Measuring ground reinforcement with fiber optics
Cement-based ground reinforcement makes soft clay safe to build on, but excess cement use generates unnecessary carbon emissions. By casting thin fiber-optic cables into the ground, Sølve Hov has now documented exactly how the ground deforms under load.
Sølve Hov is a geotechnical engineer and researcher at NGI. His doctoral work shows that fiber-optic sensors can deliver far more precise measurements of cement columns than conventional methods. ( Per Olav Solberg / NGI)
When roads, railways, or housing must be built on areas with soft clay, deep mixing is a common solution. A drilling machine blends cement, lime, or other binders with the clay to create stabilized underground columns. Engineers have used the method for decades to ensure the ground can bear the weight of structures. The challenge has been that engineers lack precise data on how these columns actually compress over time, an uncertainty that routinely leads to over-engineering.
"We have a very poor understanding of how stabilized ground behaves. It’s about building knowledge of material behavior so we can optimize the method and stop overusing cement, which carries a high carbon footprint," says Sølve Hov, geotechnical engineer and researcher at the Norwegian Geotechnical Institute (NGI).
Measuring deformation with a laser
For his recently completed doctoral work at NTNU, Hov used fiber-optic sensors to collect measurement data. Instead of relying on traditional point measurements at the top of the columns, he installed fiber-optic cables directly into the concrete mixture. The technology works on the same principle as high-speed internet: a laser signal travels through a glass core.
"We send a light signal through an extremely thin glass core inside the fiber cable, thinner than a human hair. Just as you see reflections when looking through a window, something similar happens inside the cable. When the signal reflects to the measuring unit, we process the data to measure strain, meaning deformations, temperature, and vibrations along the entire cable," explains Hov.
Detecting local weaknesses
The key advantage of fiber optics is millimeter-level resolution. Traditional monitoring of cement columns relies on instruments placed at half-meter or one-meter intervals, or on measuring surface settlement.
"With fiber optics, the measurement points sit so close together that we get data for nearly every centimeter along the cable’s full length," says Hov.
That level of detail matters because a deep underground cement column is rarely a perfect blend of clay and cement — strength and stiffness vary significantly over very short distances. The high resolution allows detection of small, local weak zones that conventional investigations tend to smooth over and miss. In one field trial at Onsøy, for example, the fiber optics revealed a 20 to 30 centimeter-wide weak zone at a depth of 5.5 meters.
"When we get such an accurate picture of how the columns actually compress and respond under load, we can improve the calculation models and significantly cut cement use across different projects," he says.
Fiber-optic sensors exploit reflections of light signals inside a glass fiber thinner than a human hair to measure strain, temperature, and vibrations along the cable's full length. ( Illustration: Dall E / Open AI)
Developing a new test method
To trust the field results, Hov first had to document the interaction between the cable and the surrounding soil — the interface that determines whether the measurements reflect reality.
"At first, I thought it couldn’t be that complicated. We measure what we measure. But are we really sure we’re measuring the soil's behavior, and not just noise in the cable?" he asks.
Engineers traditionally use pull-out tests to measure cable friction in soil. Hov’s doctoral work showed that such tests produce misleading parameters because they simulate tension rather than compression. He therefore developed a new compressive test method in the laboratory that replicates how the soil actually presses on the cable in the field.
"We ran experiments at NTNU in Trondheim, at NGI’s lab in Oslo, and at the University of California, Berkeley. It took considerable collaboration to establish that we are actually measuring correctly," he says.
Field trials cut emissions
With the test methodology established, Hov conducted full-scale field trials at Lilleby and Tiller-Flotten in Trondheim, and at Onsøy in Østfold. The measurements confirmed that the cables do not slip in the soil but deliver precise data on how the columns deform over time and respond to varying loads. The sensors captured deformation properties that had previously been difficult to quantify.
Updating calculation models with this data allows engineers to reduce cement volumes. It also opens the door to safely testing alternative, more environmentally friendly binders.
"In one field trial, we used paper ash from Norske Skog at Skogn instead of cement. It worked very well, and the fiber-optics confirmed it. This is ash that would otherwise end up in a landfill," explains Hov.
The fiber-optic cables can also detect vibrations, in principle converting the cable into a series of microphones.
"We used this in field trials to investigate the wave velocity of the columns, which we can then link to other properties through correlations established in the laboratory. That way, we could track how the properties changed over time," he says.
The knowledge from this doctoral work gives the construction industry a concrete tool for reducing material use without compromising safety. The framework for interpreting the fiber-optic signals is general and can also be used to monitor natural ground and other foundation techniques. The work itself demonstrates how research can solve practical challenges.
"What’s valuable is the combination of consultancy and research that NGI stands for and strives toward. I want to do more research, but not full-time. And I want it tied to consultancy projects, because that’s where the research delivers direct benefit," concludes Sølve Hov.
