Minimizing Roadway Embankment Damage from Flooding (2016)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

60 INTRODUCTION This chapter presents a summary of the results received from the DOT engineers. Of the 50 total responses that were received, 32 made major contributions to this study. Addi- tionally, two answers were retrieved from FHWA Colorado and FHWA Washington. In addition to the survey responses, many telephone interviews were conducted. Figure 102 shows the state DOTs that made major contributions to the synthesis. The survey questionnaire is attached in Appendix A. The results are documented under the following catego- ries: case examples, documents used in current practice, modes of failure, geological considerations, hydraulic and hydrologic considerations, geotechnical considerations, construction considerations, protection, maintenance and control techniques, probability and risk, decision-making process and funding, and current and future research areas. The relevant survey question is presented at the beginning of each section, followed by a summary of the responses. CASE EXAMPLES Q1. Please provide up to three success case examples of a highway embankment subjected to flooding and why you think it was successful. Q2. Please provide up to three failure case examples of a highway embankment subjected to flooding and why you think it failed. A total of 41 case examples were gathered through the survey. Fourteen of these studies were included in chapter three. The rest are shown in Tables 17 and 18. In Table 17, successful case examples are reported in which the design could withstand the flooding event. Table 18 present failure case examples in which the embankment was damaged in the flooding event. The cause and extent of damage are included for each case as applicable. CHAPTER EIGHT SUMMARY OF SURVEY RESULTS FIGURE 102 The participating states in the synthesis in green.

61 TABLE 17 SUCCESS CASE STUDIES GATHERED THROUGH THE SURVEY Agency Project Comments Arizona DOT Ideal Draw near Wilcox The wingwall and the box culvert were undermined. San Pedro on I-10 near Benson This is an example of bridge deck and roadway drainage effect during flooding. Laguna Wash on 163 near Kayenta The roadway embankment was affected by meandering of the river. Iowa DOT I-80 over Cedar River in Cedar County Paved shoulder mitigated damages to pavement and the damage was observed in the roadway embankment. US 65 over Des Moines River in Polk County Minimal damage was caused due to flat embankment slope. Nevada DOT Wilson Canyon This is a clearing and dredging project that placed spur dikes to protect roadway. Rainbow Canyon Reconstruction Overtopping areas constructed and have since been flooded, but remain intact. Oklahoma DOT Tropical Storm Erin 1 Blaine Co SH 33 over creek 10.8 mi east of US 281 in Watonga This was embankment failure due to contraction scour from Tropical Storm Erin. Thankfully the bridge did not fail and we were able to reopen quickly. Grady Co US 81 over Buggy Creek south of Minco This bridge had abutment wash out during tropical storm Erin. Roadway embankment was ok because highway overtopped at the fuseplug. The bridge was reopened quickly. Flooding after Tropical Storm ErinCaddo Co SH 146 over Punjo Creek and Tribs Roadway overtopped and had to be cleared of debris and silt after debris clogged RCB. A grader was used to clear roadway of silt and tree limbs. Pennsylvania DOT Spruce Street Retaining Wall This is both a failure and hopefully a successful case when construction is complete. The retaining wall protecting a rail line was severely undercut and is being repaired. Texas DOT FM 20 at San Marcos River The channel was armored approximately 4 years ago and is performing well. Loop 150 and Colorado River This is an example of slope repair and bank toe protection IH 10 at North Llano River Junction Texas This is an example of a steep bank armored with gabion faced MSE wall. Utah DOT Weber River, I-84 Stabilized roadway embankment The Weber River runs parallel with Interstate-84 within Weber Canyon. Annually during spring runoff the river experiences high flow events. The reinforced embankments along this stretch of the interstate have remained stable. TABLE 18 FAILURE CASE STUDIES GATHERED THROUGH THE SURVEY Agency Project Comments Damage Extent Cause of Damage AZDOT-F Ideal Draw near Wilcox The wingwall and the box culvert were undermined. _ Softening due to satura- tion/Underseepage/Other (s): not having concrete apron between wingwalls San Pedro on I-10 near Benson This is an example of bridge deck and roadway drainage effect during flooding. Affecting traffic Softening due to satura- tion/Other(s): water in traveling lane Laguna wash on 163 near Kayenta This is a case of settlement of bridge barrier. Partial damage- pavement damage Softening due to saturation FHWA Middle Fork Snoqualmie River Road Flow spread across debris fan crossed road in several locations and induced road surface and emabankment erosion. One low water crossing was undermined. Partial erosion Overtopping Iowa DOT IA 150 over Cedar River near Vinton Head differential caused major breach and undermining of pavement. Full breach Overtopping US 6 over Cedar River in Muscatine County Minimal damage was caused due to flat embankment slope. Partial erosion - pavement damage Overtopping/Other(s): loss of shoulder and undermin- ing of pavement/Internal erosion

62 DOCUMENTS USED IN CURRENT PRACTICE Q3. Do you have hyperlinks, hard copies, and/or soft cop- ies of the following documents? Please check the docu- ments provided and fill in the reference. A list of the documents gathered through the survey is included in the Bibliography. The documents fall under the following topics: design documents currently used by the design agency, design criteria, documents considered helpful in design, specifications, sources of hydraulic and hydrologic equations, sources of geologic and geotechnical information, documents used in assessing the embankment condition for resistance to forces generated during flooding, documents used for redesign, maintenance and repair of damaged embankments, documents that provide guidance on legal and regulatory factors included in the design, main- tenance and repair of embankments, and other documents gathered through e-mails or that were not listed under the aforementioned categories. Agency Project Comments Damage Extent Cause of Damage North Dakota DOT ND 18 from Neche to Canadian border During spring thaw after particularly heavy winter snow sea- sons in southern Manitoba/northeastern North Dakota, the Pembina River floods out of its banks. ND Highway 18 has been overtopped numerous times, with particularly serious damage in 2009. Washed out culverts and loss of pavement and soil in areas above these culverts was the primary damage. Full breach- pavement damage Overtopping New York State DOT Route 103 over Mohawk River NY Rt. 103 crosses the Mohawk River on top of the Erie Canal’s dam and lock structure at Lock 9. Just north of the bridge Route 103 continues across the north side of river some 400 ft. to shore. Back-to-back storms set up this incident in 2011. Hurricane Irene was approximately the Q100 for this location on the Mohawk (127,000 cfs). It was followed 8 days later by Tropical Storm Lee, which was close to the Q50 (116,000 cfs). Both floods resulted in tremedous amounts of trees and other floating debris. When the floating debris hit the dam, it was caught and blocked flow. The river rose up and decided to carve itself a complete new channel north of the lock structure. DOT’s bridge survived fine (abutment and pier perched on the piles) but the 40 ft high approach embankment was overtopped and washed away in hours. Subsequently rebuilt with armoring and controlled overflow section. Full breach Overtopping Oklahoma DOT 1 Blaine Co SH 33 over creek 10.8 mi east of US 281 in Watonga This was embankment failure due to contraction scour from Tropical Storm Erin. Thankfully the bridge did not fail and we were able to reopen quickly. Full breach- pavement damage Overtopping Alfalfa Co17668-04 SH 8 from Major CL N 4.0 mi April 27, 2009, flooding caused wash out of 8 cell RCB and roadway fill. Full breach- pavement damage Overtopping Underseepage Oregon DOT Little South Fork Hunter Creek This is a case of bridge abutment washout. The advantage is this bridge was designed for a washout condition and needed the approach to be rebuilt, easier than reconstructing the bridge. Full breach Bridge abutment scour Mt. Hood Highway, OR 35, Nov. 2006 Debris flows from glacial outburst clogged up culverts and bridges along a stretch of road and cut through the roadway at others. Full breach- partial erosion- pavement damage Overtopping Pennsylvania DOT Spruce Street Retain- ing Wall _ Severe erosion underneath the wall affecting the rail line Internal erosion/Softening due to saturation underseepage Texas DOT FM 787 at Trinity River This is the case of ongoing issues with bank erosion and instabilty. Pavement damage Internal erosion/Rapid drawdown US 79 at Tinity River Relief near Palestine Texas The toe erosion of the bank has resulted in slope instability. Partial erosion Failure due to sliding on foundations Utah DOT Cannonville Bridge Embankment Failure The roadway embankment on the east side of an existing bridge south of Cannonville, Utah, failed due to high flows in the Paria River. The east-side stream bank upstream of the bridge was lined with rock riprap, however high flows eroded the natural ground behind the rock riprap causing the stream bank to erode. This erosion washed the easterly bridge approach. Full breach- partial erosion- pavement damage Softening due to satura- tion/Erosion occuring behind riprap

63 MODES OF FAILURE Q4. Please specify/rank the highway embankment fail- ure modes that your agency has experienced. A range of failure modes is witnessed from flooding, including overtopping, saturation, underseepage, through- seepage, wave erosion, and sliding on foundations. The occurrence of a certain mode and the extent of damage are dependent on site-specific conditions. Figure 103 shows the common failure modes among the participating states. The DOT engineers were asked to rank the failure modes from least common (1) to most common (5). Based on the results shown in Figure 103, overtopping is classified as the most common failure mode in five states, followed by softening owing to saturation in three states and underseepage in one state. The second most common mode is overtopping in two states, followed by softening from saturation, underseepage, through-seepage, and sliding on foundations in one state each. The moderately common mode (rank 3) is overtop- ping in six states followed by saturation and underseepage in five states each, sliding on foundations in four states, and through-seepage and wave erosion in three states each. Additional notes were provided by the participants related to saturation, failing culverts, and toe erosion due floodwa- ters running in road ditches. Heavy precipitation can cause saturation. Saturation can lead to slope failures or sliding on foundations. Culvert-related issues include scour or erosion around culverts, separation between the culvert and the sur- rounding soil, and under-designed or plugged culverts by debris. Toe erosion could occur in the presence of ditches running alongside the toe, in the event water overflows from the ditches. Another cause is the lateral migration of mean- dering streams during floods eroding the toe and leading to instability and sliding on foundations. GEOLOGICAL CONSIDERATIONS Q5. What are the geologic and geomorphic factors that should be addressed in the design of highway embank- ments subjected to flooding? Why? In this section, respondents were asked to rank the geologic and geomorphologic factors from the most important (5) to the least important (1). Erosion and deposition and meander- ing potential ranked as the most important in eight states, fol- lowed by channel dimensions and topography in seven states, and floodplain and basin roughness in four and two states, respectively (Figure 104). The second most important factors were considered to be floodplain in 11 states and meandering potential and topography in nine states. It was further noted by a couple of participants that the importance of these fac- tors varies depending on the case considered, whereas others answered based on their experience. Other factors considered included the history of head cutting and the available materi- als that could be used for the purpose of the embankment fill. FIGURE 103 Common failure modes in participating states.

64 FIGURE 104 Relative importance of geologic and geomorphologic factors among states. Additional notes were included related to what factors should be largely considered in design. River meandering and sedimentation/deposition processes directly affect the design for flooding, particularly armoring, maintenance and general dimensions of the embankment. The general topog- raphy and floodplain are considered critical factors in select- ing the site location. HYDRAULIC AND HYDROLOGIC CONSIDERATIONS Q6. Please indicate the recurrence interval of the design flood/hurricane adopted for highway embankments. Based on the survey results, the recurrence interval as well as the availability of guidelines for the selection of recurrence interval for a roadway embankment vary from state to state. The most common recurrence intervals vary between 25 and 100. A number of factors were used in dif- ferent states to select a recurrence interval such as the road classification, the average annual daily traffic, urbaniza- tion, and the size of the watershed. Such factors are shown in Table 19. It was also noted that some states do not have regulations specific to roadway embankments. In such cases, available guidelines for other drainage structures are used. As for hurricane requirements, three states specified the 500-year recurrence interval. Q7. How do you predict the recurrence interval (RI) that would lead to overtopping for an existing embankment? Hydraulic analysis is commonly carried out among DOT engineers to identify the overtopping height and the flow velocities. About 70% of the participating states use HEC- RAS and 50% of them also use HY-8. TABLE 19 DESIGN RECURRENCE INTERVAL REQUIREMENTS IN DIFFERENT STATES State Guidance for Selection for Recurrence Interval Arkansas Design for a 50-year storm and check for 100- year storm. Connecticut Assumed same requirements as those for hydraulic struc- tures: 50-year storm and 100 years for drainage area > 1 mi. Check is required with 100-year and 500-year inter- vals, respectively. Florida Requirements for bridges: up to 50 years depending on the AADT (average annual daily traffic). Georgia Use a 50-year storm for the base, 100-year storm for shoulder breaking point. Idaho No overtopping of roadways allowed at the 199-year storm. Illinois Overtopping criteria are available: low point of roadway across the floodplain = Q50 + 3 ft; freeboard = 3f. Maryland Recurrence design criteria is dependent on road classification. Minnesota Minimum overtopping criteria varies with AADT and the type of the infrastructure. 100- year flood or “flood of record” have been used. North Dakota Administrative code requirements: 25-year storm for state highways, 50-year storm for interstate. Utah Different requirements (10–50 years) depending on highway classification and AADT. Wyoming 100-year storm if developed property is involved. Q8. How do you estimate the peak discharge corre- sponding to a flood recurrence interval for gaged and ungaged sites? A number of methods are used by state DOTs to compute peak discharges for gaged and ungaged sites. Generally, a combination of methods is used. In gaged sites, about 60% of the participating DOTs use direct available data, about 40% use regression equations, about 28% use HEC-HMS, and about 10% use FEMA studies. In ungaged sites, a number of methods are also used to determine the peak discharges. About 52% use regression equations, 35% use StreamStats, and 24% use HEC-HMS. Q9. When overtopping takes place, how long does it typi- cally last? The answers to this question were limited, typically because records of such data are not maintained by DOTs. Roughly, overtopping events could last anywhere between less than an hour to several weeks. The longest overtop- ping duration—2 to 3 weeks—was presented by Minnesota. Other respondents specified overtopping durations up to 48 hours. One answer was obtained relevant to the duration in coastal environments, which is 5 hours.

65 GEOTECHNICAL CONSIDERATIONS Q10. What is the soil type typically used to build embankments? Based on the survey results, all types of materials are used to construct roadway embankments. The materials requirements are generally specified in relevant state manuals. Consideration is given to use the site materials for feasibility purposes. The types of embankment materials used is summarized in Figure 105; 32% of the participating states use both coarse and fine fill material, 27% use only coarse, 27% use only fine, and 14% use coarse grained, fine grained, and zoned embankments. FIGURE 105 Embankment material. Q11. Please indicate the in situ and/or lab tests used to design embankments subjected to flooding. Limited soil testing data were received from the survey. This could be because the survey answers were generally completed by hydraulic engineers. Based on the available data, soil testing was mostly lim- ited to gradation, plasticity, and compaction. Two states speci- fied the use of strength tests (shear and triaxial). One state specified the use of Pin Hole (erodible soils) and testing for strength of erosion control mats. One state further noted that it does not carry out any tests specifically for flood design. DESIGN AND CONSTRUCTION CONSIDERATIONS Q12. Please specify the chosen design life of embankments. Different answers were given by the states. The answers ranged from as low as 5 years to as high as 50 years, 75 years, and 100 years because the embankment was expected to last as long as the structure. Q13. Are there any special pavement or subbase require- ments/techniques that are used to decrease the chance of pavement loss due to flooding? Of the 29 answers given for this question, seven states said that they have some considerations for pavements prone to flooding. Those considerations are outlined herein: • Using rockfill for the subbase to avoid scour/erosion (Arkansas Practice) • Using black base, graded aggregate base, and underd- rain (Florida) • Allowing for a freeboard of 2 ft at design flood for a culvert (Idaho) • Using recycled asphalt pavement, geogrid with vegeta- tion, riprap, or cap clay on the slope, and paving the slope (Minnesota) • Armoring the slope (Nevada) • Keeping the subbase above design flood level (Tennessee). PROTECTION TECHNIQUES Q14. What current design, maintenance, and repair tech- niques do you use to minimize roadway embankment damage during flooding? The protection systems commonly used for embank- ments are shown in Figure 106. Based on 30 answers, veg- etation and riprap are the most commonly used protection means in 19 states (about 65%), followed by 12 states (40%) that use gabions, 11 (37%) that use precast concrete blocks, and five (17%) that use armoring. FIGURE 106 Protective measure installed by DOTs. Obviously, the state practice is generally not limited to one protection measure but rather a number of measures for different applications (different applied velocities and differ- ent weather conditions when it comes to vegetation). Other

66 practices were identified by three states: rock buttresses, concrete lining of slopes, and the use of fuse plugs. One state did not specify any protection measures. Instead, the importance of the application of an internal drainage system to reduce hydrostatic pressures was emphasized. A comparison of the estimated costs and installation time between the protection systems is further presented in Figures 107 and 108, respectively. Based on the figures, the vegetation, the geotextile, and the rock buttresses are on the low cost end. The cost of riprap ranges from low to high depending on its size. Armoring and concrete lining fall in the intermediate cost range, while gabions and precast con- crete blocks are on the high cost end. As for the installation time, the vegetation, riprap, geotextile, rock buttresses, and lining can be constructed in a short amount of time. Gabi- ons and armoring require more installation time, and precast concrete blocks generally are the most time consuming. Cost Low Intermediate High Vegetation Riprap Gabions Geotextile Precast Concrete Blocks Armoring Rock Buttress Concrete Lining FIGURE 107 Comparison between the costs of protection systems. FIGURE 108 Comparison between the times of installation for the protection system. Installation Time Low Intermediate High Vegetation Riprap Gabions Geotextile Precast Concrete Blocks Armoring Rock Buttress Concrete Lining It is important to note that geotextiles are generally used in combination with other systems such as under riprap or precast concrete blocks. And even though vegetation can be installed quickly, it requires more time to grow and develop its full resistance. Q15. What type of vegetation, vegetation cover density, and height of the vegetation is used to protect highway embankments from damage during flooding? Various types of vegetation are commonly used across the states. The selection of type depends on the weather, and specific seed mixes are generally developed for different weather conditions within each state. Aside from the differ- ent types of mixes available, different densities and heights are adopted by different states. Q16. What sediment control techniques do you use upstream of highway embankments to capture sediments and minimize the potential for flooding? Silt fencing is the most commonly used sediment control technique (17 answers), followed by bales (12), filter strips (nine), and, finally, retaining walls (one). It is worth noting that such techniques are not necessarily used to minimize embankment damage during flooding. PROBABILITY AND RISK Q17. What probability of failure does your agency con- sider acceptable for a highway embankment and for a pavement? This section identified the current risk-related practice. Of the 23 states that answered this question, five indicated that they tie their design practice to risk. One state indicated that because roadway embankments are classified based on their average annual daily traffic, the higher this value the higher the protection required for the embankment. Another state indicated that informal risk evaluations are typically considered. This evaluation is tied to minimum safety fac- tors outlined in the geotechnical manual. DECISION-MAKING PROCESS AND FUNDING Q19. Who is responsible for repairing highway embank- ment damage due to flooding in your state? A number of agencies are consistently named as respon- sible for repairing damage, mainly the owner of the roadway and the DOT maintenance agency. One state clarified that a combination of forces are responsible depending on who owns the road. Q20. Please specify the source of funding available in cases of emergency repair due to damage by flooding events.

67 The source of funding was identified by 12 states as ER and FEMA funds. One state commented that the source of funding depends on whether the area is declared a “disaster area” by the governor or president. Three states identified only ER as a funding agency. Six states specified state funds. Q21. Sometimes, there may be a request by a related agency to design the embankment as a levee and this impacts the design or repair process. What are the factors that influence the design and/or repair decision process to minimize highway embankment damage during flooding? Seven states clarified that FHWA discourages the use of roadways as levees. As a result, these states do not design embankments as levees. A number of considerations that would influence the design/repair decision process were men- tioned, including considerations related to the funding agency, environmental considerations, and feasibility and time con- siderations, in addition to prioritization of other projects. CURRENT AND FUTURE RESEARCH AREAS This section aims to identify the documents currently used to minimize roadway embankment damage from flooding, as well as the areas that require future research. Q22. Sometimes, there may be a request by a related agency to design the embankment as a levee and this impacts the design or repair process. What are the factors that influence the design and/or repair decision process to minimize highway embankment damage during flooding? The survey respondents provided the following exist- ing studies: • Douglas and Krolak (2008), Highways in the Coastal Environment, HEC-25, 2nd ed. • Chen and Anderson (1987), Development of a Methodology for Estimating Embankment Damage Due to Overtopping, FHWA/RD-86/126 • Paul and Clopper (1989), Hydraulic Stability of Articulated Concrete Block Revetment Systems During Overtopping Flow, FHWA-RD-89-199 • Clopper and Chen (1988), Minimizing Embankment Damage During Overtopping Flow, FHWA-RD-88-181 The survey also identified ongoing relevant research in Minnesota: – MnDOT—Design Considerations for Embankment Protection during Road Overtopping Events, University of Minnesota—project in progress, end date March 31, 2017. – MnDOT Flash Flood Vulnerability and Adaptation Assessment Pilot Project, Philip Schaffner. This proj- ect includes investigation of slope failure, mainte- nance-identified historical overtopping, and assessed impacts of climate change on overtopping frequency. – NCHRP—NCHRP 24-36 Scour at the Base of Retaining Walls, David Reynaud—in progress. Q23. Which areas of embankment damage due to flood- ing require further research? The following issues were identified by the survey as requiring further research. General design issues • Recommendations for the design of approach embankments • Recommendations for the design of culverts placed in embankments • Guidance on designing inexpensive embankment pro- tection at locations subject to overtopping • Development of the tractive shear equation Meyer- Peter-Muller coefficients and the embankment length and time step of application • Guidance for mass wasting or slip-circle failure of embankment • The effect of tailwater level on embankment erosion • The effect of fill height on embankment erosion • The effectiveness of vegetation in resisting erosion • Design of embankment gradation Pavement-related research • Preserving the pavements during flooding through the use of paved versus gravel shoulders Risk-related guidelines • Development of a risk-based methodology for priori- tizing the implementation of climate change adaptation measures at state highway structure locations (espe- cially at locations within the coastal zone) Compliance with other agency requirements • Means and methods of ensuring proposed embankment “resilience” in compliance with Amended EO 11988 and the Federal Flood Risk Management Standard SUMMARY This chapter presented a summary of the survey results obtained from DOT engineers. The replies received on each survey question were compiled and tabulated as applicable. These results, along with the information gathered from the case examples and the literature review, were used to outline design considerations, countermeasures, and mitigation and repair measures in the next two chapters.