Earthquakes on tap
In a former service tunnel built for the Furka-Gotthard rail link, ETH geoscientists are looking to move mountains. We take a closer look at the underground lab.
The post bus stops here, but “Bedretto, Bivio per Ronco” is very much in the middle of nowhere. At first glance, all you can see is a gravel pit with huts, heavy machinery, dumper trucks and heaps of stone. It could be a construction site anywhere in Switzerland. Yet it’s far more intriguing than that: it marks the entrance to ETH Zurich’s BedrettoLab.
In early September, the lab opened its doors to the public. After several years of preparation and preliminary experimentation, the lab team is now gearing up for the main event. This is scheduled for next March, when the controlled activation of a tectonic fault zone deep inside the rock will trigger tiny quakes – an earthquake on tap, so to speak.
The researchers, all dressed in Day-Glo workwear, await the tour group. Equipped with helmets, jackets and trousers in luminescent yellow, they look very much like tunnel workers – and that, in a sense, is what they are. Visitors are prepared for the foray into the bowels of the mountain: everyone gets a helmet, torch, high-vis bib and, last but not least, breathing apparatus, which is the size of a compact shoulder bag. In an emergency, this 4.5-kilogram piece of kit is a lifesaver, providing one hour’s worth of oxygen – enough to get you back to the tunnel entrance, or to the Furka Base Tunnel.
Robots in action
This text appeared in the 25/04 issue of the ETH magazine Globe.
Our group approaches the entrance to the tunnel, where we are greeted by a banner bearing the words “BedrettoLab – ETH Zürich”. The journey into the mountain and to a unique geoscientific facility is about to begin. Deep beneath Pizzo Rotondo, the highest point of the Gotthard Massif, an international team has been busy setting up this unrivalled research lab. It is here that scientists will explore in great detail just how earthquakes occur – what happens when they are triggered, how they spread and how they finally stop. Behind this bold ambition is a project named FEAR (Fault Activation and Earthquake Rupture), funded by the European Research Council to the tune of 13.7 million euros.
Fault Activation and Earthquake Rupture (FEAR) is a project led by ETH professors Domenico Giardini and Stefan Wiemer, Florian Amann from RWTH Aachen University and Massimo Cocco from the Instituto Nazionale di Geofisica e Vulcanologia in Rome. Funded by an ERC Synergy Grant, the experiment also receives financial support from the Swiss Confederation and the Werner Siemens Foundation. The project is scheduled to run for nine years.
The tunnel is a couple of metres across, just wide enough for a vehicle to pass. To the right, a torrent rushes by, flushing away the water that continuously leaches from the mountain. Pipes and cables run along the tunnel walls, where rusty old hooks have been driven into the rock. Corroded grating is bolted to the bare tunnel roof to prevent stones from falling on passersby below. Despite this, we are advised to wear a helmet at all times.
As we proceed deeper into the tunnel, Florian Amann, one of the four project leaders, suddenly halts. Pointing to an inconspicuous crack in the rock, he moves closer and runs his finger along the groove. A blob of light-grey sludge gathers on his fingertip. This material, known as fault gouge, indicates that there has been earthquake activity here. In the past, two masses of rock rubbed against one another like millstones, producing fine granules that turned to a slurry in the presence of water.
“This tells us that a fault zone runs through the massif right here,” explains Amann. And that is exactly what the geoscientists are looking for. “This blob of gouge might look pretty unspectacular, but to us it’s worth its weight in gold,” he smiles. The plan now is to gently activate this fault zone in order to trigger tiny tremors and thereby investigate the various mechanical processes involved as an earthquake unfolds. Later, these findings should help improve earthquake prediction – both the when and the where – and consolidate what is still a very inexact science.
“As a major natural hazard, earthquakes pose a serious global risk,” explains Stefan Wiemer, head of the Swiss Seismological Service (SED) at ETH Zurich and one of the four project leaders. “Despite intensive research in recent decades, we still don’t understand enough about them.” The Bedretto tunnel offers an ideal opportunity to fill the gaps in this knowledge. Built in the 1970s as a service tunnel for the construction of the Furka Base Tunnel, it runs for five kilometres under Pizzo Rotondo, joining the Bedretto valley with the rail tunnel itself. Once the base tunnel had been completed, the Bedretto tunnel was mothballed.
By 2016, however, geoscientists were searching for a suitable location for an underground rock lab. It was then that Simon L?w, professor emeritus at ETH Zurich, recalled the Bedretto tunnel. Wiemer and his colleagues inspected the site and were extremely impressed. It seemed perfect for their needs. “It had barely been touched since the construction of the rail tunnel,” Amann says. They had also looked at tunnels in working mines. But, as he explains, doing scientific work there would have been a non-starter, as commercial operations would have always taken priority: “Here, however, we can do our research without any restrictions.”
Florian Amann explains how the experiment will proceed. (Image: Daniel Winkler / ETH Zurich)
In this cavern, water will be injected at high pressure into boreholes to set the rock masses in motion. (Image: Daniel Winkler / ETH Zurich)
The screen displays current measurement data. (Image: Daniel Winkler / ETH Zurich)
Cross section of a core sample. (Image: Daniel Winkler / ETH Zurich))
Sensitive instruments
We come to a fork in the tunnel. The cavern is brightly lit and filled with a background roar. Fresh air flows in through a yellow ventilation pipe suspended from the roof. It is also pleasantly warm – warmer than outside. “The drilling work generates a lot of heat,” Wiemer explains, “which is why it’s so nice and cosy down here right now.” It has only been a few months since the team had the side tunnel blasted into the rock. This is where they will activate the fault zone. To ensure that nothing goes unnoticed, the geoscientists have also drilled dozens of boreholes into the rock – totalling 3.6 kilometres in length – and filled them with an array of highly sensitive sensors and measuring instruments.
A small number of boreholes are used to inject water into the rock and thereby trigger earth tremors, while the rest serve to monitor the processes in the rock. Seismometers installed in the BedrettoLab can detect earthquakes with a magnitude of –5. The energy released in this process is just enough to make the rock masses move a few micrometres against one another. “There’s no other network of measuring instruments quite like this, situated directly on a fault zone, anywhere else in the world,” Wiemer says with pride. He explains that while a global seismographic network exists, most of its sensors and measuring instruments are located above ground and therefore far from where earthquakes originate.
Moving mountains
Now the researchers are eager to reap the rewards of all this groundwork. In March 2026, they plan to trigger tiny earthquakes of a very low magnitude along the fault zone and study in detail what happens before, during and after a seismic event. The team will also be on the lookout for signs that point to an impending quake. For this purpose, they will inject at high pressure hundreds of cubic metres of water into boreholes drilled along the fault zone – until the twin rock masses suddenly begin to shift, just like two tectonic plates but on a much smaller scale.
The energy released in the process will be equivalent to that of a magnitude-1 earthquake – just enough to shift the two rock masses by around 1 millimetre. “These artificially induced earthquakes will be of such a small magnitude that no one outside of the BedrettoLab will notice anything,” says Wiemer.
Terabytes of data
Within a fraction of a second, all of the instruments that have been embedded in the rock over the past months and years – such as seismometers and geophones – will then record millions of data points. Thanks to a dense network of sensors installed in and on the rock, researchers will be able to follow and capture every detail of these artificially induced micro-?earthquakes in real time. A single measurement will generate several terabytes of data.
The experiment will be continuously monitored, with real-time data sent to ETH. Back in Zur?ich, team leaders will be able to regulate the water pressure at any time to keep the situation under control – or, if necessary, increase it to make the rock masses move. But Amann considers it highly unlikely that a powerful, uncontrollable quake will occur. “It’ll be great if we manage to move anything at all,” he cautions. “There are over 1,000?metres of solid rock above the lab, weighing an estimated 35?million?metric tons.”
But can the findings from this experiment be applied to massive earthquakes – quakes that occur along a fault line extending hundreds of kilometres, where tectonic plates move against one another by several metres? Could the findings apply to an earthquake like the one in Myanmar in spring 2025, for example, which was magnitude 7.7? “That’s what we’re aiming for: we hope to be able to transfer the insights that we gain at the BedrettoLab to systems on a much larger scale,” Wiemer explains. “Whether it’s a micro-earthquake of magnitude?1 or a gigantic earthquake of magnitude?7, the physics are the same.”
Back to daylight
For the visitor group, it is time to leave the lab. Bidding farewell to the geo?scientists and their measuring equipment, we head back along the straight tunnel to the daylight of the Bedretto valley.
After two-and-a-half kilometres, the entrance appears. Outside, it is raining, the mountain peaks shrouded in heavy clouds. Despite the miserable weather, it feels good to swap the tunnel roof for the grey sky and breathe in the fresh Alpine air. Out here, and down in the valley below, no one will be any the wiser when micro-earthquakes are triggered deep inside the mountain. But once all that data has been analysed back in Zurich, the results may well lead to better forecasting – so that one day valley dwellers will benefit from advance warning of actual earthquakes.