Two years of moisture monitoring using microwave sensors in the medieval cave town of Uplistsikhe, Georgia

Deep within the stone, hidden from the eye, a silent battle is constantly being waged against time. This is the story of rock moisture, a quiet but powerful force that shapes the fate of ancient structures. Water is not merely a liquid that sits on a surface; it is a dynamic agent that seeps into the tiny pores of stone, expanding and contracting with changes in temperature and humidity. When this moisture moves in and out of the rock, it triggers chemical reactions and physical stresses that cause stone to flake, crumble, and eventually disintegrate. For the world's cultural heritage, from medieval castles to ancient cave cities, understanding where this water comes from and how it moves is the key to saving these irreplaceable treasures. Without knowing the source of the dampness, any attempt to preserve a site is often a guess, and a wrong guess can sometimes cause more harm than good.

In the heart of Georgia, the medieval cave town of Uplistsikhe stands as a dramatic example of this struggle. Carved directly into a massive cliff of soft, yellowish sandstone, the town was a thriving hub during the high Middle Ages before being abandoned centuries ago. Today, its chambers and corridors suffer from the same decay that plagues stone everywhere: surfaces are peeling away in thin layers, and the rock is turning into a granular powder. Scientists have long suspected that the culprit is the fluctuating humidity within the stone, but the exact pathways the water takes to reach the back of these deep caves have remained a mystery. To solve this, a team of researchers from Germany, Austria, and Georgia decided to listen to the stone itself. They installed a network of specialized sensors inside two of the town's most significant caves, the Grand Hall and the Long Hall, to monitor the moisture levels continuously for two full years.

The technology they used was a form of microwave sensing, a method that works by sending a pulse of energy into the rock and measuring how much bounces back. Because water holds energy differently than dry stone, the sensors can detect exactly how much moisture is present without drilling or damaging the ancient walls. The team placed eight of these sensors in total, some near the floor and others near the ceiling, and some deep inside the caves while others were positioned just outside. To ensure the readings were accurate, the researchers first took large blocks of the same type of sandstone found at the site to a laboratory. There, they dried the stones completely and then soaked them in water, calibrating the sensors to translate their electronic signals into precise measurements of water content. They also had to account for the fact that temperature changes can trick the sensors, so they developed a way to adjust the data based on the heat or cold of the rock.

Once the equipment was running, the sensors began to reveal a complex and surprising picture of life inside the stone. The data showed a clear seasonal rhythm: the rock was generally wetter during the summer months and drier in the winter. This was counterintuitive to some, as one might expect rain to make the stone wettest in the wettest seasons, but the summer heat seemed to drive moisture from the air into the rock, or perhaps pull water up from the ground. The researchers found that the stone reacted to different weather events in distinct ways. When heavy rain fell, the sensors on the exposed outer walls of the Long Hall reacted immediately, their moisture levels spiking as the rain hit the surface. However, the sensors deep inside the Grand Hall did not react instantly. Instead, they showed a delayed response, sometimes taking several days to show an increase in moisture after a storm. This delay suggested that water was not just sitting on the surface but was slowly seeping through the rock from above, traveling through the ceiling of the cave before finally reaching the sensors.

The study also uncovered that the water was arriving through multiple different channels, not just one. In some cases, the moisture appeared to be driven by condensation, where water vapor from the air settled onto the cooler rock surfaces. In other instances, particularly near the floor of the Grand Hall, the moisture levels rose in a way that suggested water was being pulled up from the ground below, a process known as capillary rise, likely triggered when the water table in the surrounding hillside rose after heavy rains. The researchers observed that the rock was not a static object but a living system that breathed, absorbing water during humid periods and releasing it during dry ones. This daily cycle of wetting and drying, which the team described as a kind of evaporative pumping, was driven by the sun warming the surface during the day and the rock cooling at night.

Perhaps the most important finding was that there is no single solution for preserving the site. Because the water enters the Grand Hall and the Long Hall through different paths—some from the sky, some from the ground, and some from the air—what works for one cave might not work for another. The researchers concluded that trying to seal the caves with waterproof coatings would likely be a mistake, as it would trap moisture inside and accelerate the decay. Instead, the best approach is to manage the water where it enters, such as by directing rainwater away from the cave entrances or lowering the water table in the hillside to reduce the upward pull of moisture. The two years of monitoring provided a rare, detailed look at the hidden hydrology of an ancient site, proving that to save the stone, one must first understand the invisible water that moves within it.