When retaining walls fail


    Advancements in design and construction of large retaining walls have transformed civil engineering practices in the last 30 years. New techniques and technologies, particularly for mechanically stabilized earth (MSE) walls, have allowed project teams to design walls that support more weight on steeper slopes and do so more cost effectively than ever before.

    Unfortunately, the risks associated with large retaining walls have grown faster than the new designs. In particular, some reports estimate that MSE walls fail at an alarmingly high rate — about 1 in every 1,000. This high failure rate has led to frequent litigation and has caused many design engineers to avoid projects with large retaining walls. If this trend continues, developers and other project owners will lose an economically valuable structural system.

    The predominant cause of failure might surprise some. In about 90 percent of the cases, the cause is a lack of communication among the project team, along with a common engineering nemesis: water.

    Although retaining walls may appear to be simple structures, their design and construction can be a complex process. MSE walls are even more complex to design, construct, and inspect, yet large retaining walls are more likely to be designed and built on a low-bid basis by a third-tier contractor. The current mindset in the industry is that retaining wall design and construction can be accomplished in a “cookbook” fashion. Plugging numbers into a software package and using off-the-shelf materials makes designing and constructing an effective retaining wall appear deceptively simple. It’s imperative that the industry treat large retaining walls as complex engineering structures and build in the inherent communication and collaboration between multiple disciplines that will drive long-lasting, reliable designs.

    An inside look
    Here’s how the typical MSE wall project works. The civil engineer decides a retaining wall is needed. The geotechnical engineer evaluates the soil or rock that will be below and behind the wall to determine geotechnical parameters for design and checks global stability. The structural engineer or wall design engineer completes detailed design, and the owner and/or architect gives input about the wall appearance.

    Meanwhile, the wall supplier provides MSE wall materials, the surveyor figures out where to build it, and the general contractor hires an earthwork contractor, who may hire a specialty contractor to construct it. Possibly, a ground improvement contractor may modify the soils below the retaining wall. Finally, the inspector and/or construction manager monitors the construction. Few other portions of a civil engineering project require the input of so many people.

    The entire process is a recipe for confusion and mistakes, which ultimately can spell failure. Take, for instance, the following case study that involved the design and construction of two large MSE walls to support a high school football stadium. The site was a hillside. To create the flat site for the field, the design incorporated one uphill MSE wall in the cut and another MSE wall downhill in the fill (simple cut-and-fill techniques). Both walls were taller than 30 feet.

    During its construction, the downhill wall experienced a global stability failure. It turns out that the wall design engineer mistakenly used the geotechnical engineer’s recommendations for the uphill wall (on bedrock) to design the downhill wall (on soft soils). In the wall design engineer’s defense, the architect did not distribute his shop drawings and design calculations; therefore, no one discovered the design errors until after the failure.

    The project team learned a valuable lesson on this project: Always have the geotechnical engineer review the wall design and have the inspector confirm that the review process has been completed before construction starts. The project manager also should make sure that a global stability analysis has been completed by the appropriate parties. Finally, all activities should be put in writing. If these steps in simple communication had been completed on this stadium project, the wall would not have failed. Most engineering and construction firms know these rules but forget to follow them.

    Use of sand instead of clean gravel as drainage fill caused hydrostatic pressure to develop during a heavy rain, which blew out the front face of this 16.5-foot-tall wall. In general, when a retaining wall fails, water should be considered guilty until proven innocent.

    Documenting change
    Similarly, lack of clear communication during the construction of a big box retail store caused problems with another retaining wall failure. In this case, the geotechnical engineer was tasked with designing a 16.5-foot-tall, 800-foot-long MSE wall. At the request of the contractor during a conference call and with the consent of the owner, the engineer allowed use of sand instead of clean gravel as drainage fill to save money. A few weeks after the wall was constructed, a heavy rain caused hydrostatic pressure along the wall, which blew out the front face of the wall.

    After the wall failure, the parties involved each had different recollections of the earlier conference call discussing the use of sand. The owner denied that it had approved the change because after the wall collapse, careful review of the manufacturer’s design guidelines for using sand as backfill had some inconsistencies. Even though the wall could be repaired, the owner considered it a failure because those inconsistencies could be interpreted to indicate the wall had not been properly designed. The problem could have been avoided had someone on the project team taken minutes during that conference call and distributed them. Those minutes would have established everyone’s participation in the decision and avoided the “blame game” afterward.

    When a cut-and-fill technique was used to create a flat site for a football stadium, the engineer mistakenly used recommendations for an uphill wall founded on bedrock to design this downhill wall on soft soils. The design errors were not discovered until after the global stability failure because the architect had not distributed shop drawings and design calculations.

    When parties communicate
    Of course, there are times when even close communication does not eliminate the chance of retaining wall failures. An excavation for a large water line for a power plant in South Florida was supported with internally braced sheet piles. At one portion of the excavation, the internal braces interfered with the construction of a large underground pump structure. The geotechnical engineer and contractor agreed to remove some of the internal bracing for a few days to simplify construction of this pump structure and help save money during construction.

    Unfortunately, a lot of rain fell during the course of those few days, and the surface runoff was directed toward the excavation. The sheet piles dammed up the water behind the wall, creating hydrostatic pressure that caused the toe to lose strength from the upward seepage gradient. The walls moved inward about 18 inches, but didn’t collapse. However, the pipes and pumps no longer fit in the excavated area. Correcting the problem would cost substantially more than the money that was saved by the contractor with the removal of the sheet piles.

    Lack of documented communication regarding design and construction changes led to a “blame game” among the project team and owner after this wall failed.

    However, because all parties in the design and construction process communicated throughout, there was no litigation or dispute. The contractor knew the risk, and the geotechnical engineer documented decisions and discussions in writing. More importantly, the project manager understood the risk of the decision to remove the bracing as well as the impact of the unexpected rain, and he encouraged everyone involved to learn from the mistake. The communication succeeded even though the wall failed.

    This project also is an excellent reminder of the power of water.

    When water flows
    About 90 percent of soil problems are really water problems. Most retaining wall failures occur during heavy rainstorms. Recently, a 43-foot-tall “big block” MSE wall failed during a heavy rain just eight months after it was built, launching 2,400-pound blocks more than 50 feet when the wall popped. This wall failed during a storm, even though it had survived earlier, heavier rains. It appears that the earlier storms caused the wall’s drains to clog, and therefore this subsequent storm contributed to the failure.

    Water can cause retaining wall failures in a variety of ways. As discussed above, drain lines might be clogged by silt from flowing water. Drain lines may be too small to handle the amount of water flowing through the system. The backfill material may not be sufficiently porous to allow flow of water. The porosity of the backfill material might vary, making tests unrepresentative. Soil samples may have been collected during a dry season, resulting in higher laboratory strength measurements.

    The amount of water flowing downhill toward the wall may be more than anticipated. The amount of water that can flow through the face of the MSE wall may be less than was assumed by the design engineer. The foundation soils or rock below the wall may soften as they are exposed to water over a period of time. Water or sewer lines behind the wall may leak. A stormwater detention basin might be constructed behind the wall or near its toe. The retaining wall may have been built over a natural spring. Drainage swales designed to divert water away from the back of a wall may clog. The wall’s drain lines may be not be installed at the correct locations. Designs may not have placed drain lines at appropriate locations. Drain lines might be crushed or cut by construction or landscaping equipment. Perhaps failure due to water is so prevalent because there are so many ways it can attack a wall. In general, when a retaining wall fails, water should be considered guilty until proven innocent.

    The project team can reduce the risk of these water-related problems by keeping the wall designer, geotechnical engineer, and civil engineer engaged with the contractors and each other throughout planning, design, and construction. They should continually ask: “Where could water come from? How much will there be? How is it going to drain? Where will it go? How will it impact the wall backfill, foundation soils, and retained soils? Are there any water or sewer lines nearby that could leak? How will this all change over time?”

    Budgets should allow for the engineer’s involvement during construction, including money for site visits.

    Rules to build by
    To reduce the risk of retaining wall failures, consider the following rules:

    • maintain good communication throughout the project team — a few e-mails are not sufficient;
    • document all communication in writing;
    • keep the engineers involved during construction, including site visits;
    • control the water; and
    • treat the retaining wall (particularly an MSE wall) as a complex engineering structure.

    If a retaining wall on your project does fail, get to the job site as soon as you can so you see, hear, and participate in everything; work with the project team (engineers, contractors, and owners) to settle the issues before litigation erupts; and understand that “fault” or “blame” is almost never black and white. The situation is almost always “gray” with uncertainties and misunderstandings.

    What is failure?
    A retaining wall does not have to collapse to fail. In fact, a failure is perhaps better defined as when the wall does not perform as expected. This definition is complicated by the fact that different people on the design team may have different expectations. It may be acceptable to you, but not to the owner. Make sure your expectations match those of the person paying for the wall.The following questions could show how those expectations differ:
    • What if the wall is perfectly stable, but continually has water seeping out of its face?
    • What if the wall appears to be leaning?
    • What if it doesn’t look as attractive as expected?
    • What if the surveyor places it a few feet from the planned location?
    • What if it has small cracks in it?
    • What if you can’t plant a big tree at the top of the wall?

    Steve Wendland, P.E., R.G., is a senior principal geotechnical engineer at Kleinfelder. He has 23 years of geotechnical engineering experience focusing on deep foundations, design of large retaining walls, and subsurface exploration. He can be contacted at 913-647-5018 or swendland@kleinfelder.com.