Jean-Philippe Avouac is a professor of geology at the California Institute of Technology and a grantee through the foundation’s Caltech commitment.
Avouac leads the Caltech Tectonics Observatory, a multidisciplinary center designed to understand and eventually predict the earth's behavior near the intersection of key tectonic plates.
The group strives to provide strategic insights into a variety of destructive phenomena such as earthquakes, volcanic eruptions, tsunamis and landslides.
What topics/problems in science are you most interested in solving?
We don’t understand earthquakes very well — it’s the only natural hazard that we have no clue how to predict at this time. We see a larger and larger fraction of the world’s population living in seismic hazard areas with no preparation, especially in developing countries.
There was a revolution forty years ago when plate tectonics was established, and we thought we could solve all the problems in earth sciences because there was a theory bringing everything together. In a nutshell, the theory states the crust is deforming only at limited places on the plate boundaries and this is where the action is taking place.
Given this information, people thought at the time, ‘OK, we’ll be able to predict earthquakes soon.’ And the truth is, 30 to 40 years later, we haven’t made much progress, and we don’t understand why: is it that any plate boundary can produce an earthquake in a random process, or can we measure something that would tell us ‘here, we have the potential to have a really large earthquake.’ This is the key question that has been driving my research.
How do you and your colleagues work toward understanding how earthquakes occur?
The way we decided to attack this problem was through new technologies to measure how the crust is deforming, to monitor key areas on the planet where we think large earthquakes might occur in the future. We measure stress building up before and stress release during earthquake.
One technology we use is GPS, which hadn’t been used for earthquakes until 15 to 20 years ago, when the cost became affordable to deploy GPS instruments in earthquake-prone regions around the world.
We also use radar interferometry, based on satellite images taken with a radar system that allows you to measure how the ground is moving toward the satellite. These techniques are complementary, as GPS can tell you displacement at one point in the ground with a high frequency (time resolution of less than one second, which allows measuring ground displacement as an earthquake unfolds), while the satellite images give you good spatial resolution but the measurement is done once every two weeks.
Our group also developed a technique using optical images. These techniques are also combined with longstanding techniques such as seismology, which carries a lot of information about fault slip during an earthquake. All of these data tells us how an earthquake starts, grows and arrests and how these characteristics determine ground motion and destruction.
What have you learned using these techniques?
We first selected Sumatra for a field deployment in 2003 before the 2004 earthquake, which was a 9.1-magnitude, really large earthquake. We collected a lot of data from this earthquake and many others, and by 2014, the only place that hadn’t had an earthquake yet was Nepal.
In all these examples, we found large earthquakes happened at places of stress build up that could be imaged from the measurement of surface deformation before they actually happened. We now have an approach to estimate where energy is accumulating and how it might be released in future earthquakes. We also made important progress regarding how to use this information for seismic hazard assessment.
How did the 2014 Nepal earthquake happen?
In Nepal, we measured the displacement of the ground over more than ten years before the Gorkha earthquake of 2014. India is moving like a rigid body to the north at about 4 cm/year, and it collides with Central Asia, at the Himalayan arc, which stretches over 2,000 km.
The strain rate map shows that the entire Himalayan arc is a place of rapid contraction, so it is no wonder why large earthquakes have occurred along the Himalaya. This tells you this behavior must be elastic strain — you are deforming the Himalayas, the fault is locked and the zone is being squeezed. At some point it will break, and that’s when you have an earthquake like happened during the Gorkha earthquake.
Our GPS measurements showed the earthquake ruptured the main fault along which India is thrust beneath the Himalaya. Given we demonstrated the fault slips at a rate of 2cm/year on average, such an earthquake would need to repeat every few centuries. This is very qualitative reasoning as successive earthquakes in an area can have very different rupture areas and slip.
Destruction during the earthquake was mostly due from seismic waves, and sliding on the main fault patch was surprisingly soft. This is why Kathmandu was relatively spared.
We are now investigating if these characteristics were due to identifiable factors. The learning would benefit our ability to predict ground motion and destruction in future earthquakes, whether in the Himalaya or elsewhere.
What are your greatest challenges as a scientist?
We have made important progress thanks to the use of space technologies. We can measure displacement at the earth’s surface and image where stress is building up in preparation of large earthquakes.
However, we have to assume that the crust is elastic and deforms like a spring. Is it really the case, or does it deform inelastically, like Play-Doh? Also, we still don’t know how to evaluate the absolute stress level — the energy available to drive future earthquakes.
I am hopeful we will continue to make progress on these questions. We need more collaboration and ambitious projects bringing together specialists in geology, seismology, remote sensing, sensor physics and mechanical engineering, who often don’t speak the same language and don’t interact much given the way academic activities are organized and funded.
Learn more about the Nepal earthquake in a video featuring Avouac here.
Message sent
Thank you for sharing.