CHAPTER 5

Summary of Findings

The objective of this synthesis was to document the experience and strategies of state DOTs in detecting, locating, and controlling water infiltration in highway tunnels across the United States. The synthesis report includes responses from states with tunnel inventories in the National Tunnel Inventory (NTI), but also includes deck-over structures with leakage issues, which may or may not be included in the tunnel inventories of the responding states. The scope of the synthesis included methods to detect and locate leaks, as well as remediation methods and acceptance criteria.

The synthesis began with a literature review to document the types of leaks encountered in highway tunnels and experience in mitigating tunnel infiltration. The literature review included sources from prior research, journal publications, conference proceedings, and documented case histories of leakage repairs. A key piece of information gained from the literature review was the design strategy of a tunnel: whether it was designed to allow water to drain into the tunnel (drained or open) or to prevent water from infiltrating (undrained or closed). This information is important when considering leak remediation methods. For drained structures not designed to resist hydrostatic forces, a means of collecting the infiltration and distributing it to the tunnel drainage system is compatible with its design strategy. Unlined rock tunnels and many mined tunnels in rock with concrete linings are designed as drained structures.

Leak mitigation measures may be used from the inside of the tunnel (negative side) or the outside (positive side). Positive side measures are difficult to implement because they involve excavating and exposing the original structure. Negative side measures are prevalent and include redirecting water, grouting cracks and joints, replacing joints, and applying cementitious coatings. The literature indicated that use of chemical grouts is most effective in stopping leaks and that selection of appropriate grout for a given condition is crucial to the success of the remediation.

A survey was developed to determine the state of the practice in identifying and controlling water leaks in highway tunnels. The survey addressed the prevalence of leaks and problems resulting from them, the methods of detecting and identifying the source of leaks, tunnel types and leak locations, mitigation methods and their effectiveness, and acceptance criteria for leak remediation projects. The survey was distributed to all 50 states and the District of Columbia and Puerto Rico. The findings are presented in Chapter 3 and are based on 33 respondents, including 26 with tunnel inventories. The complete survey is presented in Appendix A and individual state DOT responses are aggregated in Appendix B. Figure 1 shows all state DOTs that responded to the survey and those with tunnel inventories that provided input.

The results of the survey include the experience of 26 state DOTs representing an inventory of 206 of the 552 tunnels in the National Tunnel Inventory (NTI). The responding DOTs indicated that 121 of these 206 tunnels (59%) leak. The most common problems reported by state DOTs because of infiltration are structural deterioration (14 out of 26) and icicles (14 out of 26). To find

the leaks and identify their sources, DOTs rely predominantly on visual inspections, although six DOTs indicated they have used NDT tools (e.g., LiDAR scans, thermography, photogrammetry, and GPR) to locate the sources of water.

Twenty-three DOTs provided specific information about 57 tunnels and characterized the leakage within the tunnels. Table 3 summarizes this information by tunnel type. Data are provided for 21 mined tunnels in rock, 12 unlined rock tunnels, 15 cut-and-cover tunnels, 1 immersed tube tunnel, and 8 shield-driven, oval or horseshoe tunnels in mixed or soft ground. The source of water most frequently noted is groundwater and surface runoff; for cut-and-cover tunnels in urban areas, utilities, ponds, or even irrigation systems are sometimes the source of water infiltration. Joints (27 out of 57) and cracks (23 out of 57) account for the locations contributing the heaviest leakage for the 57 tunnels, but penetrations in the liner (3 out of 57) and transitions (3 out of 57) were also noted.

The leading remediation methods used by responding state DOTs are catchment systems/redirecting drainage, which have been used most often and were reported as the most effective of the remediations (63%). Some agencies noted the amount of water precluded the use of any other remediation. DOTs selecting the redirecting drainage option indicated that other remediations were not effective for a long period. The second highest response rate for remediations was replacing joint material and repairing joints, but here the effectiveness was only 30%. Injecting cracks with chemical grouts was also listed as a remediation used often by DOTs and its long-term effectiveness ranked second at 44%. Even these three most-used remediations were still only rated “effective after 10 years” 47% of the time. One explanation for this low rate of effectiveness may be related to the DOT personnel completing the surveys and staff turnover in recent years. Newer staff would not know what mitigations were implemented 10 years ago, and recent remediation work is not yet 10 years old and might not have been reported as effective. One agency provided acceptance criteria for new tunnel construction. Those criteria are outlined in Chapter 4.

The case examples were selected to obtain a cross section of different tunnel types, conditions, and DOT experience. Staff at four DOTs (i.e., Arizona, Colorado, Pennsylvania, and Washington State) were interviewed.

ADOT reported on infiltration through the deck at a structure constructed as a cut-and-cover tunnel with a public park on top of it. The tunnel roof structure consists of multiple post-tensioned box-girder units placed side by side. ADOT noted that mitigation from within the tunnel is complicated by the overall depth of the box structure and the post-tensioning. ADOT has had some success by excavating and repairing waterproofing from outside the tunnel, particularly over support spaces. This type of deck-over structure is common in states across the United States and ADOT’s experience will be of interest to many DOTs with similar structures.

CDOT observed that leakage is worse in tunnels with liners constructed of pneumatically applied concrete. Issues are predominantly at joints in liners, and CDOT staff have implemented chemical grouting and installed drainage structures, neither which have been effective in the long term. CDOT staff noted frustration that the leaks are seasonal and often do not exist when the NTIS inspections occur during summer weather.

The perceived effectiveness of the NDT methods varied. Although 73% of DOTs have not used NDT for tunnel leakage detection, most of those who have used NDT stated that it was either very effective or somewhat effective in detecting leakage issues. CDOT participated in the SHRP2 research sponsored by FHWA, and CDOT used NDT to identify leaks within tunnels. Both thermal imagery and LiDAR scans were used during the study with seemingly good results, and CDOT endorsed the use of these technologies. The results were never compared to actual field inspection data to verify the locations of water leaks.

PennDOT described significant water remediation projects for a toll agency in the state. These remediation projects use umbrella type catchment systems to collect infiltration and carry it to the tunnel drainage system.

In Washington State, icicle formation in support spaces and egress corridors was cited as a significant problem resulting from infiltration. WSDOT also described mineral formations in the drainage system that require regular cleaning. WSDOT has state-employed maintenance crews with experience stopping leaks in their large pontoon bridge structures which have zero tolerance for water intrusion. WSDOT discussed grout products they have used successfully for grouting cracks and joints.

The top challenge reported by responding state DOTs is the effective remediation of leaks. Of the 26 respondents, 23 commented that the mitigation was not effective in the long term. DOTs with tunnels in urban areas also noted that locating the point of infiltration and the source of the water can be challenging.

This synthesis provided an opportunity to identify areas for further research. The literature search found some examples of water remediation projects monitored over a multi-year timeframe and discussion of lessons learned, but, in general, there are information gaps about the methods and materials used for water remediation and their effectiveness over time. With advancement of technical staff (responsible for administering remediation projects) within DOTs to new positions, or key staff turnover in general, the history of what has been tried and what has worked is lost. Future research that can track the effectiveness and lessons learned of remediation approaches over time would be valuable information for DOTs planning water infiltration remediation. Information on specific materials, including their effectiveness and limitations on use, would be a valuable resource for tunnel owners. This information could be sorted by tunnel type, substrate, drained/undrained design, and other such factors that would allow DOTs to compare their tunnel conditions with those used in the documented remediations.