Ground beneath rapidly growing cities can move in ways that are almost impossible to see until the consequences become dangerous. A few millimeters of gradual settlement may eventually contribute to cracked structures, distorted roads, damaged utility networks or, in severe cases, sudden ground collapse. The risk is particularly serious around underground infrastructure, where excavation, tunnel construction, imperfect backfilling and hidden voids can weaken the ground without producing obvious surface warning signs. Now, researchers in South Korea have developed a multi-scale remote sensing framework that combines satellite radar, laser scanning and ground-penetrating radar to detect settlement across large urban areas and investigate what may be causing it.
The system, developed by a research team led by Tae-Yong Park of Seoul National University of Science and Technology, is designed to overcome the limitations of relying on a single monitoring technology. Its large-scale component uses satellite-based Interferometric Synthetic Aperture Radar, or InSAR, to track slow changes in the elevation of the ground surface over long periods. Its site-specific component uses laser scanning to measure deformation in an exposed structure, while ground-penetrating radar, or GPR, examines the shallow subsurface for signs of voids and disrupted soil layers. By combining the three forms of evidence, the researchers aim to transform settlement monitoring from a system that merely detects movement into one capable of investigating its likely origin.
Ground settlement occurs when the surface descends because the soil or rock beneath it changes. Such changes can result from groundwater withdrawal, compression of loose deposits, underground excavation, insufficient compaction during construction or the development of cavities below the surface. In dense cities, these processes may take place beneath roads, buildings and transit corridors, making conventional inspections difficult and expensive. Continuous, wide-area surveillance is therefore essential, but each established method has weaknesses. GPS equipment can provide highly accurate positioning at selected points, yet satellite signals cannot penetrate the ground and are unsuitable for directly monitoring underground structures. InSAR covers large regions, but its measurements can be influenced by radar viewing geometry, atmospheric delays, vegetation and surface changes.
InSAR works by comparing radar signals reflected from the Earth’s surface during repeated satellite passes. When the returning signals are processed interferometrically, very small changes in the distance between the satellite and the ground can be estimated, potentially revealing millimeter-scale movement over time. The researchers strengthened this analysis by applying seasonal-trend decomposition using LOESS, a statistical technique that separates a time series into long-term trends, repeating seasonal patterns and irregular variations. This helped distinguish persistent settlement from temporary fluctuations associated with seasonal environmental conditions. The approach was tested along more than 16 kilometers of the Seoul Metropolitan Subway Bundang Line corridor, between Suseo Station and Cheongnyangni Station, allowing the team to screen a substantial urban area for unusual deformation.
The satellite analysis identified a ventilation shaft with a distinct and progressively increasing settlement signal. Rather than treating the InSAR result as a final diagnosis, the researchers used it as a trigger for a more detailed field investigation. This “forensic” strategy is central to the framework: satellites identify where the ground may be changing, and ground-based tools then determine how the structure and subsurface are responding. The ability to prioritize specific locations could help infrastructure authorities focus limited inspection resources on areas showing the strongest evidence of long-term movement, rather than attempting to examine every part of an extensive subway network at the same level of detail.
At the selected shaft, the team carried out laser scanning using LiDAR technology. LiDAR emits laser pulses and measures the time required for them to return after striking nearby surfaces. Millions of measurements can be assembled into a three-dimensional point cloud, creating a detailed digital representation of the structure. This allows investigators to quantify subtle changes in geometry and identify deformation patterns that may be difficult to judge visually. The shaft ceiling already showed several cracks and signs of repair. Analysis of the laser-scanning data indicated that settlement became more pronounced toward the section of the ceiling directly beneath the roadside above the shaft, suggesting that the deformation was spatially organized rather than random.
The researchers then surveyed the road above the shaft with ground-penetrating radar. Unlike InSAR and LiDAR, which primarily reveal movement or geometric change, GPR investigates the condition of materials below the surface. The system sends high-frequency electromagnetic pulses into the ground and records reflections produced when the signals encounter boundaries between materials with different electrical properties. Changes in moisture, density, composition or structure can alter the reflected signal. In this case, the radar data showed features consistent with void-like structures near the shaft and reduced continuity in subsurface layer boundaries. These observations indicated that the ground was not uniform and provided a possible explanation for the localized settlement observed above and within the underground structure.
When the three datasets were visualized and interpreted together, they produced a coherent picture of the anomaly. InSAR demonstrated that the area was undergoing progressive surface settlement over time. Laser scanning confirmed that the ventilation shaft itself was deforming, with the strongest displacement concentrated beneath a particular roadside section. GPR supplied evidence of potentially unstable near-surface conditions, including void-like features and disrupted layers. Because the anomaly appeared independently in multiple measurements at different spatial scales, the researchers concluded that it was unlikely to be an artifact caused by one sensor, one viewing geometry or one processing error. The fusion of the datasets reduced uncertainty and connected surface movement with structural and subsurface evidence.
The proposed framework could make urban infrastructure management more proactive. Instead of waiting for visible cracks, road depressions or sudden sinkholes to trigger an emergency response, authorities could use satellite time series to identify developing patterns and then deploy targeted field surveys. Laser scanning could reveal whether an underground structure is bending, settling or changing shape, while GPR could help identify voids, poorly compacted zones or other construction-related defects. The method may be especially valuable in cities where subway systems, utility tunnels and deep excavations occupy crowded and geologically complex environments. Park says the long-term goal is to support timely maintenance and focused investigation of high-risk locations, reducing the possibility that gradual ground movement will evolve into structural damage or infrastructure failure.
The study does not suggest that any single technology can predict every settlement event, and the interpretation of remote sensing data still requires engineering expertise and site verification. Atmospheric effects, surface conditions, sensor resolution and the complex behavior of urban soils can all affect measurements. However, the Seoul research demonstrates how these challenges can be addressed by combining complementary observations rather than treating them in isolation. Published in Tunnelling and Underground Space Technology, the work presents a practical model for using satellite surveillance as an early-warning layer, field laser scanning as a structural diagnostic tool and GPR as a window into the shallow subsurface. Together, the technologies could help cities detect hidden ground risks earlier—before a slow, almost invisible movement becomes a highly visible disaster.
Subject of Research: Not applicable
Article Title: Satellite based forensic MSRS monitoring system for detecting settlement of urban underground
Web References: https://doi.org/10.1016/j.tust.2026.107723
References: Tae-Yong Park et al., “Satellite based forensic MSRS monitoring system for detecting settlement of urban underground,” Tunnelling and Underground Space Technology, Volume 174, published 1 August 2026. DOI: 10.1016/j.tust.2026.107723
Image Credits: Tae-Yong Park, Seoul National University of Science and Technology, Korea
Keywords
Ground settlement, urban infrastructure, InSAR, satellite remote sensing, LiDAR, laser scanning, ground-penetrating radar, subway safety, underground construction, sinkhole detection, civil engineering, disaster prevention
Tags: advanced urban geotechnical investigationcity ground stability analysisearly detection of ground movementground collapse risk mitigationground-penetrating radar for subsurface analysislaser scanning for structural deformationmulti-technology urban infrastructure monitoringmultiscale remote sensing frameworksatellite radar for urban monitoringSeoul National University researchunderground infrastructure integrity assessmenturban ground settlement detection
