Progress in seismology is very encouraging, but “we are still far away” from being able to detect the occurrence of an earthquake in advance, warn the scientists interviewed by franceinfo.
The question arises after each large earthquake, like the one that struck Morocco on Friday September 8: will we one day be able to predict earthquakes precisely? For the moment, this is not the case, insist the experts unanimously. “It’s impossible to predict the time and day of an earthquakerecalled on franceinfo Badaoui Rouhban, expert in earthquakes and natural disaster prevention, Monday September 11. “For a given area, a given country, we can say that this area can be subject to earthquakes of such magnitude. We can specify this, but we can never know when the earthquake will occur .“
The Marrakech earthquake, which killed more than 2,900 people, according to a still provisional report, was not totally unexpected, even if its power surprised experts. “The earthquake that occurred does not surprise the science of seismology. It is an area where earthquakes can occur”adds Badaoui Rouhban.
For the moment, “even within a year, we cannot”
Experts knew such an event was coming. From there to being able to determine the day and approximate time of the earthquake, a significant gap remains to be crossed. “We are far from it”comments seismologist Philippe Guéguen, member of the waves and structures team at the Institute of Earth Sciences at the University of Grenoble-Alpes (ISTerre), to franceinfo. “Even within a year, we can’t”slips Mohamed Chlieh, researcher at ISTerre.
>> Earthquake in Morocco: can technology help predict or prevent earthquakes?
However, knowledge is refined thanks to an increasing volume of information. “Our knowledge of earthquakes – their origin, the processes – is largely due to the provision of data., remarks Philippe Guéguen. However, the instruments are proving to be more and more precise, and “a lot of progress has been made on the recordings”. The scientist mentions the contribution of more efficient seismometers, or the considerable contribution of geodesy: this ancient discipline, which is interested in the shape of the Earth, has made a considerable leap in recent decades thanks to satellites. Networks that help navigate the surface of our planet (like GPS) are proving valuable for better understanding earthquakes. Indeed, when the earth’s crust moves under the effect of plate tectonics, they can detect movements with precision down to the millimeter.
The difficult search for “precursor” signs
Despite the progress made, one sign shows that earthquake prediction is not for the immediate future: during meetings between seismologists, there are no discussions focused on this subject yet. It can be addressed, but “it’s not presented as such”, notes Philippe Guéguen. By modesty ? All work in seismology can ultimately contribute to this, he believes, but it is currently premature to talk about forecasts.
At this stage, specialists are content to mention studies relating to “precursors”. Because the crux of the problem lies precisely in identifying the warning signs of an earthquake. A task that turns out to be difficult, as they are so difficult to identify and interpret.
“You don’t have the same signature in every earthquake. For each one, we see a different precursor.”
Mohamed Chlieh, seismologistat franceinfo
What are the precursors being scrutinized? Scientists suspect radon, a radioactive and carcinogenic natural gas, to be one of them. Observations showed that it had been cleared shortly before certain earthquakes. It is still necessary to succeed in detecting a real signal which differs significantly from its usual presence in the environment. Some scientists are also interested in the agitation of ants before the occurrence of an earthquake, or the abnormal behavior of animals such as cats and dogs. The efforts made have not yet been able to uncover a reliable mechanism.
The “slow slide” trail
Another indicator in the viewfinder are the “precursor earthquakes”, which would testify to the beginning of the instability of the affected area. This is the whole subject of an article published in July in the scientific magazine Science and titled “The precursor phase of major earthquakes”. Schematically, an earthquake can be presented as a “quick sliding” of two plates at the level of a fault in the earth’s crust, Jean-Mathieu Nocquet, seismologist at the GeoAzur institute and author of the article, explains to franceinfo. “Lthe idea is to say that this rapid sliding is preceded by a slower slidinghe explains. We have known for twenty years that they exist. They have been found on a large number of faults.”
Starting from a physical model linking sliding at the level of the fault and displacement on the surface, Jean-Mathieu Nocquet took the data available on 70 large earthquakes and “GPS measurements trvery precise stations”. He managed to free “a very subtle signal, of the order of a centimeter”. It suggests that powerful earthquakes could be detectable a few minutes to two hours before the event. The advantage of this indicator, he notes, lies in the fact that it is based on a “hypothesis verifiable by observations”. “It brings hope and avenues to explore and develop”continues the specialist, while showing great caution: “We do not know not if all earthquakes exhibit this shape and behavior.”
“It is possible that earthquakes obey a single law, but also that there is a diversity of behavior.”
Jean-Mathieu Nocquet, seismologistat franceinfo
To consolidate this work, Jean-Mathieu Nocquet believes that it is necessary to gain precision by observing more earthquakes. This requires investments to deploy “systematically dense sensor networks in areas where earthquakes occur”.
Valuable data collected by farmers
For his part, the geologist Philippe Vernant does not hide a certain enthusiasm. This teacher-researcher at the Géosciences Montpellier laboratory emphasizes that the price of certain devices has dropped drastically – from several tens of thousands of euros to a few hundred euros – while still producing precise and reliable data. He mentions the deployment of the Centipede network launched by Julien Ancelin, farmer in Charente-Maritime and design engineer at the National Research Institute for Agriculture, Food and the Environment.
This network allows farmers to access technology at lower cost to carry out high-precision agriculture, in particular tractor auto-guidance. To access this geolocation-based system, operators place a small GPS on a fixed point, such as a barn, and another on agricultural machinery, as explained in the video below.
In the space of three years, hundreds of receivers have been installed by private actors throughout France, he underlines. And this initiative is starting to extend to other European countries, such as Hungary. First designed for farmers, the system produces data shared in real time and constantly, which is of interest to seismologists and geologists. By treating them, “we find lots of things we can work on”, welcomes Philippe Vernant. The arrival of this additional data, coupled with the development of artificial intelligence could, according to him, speed up things and, possibly, allow the establishment of an alert system in the decades to come.
But be careful: Jean-Mathieu Nocquet points out that several physical models remain in competition, one of which assumes that earthquakes are absolutely not predictable. “In this vision, which is perhaps correct, the earthquake system is chaotic in the sense that it obeys physical details and variations of shapes so fine that they are not accessible to observation”, he explains. For him, “it is too early to make a prediction” regarding the advent of potential predictions. While being cautious, Jean-Philippe Guéguen believes that it is not unreasonable to think of achieving this one day: “One could also say that it is completely crazy to send someone to the Moon.”