History, performance, and design of geotextiles in levees: A report from New Orleans.
By Jody Dendurent and Mark Woodward
Geotextiles are a key element in building levees that will survive catastrophic storms. The geotextile-reinforced earthen levees in New Orleans performed remarkably well during Hurricane Katrina in August 2005. With the current focus on updating its levee system to protect New Orleans from a 100-year storm event, the U.S. Army Corps of Engineers (USACE) relies heavily on geotextiles in its designs.
In addition to reinforcing levees and allowing existing levees to be built taller and more robust, geotextiles help reduce construction costs and reduce the size of — or even eliminate — stability berms. Geotextiles are also used in innovative ways for recovery and immediate repair of damaged levees.
This article focuses on the history, performance, and future use of geotextiles in levees. It will include a review of literature and design methods, an examination of the performance of the geotextile-reinforced levees that survived Hurricane Katrina, and USACE design improvements.
Most of the engineering focus in the aftermath of Hurricane Katrina has been on the failed levees, the causes of failure, and how to improve levee design so similar calamities are prevented in the future.
Little attention from the media, technical investigators, and the general public has been focused to the levees that successfully withstood the similar storm surges and conditions as the levees that failed. For example, 9mi. (14.5km) of the St. Charles Levee and 7mi. (11.3km) of the Jefferson Parish Lakefront Levee on the south shore of Lake Pontchartrain are reinforced with geosynthetics (Figure 1). In a press release from the Industrial Fabrics Association International (IFAI) in April 2008, the USACE-New Orleans District (NOD) stated that “both the St. Charles and Jefferson levees were loaded (filled by the storm) during Katrina and performed exceptionally. They were stable and the geosynthetic was inherent to their strength” (Aho, 2008a).
USACE engineers have been among the pioneers in levee design who saw the merits of using high-strength geotextiles to improve the stability of levees and in the process save billions of dollars in construction costs, thousands of acres of land, and years of construction time. USACE has continuously been at the forefront of innovation in the design of geosynthetic-reinforced embankments and levees on soft soils. As the USACE works to update the New Orleans levee system to withstand future storm events, it will continue using design methodologies that include geotextiles; and the USACE also continues its partnering with industry leaders to improve public safety and to refine the design of reinforced levees.
2. History of geotextiles in levees
In the 1980s, the geosynthetics industry was small but established. Forward-thinking design engineers were looking at conventional designs in many areas of civil engineering and starting to see where geosynthetics could be used to improve public safety, project lifecycles, constructability, feasibility, and construction costs.
USACE engineers were among the first to use geosynthetics in a soft-soil embankment project. They designed and built a 26ft- (8m)-high embankment at the USACE’s dredged disposal site at Pinto Pass in Mobile Harbor, Ala., in 1980 (Holtz, 2004). The foundation soils under the embankment had cohesions ranging from 50psf (2.4kPa) to 150psf (7.2kPa) (Fowler, 1981). As per Holtz (2004), this is an important case history because the USACE documented and verified its design assumptions and procedures. It also emphasized that proper construction methods are absolutely crucial for successful construction of embankments on soft soils.
The USACE-NOD is credited with working with the geosynthetics industry to develop the high-strength, woven, polyester geotextiles that USACE now uses routinely to reinforce hurricane levees. These geotextiles were first proposed for use on the New Orleans-to-Venice Hurricane Protection Project, which “was stalemated because they could not raise the levee by conventional methods. Raising would have made the levee fail into a drainage canal and they would have been forced to ruin a whole lot of wetlands” (Hall, 2003).
Prior to beginning work on a 13mi. (21km) of geotextile-reinforced hurricane levee (see 2.1.1), the USACE-NOD constructed the first full-scale test section of a geotextile-reinforced hurricane levee to test the performance of the proposed design. The test section performed better than expected (Duarte et al., 1989) and the levee has performed well to date.
Since this first test section, USACE-NOD has been constructing and monitoring levee test sections and stretches of geotextile-reinforced levees, employing what it has learned to improve design methodology and construction techniques.
2.1 USACE-NOD test sections
To better understand and improve the performance of geosynthetic-reinforced levees, USACE-NOD designed and monitored four major levee test sections, three of which are examined in this article.
2.1.1 ‘Reach A’ test section
As mentioned above, a geotextile-reinforced levee test section constructed in 1986 between the towns of Nairn and Empire in lower Plaquemines Parish in far southeastern Louisiana was a prototype for a proposed 13-mi. (21-km) enlargement of an existing levee between St. Jude and Tropical Bend (Figure 2). This particular levee segment — part of the New Orleans-to-Venice Hurricane Protection Project — is typically referred to as “Reach A” (see Figures 2 and 3).
In this case, the geotextile-reinforced alternative was considered because raising the levee by conventional methods, to the designed storm height, would have required a large footprint and relocating the levee 120ft (46.6m) toward the Gulf side and into a marsh (Bakeer et al., 1988).
The geotextile-reinforced design showed that such a levee could be built by degrading and raising the levee on its existing alignment, thus saving time, money, and land (Duarte et al., 1989).
Prior to this test section, it was well-known that geotextiles could reinforce a levee embankment, reduce its footprint, control deformation, and increase its stability. However, the actual design methods for geotextile-reinforced embankments still consisted mostly of conventional concepts of earth pressure and slope stability, with minor modifications for the effect of the geosynthetic materials. They also had not been sufficiently field verified to give USACE-NOD the confidence needed to build a large, reinforced, hurricane-protection levee (Bakeer et al., 1988).
The potential benefits of building and monitoring the performance of a geotextile-reinforced embankment were significant enough to justify the expense of such a test section. The Reach A test section (Figure 3) was successfully instrumented with inclinometers, settlement plates, piezometers, and foil strain gauges. The instrumentation monitoring continued for two years after construction began and verified the assumption that this geotextile-reinforced levee design was feasible, safe, and economical (Duarte et al., 1989).
The economic benefits of the geotextile-reinforced design on the 13mi. (21km) of raised levee were (Bakeer et al., 1988):
- a 35% savings on the overall cost compared to the original design ($30.8 million savings).
- reduction in construction time from 13 to 6 years (resulting in significant insurance savings to residents).
- a 97% reduction in marshland used for the levee — original estimates of 4,000 acres (1,619 ha) were reduced to 100 acres (41 ha).
- a 60% reduction in required construction materials.
The contributions of the test section to ongoing reinforced soft-soil embankment design research were:
- Measured (mobilized) strains in the geotextile were less than half of the design strains (Bakeer et al., 1988).
- Reinforced levee design resulted in a smaller cross section, which reduced the destabilizing (driving) forces and reduced vertical settlements (Bakeer et al., 1988).
- Observed maximum stresses as captured by the field instrumentation were not in the same location as the postulated failure plane during design (Duarte et al., 1989).
- Well-thought-out instrumentation installation essential for capturing and presenting meaningful (reliable and realistic) displacement and deformation profiles (Duarte et al., 1989).
- Provided data that helped refine and calibrate finite element analysis of geotextile-reinforced embankments (Bakeer et al., 1988).
- Provided data used in the development of new guidelines for future reinforced-embankment design.
- The settlement and deformation pattern at the fabric level followed that of the generation and dissipation of the observed pore water pressures (Bakeer et al., 1988).
After USACE-NOD completed the Reach A test section, Plaquemines Parish officials reported that, “The Plaquemines Parish Government wholeheartedly supports the U.S. Army Corps of Engineers in the use of the geotextile fabric to bring the New Orleans-to-Venice Hurricane Protection Levee up to grade and feels that the cost and time saved to complete this portion of the project (Reach A) was excellent” (Petrovich, 1987).
2.1.2 Bonnet Carre test section (Chiu et al., 1988)
The Bonnet Carre Spillway test section was constructed in 1988-89 in St. Charles Parish prior to the construction of 7mi. (11.3km) of the Jefferson Parish Lakefront Levee.
The goal of this test section was to verify USACE’s newly revised geotextile-reinforced levee design procedures. Similar to the Reach A levee, weak foundation soils and right-of-way concerns made the cost of constructing an unreinforced levee prohibitive. The cost savings already realized in Reach A and the potential cost savings in the Jefferson Parish Lakefront, St. Charles Parish, and West Bank levee projects provided justification for further full-scale research.
The Bonnet Carre test section consisted of an all-earthen, unreinforced section of levee (UI), a reinforced levee section with one layer of geotextile (RI), and a reinforced levee section with two layers of geotextile (RII), all built to a typical earthen levee height of 19ft (5.6m) National Geodectic Vertical Datum (NGVD). Full-scale field pullout tests were also conducted at the test site using the borrow fill that would be used to construct the St. Charles and Jefferson Parish Lakefront levees.
USACE-NOD wanted to push all three test sections to failure, which it tried to do by excavating immediately adjacent to the levee toe. Table 1 summarizes the design geotextile strengths and the failures of the three sections.
Chiu et al. (1988) observed and Napolitano (1994) stated that the “analyses indicate that these two levees (RI and RII) may have experienced a bearing capacity failure, or excessive lateral movement, and not the conventional rotational shear failure.” This is important because it reinforced that all failure modes should be analyzed carefully.
The economic savings and benefits realized from the Bonnet Carre test section for the mainline St. Charles Parish and Jefferson Parish Lakefront levees were similar to the benefits realized on Reach A.
The contributions of the Bonnet Carre test section to ongoing reinforced soft-soil embankment design research were:
- Two layers of reinforcement appear to be a more efficient “reinforcing pattern” than one layer.
- Levees with two layers of geotextile can be easily and effectively repaired.
- Often, the assumed failure mode may not be the most critical one.
- Failures of reinforced levees involve less-significant consequences than failures of unreinforced levees under similar conditions.
- Particular attention must be paid to the stress-strain characteristics of the soil and the geotextile to ensure that the two materials are compatible, as large movements in the soil may cause failures at correspondingly small strains in the geotextile.
2.1.3 Westminster east-west test section(Varuso et al., 2005)
The Westminster east-west test section was constructed in 2004 south of the Mississippi River between the towns of Westwego and Harvey in Jefferson Parish, to help USACE-NOD determine how to efficiently utilize geosynthetic reinforcement in earthen levee embankment design and construction.
This test section was part of a levee project approximately 1mi. (1.61km) long. An unreinforced section for this levee was analyzed in design but was determined to be expensive and difficult to construct.
USACE-NOD wanted to use monitoring data from this test section to derive a new design methodology that would account for the anticipated gains in shear strength of the foundation soil due to consolidation during and immediately following construction. USACE-NOD had observed in the past that consolidation, and thus subsequent shear strength gains, were more uniform underneath reinforced test sections. This shear strength gain was attributed to the rapid consolidation of upper strata resulting in less tensile force being transferred to the geotextile. USACE-NOD wanted to verify all its design assumptions using the field instrumentation data.
The test section had the same cross section as the mainline reinforced levee but used reinforcement with a 5% strain strength of 5,822lbs/ft (85kN/m) instead of the 11,644lbs/ft (170kN/m) reinforcement that was used in the mainline levee. Instrumentation for this test section was designed to provide data needed to develop a design methodology that would result in optimizing the use of the geotextile reinforcement’s tensile strength. Soil samples were also taken six months after construction to further analyze gain in shear strength.
The economic savings and benefits realized from this test section were similar to previous geotextile-reinforced levees.
Its main impacts to ongoing reinforced soft-soil embankment design were:
- quantification of the magnitude of shear strength increase in the foundation of this test section and formulation of a method to account for it in design.
- verification that the increase in cohesive shear strength resulted in an increased factor of safety from 1.0 to greater than 2.0.
- verification that second lift construction costs could be reduced by up to 75% if the geotextile reinforcement is initially designed to support the loading conditions.
2.2 Design guidance
As in other areas of civil engineering, design concepts using geosynthetics are still not included as standard design topics in many foundation and soft-soil embankment textbooks or design manuals. Likewise, the USACE-issued engineering manual, EM 1110-2-1913 Design and Construction of Levees (2000), does not include any discussion on designing or constructing levees with geosynthetics.
As with other agencies, such as the Federal Highway Administration (FHWA), USACE references a separate design manual, the Unified Facilities Critieria (UFC) Engineering Use of Geotextiles (USACE et al., 2004). The UFC Engineering Use of Geotextiles manual (2004) has one chapter dedicated solely to the design of geotextile-reinforced embankments on soft foundations, following the basic design methodology of Holtz et al., (1997) and other industry-accepted, geotextile-reinforced embankment design procedures. The following topics are addressed in the UFC Engineering Use of Geotextiles manual (USACE et al., 2004):
- overall bearing capacity
- slope stability
- sliding wedge analysis for embankment spreading/splitting
- analysis to limit geotextile deformation
- determination of geotextile strength parallel to the centerline of the levee
- analysis of embankment settlements due to primary consolidation and plastic flow
The UFC Engineering Use of Geotextiles manual (2004) is currently under revision and is scheduled to be published by 2010. It is the authors’ understanding that the basic design methodology on geotextile-reinforced embankment on soft soil foundations has not been significantly modified, but it has been updated to reflect more current constructability criteria.
USACE-NOD currently uses a design procedure for reinforced levees similar to design methodologies of the UFC Engineering Use of Geotextiles manual (2004) with updates based on technology advances, improvements to industry-accepted design procedures, experience accumulated from test section monitoring data analyses, and advances in construction methods.
The most significant design development that USACE-NOD has implemented is incorporating (or quantifying) the foundation shear strength gain during construction of a new geotextile-reinforced levee. This development is not yet included in any of its technical manuals. This reality will increase the calculated stability factor of safety of a levee (Varuso et al., 2005) and will reduce the required geotextile tensile strength, the size of the levee stability berms, and right-of-way requirements.
Other design procedures that USACE-NOD has incorporated into its levee designs as a result of lessons learned from Katrina and the intense reviews that followed, are:
- Designing the geosynthetic reinforcement to support subsequent lifts that sometimes have to be constructed to reach long-term elevations. This has significantly reduced the cost, construction time, and settlement of the subsequent lifts.
- Performing the design procedures using a number of different analytical methods to determine the possible failure modes of a reinforced levee and the needed geotextile modulus. This has increased the confidence in newly designed geotextile-reinforced levees by ensuring that possible failure modes are recognized and the geotextile is designed to withstand loading associated with critical failure planes.
Jody L. Dendurent, P.E., TenCate Geosynthetics, Wichita, Kansas and Mark L. Woodward, P.E., New Orleans District, U.S. Army Corps of Engineers.
Aho, A. (2008a), Levees reinforced with geosynthetics perform exceptionally well, Industrial Fabrics Association International, News Release.
Aho, A (2008b), Language specifying use of geosynthetics submitted for water resources development act 2008, Industrial Fabrics Association, News Release
Bakeer, R.M., Hadj-Hamou, T.A., Duarte, F.M., Satterlee, G.S. (1988), Field test of a geotextile-reinforced levee, Journal of Geotechnical Engineering, 26: 90-101.
Chiu, S.K., Napolitano, P.J., Duarte, F.M., (1988), Bonnet Carre spillway geotextile-reinforced test section and pullout tests report, United States Army Corps of Engineers-New Orleans District, 1-16.
Duarte, F.M., Satterlee, G.S. (1989), Case Study of a Geotextile Reinforced Levee on a Soft Clay Foundation, Geosynthetics ‘89 Conference, San Diego, USA: 160-171.
Fowler, J. (1981), Design, Construction and Analysis of Fabric-Reinforced Embankment Test Section at Pinto Pass, Mobile, Alabama, Technical Report EL-81-7, U.S. Army Engineer District, Mobile Alabama and Office, Chief of Engineers, U.S. Army, Washington D.C.
Haliburton, T.A., Fowler, J., Langan, J.P. (1980), Design and Construction of Fabric-Reinforced Embankment Test Section at Pinto Pass, Mobile, Ala., Transportation Research Board, Transportation Research Record 749: 27-34.
Hall, J. (2003), Satterlee: Corps must rise to challenges, Riverside magazine, United States Army Corps of Engineers (USACE), New Orleans District (NOD), January-February 2003: 3, 7.
Holtz, R.D. (2004), Geosynthetics R&D-The “Early” Days (1960s to circa 1985), Proceedings of the Koerner Research Symposium, Drexel University, Philadelphia, Pa.
Holtz, R.D., Christopher, B.R., and Berg R.R. (1997), Geosynthetic Engineering, BiTech Publishers Ltd., Richmond, British Columbia, Canada.
Mylleville, B.L.J. and Rowe, R.K. (1991), On the design of reinforced embankments on soft brittle clays, Geotechnical Fabrics Report, May/June 12-24.
Napolitano, P.J. (1994) Levees Constructed on Soft Soils Using High Strength Geotextiles, Proceedings of REMR Workshop on Levee Rehabilitation, USACE, 105-129.
Petrovich, L.A. (1987), Letter to the United States Army Corps of Engineers, July 1987.
Pinner, R.P. (1993), St. Charles parish geotextile-reinforced levee instrumented section, USACE New Orleans District, 2-14.
Nicolon Mirafi Group (1989), “Reach A” Hurricane Protection Levee, case study.
Sprague, C.J. and Koutsourais, M. (1993), The Evolution of Geotextile Reinforced Embankments, Proceeding from Grouting, Soil Improvement and Geosynthetics, ASCE Geotechnical Specialty Conference, New Orleans, La. 1129-1141.
USACE, Naval Facilities Engineering Command, Air Force Civil Engineer Support Agency (2004), United Facilities Criteria (UFC) 3-220-08FA Engineering Use of Geotextiles.
USACE (1989), Design Memorandum No. 18 St. Charles Parish North of Airline Highway, Serial No. 23, Volume I.
USACE (1987a), Design Memorandum No. 1 Reach A City Price to Tropical Bend.
USACE (1987b), Design Memorandum No. 17 Jefferson Parish Lakefront Levee, Volume I.
USACE (2000), Design and Construction of Levees, EM 110-2-1913.
Varuso, R.J., Grieshaber, J.B., Nataraj, M.S. (2005), Geosynthetic reinforced levee test section on soft normally consolidated clays, Geotextiles and Geomembranes 23, Elsevier Ltd.: 362-383.