Highway Hydraulic Engineering State of Practice (2020)

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.

55 Overview Interviews were conducted with four state hydraulics DOT representatives, selected by examining the survey responses to those who replied “yes” to the survey question asking if they would be willing to participate in a follow-up interview to collect case example informa- tion and via literature review of published materials. Case examples were to illustrate different policy changes, diversity in topics, and early adopters of new/innovative/successful practice. Interviews via email and phone conversations were conducted to acquire information and documentation. Case examples from the selected states—Michigan, Georgia, North Carolina, and Delaware— are included in the report. Table 18 identifies which subtopic is addressed and highlights of the practice that the case example illustrates. Data Challenges in 2D Advanced Hydraulic Modeling (Michigan DOT) Recently, Michigan DOT has begun participating in FHWA’s Every Day Counts initiative, “Advance to the Next Generation of Engineering [EDC-5: Collaborative Hydraulics: Advanc- ing to the Next Generation of Engineering (CHANGE)],” by exploring the use of 2D modeling (Carlson 2019, FHWA 2019). The FHWA Every Day Counts initiative includes use of next- generation hydraulic applications to improve understanding of interactions between transpor- tation assets and river environments (FHWA 2019). To advance toward policy change and the use of 2D hydraulic models for design, Michigan DOT has completed several test models to develop and validate 2D modeling components. For example, Michigan DOT is currently applying 1D HEC-RAS and 2D SRH-2D via SMS for a proposed culvert replacement project on M-35 over the Carp River in Marquette County, Michigan. Two reasons for choosing this site to experiment with 2D modeling are that the Carp River meanders upstream and downstream of this location, and LIDAR data was available for the floodplain. Michigan DOT deemed the LIDAR data to be relatively reliable as compared to LIDAR data for previous 2D models (Figure 18). LIDAR data must accurately depict topogra- phy, and Michigan DOT had noticed data discrepancies in highly vegetated areas as compared to conventional survey methods. Discrepancies in LIDAR data can cause inaccurate representation of natural conditions in models (Carlson 2019) (Figure 19). Following the application of 1D and 2D models, the model results must be carefully examined and compared to observations to determine the reliability of simulated water surface elevations. Michigan DOT is also particularly interested in the level of survey data necessary for 2D models, as well as differences in scour and C H A P T E R 4 Case Examples

56 Highway Hydraulic Engineering State of Practice the effectiveness of scour countermeasure designs as predicted by 1D and 2D models (Carlson 2019). At the end of the study, Michigan DOT anticipates FHWA Resource Center’s Hydraulic Engineering Team reviewing the 2D results and providing comments. Through the development and evaluation of test models, Michigan DOT has identified several challenges to 2D modeling (Carlson 2019). First, 2D models seem to be more time consuming to construct than typical 1D methods, which may be due, in part, to a lack of experience with 2D modeling. To address this, modeling time will be monitored over more test cases. Second, it can be difficult to determine the amount of survey detail required for a 2D analysis and the cost- effectiveness of additional data collection. Full bathymetry data, which consists of multibeam and mobile LIDAR obtained by using a boat, is more expensive to collect than the data neces- sary for a 1D model. Michigan DOT has learned that other states are using a procedure known as “channel stamping” to define their design bathymetry, which may be a more cost-effective approach. Third, the file size of 2D models (including the survey data) is much larger than a conventional 1D model, which can be costly for governmental agencies such as Michigan DOT Agency Interviewed Subtopic Highlights of the Practice the Case Example Illustrates Michigan DOT Advanced Hydraulic Modeling: 2D Challenges: time requirements, data adequacy, file size, validation of roughness coefficients Georgia DOT Advanced Hydraulic Modeling: 2D Challenges: determination of model type, FEMA acceptance, calibration North Carolina DOT Regulatory Requirements; Floodplain Impacts and Mitigation -Modernization of flood mapping -Establishment of the publicly available North Carolina Flood Risk Information System or FRIS -Programmatic Memo of Agreement (MOA) facilitated a streamlined process through North Carolina DOT: Transfer of compliance FEMA responsibilities to North Carolina DOT Delaware DOT Coastal Hydraulics Green infrastructure design: raising and reshaping the dunes, preserving existing and creating additional marshes, and protecting the eroding shoreline using of oyster shell bags, new tide gate system Table 18. Agency interviewed, subtopic, case example highlights. Figure 18. Carp River Aerial: Low vegetation thus minimal discrepancies in LIDAR data recognition in the project area. Courtesy of Michigan DOT. 2019.

Case Examples 57 who pay for data storage. Large files can also be challenging for file sharing. Fourth, 2D model results show that more research is needed to define Manning’s roughness coefficient (“n”) values for 2D models, since they directly impact water surface elevations. Preliminary modeling results show that water surface elevations in 2D models are slightly higher than in 1D models with the same Manning’s “n” values. Finally, regulatory agencies still use 1D models for floodplain accep- tance, which can require parallel modeling if 2D models are to be applied. Currently, the 2006 Michigan DOT Drainage Manual states that physical and 2D modeling is not required but “shall be used in specialized applications” (Michigan DOT 2006); however, Michigan DOT anticipates updates to the state drainage manual as 2D modeling studies are completed and acceptance of 2D model results progresses. Model Type Determination, FEMA Acceptance, and Calibration in 2D Advanced Hydraulic Modeling (Georgia DOT) According to criteria outlined in the State of Georgia DOT: Drainage Design for Highways Revision 3.4 (Georgia DOT 2018a), a highway crossing of Pachitla Creek in Calhoun County, Georgia, was well suited for 2D hydraulic modeling. Georgia DOT incorporated 2D modeling as policy for this project since traditional 1D modeling would not have accurately represented hydraulic conditions at the site, which included three parallel openings, a wide floodplain, and transverse flow characteristic of a meandering channel (Georgia DOT 2018a). The total width of the (undesignated) floodplain is approximately 1,500 ft, as shown in Figure 20. The three proposed prestressed concrete replacement bridges will have lengths of 240 ft, 180 ft, and 135 ft, respectively. Each of the structures will have spill-through abutments and will be 40 ft wide and oriented 90 degrees to the roadway centerline. Approximate distances between the bridges will be 200 ft between the first and second bridges and 300 ft between the second and third bridges. As modeled for the proposed condition, no overtopping occurs for Figure 19. Carp River model mesh produced with reliable LIDAR data. Courtesy of Michigan DOT. 2019.

58 Highway Hydraulic Engineering State of Practice the 50-year flood frequency, and even though overtopping is shown to occur for the 100-year flood event, this is predicted to occur at a low point on the roadway outside the project limits (Georgia DOT 2018b). The project was modeled with U.S. Department of the Interior Bureau of Reclamation SRH-2D Surface Water Modeling (SMS) Version 11.1.1. This model was chosen primarily because Georgia DOT had received licensing via FHWA and had worked with FHWA on a proj- ect. Georgia DOT also participated in FHWA 2D training and then joined FHWA Every Day Counts (FHWA 2019). The SRH-2D model originally did not have pressure flow or structure- modeling capabilities. However, it now has these capabilities and many others including weir, gates, culverts, etc. (Nguyen 2019). During the project proposal and award process, Georgia DOT evaluates hydrologic and hydraulic complexity and determines which modeling type is the most appropriate for a given project. In some cases, 2D modeling has been suggested only after finding out the 1D models were inadequate in modeling the existing condition. For projects completed by consultants, Georgia DOT will suggest the use of 1D or 2D modeling; however, the consultant is not restricted to specific modeling software. Georgia DOT is available to work closely with the consultants to ensure the study results are adequate prior to submission for review, regardless of which software the consultant uses (Nguyen 2019). For floodplain permit applications, FEMA does accept 2D models as long as the modeling follows specified guidelines and meets the minimum requirement of the National Flood Insur- ance Program. Despite that fact, Georgia DOT has yet to submit any 2D study to FEMA. Typi- cally, the model received from FEMA is in 1D; therefore, Georgia DOT would convert 2D model results back to a 1D model and calibrate the 1D model for submission (Nguyen 2019). The primary challenge in 2D hydraulic modeling is obtaining data for the model. Georgia state- wide LIDAR is available for only a handful of counties. The Georgia Geospatial Information Figure 20. Velocity distributions depicted in a 2D model of three structures spanning the 1,500-ft wide floodplain. Courtesy of Georgia DOT (2019).

Case Examples 59 Office is in the process of acquiring LIDAR data for 89 more counties, but this data will not be available anytime soon. Thus, obtaining useful surface data for 2D modeling can still be costly (Nguyen 2019). Programmatic Memorandum of Agreement (MOA) Between North Carolina Floodplain Mapping Program (NCFMP) and North Carolina DOT Over the previous 100 years, approximately 20 Category 3 or greater storms passed within 50 nautical miles of North Carolina, and North Carolina has experienced major hurricane events about every 5–7 years. Impacts to eastern North Carolina from Hurricane Floyd in 1999 included thousands of square miles of flooded land and North Carolina DOT assets. With North Carolina DOT being the largest stakeholder of assets in flood risk areas (Figure 21), there is a need to accurately assess risk and damages, and to formulate a plan for reliable compliance procedures by proper design, construction, and floodplain mitigation (Snead 2018). Federal Aid Projects are required to comply with FHWA regulations, which are consistent with National Flood Insurance Program regulations under a 1982 Memorandum of Understanding (MOU) between FEMA and the FHWA, Procedures for Coordinating Highway Encroachments of Floodplains with FEMA [supplement to the Federal Aid Policy Guide under 23 CFR 650A Loca- tion and Hydraulic Design of Encroachments on Flood Plains] (Snead 2018). In 2000, FEMA designated North Carolina as a Cooperating Technical Partner State, and this allowed for the state to assume primary ownership and responsibility for updates and modernization of the state’s flood maps. This led to the establishment of the North Carolina Floodplain Mapping Program (NCFMP), which was charged with remapping FEMA FIRMs for all North Carolina communities as part of the NFIP. This includes conducting flood hazard analyses and producing Figure 21. North Carolina DOT Current 2017–2027 Statewide Transportation Improvement Program Map. Courtesy of North Carolina DOT. 2018.

60 Highway Hydraulic Engineering State of Practice updated DFIRMs. These maps and other flood risk information is publicly available on the North Carolina Flood Risk Information System (FRIS) (Snead 2018). An example inundation map is shown in Figure 22. Many of the state’s formerly effective FEMA flood studies and maps prior to 2000 were based on decades-old hydrologic and hydraulic data derived from approximate study methods, old HEC-2 (or other) computer models, and topographic data from older USGS topographic quad- rangle mapping. The LIDAR and GIS technologies that emerged in the late 1990s were not available for those earlier studies. NCFMP was very proactive in launching a statewide flood- plain remapping effort leveraging these new technologies. When NCFMP began the statewide remapping effort in 2000, they contacted North Carolina DOT to request current records on hydraulic structures for flood study updates. This new program provided an opportunity for North Carolina DOT to begin working with NCFMP to ensure NFIP regulatory compliance and to provide FEMA with current project data to ensure that the new flood maps would accurately reflect the latest changes made by North Carolina DOT to hydraulic structures on roadways crossing FEMA-regulated streams. Given that North Carolina DOT is the largest stakeholder with regard to assets in flood risk areas, and that the new statewide floodplain remapping effort resulted in nearly 85% of the state’s streams being designated as Special Flood Hazard Areas on FEMA flood maps, it was essential for North Carolina DOT to be able to accurately assess risk and damages, and formulate a plan for reliable compliance procedures for design and construction to mitigate floodplain risk. In August 2006, representatives from FHWA, FEMA Region IV (Atlanta), and NCFMP met with the North Carolina DOT Hydraulics Unit to discuss floodplain management concerns. FHWA emphasized that Federal Aid Projects were required to be in compliance with FHWA regulations, which are consistent with NFIP regulations, as outlined in the 1982 MOU noted Figure 22. Flood inundation mapping FRIS. Courtesy of North Carolina DOT. 2019.

Case Examples 61 previously. Further, this policy was reinforced by both Federal and State Executive Orders and was reaffirmed in the commitment by the North Carolina DOT Highway Administrator at that time (2006) to ensure NFIP compliance moving forward. Therefore, in 2008, North Carolina DOT entered into a formal Memorandum of Agreement (MOA) with NCFMP to streamline FEMA NFIP compliance approvals for projects with no rise or minor decreases in BFE and to ensure flood hazard maps are updated statewide to reflect changes by North Carolina DOT, the state’s largest developer. The MOA thus facilitated a streamlined LOMR approval process for all North Carolina DOT projects impacting FEMA-regulated floodplains throughout North Carolina (Snead 2018). Currently, the North Carolina DOT State Hydraulics Engineer and NCFMP Director meet for ongoing MOA oversight, and representatives of both agencies hold monthly coordination meet- ings to discuss specific projects and issues of concern to ensure timely FEMA NFIP compliance approval and keep project delivery on schedule. The efficient all-digital process developed under the MOA for submission, comments, responses, and approvals has benefitted North Carolina DOT by: • providing a consistent review process by a single agency instead of reviews by approximately 650 individual communities (i.e., counties and municipalities); • cutting approval time by 75% for no-rise or decrease in base flood elevation submittals, compared to that for the conventional CLOMR process; • meeting project let date deadlines; • advocacy for North Carolina DOT by NCFMP to both FEMA and to all North Carolina communities; and • participation with NCFMP in ensuring that NFIP flood studies, maps, and associated hydro- logic and hydraulic data for North Carolina DOT roadways are kept accurate and current statewide. Many successful projects have resulted as an application of the MOA, and North Carolina DOT has participated with NCFMP in developing mitigation strategies flood studies in areas devastated by recent hurricanes Matthew and Florence as part of the Governor’s Rebuild North Carolina program (Snead 2018). Moving forward, North Carolina DOT sees many opportunities both for improving current processes and for further collaborative efforts with NCFMP to achieve mutually beneficial goals. Further process efficiencies will be achieved by modifying the current funding mechanism for project reviews from a fee structure on a project-by-project basis to a programmatic annual funding approach through a recently established North Carolina DOT Highway Floodplain Program as a statewide program of the North Carolina DOT’s Hydraulics Unit. While NCFMP has been very accommodating to North Carolina DOT, NCFMP is also accountable to FEMA for regulating NFIP compliance for all public and private development across the state, and there are some inherent limitations to the workload and staff resources they have been able to dedicate to North Carolina DOT projects. Through a separate, but similar, interagency agreement which was recently signed, North Carolina DOT will soon begin funding two individual positions within NCFMP—one engineer and one GIS specialist—who will be dedicated to North Caro- lina DOT projects and concerns. Some additional collaborative efforts that are also planned or underway through the MOA include: • Augmenting the state’s stream gage network for NCFMP’s Flood Inundation Mapping and Alert Network, which North Carolina DOT relies upon heavily during major storm events; • Studying ways to improve strategic corridor routes in the state that were closed during recent hurricanes, costing billions of dollars in economic loss, by implementing more resilient design standards considering anticipated sea level rise and increasing intensity of extreme events;

62 Highway Hydraulic Engineering State of Practice • Continued planning and prioritization of approximately 130 coastal bridges to provide counter measures for wave and surge forces in extreme events; and • Development of LIDAR and advanced hydraulic modeling technology to better predict water surface elevations and optimize design to minimize litigation risk associated with North Carolina DOT projects. Improving Infrastructure on a Case-by-Case Basis at Delaware DOT Delaware DOT faces many challenges due to sunny day nuisance flooding and flooding associated with wave impact and tidal surge. Prior to Hurricane Sandy in October 2012, raising vulnerable roads had been suggested to alleviate key infrastructure from flooding. However, major differences exist between risks associated with sunny day nuisance flooding and risks associated with wave impact and a tidal surge. Depending on the setting, raising coastal struc- tures may increase vulnerability to wave attack, resulting in more costly repairs and longer service outages (Sisson 2019). As one aspect of the state adaption policy, Delaware DOT evaluates projects impacted by sea level rise on a case-by-case basis. In 2013 Governor Jack Markell signed Executive Order 41, “Preparing Delaware for Emerging Climate Impacts and Seizing Economic Opportunities from Reducing Emissions” (Georgetown Climate Center 2013, Markell 2013). This order established a Governor’s Committee on Climate and Resiliency and charged all state agencies to form plans to adapt to sea level rise and increases in flood level in the designs of state-funded projects including new construction, reconstruction, and renovation (Georgetown Climate Center 2013, Markell 2013). Under Executive Order 41, 150 actions were assigned to state agencies; 19 of these to Delaware DOT. In response, Delaware DOT completed the Strategic Implementation Plan for Climate Change, Sustainability and Resilience for Transportation in 2017, listing the prioritization of the 19 actions. Four tier-one recommendations and a timeline to completion were as follows: 1. Development of geospatial data sets to identify vulnerable areas – 2020; 3. Integrate climate resilience into bridge and highway design manuals – 2019; 9. Incorporate climate change into infrastructure investments – 2018; 17. Support local governments with land use assessment tools – 2021. Delaware DOT has implemented this plan by inventorying roadway assets using LIDAR and developing geospatial data sets. Data is analyzed and processed to produce rainfall and storm surge models for assessment of infrastructure impacts from extreme events and sea level rise. Delaware DOT has initiated updates to mitigate for climate change and sea level rise impacts in its draft Road Design Manual. Delaware DOT continues to review progress on the Strategic Implementation Plan annually, update the plan, and seek additional resources, such as FEMA, National Oceanic and Atmospheric Administration (NOAA), EPA, etc. (Delaware DOT 2017b). Prior to Hurricane Sandy, Delaware DOT had completed preliminary estimates to raise most threatened roadways, including coastal portions of SR 1 (Delaware DOT 2017a). During Sandy, coastal dunes washed over the road and protected it from being undermined during the surge and withdrawal. The urgency to raise the corridor was alleviated (Sisson 2019). Had the corridor been raised, the elevated section may have been vulnerable to wave attack or weir-flow damage (Figure 23). The weir-flow damage mechanism and subsequent road loss is well understood and is described by the example of the raised Florida 292 on Perdido Key, Florida. At this location, the road was elevated +8 ft North American Vertical Datum (NAVD), but the storm surge was +11 ft NAVD, and the road was destroyed. The cyclic mechanism of destruction is such that the roadway forms a broad-crested weir, with the pavement crown as the crest; the surge elevation exceeds the crest and the landward flow is supercritical, thus scouring the shoulder; then water retreat causes the same action and damage in reverse (U.S. DOT FHWA 2008).

Case Examples 63 A significant amount of national-level research is being conducted to help shape future policy, including the following: • NCHRP Project 20-59(53): “FloodCast” (report pending) uses the NOAA National Water Model; see https://water.noaa.gov/about/nwm. • NCHRP Project 15-61: “Applying Climate Change Information to Hydrologic and Hydraulic Design of Transportation Infrastructure.” • NCHRP Project 20-05, Topic 51-10: “Flood Resiliency and Adaptation Through Integrated Storm Prediction and Response Systems” (complements Floodcast). Green infrastructure design is an example of a technique Delaware DOT has used to protect roadway assets from flooding; since under extremely high tides and sea level rise, the inland bays push toward SR 1. The SR 1 living shorelines project analyzed the applicability of green infrastructure design techniques that can help lessen tidal intrusion into the bay (Delaware DOT 2017a). Figure 24 shows a living shoreline consisting of natural material, including plants, sand, and a rock wall. Figure 23. SR 1 South Bound Indian River Inlet Bridge Approach Protected by Combination of Beach Nourishment and Sheet Piles. Photo Courtesy of Delaware DOT South District. 2019. Figure 24. Living Shoreline. Reprinted from Wikimedia Commons. 2014.

64 Highway Hydraulic Engineering State of Practice Road elevation is still an appropriate strategy in some cases. As an example of road elevation, SR 54 outside of Fenwick Island is a major artery to the beach, as well as a hurricane evacua- tion route that was affected by tidal flooding. The road needed operational improvements and protection from both nuisance flooding and tidal surge. The corridor runs perpendicular to the shore line and is not subject to direct ocean wave attack, but given the vital nature of the corridor, the most vulnerable portions were elevated. A viaduct consisting of two separate structures totaling 2,400 ft in length was completed in 2001 (Delaware DOT 2017a). There were valuable lessons learned after Hurricane Sandy occurred, and Delaware DOT plans to use data collected from events like Sandy to update risk assessments and design resiliently (Delaware DOT 2017a).