Earthquakes account for five of the ten worst catastrophes on Wikipedia's list of natural disasters by death toll.
“Prediction of earthquakes is one of the most important questions in Earth sciences”, says Professor Francois Renard at NJORD – Centre for Studies of the Physics of the Earth.
This is not an easy task.
“One important difficulty is that most earthquakes start at a depth of 15 to 20 kilometres before propagating toward the surface. Nobody has been down there to see what happens”, Renard says to Titan.uio.no.
Today, earthquakes are studied by installing sensors, called seismometers, at the surface of the Earth. These sensors can “listen” to the earthquakes by recording the ground motion they produce.
Renard will try a different approach.
“We will take small pieces of rock, put them at conditions of 15 kilometres depth, and deform them to create a tiny earthquake in the sample. We will measure the evolution of the sample before and during the earthquake”, Renard says.
To do this, he has to build a new instrument that he will install at the European Synchrotron Radiation Facility in Grenoble in France. Today Renard was told he will receive funding through an Advanced Grant from the European Research Council (ERC). This grant supports researchers who have established themselves as leading researchers in their field.
“What I propose is to build an instrument where we will be able to see inside the sample in three dimensions at a micrometre resolution and through time. Pressure and temperature will be the same as it is at 15 kilometres depth”, Renard says.
Like a big telescope
He will collaborate with researchers from California, Maryland and Oxford, and of course the synchrotron in Grenoble.
“The synchrotron is a powerful x-ray source used in material sciences, in geophysics, and in biology to understand fundamental states of matter. A synchrotron in my field is like navigating on a research vessel for an oceanographer or observing the universe with telescope for an astrophysicist”, Renard says.
He has already built a smaller instrument that can reproduce earthquakes at two to three kilometres depth.
“I installed this instrument at the synchrotron in Grenoble in 2016 and I started to produce results. Experimental conditions were still a bit far from real earthquakes, but this machine showed that it was possible to look inside rock samples.”
“The next apparatus I will build will be much more powerful. It will not only be possible to see inside the sample by using the synchrotron x-rays, we will also be able to listen to the sample”, Renard says.
Listening is very important when he wants to study what kind of earthquakes that occur. While some earthquakes cause major damage at the surface of the Earth, there are others moving quietly deep down.
“30 years ago a major discovery in geophysics was the observation of what we call slow earthquakes. Instead of occurring in tens of seconds, as the largest earthquakes do, a slow earthquake will last several weeks or months. Over such a period, two tectonic plates may move by several decimetres, but slowly, without creating damaging seismic waves”, Renard explains.
He wants to understand why some earthquakes are slow and others are fast, big and cause lots of damage.
“That’s why I need these sensors to listen to the sample. We will be able to hear if the sample breaks suddenly, as during a fast earthquake. Conversely, the absence of noise will indicate a slow earthquake.”
To illustrate this, Renard uses a sheet of paper. When he tears it apart quickly, you clearly hear the noise. When he tears slowly, very slowly, there is almost no sound.
“With my experiments, I will be able to separate in the rock sample some places where it is breaking fast and some places where it is breaking very slowly. In the end, the sample will be broken and produce fractures, like in an earthquake, but it will happen either with a big sound or with almost no sound.”
A step toward prediction of earthquakes
Renard will work on this project, called BREAK, for five years. That is not enough to reach the ultimate goal of knowing in advance when an earthquake will happen.
“I will not be able to predict earthquakes, but I will make progress in that direction. I will be able to measure what happens before the earthquake starts. What I expect from the project is to be able to detect precursors in the samples”, Renard says.
“I think I will be able to demonstrate that what happens in the volume of rock around faults, not just the interface where earthquake propagate, is very important. Tiny little fractures in the rock volume may open and close in the months before an earthquake and control the preparation process of large earthquakes.”
The fractures they study in the samples are the size of a few grains of rock.
“An important difficulty is to upscale processes that occur on the scale of one grain, one millimetre, to the size of the largest earthquake which can propagate over hundreds of kilometres. That is a factor of 100 million and it is an outstanding challenge”, Renard says.
This summer he will take some of his colleagues and students to Lofoten in Northern Norway to collect rock samples where fossil earthquakes are recorded. He will compare them to rocks from the San Andreas Fault in California.
These samples will have a very different history of earthquakes. Rocks from the San Andreas Fault have recorded traces of hundreds of earthquakes. From Lofoten they will only have experienced one.
“We will look for differences in the rocks and if they have seen fast or slow earthquakes.”
“My instrument will be the first in the world that can do that”, Renard says.