Overburden pressure, adverse geology, sensitive structures nearby: every tunnel in the world is a different challenge (or series of challenges). Whether you’re rehabilitating a mountainside cut-and-cover or driving a tunnel-boring machine underneath critical civic infrastructure, tunnel deformation monitoring helps mitigate the risks of today’s increasingly sophisticated projects.
In this post we’ll compare different methods of monitoring tunnel deformation and convergence. The right monitoring tools provide data that can help protect a project’s timeline, economics, and safety—as well as ensure third-party assets are left undisturbed.
Tunnel convergence is a type of deformation that occurs when A) the surrounding ground is displaced (i.e., overburden pressure, erosion, freeze-thaw), or B) the tunnel’s structural elements (i.e., concrete lining) move unexpectedly because of defect or error . Deformations may be brief or prolonged, minor or major.
During construction, deformation most frequently occurs close to the tunnel face, shortly after advancement , which means it’s important that deformation monitoring begin as quickly as possible. To support this goal, monitoring tools should be unobtrusive enough that they don’t interfere with construction, and also robust enough to withstand a work environment that may be hazardous or polluted with rubble, dust, moisture, etc.
Simultaneously, it’s often important for surrounding properties or assets to be monitored for settlement, displacement or deformation due to the construction process. A successful tunnel project in a high-density urban areas requires that neighbouring infrastructure and utilities aren’t affected by settlement or deformation themselves.
For existing tunnels, convergence and deformation monitoring may be used in the long-term as an early warning system against potential environmental hazards and structural failures. Monitoring is also crucial in the short-term, during rehabilitation work on or near the tunnel.
Finally, tunnel designers can benefit from performance measurement of existing tunnels, obtaining valuable data that helps calibrate models for new and subsequent projects.
3D geodetics using total stations has become one of the dominant solutions for measuring tunnel convergence. The total station may be manned by a surveyor, or else installed and automated for long-term use. The instrument measures the positions of optical targets affixed to the tunnel walls and places them within an absolute coordinate system. 
One drawback, however, is that total stations require an unobstructed line of sight within the tunnel. According to the team who drove the 4.8km Weinberg rail tunnel underneath Zurich in 2011, total stations could not be used to monitor convergence because—at 11 meters wide and 150 meters long—the tunnel-boring machine took up most of the tunnel’s cross-section.  Corners are also a limitation for these instruments, and according to some research, 3D coordinate measurement can produce an error when measuring crown deformation in tunnels. 
Total stations were considered for use ensuring the integrity of the Victorian-era brick Ranelagh sewer during the construction of the Crossrail passenger line underneath central London. However, human operators entering the sewer in an ongoing way was determined to be an unacceptable risk. While an automated total station was an option for the sewer, ultimately the design team chose to install remote sensors, instead. 
In the last decade, improvements in technology have seen the rise of another optical method of tunnel monitoring: laser scanning, or LiDAR. Whereas total stations measure a fixed set of targets, laser scanners deliver measurements on millions of 3D points, forming a point cloud that can be compared over time.
Laser scanners have a lot of potential for use in tunnel construction, especially compared to total stations.  When it comes to deformation monitoring, however, the chief problem is that “the point accuracy of the original data is generally the same magnitude as the smallest level of deformations that are to be measured.”  This results in a need for statistical and 3D picture processing techniques.
Laser scanners generally have similar drawbacks to total stations: they require a clear line of sight, and while they can be automated,  the need for human operators opens up opportunities for human error and reduces the instruments’ usefulness for long-term monitoring, and monitoring in remote or hazardous locations.
Tape extensometers are portable instruments designed to manually measure the relative distance between reference points (e.g., anchors bolted into the tunnel lining).
This instrument is useful for monitoring tunnel convergence, but its reliance on human operators and open line between reference points make it unsuitable for use in live tunnels or with tunnel-boring machines, as we’ve previously seen. Meanwhile, data is neither real-time nor continuous.
Unlike total stations, this system of articulated arms and tilt sensors (a.k.a. MEMS accelerometer sensors) was purpose-built for rail tunnels and has no line-of-sight requirements. When mounted to the intrados of the tunnel lining, sensors log any X and Y displacement for each reference point, and transmit them to proprietary software. 
The system’s relatively low profile, near-real-time data collection, and rugged construction made it popular in rail tunnel monitoring projects for many years. However, its profile is not low enough for use during construction with a tunnel-boring machine , and this type of sensor has been eclipsed by newer technologies.
ShapeArray is an automated, remote MEMS accelerometer array that—like the Bassett Convergence System—can be affixed to a tunnel’s interior. It measures tilts relative to gravity in order to plot a 3D shape and monitor for any changes. Unlike a total station, ShapeArray’s data is relative to its initial starting reference (rather than absolute coordinates.)
ShapeArray can be automated after installation, with remote data collection available at the desired sample rate. Automation and a low profile (maximum 46 mm, including clamps) make this instrument well-suited for long-term use in remote, difficult-to-access stretches of tunnel, or areas with high traffic where disruption is costly. Because of its ease of installation (which can be as short as 25 minutes from deployment to data logging), ShapeArray is also suited for temporary monitoring needs during construction. With the convergence installation kit available for SAAV-model ShapeArrays, the instrument comes pre-assembled inside a flexible 21 mm ID PVC flex conduit, so that the user can easily deploy the system directly from its shipping reel.
Monitoring deformation and convergence with the right instruments is crucial for stakeholder confidence in the performance and health of every tunnel project, as well as the structures and property around it.
At Measurand, we believe the future of tunnel deformation monitoring instrumentation is faster installation and seamless, reliable data collection. In other words, giving decision-makers more time to look at data by helping them spend less time collecting it.
 A. Strauss, J. Bien, H. Neuner, et al. “Sensing and monitoring in tunnels testing and monitoring methods for the assessment of tunnels,” Structural Concrete, vol. 21, pp. 1356– 1376, February 2020.
 B. Bradley and R.P. Garcia, “The use of Shape Accel Array for monitoring utilities during urban tunnel drives,” ICE Publishing, Tech. Rep., 2014.
 C. Meyer and S. Schutz, “Risk control – online deformation monitoring inside an urban TBM-driven tunnel using the tShape measurement system,” in World Tunnel Congress 2013 Geneva Underground – the way to the future! G. Anagnostou and H. Ehrbar, Eds. London: Taylor & Francis Group, 2013. pp. 1395-1402.
 T. Nuttens, A. Wulf, L. Bral, et al, “High resolution terrestrial laser scanning for tunnel deformation measurement” in Facing the Challenges – Building the Capacity, FIG Congress 2010, Sydney, Australia, April 11-16, 2010 .
 W. Wang, W. Zhao, L. Huang, et al, “Applications of terrestrial laser scanning for tunnels: A review,” Journal of Traffic and Transportation Engineering vol. 1(5), pp. 325–337, October 2014.
 Y. Luo, J. Chen, W. Xi, et al, “Application of a total station with RDM to monitor tunnel displacement,” Journal of Performance of Constructed Facilities, vol. 31, issue 4, August 2017.
U.S. Patent 5,321,257
U.S. Patent 5,633,494
Canadian Patent 2,073,162
U.S. Patent 6,127,672, 6,563,107
U.S. Patent 6,127,672, 6,563,107
U.S. Patent 7,296,363
Canadian Patent 2,472,421
U.S. Patent 7,296,363
Canadian Patent 2,472,421
Canadian application 2,815,199 & 2,815,195
Cyclical Sensor Array, Canadian application 2,815,199 & 2,911,175