ETH researchers look deep into the earth"There was a huge anomaly - I thought: This can't be"
Philipp Dahm
2.2.2025
Using a new method, ETH researchers have looked deeper into the Earth's interior than ever before. The big surprise is that they found an anomaly in the western Pacific Ocean that was previously only associated with continental plates.
Google Earth
Researchers at ETH Zurich have developed a new model that provides a clearer view of the Earth's interior and have found anomalies that nobody expected. Thomas Schouten explains what's going on.
02.02.2025, 21:00
03.02.2025, 07:46
Philipp Dahm
No time? blue News summarizes for you
ETH researchers have developed a new model that allows a closer look inside the Earth.
They found anomalies in tunnels in the Earth's mantle where scientists had not previously suspected them.
Thomas Schouten, who is involved in the project, explains how the model works and what the anomalies might be all about.
"Sensational discovery: Is there a lost world beneath the oceans?", headlines "Bild". The reason: researchers at ETH Zurich have succeeded in gaining an unprecedented insight into the Earth's interior. It is the "first big research surprise of the year", the Germans cheer.
Thomas Schouten is part of this team of researchers whose work has been published in "Nature": the doctoral student explains what they have discovered in an interview.
Mr. Schouten, why do we know more about space and the deep sea than about the interior of the Earth?
The problem is that we cannot look directly into the Earth's interior. The deepest hole that man has drilled is in Russia. After 12 kilometers of drilling, the machines broke down because the pressure and temperature were too high. The difficult thing is that the Earth is over 12,700 kilometers in diameter, so there is quite a lot that we cannot see.
A drilling rig on the Kola Peninsula in 2007. The Kola borehole is considered to be the only over-deep borehole in a continental shield.
Commons/Andre Belozeroff
But you were still able to take a deeper look.
It's like a doctor taking an ultrasound scan: we use waves that have a certain energy. Depending on how these waves pass through a body, we can make statements about its structure. This applies to both humans and the earth. The waves are accelerated, slowed down, deflected or reflected by structures in the body. These changes in wave speed are called "seismic anomalies".
What problems do geologists have when measuring these anomalies?
When you go to the doctor and have an ultrasound of the heart, you theoretically get a uniform image. You have the same high resolution of the interior everywhere because there is a uniform distribution of sources and receivers in the ultrasound device. The problem with the earth is that the seismographs that register the waves are not as well placed as with ultrasound. They are usually only found on land: there are also some on the seabed, but only a few and only temporarily.
Where are seismographs usually located?
They are usually located where there are earthquakes. In South America, most of them are located on the western coast, where the Pacific plate is pushing under the South American plate. For a country like Chile, it is therefore much more important to have a good network of seismographs than for Brazil, where there are far fewer earthquakes.
An ultrasound device produces the same high quality images everywhere, and with the earth it depends on where there are how many seismographs?
Exactly, it depends on where you are and what is being recorded. Another problem so far has been that an earthquake generates a lot of data - actually too much to process. That's why you limit yourself to a selection of waves that you measure. You take the ones that are easiest to recognize: These are the first two waves you see. That has already worked well: We were able to take pictures of the Earth's interior.
But?
The resolution varies greatly. In some areas - the Pacific Ocean, for example - we only see a very blurred image. It's an ocean, there are very few seismographs.
What has changed?
In my department, seismology professor Andreas Fichtner and his doctoral students Sebastian Noe and Sölvi Thrastarson have been working on a method called full waveform inversion. This involves using not only the first two waves of the seismogram, but also much more information from waves that arrive later. There are actually too many. Andreas Fichtner and his team looked at this and considered how to process this information as efficiently as possible. The result is a new, high-resolution model
Waves on a seismograph at the Sea Lab in New Bedford, Massachusetts.
Archive image:IMAGO/USA TODAY Network
What did you want to use this model for?
I reconstruct plate paths. This is difficult because large parts of the earth's plates disappear into the earth's interior through subduction - as is the case with the west coast of South America. Previous images of the Earth's interior have always shown some kind of anomalies under these subduction zones, and the common theory is that such anomalies are simply the plates that have disappeared.
What did the new method show?
When I looked at the western Pacific in the new, high-resolution model, I thought: This can't be right. There was a huge anomaly that no model had ever seen before. At first we thought we had found a new plate.
Why is that not the case?
We found the anomalies not only under the western Pacific, but everywhere - especially under stable continents and oceans. We thought we would only find the anomalies where plates slide under other plates. Now we see that this type of anomaly is much more widespread than we thought. Part of the reason these anomalies were always interpreted as tectonic plates is that they were mostly seen where they were already expected.
Earthquake waves move slower in the red zones and faster in the blue ones: a new ETH model suddenly shows a blue zone in the western Pacific that nobody expected.
Graphic:Sebastian Noe / ETH Zürich
Can you explain the image you published again?
The image shows wave speeds: They are slightly faster in the blue zone than on average and slightly slower in the red zone. On the one hand, this is influenced by temperature: blue means slightly colder, red is warmer. The idea is that a plate that is cold sinks into the mantle. However, there are other factors besides temperature that influence the waves.
What are they?
It's about the chemical composition, which can indicate that it's a different rock. Seismic waves move faster through earth material if it contains more iron or silicon. Silicon is one of the most important chemical elements in the Earth's mantle, along with the main component of sand.
Is that surprising?
No, it is not so surprising that it has different compositions in different mantle regions. We know this because geochemists have studied the material from volcanoes. There have been convection movements in the Earth's interior for four billion years, but they don't work in such a way that everything is completely mixed. In some regions there is less mixed material, where there may be more iron or more silica.
Are there any other theories about the anomalies?
There is a little more iron in the plates than in the mantle and the iron is dissolved during subduction. The plate sinks and dissolves, but the iron can collect in certain zones in the mantle due to the convection movements - that is also a possibility. In any case, there is a lot for science to do!