IRD researchers publish a study on May 11, 2022 in the journal Nature showing that gravitational signals can be exploited to instantly estimate the magnitude of large earthquakes. These signals, which are very weak and propagate at the speed of light, were discovered in 2017 in data from the 2011 Fukushima earthquake. This study could eventually lead to more reliable and faster tsunami warning systems.

Earthquakes and the tsunamis that they generate caused almost one million casualties in the last thirty years. Alert systems have been developed to limit the human and material cost of these natural disasters. 

Up to now, these systems have used seismic waves to warn the population seconds before shakings. Tsunamis propagate slower, giving more time to act (tens of minutes). However, alert systems have difficulties to estimate quickly and accurately the magnitude of very large earthquakes. For instance, the Japanese system estimated a Magnitude 8 event instead of a Magnitude 9 during the 2011 earthquake, and consequently a 3-meter wave instead of 15-meter one, a mistake of dramatic consequences in Fukushima.

Illustration of the AI algorithm estimating large earthquakes magnitude based on gravitational signals (PEGS) traveling at the speed of light, much faster than seismic (P and S) waves.

© DR

An international research team (IRD – CNRS – Université Côte d’Azur – Observatoire de la Côte d’Azur – Los Alamos National Laboratory – Kyoto University) publishes in Nature a study using Artificial Intelligence (AI) to estimate instantaneously the magnitude of large earthquakes based on Prompt Elasto-Gravity Signals (PEGS). PEGS are gravitational perturbations generated by the motion of a large mass of rocks during the earthquake. They propagate at the speed of light, carrying earthquake information much faster than seismic waves, just like lightning warns of impending thunder. While in principle PEGS could help speed up warnings, their very weak amplitude prevented, so far, their use in alert systems. The researchers overcame this limitation thanks to an AI algorithm. They showed that the magnitude of large earthquakes can be accurately estimated based on PEGS seconds after the earthquake starts, and tracked as the earthquake grows, beating existing warning methods.

“Tested in Japan, this algorithm has proven capable of estimating the Fukushima earthquake magnitude faster and more accurately than any existing system, without using any seismic waves” Andrea Licciardi – geophysicist and leading author of the study – said. Quentin Bletery, who initiated the project, added that: “The implementation of the algorithm in operational alert systems remains to be done, but our results indicate that PEGS could significantly improve tsunami alert systems”.

Reference: Instantaneous tracking of earthquake growth with Elasto-Gravity Signals. A. Licciardi, Q. Bletery, B. Rouet-Leduc, J.-P. Ampuero et K. Juhel. Nature, May 11, 2022.

Researcher contacts:

  • Andrea Licciardi, IRD post-doctoral researcher at GEOAZUR (IRD/CNRS/Observatoire de la Côte-d’Azur/Université Côte d’Azur) 
  • Quentin Bletery, IRD research fellow at GEOAZUR (IRD/CNRS/Observatoire de la Côte-d’Azur/Université Côte d’Azur)