHAMMED2SAID), a science writer

the quote: Al-Saeed m. (2022) Earthquake detection at the speed of light. Eos, 103, https://doi.org/10.1029/2022EO220261. Posted on June 2, 2022.
Text © 2022. The authors. CC BY-NC-ND 3.0.0 Update
Unless otherwise noted, images are subject to copyright. Any reuse without the express permission of the copyright owner is prohibited.