National Academies Press: OpenBook

Practices for Integrated Flood Prediction and Response Systems (2021)

Chapter: Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems

« Previous: Chapter 2 - Literature Review
Page 33
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 33
Page 34
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 34
Page 35
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 35
Page 36
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 36
Page 37
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 37
Page 38
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 38
Page 39
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 39
Page 40
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 40
Page 41
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 41
Page 42
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 42
Page 43
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 43
Page 44
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 44
Page 45
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 45
Page 46
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 46
Page 47
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 47
Page 48
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 48
Page 49
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 49
Page 50
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 50
Page 51
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 51
Page 52
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 52
Page 53
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 53
Page 54
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 54
Page 55
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 55
Page 56
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 56
Page 57
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 57
Page 58
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 58
Page 59
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 59
Page 60
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 60
Page 61
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 61
Page 62
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 62
Page 63
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 63
Page 64
Suggested Citation:"Chapter 3 - Survey of State Practices for Integrated Flood Prediction and Response Systems." National Academies of Sciences, Engineering, and Medicine. 2021. Practices for Integrated Flood Prediction and Response Systems. Washington, DC: The National Academies Press. doi: 10.17226/26330.
×
Page 64

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.

33   Introduction A survey was distributed to the state hydraulic engineer (or equivalent position) at the 50 state DOTs and the District of Columbia DOT. Forty-eight DOTs responded (a survey response rate of 94%) and provided input on state practices for integrated flood prediction and monitoring systems. The survey questions and a summary of the survey results are presented in Appendix A and Appendix B, respectively. This chapter is organized into five general topic areas: (a) current status of flood event management, (b) flood monitoring, (c) flood prediction, (d) flood warning systems, and (e) flood response systems. Current Status of Flood Event Management Forty-four DOTs (92%) reported that they experienced recurring substantial infrastructure and economic damages attributable to flood events in the past 15 years. Only four state DOTs (8%) reported that their state has not experienced sustained substantial infrastructure and eco- nomic damages attributable to flood events in that time. Figures 9 and 10 describe reported flooding types and flooding causes, respectively. Success of systems: When asked how successful—on a scale of 1 to 5 (where 1 = least effec- tive and 5 = very effective)—their DOT integrated system for flood prediction and response system is viewed from the perspective of their office, 25 DOTs self-rated their systems with a ranking of 3, 4, or 5, which is considered successful. This question, together with other self- ranking questions in the survey, was used as one of the criteria in selecting case example states in Chapter 4. The 25 state DOTs that ranked their integrated system for flood prediction and response system as successful (rank 3, 4, or 5) provided information about plans or initiatives that have contributed to the successful systems as described in Table 3. Observed benefits: The survey received 25 state DOT responses about observed benefits of the DOT’s successful integrated system. Twenty-one DOTs reported that overall safety enhance- ment (e.g., reduction in death toll attributable to flood events, decreased crash frequency during heavy rain season) was an observed benefit of their successful integrated system. The distribu- tion of observed benefits for successful integrated systems is shown in Figure 11. Forty-six DOTs provided a response when asked to identify their top three learned lessons from their flooding over the past 15 years. Common lessons learned include making sure a plan is in place and being prepared for a flood event. Other common responses were ensuring good communication between agencies and being proactive rather than being reactive. C H A P T E R 3 Survey of State Practices for Integrated Flood Prediction and Response Systems

34 Practices for Integrated Flood Prediction and Response Systems 6 (13%) 24 (50%) 44 (92%) 47 (98%) 0 5 10 15 20 25 30 35 40 45 50 Other (e.g., ice jams, storm surge) Coastal flooding Surface/overland flooding (e.g., due to the poor drainage) River flooding (e.g., overtopping banks) Number of DOTs: Total Number of DOT Responses: N = 48 Figure 9. Reported types of flooding experienced by states. Survey respondents were allowed to select multiple answers. Total Number of DOT Responses: N = 48 8 (17%) 13 (27%) 14 (29%) 26 (54%) 28 (58%) 47 (98%) Other (e.g., ice jams, riverine) Flooding due to king tide events Tidal flooding due to sea level rise and/or subsidence Flooding due to snow melt Flooding due to storm surge Flooding due to heavy rain events Figure 10. Reported causes of flooding experienced by states. Survey respondents were allowed to select multiple answers.

Survey of State Practices for Integrated Flood Prediction and Response Systems 35   State DOT Response California • In-house monitoring system for rainfall, flash flood, etc. • Some instrumentation on bridges Delaware • DelDOT Flood Monitoring System brochure Georgia • BridgeWatch—flood monitoring website • USGS rapid deployment gages Idaho • Idaho Transportation Department (ITD) scour committee team to evaluate bridge to be monitored on BridgeWatch system Illinois • Extensive USGS gage system • BridgeWatch Iowa • BridgeWatch program to monitor real-time rainfall • USGS gages to provide alerts when flooding could impact infrastructure Kansas • Monitor NWS information/warnings and USGS streamflow gage data regularly • Field personnel and bridge inspectors respond quickly to events Louisiana • An excellent emergency plan and management • Conduct exercises and work to always improve our response Maine • Biggest effort is for snowmelt season • Coordinated effort with many state, federal, and county agencies, as well as private hydrology Nebraska • Nebraska DOT doesn’t have an “integrated” system • Use information from the weather service, stream gage data, and direct information from District personnel Nevada • Partners with the USGS to fund stream gages • Partners with local regional flood control districts to identify projects that have multi-agency benefit and can be integrated into transportation projects • Nevada DOT (NDOT) Hydraulic Division and Maintenance forces monitor the NWS website and the California Nevada River Forecast Center (using USGS data) to determine when and where NDOT maintenance forces are most needed during an event New Hampshire • WebEOC Incident Management managed by Homeland Security and Emergency Management • NH Silver Jackets • New Hampshire Department of Environmental Services Dam Bureau Operations Section • Transportation Management Center under the New Hampshire DOT Operations Division & Bridge inspection and monitoring during flood events New Jersey • New Jersey DOT does not have an “integrated” system—process is two separate components • Items contributing to prediction success include the NWS–Riverine Gauge readings and coastal forecasting • Long-serving operations personnel who know the locations that historically flood New York • Statewide Transportation Information and Coordination Center (STICC) watching for weather events and preparing responses to weather events • When upcoming events are anticipated people are deployed to address the situation, and others are on call if needed North Carolina • NCDOT Infrastructure Flood Data • Road flooding from storm surge: Data layers show the predicted amount of water over the roadway. • Flood Inundation Mapping and Alert Network for Transportation (FIMAN-T): A web-based tool based on forecasts from NWS that allows impacts to be visualized at the predicted peak flooding. • Flood Inundation Mapping and Alert Network (FIMAN): Website that provides real-time data on stream elevation, rainfall, and weather parameters from over 550 gages across North Carolina. • Multi-Frequency Flood Studies Road Inundation Mapping (East of 95): Identification of road flooding areas based on flood study areas. • NCDOT bridge span vulnerability: Identification of NCDOT bridge spans that are vulnerable to surge and wave forces. • Hurricane Briefs, Hurricane Tracking, Surge Data, Rainfall Data, and River Forecast • The full version of North Carolina’s response is in Appendix B. Ohio • Strong relationship with NWS and a paid weather forecasting service • Emergency management center that coordinates efforts Table 3. Reported plans or initiatives contributing to successful integrated systems. (continued on next page)

36 Practices for Integrated Flood Prediction and Response Systems State DOT Response Oklahoma • Our communications and cross-over people with Office of Emergency Management, U.S. Army Corp of Engineers, and cooperative gages with the USGS • An extensive Mesonet system of precipitation stations monitored by NOAA Pennsylvania • PennDOT’s data and information South Carolina • Improving pre-storm planning, documenting institutional knowledge from previous events, and meetings that include all parts of the DOT involved in pre- /during/post-response • Improve communication between DOT units, with other state agencies, and with federal agencies • Keep dedicated employees and the ability to quickly understand information and apply it • Expand use of rapid deployment gages (RDGs), USGS Gaging System, BridgeWatch, USGS river cams, traffic cams Texas • TxDOT is currently working on improving the National Water Model with research on the densification of the gage network Utah • Use UDOT application on phones to warn of flooding on the roads Vermont • Training and knowledge of existing systems • Have a few gage height predictions and inundation maps, but the general understanding and knowledge of these simple tools go a long way in being prepared Washington • WSDOT Office of Emergency Management’s website is tied to the NOAA gages to predict flood peaks • An email distribution is sent internally West Virginia • Frequent progress meetings with field offices • Constant communication with traffic management, field offices, and other agencies Wisconsin • Command Structure Emergency Transportation Operations Plan Table 3. (Continued). Total Number of DOT Responses: N = 25 1 (4%) 8 (32%) 9 (36%) 10 (40%) 12 (48%) 17 (68%) 17 (68%) 21 (84%) 0 5 10 15 20 25 Does not have an integrated system Enhanced common database Better understanding of the performance of flood conveyance systems Reduction in economic loss (e.g., reduced infrastructure repair and maintenance cost) Positive public feedback/trust (e.g., enhanced reliability on the state DOT issued warning, etc.) Improved and reliable relationship within state DOT offices Streamlined and collaborative inter-agencies communication Overall safety enhancement (e.g., reduction in death toll due to flood events, decreased crash frequency during heavy rain season, etc.) Number of DOTs: Figure 11. Reported observed benefits of DOT successful integrated systems. Survey respondents were allowed to select multiple answers.

Survey of State Practices for Integrated Flood Prediction and Response Systems 37   Flood Monitoring Figure 12 illustrates the reported DOT offices that are responsible for flood monitoring. Some of the DOTs noted that their flood monitoring is a joint effort with no one office or person in charge. Methods and practices: When asked about the methods and practices they use for flood monitoring, 24 DOTs (56%) reported using USGS’s National Water Information System for flood monitoring. Further description on methods and practices DOTs used for flood moni- toring is presented in Figure 13. Other methods and practices DOTs apply for flood monitoring are presented in Table 4, while additional information on and links to dynamic flood inundation maps or state models are presented in Table 5. 1 (2%) 3 (6%) 3 (6%) 4 (8%) 4 (8%) 5 (10%) 7 (15%) 8 (17%) 9 (19%) 9 (19%) 10 (21%) 11 (23%) 16 (33%) 0 2 4 6 8 10 12 14 16 18 Research office staff Consultants contracted to flood monitoring and deliver to DOT office Planning and/or programming office staff Bridge Division National Weather Service Our state does NOT monitor flooding at all Other (e.g., Construction, Division of Water, RWIS) Emergency management Maintenance Operations Local/district office Asset management office staff Hydraulics/hydrology office staff Number of DOTs: Total Number of DOT Responses: N = 48 Figure 12. Reported state DOT offices that are currently responsible for flood monitoring. Survey respondents were allowed to select multiple answers. 3 (7%) 3 (7%) 4 (9%) 6 (14%) 8 (19%) 8 (19%) 12 (28%) 24 (56%) 0 5 10 15 20 25 30 USGS, USGS gages State model BridgeWatch National Water Model NWS Dynamic Flood Inundation Map Other National Water Information System Number of DOTs: Total Number of DOT Responses: N = 43 Figure 13. Reported methods and practices DOTs apply for flood monitoring. Survey respondents were allowed to select multiple answers.

38 Practices for Integrated Flood Prediction and Response Systems Instruments and tools: The most commonly observed instrument and tool that DOTs use for flood monitoring are federal stream gages, which are often USGS stream gages. The distribu- tion of instruments and tools used for flood monitoring is shown in Figure 14. Table 6 provides information on other reported instruments and tools DOTs use for flood monitoring. Data type and storage: Forty-three state DOTs provided a response when asked what type (format) of data is collected to support flood monitoring. Gage/point data, staff reports, and images were the most commonly reported type (format) of data. Figure 15 presents the distribu- tion of the reported type (format) of data collected to support flood monitoring. State DOT Response California California Nevada River Forecast Center Connecticut News reports Minnesota Field monitoring Nebraska Nebraska Emergency Management Agency New Hampshire Operations Division monitoring and response New York Maintenance engineers watching hydraulically vulnerable bridges North Carolina Southeast River Forecast Center, multi-sensor precipitation estimates Pennsylvania Internal systems South Carolina Similar event data, Coastal Emergency Risks Assessment, river cams Table 4. Other reported state DOT methods and practices for flood monitoring. State DOT Dynamic flood inundation map State model Alaska Through the Silver Jackets Program, Alaska has seen some flood inundation mapping efforts in a few communities (Juneau, Matanuska-Susitna Borough, maybe others) Delaware Delaware Coastal Flood Monitoring System Idaho BridgeWatch Louisiana Use NOAA and USGS data North Carolina Google FIMAN NC Ohio Third-party platform Oklahoma Produced inundation maps this year with USACE; no link Pennsylvania Internal system South Carolina South Carolina Department of Natural Resources (SCDNR) will have a website in December 2020. Table 5. State information and links for dynamic flood inundation map or state model. 8 (19%) 11 (26%) 11 (26%) 14 (33%) 14 (33%) 16 (37%) 17 (40%) 21 (49%) 25 (58%) 41 (95%) 0 5 10 15 20 25 30 35 40 45 UAV/UAS Remotely sensed data Video Rain gages (non-federal) Other Radar Geographic information system (GIS) Stream gages (non-federal) Rain gages (federal) Stream gages (federal) Number of DOTs: Total Number of DOT Responses: N = 43 Figure 14. Reported instrument and tools DOTs use for flood monitoring. Survey respondents were allowed to select multiple answers (UAV = unmanned aerial vehicle).

Survey of State Practices for Integrated Flood Prediction and Response Systems 39   State DOT Response Alaska Local reports, local observations, satellite data Connecticut BridgeWatch Hawaii National Weather Service Massachusetts Tide gages Minnesota Flights Mississippi Information from local DOT maintenance personnel Nebraska NDOT cameras Ohio Third-party platform Oklahoma Mesonet Oregon Preventative, used in design, not during a flooding event Pennsylvania Internal systems South Carolina On-site measurements, RDGs, river cams, BridgeWatch, on-site pictures Wisconsin GIS Division of State Patrol (DSP) Table 6. Other reported instruments and tools used for flood monitoring. 8 (19%) 8 (19%) 17 (40%) 22 (51%) 24 (56%) 26 (60%) 28 (65%) 30 (70%) 31 (72%) 0 5 10 15 20 25 30 35 Other (e.g., internal systems, social media) LiDAR data Citizen reports Bridge scour Videos Emergency services reports Images Staff reports Gage data/point data Number of DOTs: Total Number of DOT Responses: N = 43 Figure 15. Reported types (formats) of data collected to support flood monitoring. Survey respondents were allowed to select multiple answers. The most common way to log and store flood data according to the 40 state DOTs that responded (83%) is in a database shared between state DOT divisions and offices (11 DOTs) followed by in a single unshared repository (9 DOTs). This information is presented in Figure 16 and Table 7. In Figure 17, the practices on flood data sharing across federal, state, and local organizations are presented. Data access: Of 40 state DOTs, 14 (35%) reported that a public website was available to view the DOT’s flood monitoring data. Table 8 presents the details provided by 13 states. The four DOTs that provide public access to the guidelines on flood monitoring are presented in Table 9. Success of systems: When asked about the DOT’s flood monitoring program assessment, 6 state DOTs (15%, of 35 DOTs responded) reported that they assess their flood monitoring program relative to its performance and accuracy. Both observation and user feedback are used almost equally to assess flood monitoring programs. Of the state DOTs that responded, 41 ranked the success of their state’s flood monitoring program from the perspective of their office. On a scale of 1 to 5 (where 1 = least effective and

40 Practices for Integrated Flood Prediction and Response Systems 2 (5%) 3 (8%) 5 (13%) 7 (18%) 9 (23%) 11 (28%) 12 (30%) 0 2 4 6 8 10 12 14 In an integrated structured database Multiple locations/databases Not stored In a database shared between other state and/or federal agencies In a single repository, unshared In a database shared between state DOT divisions and offices Other Number of DOTs: Total Number of DOT Responses: N = 40 Figure 16. Reported ways flood data are logged/stored. Survey respondents were allowed to select multiple answers. State DOT Response Alaska On bridge and culvert plans Connecticut BridgeWatch logs alert data in system accessible to department personnel; proposed integrated database to be developed in the future Hawaii On reports from bridge inspections Massachusetts State Emergency Management Minnesota Database in development New Hampshire Operations servers Ohio Enterprise Information Management System Oregon No dedicated system; shared as needed Pennsylvania Paper files South Dakota Local electronic files Table 7. Other reported ways flood data are logged/stored. 2 (5%) 3 (8%) 4 (10%) 5 (13%) 6 (15%) 8 (21%) 13 (33%) 19 (49%) 0 2 4 6 8 10 12 14 16 18 20 Better system needed Shared if requested Intranet platform Other (e.g., orally, data are not collected) Reporting through transcripts Online platform (please provide the website link) Data are not shared Email Number of DOTs: Total Number of DOT Responses: N = 39 Figure 17. Reported ways flood data are shared across federal, state, and local organizations. Survey respondents were allowed to select multiple answers.

Survey of State Practices for Integrated Flood Prediction and Response Systems 41   5 = very effective), 35 of 42 DOTs responded with a self-ranking of 3, 4, or 5, which is considered successful. Figure 18 illustrates the reported observed benefits of state DOT flood monitoring systems where improved emergency response and improved understanding of maintenance needs are ranked as top benefits. The survey also asked about challenges and weaknesses in implementing a successful flood monitoring system. As illustrated in Figure 19, limited state resources (e.g., staffing, funding, turnover) was the most reported factor. Gage coverage: Table 10 displays the state DOT reported gage coverage status of streams crossing roads. State DOT Response Arizona NOAA website Colorado Colorado Hazard Mapping & Risk MAP Portal Delaware DelDOT MAPS* Illinois USGS Current Water Data for Illinois Indiana USGS gage data Louisiana Our 511 site as it applies to road closures* Nebraska Nebraska 511—Traveler Information* North Carolina Flood Inundation Mapping and Alert Network (FIMAN) North Carolina (NC) North Dakota ND response Rhode Island NWS Weather Forecast Texas DriveTexas* Washington Flicker for photos West Virginia West Virginia 511* *Website managed by state DOT. Table 8. Reported links to public websites to view flood monitoring data. State DOT Response Alaska No policy and procedure exist to cover all flood risks and scenarios; however, Department of Transportation & Public Facilities (DOT&PF) has an “Incident Field Operations Guide” publication that attempts to cover multiple hazards, not just flooding. Minnesota Minnesota DOT Hydraulics Pennsylvania Pennsylvania DOT (PennDOT) Publications: Pub 238 and Pub 23 Texas TxDOT Inclement Weather Table 9. Reported documents and links to public access to guidelines on flood monitoring. 7 (23%) 12 (40%) 17 (57%) 20 (67%) 23 (77%) 23 (77%) 0 5 10 15 20 25 Other (e.g., better communication) Improved emergency evacuation Improved understanding of system performance Improved emergency planning Improved understanding of maintenance needs Improved emergency response Number of DOTs: Total Number of DOT Responses: N = 30 Figure 18. Reported observed benefits of the flood monitoring system. Survey respondents were allowed to select multiple answers.

42 Practices for Integrated Flood Prediction and Response Systems 4 (9%) 13 (28%) 14 (30%) 16 (35%) 17 (37%) 18 (39%) 19 (41%) 20 (43%) 28 (61%) 41 (89%) 0 10 20 30 40 50 Other Data timeliness Data accessibility Data format consistency Data completeness (contents) Data accuracy Technical expertise Local resources (staffing, funding, turnover, etc.) Data coverage (e.g., lack of stream gages) State resources (staffing, funding, turnover, etc.) Number of DOTs: Total Number of DOT Responses: N = 46 Figure 19. Reported challenges and weaknesses in implementing a successful flood monitoring system. Survey respondents were allowed to select multiple answers. Response Response rate Less than 10% 37 DOTs (80%) 10%–25% 6 DOTs (13%) 25%–50% 1 DOT (2%) 50%–75% 2 DOTs (4%) More than 75% 0 DOTs Table 10. Reported coverage status of the streams crossing roads equipped with gages. Assessment methods: The survey also sought information about the methods used in assess- ing roadway segments and bridges after a flood event. Figure 20 and Table 11 describe reported methods for assessing roadway segments, while Figure 21 represents the methods for bridge assessment. As shown in both figures, visual inspection was the most applied method. Flood Prediction A total of 48 DOTs responded regarding which state DOT offices are currently responsible for flood prediction as presented in Figure 22. Models and data: The survey sought to obtain the flood prediction models used by state DOTs. On the basis of 35 responses, the most common response to the type of model used is a hydrologic/hydraulic model as shown in Figure 23. Table 12 presents state DOTs that responded with other types of models. In addition, Figure 24 provides details on how a state DOT’s flood prediction model is devel- oped and managed; 31 DOTs responded. Information about the platform/model of the flood prediction model used by the state is shown in Figure 25. When asked about observations incorporated in the flood prediction model, stream stage/ discharge was the most frequently observed followed by precipitation, as shown in Figure 26. Details of the state DOTs that responded with “Other” observations are presented in Table 13. On the basis of the 27 DOTs responses, a timestep of 24 hours (10 DOTs) was the most fre- quently applied timestep of the precipitation input, while 7 DOTs reported 1-hour timesteps.

Survey of State Practices for Integrated Flood Prediction and Response Systems 43   2 (4%) 3 (6%) 5 (11%) 7 (15%) 8 (17%) 18 (38%) 22 (47%) 47 (100%) 0 10 20 30 405 15 25 35 45 50 UAV/drone Other Satellite image comparison LiDAR GIS External agency reporting Public reporting Visual inspection Number of DOTs: Total Number of DOT Responses: N = 47 Figure 20. Reported ways that roadway segments are assessed after a flood event. Survey respondents were allowed to select multiple answers. State DOT Response Florida Flooding can not only cause significant damage to roadway assets but also pose a safety threat to the road users. Thereby, flooded pavements are closed to traffic to restrict further deterioration of pavements and to ensure safety of the traveling public. Washington Maintenance staff Table 11. Other reported ways that roadway segments are assessed after a flood event. 3 (6%) 4 (9%) 4 (9%) 6 (13%) 14 (30%) 15 (32%) 31 (66%) 47 (100%) 0 5 10 15 20 25 30 35 40 45 50 Other (sonar, drone footage, bridge inspectors) Satellite image comparison LiDAR GIS Public reporting External agency reporting Bridge scour Visual inspection Number of DOTs: Total Number of DOT Responses: N = 47 Figure 21. Reported ways that bridges are assessed after a flood event. Survey respondents were allowed to select multiple answers.

44 Practices for Integrated Flood Prediction and Response Systems 1 (2%) 3 (6%) 3 (6%) 3 (6%) 4 (8%) 4 (8%) 8 (17%) 18 (38%) 18 (38%) 0 2 4 6 8 10 12 14 16 18 20 Research office staff Asset management office staff Bridge Section Operations Division Consultants contracted for flood prediction and delivery to DOT office Emergency Management and Maintenance Other (e.g., consultants, weather department) Hydraulics/Hydrology office staff Our state does NOT perform flood prediction/modeling at all Number of DOTs: Total Number of DOT Responses: N = 48 Figure 22. Reported State DOT offices currently responsible for flood prediction. Survey respondents were allowed to select multiple answers. 2 (6%) 2 (6%) 2 (6%) 4 (11%) 9 (26%) 18 (51%) 0 2 4 6 8 10 12 14 16 18 20 Hydrologic/hydraulic and threshold model BridgeWatch NOAA Threshold model Other Hydrologic/hydraulic model Number of DOTs: Total Number of DOT Responses: N = 35 Figure 23. Reported type of flood prediction model used by state DOTs. Survey respondents were allowed to select multiple answers. State DOT Response Massachusetts Currently being developed Nevada Models are project related rather than predictive Ohio Third-party platform Oklahoma Common sense and experience; no modeling will be done South Carolina Comparison to similar events South Dakota Monitor other agency information Utah Weather forecast model Table 12. Other reported flood prediction models used by state DOTs.

Survey of State Practices for Integrated Flood Prediction and Response Systems 45   3 (10%) 5 (16%) 6 (19%) 8 (26%) 13 (42%) 16 (52%) 0 5 10 15 20 By local governments By other state agency (please provide the agency) Other (e.g., Coastal Emergency Risk Assessment, National Flood Insurance Program, Delaware Coastal Flood Monitoring System, N/A) By an external consultant By a federal agency (e.g., USGS, FEMA, NOAA, NWS) In-house (by the state DOT’s office/division overseeing flood prediction) Number of DOTs: Total Number of DOT Responses: N = 31 Figure 24. Reported ways a state DOT’s flood prediction model is developed and managed. Survey respondents were allowed to select multiple answers. 4 (13%) 4 (13%) 6 (20%) 6 (20%) 14 (47%) 17 (57%) 0 5 10 15 20 Other Other state agency platform External platform hosted by a consultant In-house (by the state DOT’s office/division overseeing flood prediction) platform A national platform (e.g., HEC-RAS, SRH-2D) Federal agency (e.g., USGS, FEMA, NOAA, NWS) platform Number of DOTs: Total Number of DOT Responses: N = 30 Figure 25. Reported platform/model of the flood prediction model used by states. Survey respondents were allowed to select multiple answers. 6 (19%) 6 (19%) 9 (29%) 14 (45%) 16 (52%) 16 (52%) 19 (61%) 23 (74%) 0 5 10 15 20 25 Snowmelt depth Other Integrated (combining above attributes) Bridge scour Inundation area/depth Runoff Precipitation Stream stage/discharge Number of DOTs: Total Number of DOT Responses: N = 31 Figure 26. Reported observations incorporated into the flood prediction model. Survey respondents were allowed to select multiple answers.

46 Practices for Integrated Flood Prediction and Response Systems State DOT Response Massachusetts Standard federal products Nevada Models are projects, not flood prediction Oklahoma People and computers; we do not run models to predict floods, we design to withstand them. Table 13. Other reported observations incorporated into the flood prediction model. The survey asked for information about data being applied in state DOT flood prediction models according to different data sources (i.e., state, federal, and local and other state agen- cies). Details of the data application per data source are presented in Figures 27, 28, and 29, as well as in Table 14. Of the 30 state DOTs that responded to how often the model input data are updated, 9 DOTs stated that their model input data are updating continuously/ongoing (30%). The other reported responses to how often the flood prediction model input data are updated are presented in Table 15. Figure 30 presents the distribution of responses when asked about the flood prediction model update frequency. The other DOT responses to how often the prediction model is updated are presented in Table 16. Assessment and success: Regarding the flood prediction model assessment, the seven DOTs that reported that the flood prediction model is assessed relative to its performance and accuracy also reported on the success of the tool assessed. The most commonly used was observation (five DOTs) followed by user feedback (one DOT). One DOT reported using both observation and user feedback. When asked how often the tool is assessed, five DOTs (83%, of six DOTs responding) indicated that they assess the tool for every major storm event, while one DOT (17%) reported that it assesses the tool in years. Figure 31 presents the elements DOTs use to assess the flood prediction models. The most common responses were accuracy (seven DOTs) and applicability (four DOTs), followed by timeliness (three DOTs) and usability (two DOTs). 6 (20%) 7 (23%) 8 (27%) 9 (30%) 10 (33%) 15 (50%) 23 (77%) 0 5 10 15 20 25 Model flood outputs Images (e.g., inspector or stationary camera images) Remotely sensed data (e.g., satellite or aerial images, land cover data, etc.) LiDAR data Other Topographic data Gage data/point data Number of DOTs: Total Number of DOT Responses: N = 30 Figure 27. Reported federal data being applied in state DOT’s flood prediction models. Survey respondents were allowed to select multiple answers.

Survey of State Practices for Integrated Flood Prediction and Response Systems 47   6 (20%) 8 (27%) 9 (30%) 9 (30%) 9 (30%) 10 (33%) 12 (40%) 13 (43%) 14 (47%) 16 (53%) 0 5 10 15 20 Videos (e.g., inspector or stationary camera videos) Land use data Land cover data Citizen reports Other (e.g., survey data, N/A, unsure) LiDAR data Agency reports Images (e.g., inspector or stationary camera images) Gage data/point data Topographic data Number of DOTs: Total Number of DOT Responses: N = 30 Figure 28. Reported state DOT data being applied in state DOT’s flood prediction model. Survey respondents were allowed to select multiple answers. 4 (14%) 6 (21%) 6 (21%) 7 (24%) 7 (24%) 9 (31%) 10 (35%) 10 (35%) 11 (38%) 13 (45%) 0 2 4 6 8 10 12 14 Topographic data Citizen reports Videos (e.g., inspector or stationary camera videos) Land cover data Land use data Images (e.g., inspector or stationary camera images) Other (e.g., depends on local agency, N/A, unsure) Agency reports LiDAR data Gage data/point data Number of DOTs: Total Number of DOT Responses: N = 29 Figure 29. Reported data from local agencies and other state agencies (not DOT) being applied in DOT flood prediction models. Survey respondents were allowed to select multiple answers. State DOT Response Colorado State DOT does not have function but state agency Colorado Water Conservation Board (CWCB) might per survey respondent Iowa Iowa Flood Center sensors Oklahoma Inundation maps have been used South Carolina Uses USGS gage data, NWS rainfall data, NRCS [Natural Resources Conservation Service] soil data, and other data sources South Dakota Does not have a DOT model but uses other agency information Table 14. Other reported federal data being applied in DOT flood prediction models.

48 Practices for Integrated Flood Prediction and Response Systems State DOT Response Colorado CWCB Delaware University of Delaware working with Department of Natural Resources and Environmental Control Hawaii Usually done in the design process and not usually updated Maryland As needed Nebraska When site conditions change Nevada, Oklahoma Models are not used for flood prediction New York When bridges are replaced North Dakota Project-by-project basis South Carolina SCDNR Utah No data are being inputted for the DOT Table 15. Other reported responses to how often the flood prediction model input data are updated. Ongoing 12 DOTs Needs to be updated but no formal plan for update 3 DOTs Other 15 DOTs Ongoing Needs to be updated but no formal plan for update Other Figure 30. Reported response to how often the flood prediction model is updated (30 responses). State DOT Response California Model is updated only once Colorado CWCB Hawaii Usually done in the design process and not usually updated Iowa As new versions are released Maryland As needed Massachusetts, Utah Model is not updated Nebraska When site conditions change North Dakota Project-by-project basis Oklahoma, Nevada Models are not used for flood prediction South Carolina SCDNR Table 16. Other reported responses to how often the flood prediction model is updated.

Survey of State Practices for Integrated Flood Prediction and Response Systems 49   2 (29%) 3 (43%) 4 (57%) 7 (100%) 0 1 2 3 4 5 6 7 8 Usability Timeliness Applicability Accuracy Number of DOTs: Total Number of DOT Responses: N = 7 Figure 31. Reported elements the tool assessment reviews. Survey respondents were allowed to select multiple answers. When asked about the state DOT’s self-ranking regarding its flood prediction model success from the perspective of its office on a scale of 1 to 5 (where 1 = least effective and 5 = very effective), 18 of 30 DOTs responded with a ranking of 3, 4, or 5, which is considered successful. Eighteen DOTs responded by providing information about what makes their flood predic- tion model a success as presented in Figure 32. The type of information output/usability was referred as the most commonly observed flood prediction model success factor, followed by the accuracy of the model output. The Pennsylvania DOT reported that the accuracy of the output is dependent on continually refining input data. Limitations: Forty-six DOTs responded about which issues (e.g., limitations, constraints) have been observed in implementing the flood prediction model. The most common responses include state resources (e.g., staffing, funding, turnover), followed by technical expertise and data coverage. The distribution of responses is presented in Figure 33. Table 17 presents the other issues (e.g., limitations, constraints) observed in implementing the flood prediction model provided by DOTs. 2 (11%) 7 (39%) 7 (39%) 13 (72%) 0 2 4 6 8 10 12 14 Other Ease of use Accuracy of output Type of information output/usability Number of DOTs: Total Number of DOT Responses: N = 18 Figure 32. Reported information that makes flood prediction models a success. Survey respondents were allowed to select multiple answers.

50 Practices for Integrated Flood Prediction and Response Systems 4 (9%) 6 (13%) 6 (13%) 7 (15%) 10 (22%) 11 (24%) 12 (26%) 14 (30%) 15 (33%) 18 (39%) 36 (78%) 0 10 20 305 15 25 35 40 Data timeliness Difficulty in integrating federal resources Data completeness (contents) Data accessibility Data accuracy Other Data format consistency Local resources (staffing, funding, turnover, etc.) Data coverage Technical expertise State resources (staffing, funding, turnover, etc.) Number of DOTs: Total Number of DOT Responses: N = 46 Figure 33. Reported issues (e.g., limitations, constraints) observed in implementing the flood prediction model. Survey respondents were allowed to select multiple answers. DOT Response District of Columbia In the District context, it is the Department of Energy and Environment that is delegated to be flood managers Kansas Do not have a flood prediction model due to having technical expertise, data format consistency, data coverage, and data timeliness Maine, New Hampshire, Oklahoma Does not have flood prediction model Massachusetts Any federal limitations to flood prediction Nevada Models are for projects, not flood prediction Table 17. Other reported issues (e.g., limitations, constraints) observed in implementing the flood prediction model. Flood Warning Systems All 48 DOTs (47 state DOTs and the District of Columbia) responded to which state DOT offices are currently responsible for their flood warning system, as shown in Figure 34 and Table 18. Guidelines and thresholds: Seven DOTs reported having an established flood warning protocol, published guidelines, or both in place, six of which provided more information or links to those established flood warning protocol and published guidelines, as shown in Table 19. Table 20 provides a description of the threshold parameter/data that state DOTs apply to issue a flood warning. Regarding which state DOT offices are engaged in deciding when the flood warning threshold is met, 36 DOTs responded. The most common response by 15 DOTs was operations office staff, as shown in Figure 35, with descriptions of reported external offices/agencies presented in Table 21. Information: State DOTs were also asked to rank the importance of information used within their state’s flood warning system, using a scale of one to eight stars, with eight stars being of highest importance. Results of the survey show that DOTs rank observed stream stage, observed stream inundation, and bridge scour as the most important ranked information used in their

Survey of State Practices for Integrated Flood Prediction and Response Systems 51   1 (2%) 1 (2%) 2 (4%) 2 (4%) 2 (4%) 3 (6%) 4 (8%) 4 (8%) 4 (8%) 6 (13%) 13 (27%) 15 (31%) 15 (31%) 0 2 4 6 8 10 12 14 16 Planning and/or Programming office staff Research office staff Safety office staff Bridge Section Hydraulics Section Risk management office staff Emergency Management Asset management office staff Consultants contracted to manage the flood warning system Design office staff No office, no implementation of flood warning system at all Other Operations office staff Number of DOTs: Total Number of DOT Responses: N = 48 Figure 34. Reported state DOT offices currently responsible for flood warning system. Survey respondents were allowed to select multiple answers. DOT Response Delaware Combination of Delaware DOT (DelDOT), Department of Natural Resources and Environmental Control, and University of Delaware District of Columbia Department of Energy and Environment Florida NWS and NOAA gages Kentucky Kentucky Emergency Management (KYEM) and Division of Water (DOW) to public and elected officials Louisiana State homeland security Nevada Maintenance staff in districts Oregon Region Rhode Island Natural Resourced Unit Utah Weather Department Wisconsin Technical Oversight Committee (TOC) Table 18. Other reported state DOT offices currently responsible for flood warning system. State DOT Link to website Delaware Delaware Coastal Flood Monitoring System Hawaii Hawaii DOT has an emergency response plan. Kansas We have Bridge Scour Action Plans for scour critical bridges. Louisiana It’s in our state’s emergency plan. Maine Scour critical bridge program New York New York State (NYS) Hydraulic Vulnerability Manual Table 19. Reported links to established flood warning protocol and published guidelines.

State DOT Link to website Alaska I believe many Department of Transportation & Public Facilities (DOT&PF) staff rely on the Alaska Pacific River Forecast Center’s website for tracking floods. The presentation of hazard level is based upon locally defined thresholds. California, Georgia, Kansas, Oklahoma Varies by site/case Delaware Flood inundation “potential” map generated by assuming the water level at the coast is projected inland to the community region and comparing that to the digital elevation model (DEM) elevation. This is typically called a “bathtub” model, as the water depth is based on a constant water level surface, analogous to filling a bathtub. We use the term “potential” as this method tends to over- estimate the inundated areas. In addition to generating the flood inundation map for a community, also create a time series of forecasted tidal predictions. Likewise, overlay the maximum forecasted water level for a community on road elevation profiles for transportation routes in the area. These routes were selected because they were either a Delaware DOT (DelDOT) designated evacuation route during times of coastal flooding or a typically heavily trafficked road. Between one and three roads were selected for each community. Hawaii, Iowa, Nebraska, New Hampshire, North Dakota, Ohio, Pennsylvania, Utah NWS Idaho Discharge of 25% and up Louisiana The biggest one we have is our plan for the New Orleans area. It is triggered by a Category 3 or greater. Maine HEC-RAS Massachusetts State Emergency Management provides data Missouri Comparison of historic data to anticipated impacts. Additionally, research project partners provide new innovations. Nebraska Nebraska Emergency Management Agency New York Stage height based on gage or observation North Carolina, South Carolina BridgeWatch Texas Stage to low chord. Observed by the districts. Washington, West Virginia NOAA West Virginia U.S. Army Corps of Engineers and other agencies Wisconsin Water approaching roadway traveled lanes, bridges where bridge superstructure is close to inundation, or washed-out roadways and culverts. Table 20. Reported threshold/parameter data that DOT applies to issue a flood warning. 1 (3%) 1 (3%) 2 (6%) 2 (6%) 2 (6%) 2 (6%) 3 (8%) 3 (8%) 5 (14%) 7 (19%) 9 (25%) 15 (42%) 0 2 4 6 8 10 12 14 16 Risk management office staff Asset management office staff Safety office staff Hydraulics Section Emergency Management Local district office Bridge Maintenance Other (Case by case, weather forecast, coordinate and posted on 511) Design office staff N/A, no contact External office/agency (e.g., Governor’s Office) Operations office staff Number of DOTs: Total Number of DOT Responses: N = 36 Figure 35. Reported DOT offices engaged in deciding when the threshold is met. Survey respondents were allowed to select multiple answers.

Survey of State Practices for Integrated Flood Prediction and Response Systems 53   DOT Response Alaska Public interests may dictate the threshold Colorado Colorado Water Conservation Board (CWCB) District of Columbia Homeland Security and Emergency Management Agency Louisiana Governor’s Office of Homeland Security Massachusetts Governor/state emergency management office Michigan State police North Carolina Emergency Management North Dakota National Weather Service Pennsylvania Governor’s Office Table 21. Reported external office/agency engaged in deciding when a threshold is met. DOT’s flood warning system. In contrast, model precipitation and model inundation informa- tion are typically ranked as relatively less important than the aforementioned factors by the DOTs. Figure 36 presents the distribution of importance rankings for information used in DOT flood warning systems that ranked between Rank 5 and Rank 8. For further detailed informa- tion, Table 22 provides individual state’s ranking status. Success of systems: The survey also sought to gather information about the state DOT’s self- rated flood warning effectiveness and accuracy, respectively, on a scale of 1 to 5 (where 1 = least effective and 5 = very effective). For each ranking case, approximately 60% of the DOTs responded (36 DOTs) with a ranking of 3, 4, or 5 for their flood warning system, which is considered successful. 1 (2%) 2 (5%) 3 (8%) 1 (2%) 1 (2%) 1 (2%) 3 (8%) 4 (11%) 3 (8%) 4 (11%) 4 (11%) 3 (8%) 2 (5%) 6 (16%) 6 (16%) 6 (16%) 3 (8%) 2 (5%) 2 (5%) 2 (5%) 2 (5%) 2 (5%) 5 (14%) 4 (11%) 16 (34%) 5 (14%) 2 (5%) 15 (32%) 19 (40%) 11 (30%) 9 (24%) 12 (32%) 0 5 10 15 20 Bridge scour Model inundation Model precipitation Observed stream inundation Observed stream stage Observed precipitation Forecast stream stage Forecast precipitation Number of DOTs: Rank 8 Rank 7 Rank 6 Rank 5 Total Number of DOT Responses: N = 37 Figure 36. Reported importance rankings (where 1 = least important and 8 = most important) of information used in DOT’s flood warning system.

54 Practices for Integrated Flood Prediction and Response Systems DOT Forecast precipitation Forecast stream stage Observed precipitation Observed stream stage Observed stream inundation Model precipitation Model inundation Bridge scour Alaska 6 6 6 8 8 5 6 8 California 7 6 4 4 2 2 N/A 5 Delaware 6 6 4 8 8 4 8 6 District of Columbia 7 7 6 4 7 4 5 6 Georgia 8 8 8 8 8 N/A N/A 8 Hawaii 6 N/A 6 6 6 N/A N/A 4 Idaho 4 N/A N/A 8 N/A N/A N/A 6 Iowa 5 7 2 8 8 6 N/A N/A Kansas 7 8 7 7 7 N/A N/A 7 Louisiana 8 8 8 8 5 5 5 8 Maine 6 2 6 8 4 N/A N/A 8 Maryland 6 4 4 6 6 7 7 7 Massachusetts 8 8 8 8 8 8 8 8 Michigan N/A N/A N/A N/A N/A N/A N/A 8 Missouri 8 7 8 8 8 6 6 8 Nebraska 8 8 8 8 2 2 8 8 New Hampshire 7 7 7 7 8 1 1 8 New York 6 6 6 8 8 4 4 7 North Carolina 8 8 8 8 8 7 8 8 North Dakota 8 8 8 8 8 8 8 8 Ohio 8 6 6 5 4 3 4 3 Oklahoma 8 8 8 8 8 N/A N/A 4 Pennsylvania 5 5 8 8 8 6 6 8 Rhode Island 5 5 N/A N/A N/A N/A N/A N/A South Carolina 4 6 4 8 6 3 3 N/A Texas 5 5 8 8 8 6 6 8 Utah 8 2 4 1 1 3 2 N/A Washington 8 8 5 8 8 N/A N/A 8 West Virginia 8 7 8 8 8 5 7 8 Wisconsin 8 8 8 8 8 N/A N/A 8 Table 22. Reported importance rankings system (1 = least important and 8 = most important) of information used in DOT’S flood warning system. When asked about the key attributes and factors that have led to the success of the flood warning system at their agency, the most common key attribute/factor was an accurate flood monitoring system. Figure 37 presents the distribution of 23 DOT responses. To further obtain information about state DOTs’ effective communication systems (i.e., com- munication, internal communication, inter-agency communication, and public communica- tion), DOTs were asked to elaborate on such systems, as presented in Table 23. Observed benefits: The survey sought to identify the observed benefits of having a state DOT flood warning system. As indicated in Figure 38, the most common was improved emergency response, followed by improved emergency evacuation and flood damage mitigation (reducing the amount of damage due to a flood).

Survey of State Practices for Integrated Flood Prediction and Response Systems 55   4 (17%) 4 (17%) 5 (22%) 8 (35%) 10 (43%) 10 (43%) 13 (57%) 0 2 4 6 8 10 12 14 Effective inter-agency communication system Effective public communication system Other (e.g., people and computers, diligent staff) Effective internal communication system Effective communication system Accurate flood prediction system Accurate flood monitoring system Number of DOTs: Total Number of DOT Responses: N = 23 Figure 37. Reported key attributes and factors that have led to the success of the flood warning system. Survey respondents were allowed to select multiple answers. Issues: The main issues observed in the state flood warning system that DOTs reported were the availability of state resources (e.g., staffing, funding, turnover), followed by technical exper- tise. The distribution of reported observed issues in the flood warning system are presented in Figure 39. Flood Response Systems A total of 48 DOTs provided answers to which offices are currently responsible for the flood response systems. Figure 40 shows collected responses from 48 DOTs. When asked about other agencies involved when dealing with major flood events, the majority of DOTs included the State Emergency Management Agency. The distribution of agencies involved when dealing with major flood events is presented in Figure 41. Structure and effectiveness: Additionally, the survey asked about the way the internal commu- nication system is structured to respond to flood events and its effectiveness. Figure 42 presents a distribution of the ways that the state DOT’s internal communication system is structured to respond to flood events, on the basis of 47 DOT responses. For the self-rated internal communi- cation system effectiveness, on a scale of 1 to 5 (where 1 = least effective and 5 = very effective), approximately 70% responded with a ranking of 3, 4, or 5, which is considered successful. Figure 43 presents the trends in DOT practices and approaches for communicating and working with other state agencies when responding to flood events; Table 24 provides detailed information. Similar to the internal communication effectiveness ranking result, for the inter-agency commu- nication effectiveness, approximately 70% responded with a ranking of 3, 4, or 5, which is consid- ered successful. In addition, 30 DOTs indicated that both internal and inter-agency communication systems are considered effective. Tool and methods: For the tools and methods available to alert the public in response to flooding events, 47 DOTs provided responses, which are summarized in Figure 44. For the public communication system experience self-ranking, approximately 80% of DOTs indicated a successful system. The overall successful communication system effectiveness across various entities is summarized in Table 25.

56 Practices for Integrated Flood Prediction and Response Systems State DOT Effectiveness of general communication system Effectiveness of internal communication system Effectiveness of inter-agency communication system Effectiveness of public communication system California Email to maintenance. Georgia System attributes lead to quicker response times. System attributes lead to quicker response times. Iowa Requires communication between field and design staff. Louisiana Critical to responding. System coordination is critical. Critical for travelers and their safety. Maine An established communication system between headquarters ops center, Bridge Maintenance, and crews on the road. Maryland Provided by the Coordinated Highways Action Response Team. Massachusetts Ensuring local understanding (re: evacuation). Includes the use of mobile phones. Includes activation of the Emergency Operations Center. Includes public broadcast and social media. Nebraska Relationships have been maintained between weather service and Nebraska Emergency Management Agency and district staff. New York Done by maintenance (field engineers) who report back to regional hydraulic engineers. Done by maintenance (field engineers) who report back to regional hydraulic engineers. North Carolina Done by partnership with Emergency Management and monthly communication. North Dakota Open communication lines. Effective communication system: North Dakota WebEOC. Effective public communication system: North Dakota WebEOC. South Carolina Communication between different DOT offices and state agencies. Requires making sure information is provided across the DOT. Being able to communicate with the correct personnel at other state and federal agencies. Being able to communicate evaluations and closures. Washington A part of the emergency response center. Table 23. Elaboration of communication system as a key attribute and factor that has led to the success of flood warning system.

Survey of State Practices for Integrated Flood Prediction and Response Systems 57   2 (9%) 2 (9%) 3 (13%) 6 (26%) 10 (43%) 19 (83%) Public safety Local knowledge Other (compliance with scour program, computer programs, increased response time and implementation) Flood damage mitigation (reducing the amount of damage due to a flood) Improved emergency evacuation Improved emergency response 0 5 10 15 20Number of DOTs: Total Number of DOT Responses: N = 23 Figure 38. Reported observed benefits of having flood warning system. Survey respondents were allowed to select multiple answers. Total Number of DOT Responses: N = 46 4 (9%) 4 (9%) 5 (11%) 7 (15%) 7 (15%) 8 (17%) 8 (17%) 9 (20%) 11 (24%) 13 (28%) 14 (30%) 14 (30%) 31 (67%) 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 Communication among other state… Communication within the state DOT Difficulty in integrating federal resources Data format consistency Data timeliness Data completeness (contents) Data accessibility Data accuracy Local resources (staffing, funding, turnover, etc.) Data coverage Other (e.g., N/A, unknown) Technical expertise State resources (staffing, funding, turnover, etc.) Number of DOTs: Figure 39. Reported issues (e.g., limitations, constraints) observed in the flood warning system. Survey respondents were allowed to select multiple answers.

58 Practices for Integrated Flood Prediction and Response Systems 4 (8%) 4 (8%) 4 (8%) 4 (8%) 4 (8%) 5 (10%) 6 (13%) 24 (50%) 28 (58%) 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 Risk Management Office Safety Office Staff Asset Management Office Staff Bridge Maintenance Other (e.g., State Police, Field Offices) Maintenance District/Regional Offices Emergency Management Office Operations Office Staff, Transportation Management Center Staff Number of DOTs: Total Number of DOT Responses: N = 48 Figure 40. Reported state DOT offices that are currently responsible for flood response system. Survey respondents were allowed to select multiple answers. 2 (4%) 4 (9%) 5 (11%) 15 (32%) 20 (43%) 23 (49%) 29 (62%) 33 (70%) 35 (75%) 37 (79%) 41 (87%) 41 (87%) 45 (96%) 0 5 10 15 20 25 30 35 40 45 50 FHWA USGS Other (e.g., NWS, USACE, County Police, Division of Fire, etc.) Division of Fish and Wildlife Public Health Department of Water Resources, Bureau of Water Department of Environmental Protection National Guard Governor's Office FEMA State Police Local Government State Emergency Management Agency Number of DOTs: Total Number of DOT Responses: N = 47 Figure 41. Reported other agencies involved when dealing with major flood events. Survey respondents were allowed to select multiple answers.

Survey of State Practices for Integrated Flood Prediction and Response Systems 59   2 (4%) 4 (9%) 4 (9%) 11 (23%) 18 (38%) 26 (55%) 26 (55%) 32 (68%) 0 5 10 15 20 25 30 35 WebEOC Internal hotline Other Automated text among internal members Existing internal emergency contact tree Emergency management center Conference calls Email list Number of DOTs: Total Number of DOT Responses: N = 47 Figure 42. Reported ways state DOT internal communication system is structured to respond to flood events. Survey respondents were allowed to select multiple answers. 4 (9%) 10 (21%) 10 (21%) 15 (32%) 0 5 10 15 20 Emergency response team Emergency operations center Agency sharing (i.e., homeland security, police, etc.) Calls and emails Number of DOTs: Total Number of DOT Responses: N = 47 Figure 43. DOTs’ reported practices and approaches for communicating and working with other state agencies when responding to flood events. Guidelines: Table 26 shows the 15 DOTs that provided established guidelines information related to flood responses. When asked about the self-rated flood event response system effec- tiveness, approximately 72% of DOTs indicated a successful one. Success of systems: The survey asked for information on key attributes and factors that have led to the success of the flood response system at the state DOT. The majority of state DOTs identified cooperation from local agencies in the state, adequate state resources (in staffing and funding), and a continuous public awareness campaign regarding the emergency response and warning systems as key factors and attributes, as shown in Figure 45. In addition, for DOTs that indicated having a successful emergency response system, the survey sought to obtain methods used to facilitate the emergency response system. This infor- mation is summarized in Figure 46. Reported issues: When asked about the observed issues regarding the flood response system, lack of resources (e.g., staffing, funding, technical expertise) was identified as a main issue. Figure 47 and Table 27 provide further details of survey responses. Effectiveness and future programs: Regarding the effective flood event response system, 21 DOTs indicated that such a system is already in place, whereas 10 DOTs reported their future program development status. This information is presented in Table 28.

60 Practices for Integrated Flood Prediction and Response Systems DOT Communication practices and approaches Alaska For large area-wide emergencies (e.g., earthquakes, extreme floods), the state of Alaska would mobilize a command center to which all engaged department staff would report. Colorado International control systems Connecticut Information and agency expertise sharing Delaware DelDOT has an integrated response system that continually coordinates with surrounding states and internal state agencies through Transportation Management Centers based on the Transportation Incident and Event Management Plan. District of Columbia In cases of major floods, the District of Columbia (DC) Homeland Security and Emergency Management Agency coordinates response efforts. In smaller flooding conditions, District Department of Transportation coordinates with DC Water—which manages and maintains the drainage network—Metropolitan Police Department, Washington Metropolitan Area Transit Authority, Fire and Emergency Medical Service (EMS) Department, VDOT. Florida Emergency Operations Center, state emergency response team calls and briefings, water management district and NWS briefings, and USACE briefings. Georgia State Operations Center activation with multiple agencies under direction of Emergency Management Agency. Hawaii DOT would work with the other agencies, such as the emergency response (county) and state emergency response teams. Idaho Telephone Iowa Collaboration between districts; Design and Emergency Operations to provide a common line of communication and decision making regarding flood events. Kansas Contact those needed Louisiana Conference calls and WebEOC to track all efforts in the response. Maine Email, text, dedicated contacts Maryland Emails and multi-agency meetings Massachusetts Contacts have been established using the structure of the Emergency Management support teams. Michigan Some staff that are part of the emergency response team. Minnesota Rarely work with others in response. Missouri Collaborative situation awareness calls, joint EOC (Emergency Operations Center) operations Nebraska Communicate with adjacent DOTs to work out emergency plans and detour routes. New Hampshire Email notifications, text messaging, WebEOC incident management system, media. New Jersey Excellent collaboration and communication, conference calls. New Mexico Email, personal relationships (if available). New York Works with Department of Environmental Conservation (DEC) for flood mitigation and flood studies. North Carolina Emergency Event Command Center if extreme event—hurricane. North Dakota Resource request system from individual agencies. Regular situational updates. Ohio We open our emergency management center and communicate there. Pennsylvania Internal protocols under the emergency management plan. Rhode Island News website; email South Carolina All state agencies have representatives at the South Carolina (SC) Emergency Management Division command center; direct communication between personnel; email. South Dakota In extreme conditions that are impacting a wide area across the state, the Office of Emergency Management and the Governor’s Office will activate the EOC (Emergency Operations Center). EOC will manage and coordinate resources, including the needs from the local entities. During less severe or regional events, email and conference calls will be used to coordinate response. Texas Participate and communicate with the Emergency Response Command Center. Utah Shared information and communication to reduce damage and prevent harm to people. Vermont Email lists and conference calls Washington WSDOT Office of Emergency Management West Virginia West Virginia Division of Homeland Security serves as a hub with a great deal of information supplied by the DOT Traffic Management staff. Wisconsin Agency-to-agency communication happens at the State Emergency Operations Center (EOC). Wyoming Phone calls, email Table 24. Reported DOT practices and approaches for communicating and working with other state agencies when responding to flood events.

Survey of State Practices for Integrated Flood Prediction and Response Systems 61   2 (4%) 5 (11%) 7 (15%) 9 (19%) 10 (21%) 14 (30%) 20 (43%) 21 (45%) 26 (55%) 33 (70%) 36 (77%) 37 (79%) 38 (81%) 0 5 10 15 20 25 30 35 40 TADD Gates RSS feed Other (e.g., Road Info, TripCheck) Dispatch lines Turn Around Don’t Drown… Emails Text messages Mobile application 511 systems Signs Television Social media (e.g., Facebook) Radio Number of DOTs: Total Number of DOT Responses: N = 47 Figure 44. Reported communication tools and methods available to alert the public in response to flooding events. Survey respondents were allowed to select multiple answers. Effectiveness scale* Number of DOTs: internal communication Number of DOTs: inter- agency communication Number of DOTs: public communication 3 7 9 7 4 19 16 20 5 7 7 11 * On a scale of 1 to 5 (where 1 = least effective and 5 = very effective). Table 25. Comparison of rankings of experience with internal, inter-agency, and public communication systems. DOT Document or link Alaska See DOT&PF’s Incident Field Operations Guide. Delaware When alerted or notified a road is approaching flood stage, the road is closed. Some floods are tidal, subject to lunar cycle and storm surge. In other places, residents just know to be aware of water encroachments at certain times of the day or cycle. At this time, Delaware does not have any scour critical bridges that may need to be closed well in advance of cresting. Delaware also has hurricane evacuations that occur days ahead of the storm making landfall. District of Columbia In draft form Florida State Comprehensive Emergency Management Plan Hawaii Hawaii DOT has guidelines for scour critical bridges. Iowa Flooding Standard Operation Procedure Louisiana State emergency response manual Michigan Scour action plans New Hampshire DOT work instructions and WebEOC incident management North Carolina Would need to get from Emergency Management. Pennsylvania Pub 238, Pub 23 South Dakota State Emergency Operation Plan is a binder hard-copy document. Texas DriveTexas Washington Washington DOT (WSDOT) Office of Emergency Management Wisconsin Emergency Transportation Operations Table 26. Reported guidelines related to flooding event responses.

62 Practices for Integrated Flood Prediction and Response Systems 5 (15%) 8 (24%) 10 (29%) 12 (35%) 13 (38%) 16 (47%) 17 (50%) 17 (50%) 19 (56%) 22 (65%) 0 2 4 6 8 10 12 14 16 18 20 22 24 Other (e.g., experience, N/A) Gained public trust through successful and accurate flood prediction system Provision of relevant training for practitioners (state, local, tribal nations, planning… Having real-time or near real-time model results/observations Holding peer exchanges for practitioners (state, local, tribal nations, planning… Establishment and distribution of a clear emergency… Consensus from stakeholders regarding the importance of the initiative Continuous public awareness campaign regarding the… Adequate state resources (in terms of staff and/or funding) Cooperation from local agencies in the state Number of DOTs: Total Number of DOT Responses: N = 34 Figure 45. Reported key attributes and factors that have led to the success of the flood response system. Survey respondents were allowed to select multiple answers. 6 (18%) 6 (18%) 7 (21%) 10 (29%) 13 (38%) 19 (56%) 23 (68%) 0 5 10 15 20 25 Workshops developed and delivered to practitioners (through LTAP) Other Joint funding for emergency response efforts Peer exchanges held with other states Committee established that oversees and contributes to effective emergency response initiatives Involvement from the FHWA Division office or other FHWA assistance Frequent coordination meetings among the stakeholders Number of DOTs: Total Number of DOT Responses: N = 34 Figure 46. Reported methods used to facilitate the emergency response system. Survey respondents were allowed to select multiple answers (LTAP = local technical assistance program).

Survey of State Practices for Integrated Flood Prediction and Response Systems 63   2 (4%) 4 (9%) 6 (13%) 8 (17%) 13 (28%) 16 (35%) 24 (52%) 0 2 4 6 8 10 12 14 16 18 20 22 24 26 Data security Lack of stakeholder commitment/willingness to invest and/or engage Lack of prioritizing from state leadership Lack of concern or understanding of flood risk Lack of commonality of databases Other Lack of resources (staffing, funding, technical expertise, etc.) Number of DOTs: Total Number of DOT Responses: N = 46 Figure 47. Reported reasons for ineffectiveness of response system to flood events. Survey respondents were allowed to select multiple answers. DOT Response Alaska “Our informal ‘system’ may be imperfect, but it seems to be effective. Best to have well-funded Maintenance & Operations personnel available for immediate flood response.” Colorado Not a DOT priority. Delaware “I am sure we could use more gages. As sea level rise (SLR) continues more road sections will need to be monitored.” District of Columbia System has not been tested yet. Kansas Lack of in-house statewide system. Maine Focus on snowmelt season seems to be working. Table 27. Other reasons reported for ineffectiveness of response system to flood events. DOT Comment Connecticut Still in planning. District of Columbia District Department of Transportation Flood Emergency Response Plan. Hawaii Hawaii would like to incorporate real-time scour monitoring for scour critical bridges. A contract is programmed in the budget for 2022. Minnesota Working to establish response program. Nebraska Emergency Response Manual. North Carolina FIMAN-T (FIMAN for Transportation), BridgeWatch. Oregon In progress. Pennsylvania PennDOT’s emergency response program is continually being reevaluated and upgraded; it is not a finished product and will continue to evolve as possible. Texas Streamflow measurement at TxDOT Bridges to Improve the National Water Model. Utah Utah is trying to create a research project to try to predict debris flows from burn scars. Table 28. Reported future programs under development to improve effectiveness of flooding event response system.

64 Practices for Integrated Flood Prediction and Response Systems Summary The high response rate of this survey (94%) allowed a comprehensive view of DOT flood management practices across the United States. The 48 DOTs that responded to this survey pro- vided input on their DOT’s current status of (a) flood event management, (b) flood monitoring, (c) flood prediction, (d) flood warning systems, and (e) flood response systems. Approximately half of the DOTs that responded reported that they consider their integrated system for flood pre- diction and response successful. Common plans or initiatives that have contributed to successful systems are the use of USGS gages, the importance placed on communication between agencies, and BridgeWatch. The observed benefits of successful integrated systems were an overall safety enhancement (e.g., reduction in death toll due to flood events, decreased crash frequency during heavy rain season), streamlined and collaborative inter-agency communication, and an improved and reliable relationship within state DOT offices. On the basis of the 43 DOTs’ responses, approximately half of the state DOTs reported using USGS’s NWIS for flood monitoring, and nearly all of the DOTs (41 of 43 DOTs, 95%) indicated that they use federal stream gages as an instrument and tool for flood monitoring. The other most common instruments and tools DOTs used for flood monitoring are federal rain gages, non-federal stream gages, and GIS. The DOTs provided information on risk thresholds that trigger action. Common responses from DOTs indicated that NWS issues the warnings and parameters that DOTs apply to issue a flood warning. According to a self-ranking, the most important information used in a state’s flood warning system is observed stream stage, observed stream inundation, and bridge scour. The most common method DOTs use in determining the extent and severity of flooding is visual inspection, according to 47 DOTs responses. Public reporting, external agency reporting, and application of sonar sensor and drone footage are also identified as methods used in assessing the extent and severity of flooding on infrastructure. For flood prediction methods, a hydrologic/hydraulic model was identified as the most com- mon type of flood prediction model used by DOTs. For the platform/model of the flood predic- tion model, most DOTs reported that the flood prediction model is produced by a federal agency (e.g., USGS, FEMA, FHWA, NOAA, NWS, USACE, Bureau of Reclamation). The survey showed that the most widely used communication tools and methods to alert the public in response to flooding events are radio, social media (e.g., Facebook, Twitter, Instagram), television, signs, and 511 systems. For communicating and working with other state agencies when responding to flooding events, direct calls and emails, agency sharing (i.e., homeland security, police), and emer- gency operations centers were identified practices and approaches applied by the DOTs. Email lists, conference calls, and an emergency management center were identified as being frequently applied in DOT internal communication systems that are structured to respond to flood events. Cooperation from local agencies in the state, adequate state resources (in staffing and funding), and a continuous public awareness campaign regarding the emergency response and warning systems were reported as key factors and attributes that have led to the success of flood response systems. The DOTs identified the most common methods used to facilitate emergency response sys- tems: (a) frequent coordination meetings among the stakeholders, (b) involvement of the FHWA division office or other FHWA assistance, and (c) a committee established to oversee and contribute to effective emergency response initiatives. Many DOTs indicated state resources (e.g., staffing, funding, turnover), data, and technical expertise as common issues (e.g., limitations, constraints) in implementing a flood prediction model and a flood warning system. Reported challenges and weaknesses to implementing a successful flood monitoring system were state resources (e.g., staffing, funding, turnover), data coverage (e.g., lack of stream gages), and local resources (e.g., staffing, funding, turnover).

Next: Chapter 4 - Case Examples of Integrated Flood Prediction and Response Systems »
Practices for Integrated Flood Prediction and Response Systems Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

State departments of transportation (DOTs) and other state and local agencies have implemented integrated flood warning and response systems to mitigate the effects of floods. These systems are critical for staging personnel, deciding when to close roads, inspecting bridges, tracking floods throughout the state, and planning recovery.

The TRB National Cooperative Highway Research Program's NCHRP Synthesis 573: Practices for Integrated Flood Prediction and Response Systems documents an overview of the state of the practice from agencies involved in finding new or innovative ways to improve flood management and response systems.

Supplementary to the report is Appendix F, which includes sample documents of practices related to integrated flood prediction and response systems.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!