How Can You Tell Whether a Road is Safe After a Flood?
Hurricane Helene brought severe flooding to much of western North Carolina, submerging many roadways – and washing some away altogether. How can engineers tell whether a road is safe after the floodwaters recede? How do you test the safety of dams and bridges? And what do engineers have to consider when repairing or rebuilding transportation infrastructure?
To learn more, we spoke with three experts from NC State University. Shane Underwood and Brina Montoya are professors in NC State’s Department of Civil, Construction, and Environmental Engineering; Mo Gabr is a Distinguished Professor of Civil Engineering and Construction in the same department.
The Abstract: How do you test a road after a flood to see if the underlying foundation is safe? How long does that take?
Shane Underwood: We would check structural capacity using a falling weight deflectometer (FWD). Imagine a hammer striking the surface while we measure how much the pavement deflects, meaning how much it deforms and then bounces back. If a pavement deflects too much then we could see that the structural adequacy is pretty low. There are no hard and fast numbers for what is acceptable as a number of factors affect judgments about structural capacity, such as how thick the pavement is and the temperature.
In more general terms a road’s structural safety is the subjective assessment of the resident/division/maintenance engineer. They are going to be looking for signs of erosion on the roadside, perhaps some visual observations of what happens when they drive a truck across the pavement (do they see water squeeze out from the soil or aggregate at the pavement edge, visible signs of pavement movement, etc.), and a healthy dose of their own judgment. They may also consider the age of the pavement or the presence of a pipe/culvert. A visual inspection will vary by the complexity of the site but could take anywhere from five minutes to perhaps several hours and involve multiple people.
TA: How frequently do the intervals need to be, in terms of distance? For example, do you need to test every 20 feet? Every 100 feet? Every mile?
Underwood: In North Carolina, FWD analysis would typically be done about every 500 feet for a long section and every 200 feet for a short section (less than a mile). This is at the discretion of the engineer. If the engineer feels shorter spacing is relevant – perhaps because they see something of concern or know something about the site and the underlying engineering (for example, a fill versus cut section) – then that will factor into their decision.
TA: Is the testing the same for opening a roadway in a limited capacity (in the short term) as compared to fully opening a roadway for long-term, regular use?
Underwood: Probably not. The triage phase is probably going to be dominated by visual inspection. The FWD analysis might come later, when assessing the potential long-term impacts of a flooding event. In the end, if a flood and perhaps heavy recovery traffic cause more damage to a pavement because the foundation is saturated, but it means faster recovery for the people in an area, then the calculus probably works in the favor of opening the roadway. Humanitarian concerns will take precedence over engineering concerns.
TA: How long does it take to remove and replace a damaged roadway if the underlying foundation is stable? And is it different for a short-term “fix” as opposed to opening a roadway for regular, long-term use?
Underwood: This depends on many factors, including the extent of the damage and how ‘unstable’ the foundation really is. It also depends on how complex the site is.
If the pavement sits on a large embankment and the embankment itself has failed, then it could take a very long time because that embankment has to be repaired and it is possible it may be deeply compromised. It is also possible that the embankment itself needs to be updated to current design standards. With some large facilities there can also be environmental regulations which need to be dealt with, particularly if the site had been grandfathered into newer environmental regulations and the site is located near an environmentally sensitive area. Since many times there are waterways (creeks, rivers, streams, etc.) involved – and those waterways are the source of the erosion – these issues may creep up. For any number of related reasons a large repair can take a year or more to fix. The AZ89 landslide in northern Arizona took two years. In the case of the I-40 repair at the Tennessee border, the NC Department of Transportation secretary has said that it may take months at best to repair.
However, if it is a simple site without excessive damage in the surrounding areas then a roadway can be repaired in just a few days. If the site requires replacing pipes then it can take a month or more depending on the complexity.
Brina Montoya: The ability to repair a roadway quickly may also be affected by the availability of supplies, such as aggregate and pipes. With so many roads needing repair in western North Carolina right now, we could see some delays related to accessing enough materials.
Underwood: Building on Brina’s comment, there can also be issues related to the availability of personnel to handle the repairs. Personnel may need to come from other parts of the state and their repair practices can be different. Sometimes this is good and brings new, better ideas to an area. It could also slow down the repairs relative to what could be done with personnel who have familiarity with local conditions.
How long it takes to make short-term fixes will also depend on a number of variables. If it is a bit of shoulder erosion, then a crew with a pickup and a shovel will fix multiple sites in a day. Substantial shoulder erosion will require heavy machinery to unload material and prepare the site. This could be a day job up to a week job or so if necessary. The variables here are significant.
TA: What about if a stretch of road has been completely destroyed or washed away? In other words, if the road collapsed, what do you need to consider when determining whether to rebuild in the same spot versus identifying a new path for the roadway?
Underwood: I think an agency will most likely build back in the same spot. There are many legal hurdles that have to be overcome to move the roadway – land to acquire, environmental regulations, NEPA [the National Environmental Policy Act] to navigate, etc. It is possible that a short-term/medium-term fix would be put into place and if the site had been damaged multiple times the division might look at programming a long-term plan (using the state’s long-term planning procedure) to find a more meaningful solution.
It is possible, I suppose, that where a plan was already in place to move a roadway and it was damaged, it could be accelerated, but the situation would have to be pretty special (maybe the land was already acquired and preparations were underway and the damage was such that travelers had reasonable alternatives, etc.).
Montoya: Even if the same path is used to rebuild, the infrastructure design may change. After Hurricane Florence [in 2018], culverts often became much larger in size, and a few times were replaced by bridges. These changes will make future flooding events at the roadway less likely.
TA: What do you folks need to consider when identifying a new pathway for the road?
Underwood: I think what I said above gives some idea. Typically, where I have seen roadway alignments change it is because of safety issues (think about some of the rural offset intersections that used to be popular, but may become a less than ideal solution once traffic volumes became high), congestion issues associated with the alignment or inability to expand as needed with the current alignment (i.e., right of way only permits two travel lanes, but the traffic volume now requires four).
I don’t know of a specific example where flooding or storm related damage was a strong motivator. Perhaps because in many places any flooding would be widespread enough that a reasonable realignment might not actually help that much. I’m sure that there are some of these out there, but again, it would be a unique situation in my opinion. Realignment in general would have to consider the availability of the land, the topography of the land and surrounding drainage issues, etc.
Mo Gabr: Another major consideration is the geology of the region, particularly in areas with shallow rock depth such as in regions like the Appalachian Mountains in the western part of the state.
For example, rock outcrops require a significant level of blasting and excavation for establishing the road alignment. Additionally, if the area is known to have karst formations, as seen in parts of eastern North Carolina, this can lead to the development of sinkholes, which can undermine roads, as has occurred recently on I-40 East near Rocky Point.
TA: What sort of things do engineers need to consider when repairing a damaged bridge? Does it depend on whether you’re doing a temporary assessment and repair, versus building for the long term?
Montoya: A bridge consists of two primary components: the superstructure, which includes the bridge deck and piers; and the substructure, which encompasses the foundation support system. For bridges that cross waterways, particularly after a severe storm, it is essential to evaluate their safety, especially if they serve as lifelines during and immediately after the storm. One critical aspect of this evaluation is assessing soil erosion and scour, which can compromise the stability of the foundation system that supports the bridge superstructure.
During a storm, the riverbed undergoes hydrodynamic changes, including sediment erosion, scour around the foundation support system (such as pilings), and the deposition of sediment into newly formed scour holes. Excessive scouring around the bridge piers or abutments, caused by an increase in water flow and velocity, can undermine the foundation support. This can lead to severe structural damage or even collapse of the superstructure. For this reason, it is imperative to assess the stability of bridges used as lifelines after severe storms to ensure they remain safe for use to serve first responders and recovery efforts.
Such assessments must determine whether scour has occurred around the foundation support system and the extent of such scour on the capacity of the foundation support system. A visual inspection, with the use of underwater cameras for example, is often insufficient. As mentioned earlier, storm-induced bed sediment movement includes both erosion and deposition. This deposition process may fill scour holes with freshly deposited sediment. This freshly deposited sediment may make the bridge appear stable, but it does not provide adequate vertical and lateral support to the foundation system. Unlike sediments established over geologic time spans, these freshly deposited materials are not compacted enough to support the bridge. As a result, even if no significant visible scour is detected, the bridge may still be structurally unsound and unfit to serve as a lifeline.
Proper post-storm assessments should involve more advanced techniques, such as sonar or geotechnical testing, to evaluate the integrity of the underlying support system. Only a thorough investigation can ensure the safety of the bridge, particularly when it is critical to the recovery and relief efforts following a storm.
TA: There were quite a few news stories related to fears that dams would fail. How do authorities test the stability of dams in the wake of flooding? And if a dam has been damaged, how does one repair it?
Gabr: In the wake of extreme flooding, such as that induced by Hurricane Helene, the stability of dams must be rigorously evaluated to assess both short-term and long-term safety; this is especially important for high hazards dams for which loss of life may occur if the dam fails. These investigations typically occur in two phases.
The first phase involves performing inspections to assess structural integrity of the dam and its foundation soils, spillway function, the functionality of any installed instrumentation, and the condition of emergency equipment.
In this phase, a team of qualified dam inspectors and engineers look for evidence of cracks or fissures, especially near the foundation and spillway, checks for erosion of the dam’s slopes, and examines the dam’s foundation for scour or the migration of soil particles downstream caused by the high flows during the storm. Additionally, the functionality of monitoring instruments, which provide data on sudden changes in water pressure or dam component deformation, is assessed. Emergency equipment, such as pumps, and emergency response plans should also be reviewed and updated.
If the first phase of inspection reveals signs of distress in the dam body and/or in its foundation support system, the second phase includes a geotechnical investigation. This second phase may involve borehole drilling, soil sampling, testing and analyses. The sampling and testing results, along with engineering analyses, are used to identify potential issues including, for example, weak seams or impending embankment slope landslides, and seepage levels under high water, as the flow of water through and under the dam is a key concern. In addition, the data are used to evaluate erosion or piping (the washing out of fine soil particles), which can also lead to dam failure.
If testing and analyses show that the dam has been damaged, depending on the severity of the damage, various remedial measures can be implemented. The repair process focuses on stabilizing the dam body and foundation and controlling through-seepage and underseepage. Through-seepage is when water is seeping through the actual structure of the dam, which can be a sign of structural problems; underseepage is when water seeps out from under the dam’s structure, which can be a sign of problems with the dam’s foundation. The appropriate remedial measures will depend on the type and extent of the damage. The measures may include, for example, compacted clay or concrete cutoffs (subsurface trenches filled with concrete or clay), well-placed drainage systems to relieve excess water pressure, installing compaction or permeation grout to stabilize the foundation soil, installing upstream impervious blankets, or building downstream berms to stabilize the structure.
Once necessary repairs or rehabilitation measures are completed, a long-term monitoring program using sensors and instrumentation is implemented to track the dam’s performance with the installed remedial measures. Emergency response plans should also be updated to address any new risks or vulnerabilities.
