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.
69 APPENDIX C Scenario 2: Minimizing Disruption During Extreme Weather Events This case scenario considers a set of extreme rainfall events drawn from events in the United States (2011, 2012, 2015, and 2016) and Central Europe (2002 and 2013) and the nature and scale of the effects of these events on transport infrastructure and services. The scenario highlights the vul- nerabilities identified during a recent series of devastating floods. The central case is modeled on riverine flooding in mountainous regions in the United States caused by intense late summer storms, with its scope broadened to include considerations of longer-duration flooding. The U.S. states are each responsible for building and maintaining more than 15,000 miles of roadway, with individual towns or counties maintaining many more local roadways. The states also oversee hundreds of miles of state-owned railways and numerous small airports. The case study describes a large-scale flooding event that affected a region of several hundred miles including sev- eral large towns and small, remote villages in mountainous regions and is enriched by a brief description of the central European floods in 2002 and 2013. Box 1 describes the causes and locations of the European floods, and Box 2 examines Germanyâs response to the floods. The condi- tions (geographical, climate, extreme weather, and storm surges) in Europe differ considerably from those in the United States, but ways of preparing for, and dealing with, extreme events can be of use for both locations. 1 introduCtion Our built infrastructure is routinely put in harmâs way from extreme weather, whether from coastal and inland flooding, wildfires, strong winds from hurricanes and tornados, or extreme snow loads. The current practice in design codes is to examine historical extreme events. With climate change and changing extremes, this approach is similar to driving while looking in the rearview mirror. âInstead of preparing ourselves for past disasters, a com- mon error, we should use existing climate research and risk studies to prepare for future disasters.â In the United States, the frequency of extreme precipi- tation is increasing across the mid latitudes, and storms in those areas will become more intense. Intensification of precipitation is often associated with an increased flooding risk in places such as the northeast United States. There is also high confidence that the strongest hurricanes are becoming more frequent. A recent study of the European Joint Research Centre estimates that under high levels of global warming: [T]he population affected and direct flood dam- ages indicate that by the end of the century the socio-economic impact of river floods in Europe is projected to increase by an average 220% due to climate change only. A larger range is foreseen in the annual flood damage, currently of 5.3 Bâ¬, which is projected to rise at 20â40 Bâ¬â¬ in 2050 and 30â100 Bâ¬â¬ in 2080, depending on the future eco- nomic growth. (1) Changing rainfall and flooding patterns are poten- tially important factors to consider in transportation infrastructure design, planning and operations, and maintenance. Todayâs infrastructure was built and new
70 t r a n s p o r t a t i o n r e s i l i e n c e BOX 1 Central euroPean floods in 2002 and 2013 Causes and Locations of Flooding Both floods in Central Europe were caused by heavy rainfall. In 2013, rainfall intensities of up to 250 mm were registered in just a few days. The 2013 flood was caused by a low-pressure system that was locked into place by a disturbance in the global wind patterns. Figure 1. Amount of rainfall. Figure 2. Obstruction of road traffic in 2013 in Germany per county (âLandkreisâ) in number of weeks (âWochenâ). In June 2013, large-scale flooding occurred in many Central European countriesâSwitzerland, Austria, the Czech Republic, Slovakia, Poland, Hungary, Croatia, Serbia, and particularly in Germany. By the end of May 2013, rainfall totaled 178% of the average monthly amount, and record-breaking soil moisture was observed in 40% of the German territory. High initial stream-flow levels were reported in the river network. At several locations, embankments were unable to withstand the floodwater, resulting in dike breaches and inundation of the hinterland. Costs and Impact of Damage For the 2002 event, total costs of damage in Germany were e11.6 billion, of which e1.8 billion was covered by insurance (19% of all houses). In the Dresden area alone, the damage was e1 billon (due to a return period 1/200 per year event). The direct damage to railways was e1 billion, but this amount does not account for the damage due to the obstruction of railway traffic and loss of consumer satisfaction. In 2013, the total damage costs were e6.5 billion for the provinces. In 2002, e1.8 billion was covered by insurance (34% of all houses). The total damage in Germany mounted to e10 billion, with the damage to railways totaling e0.1 billion. For example, the important long-range line between Hannover and Berlin was obstructed, leading to detours, which forced people to use planes, cars, and other means of transportation. This flood and other weather impacts have made author- ities responsible for the Deutsche Bahn increase investment in maintenance and preparation for extreme weather. The damage to other infrastructure in Germany in 2013 was e0.3 billion. In large parts of the country road traffic was obstructed because of actual flooding of the roads and because of other causes such as landslides.
71A P P E N D I X C : s C E N A r I o 2 : m I N I m I z I N g D I s r u P t I o N infrastructure is still sized using the principle of station- arity, which assumes that rainfall statistics will remain constant across time. However, as mentioned previously, the data suggest otherwise. Current design standards may need to be altered to account for the nonstation- ary nature of hydrological statistics. In the Netherlands, for example, climate scenarios since 2009 show that increased precipitation and rising sea levels must be considered in the designs for infrastructure to withstand higher water levels. Because existing infrastructure may be more vulner- able to future extremes than in the past, disruptions are likely to be more frequent and over larger regions, with large impacts on the transport system. Not only are floods disruptive to the transport system during the event, but the recovery after the event requires a major effort. Floods influence all stages of event prepared- ness, from preparation for the event, managing during the event, and recovery after the event. In this paper we focus on management during the event. 2 desCriPtion of the sCenario This scenario is drawn from the flooding in Vermont (2011), Colorado (2013), South Carolina (2015), and Houston, Texas (2016), as well as Central Europe (2002 and 2013). Just weeks prior to this hypothetical stateâs biggest tourist event, which brings in an annual $3 billion from visitors, an intense flood closed over 500 miles (800 BOX 2 issues raised by Central euroPean floods in 2002 and 2013 What Measures Were Taken after the German 2002 Floods? What Was Their Effect in 2013? After the 2002 floods in Germany, risks were assessed, flood maps were developed, and flood risk manage- ment plans were prepared to prevent damage in the future. The experiences in 2002 in the flood areas helped to prevent and reduce costs and damage in 2013. In 2013, the Bavaria and eastern Germany water levels significantly exceeded those of 2002 in many places on the Danube and Elbe. In Dresden, by contrast, the old city center was largely spared, unlike in 2002. Thanks to better flood control, fewer dykes on the upper reaches of the Elbe broke than in 2002, but this meant that the flood wave farther downstream was all the higher. In Magdeburg, floods reached a record level. The Nationale Hochwasserschutzprogramm (National Flood Protection Program) was launched in the after- math of the 2013 flood and is run in part by the Bundesanstalt fÃ¼r GewÃ¤sserkunde (Federal Institute of Hydrol- ogy) (http://www.bafg.de/EN/Home/). Measures Taken with Stakeholders Along the River Measures such as strengthening levees were taken individually by the states. Some states spent a lot of money in a short time, others spent less money. Examples of good coordination of measures are â¢ Treaty for flooding of a polder (Havelpolder); â¢ Coordination of different measures along the rivers and monitoring the collective impacts of measures; and â¢ International commissions to protect rivers (Mosel, Rhine, Saar, Elbe). Crisis and Emergency Organization for Flood Events in Germany Germanyâs civil protection is based on fire departments (professional and volunteer), technical public aid (vol- unteer), and other public aid organizations financed by the government. Responsibility for flood prevention and civil protection is conducted by the states and country, which in certain cases can cause a problem for coordination. In 2013, 1.7 million volunteers from fire departments and technical public aid assisted with the flood event. A total of 5.15 million sandbags were used, and an extra e60 million was spent by civil protection organizations. In principal the emergency response is organized so that the chain of responsibility begins with the lowest level of government. In Germany, this level is the Kommune, or town council. If an emergency gets worse, the Kom- mune asks the following level for help, and so on.
72 t r a n s p o r t a t i o n r e s i l i e n c e kilometers) of state highways, destroyed dozens of state bridges, and closed numerous railroad bridges, making 200 miles (300 kilometers) of railroad impassable. Days prior to the event, the state had tracked a warm humid southerly air mass as it moved toward the region. The region was already on high alert due to high water levels as a result of prior storm events and soils saturated well into the 90th percentile compared with long-term averages. These conditions are ideal for rapid runoff, flooding, and uprooting trees. Despite forecasts of large, slow-moving systems, the storm track, timing, and pattern as well as rainfall inten- sities and locations were still quite uncertain. The tall mountains in the region make it difficult to predict the side of a mountain on which any expected rainfall will flow. Before the stormâs arrival, the National Weather Service (NWS) worked directly with the Governorâs office. A day before the storm hit the Governor proac- tively declared a state of emergency, and the state govern- ment initiated emergency management procedures. The state department of transportation (DOT) carefully mon- itored the NWS projections and prepared for flash flood- ing. The DOT expected impacts across a large part of the state and prepared equipment and resources. However, NWS predictions do not readily predict destruction from floodwaters to roadways, culverts, bridges, and railways. Rainfall Picture and Runoff Response As the storm approached and pushed up the mountains, it stalled. It began to rain steadily in the afternoon, but late in the evening the wind picked up and torrential rains hit. In the first 12 hours, over 9 inches (23 cen- timeters) of rain fell. Because the ground was already saturated from previous rains, runoff in streams and rivers led to catastrophic flooding across two-thirds of the state. The first reported mudslides occurred shortly after midnight, and a widespread deluge of flood impacts followed throughout the region, particularly across the central part of the state, where rockslides, landslides, mudslides, and washouts destroyed resi- dences, roadways, and local-access bridges. Flooding was so severe that it fully rerouted creeks and rivers by more than 500 feet in some locations. The state depart- ment of environmental services had its entire complex flooded and all computer systems rendered useless when a nearby river breached its banks. The storm also spawned scattered tornados that uprooted countless trees and downed power lines across the access roads to a major state DOT maintenance yard. By the time the storm had passed, more than 18 inches (46 centimeters) of rain had fallen in a little more than 2 days, making it a 1-in-1,000-year event. Transport Effects As a result of the storm, 500 miles (805 kilometers) of state highways were closed, more than 100 state bridges were closed, 30 railroad bridges were damaged, and 200 miles (322 kilometers) of railroad lines were impassable. More than 200 (more than 90%) of the stateâs towns had to rebuild damaged roads, bridges, and culverts. The storm damaged thousands of town culverts and dam- aged or destroyed nearly 300 town bridges The entire state was at a standstill. Dozens of towns were entirely cut off from the outside with no way in or out. Figure 1 shows the Vermont transport corridor response before, during, and after Hurricane Irene in 2011. 3 issues raised by the Case The main issues related to levels of preparedness before, during, and after the event raised by the case are dis- cussed below. Before the Event The agency studied NWS projections and prepared for flash flooding. The DOT expected impacts across a large part of the state and prepared equipment and resources. Unlike large, slow-moving systems such as hurricanes that allow timely evacuations, large rainfall events in mountain- ous regions can confound preparedness efforts because no one can predict on which side of a mountain any expected rainfall will flow. Before the stormâs arrival, NWS worked directly with the Governorâs office. The Governor proac- tively declared a state of emergency, and the state govern- ment initiated emergency management procedures. The DOT was aware of the storm and established crews who readied equipment and other resources. In hindsight, it is believed that distributing resources even farther would have made the situation worse. Another level of preparation is the preparation of traf- fic management strategies and evacuation routes in case of obstructions and extreme events, which is a no-regret, multievent preparation. Preparation can also consist of the design and build- ing of (multimodal) extreme weatherâresilient infra- structure. During the Event Simply put, the event was at a scale never experienced, expected, or planned for at the DOT, and the personnel
73A P P E N D I X C : s C E N A r I o 2 : m I N I m I z I N g D I s r u P t I o N Plymouth (below: before, during, after) Plymouth Residents â Rte 100 below Stat: BraintreeâNE Central 327,024.743 cubic yards of materials (riprap, bank and crusher run, subbase, stone, etc.) 370,015.8947 tons of materials (stone grits, scalping, riprap, quarry run, fill, etc.) (Cubic yards vs. tonnage varies by supplier.) FIGURE 1 The Vermont transport corridor response before, during, and after Hurricane Irene in 2011. The devastation from the hurricane exceeded that of the 1927 flood.
74 t r a n s p o r t a t i o n r e s i l i e n c e it needed were too many and too scattered to provide a meaningful foundation for delivery of services. An incident command system (ICS) is defined by the U.S. Federal Highway Administration as âa systematic tool used for the command, control, and coordination of emergency response.â The DOT organized two ICSs that worked with the stateâs unified command. Under a unified command, a single, coordinated incident action plan directs the emergency response effort and supports agencies or divisions with different legal, geographic, and functional authorities and responsibilities to work together effectively without affecting individual agency authority, responsibility, or accountability. The DOTâs ICS focused on reopening state and local roads, pro- viding access for emergency relief efforts to stranded communities, and enhancing communication for the recovery effort. The incident command centers were the home base to over 1,000 people including nearly 300 DOT employees, hundreds of National Guard members, and DOT workers from neighboring states. Recognizing the need for a multitude of resources such as engineering services, materials, contractors, and equipment, the DOT created and maintained a âone- stopâ shopping list. This list acted like a clearinghouse in which to collect names of private contractors, consul- tants, bridge inspectors, trucking companies, surveyors, utilities, quarry owners, and others. Response Staffing For half a day, the DOT operations director sought to establish the facts and conditions that the state would address. Although radio contact was available with eight of the nine maintenance districts, there was initial diffi- culty in establishing contact with some employees. Some crews spent the night in their trucks, and one employee hiked five miles through the woods to get to a location where he could contact his supervisor. Although the DOT employees had hastened to imme- diately support storm response and recovery despite road and bridge closures and other physical barriers, they were not prepared for what turned out to be, in some cases, a 3-month separation from their homes. Communications The DOT emergency transportation information system had been âbrought to its knees.â With the system down, Google reached out to set up a system for real-time mapping of closed roads, with public updates twice daily. By this time, the storm had passed and communities were stabilizing and assessing impacts. The mapping tool was widely used to counsel travel- ers to the state as well as state residents. The stateâs emergency management division issued a statement early that morning via social media stat- ing: âRemain. Where. You. Are. Dangerous flood- ing conditions through the state for most of the day.â However, drivers still moved or drove around barri- cades onto flooded roads (https://www.youtube.com/ watch?v=LtlsIwhZHLo). Social media provided valuable updates throughout the event, and the DOT website began to list daily release times for storm event information. The DOT also used a mobile phone microsite to allow for easy access to infor- mation and used social media to communicate condi- tions. Media outlets followed the DOT Facebook and Twitter accounts; at one point, five DOT administrators worked full-time maintaining the Facebook page. After the Event Demobilization The demobilization phase is sometimes referred to as the âforgotten phaseâ in emergency management. As they closed out the ICS and people returned to their usual jobs, the ICS Logistics Section made a point to tell employees about the typical feelings experienced after a traumatic event. Subsequently, the DOT held informal lunches and offered a counseling program. The state also developed a commemorative coin recognizing the assis- tance of major stakeholders, and the coin was presented to everyone, including those who had kept on with day- to-day activities. The DOT also sent thank-you letters to everyone involved, including their families. River Management This event clearly showed that although roads and rivers compete for the same space, there are numerous loca- tions where we have failed to build roads in a way to ensure they âget alongâ with rivers. In all cases, if a bal- ance is not struck in this game, the river will win eventu- ally. It is clear that during an emergency, field staff who have not been properly trained in river dynamics need to make decisions. This storm raised awareness for a need for all field staff to have some basics in river mechanics to lessen impacts to rivers and decrease the potential for future damages. Recovery Phase â¢ Within 1 month of the storm, over 75% of the closed bridges had reopened, and more than 96% of the state highway road segments had been reopened.
75A P P E N D I X C : s C E N A r I o 2 : m I N I m I z I N g D I s r u P t I o N â¢ In 4 months, all of these state assets were ser- viceable again. Town bridges, culverts, and highway segments recovered in the 4-month period following the event. To accomplish this, the DOT expedited and streamlined procedures, which resulted in a reduction in the initial estimate of transportation system damages from $700 million (â¬625 million) to $175â$250 million (â¬156ââ¬223 million). â¢ The longer phase included continued efforts a year later to close out certain federal reimbursement issues, plan and design permanent repairs, and watch for sink- holes and riverbank landslides. 4 imPliCations for researCh The case study presented suggests several leads for fur- ther research as listed below. In Annex 1, the current approach in the Netherlands is presented as an example. â¢ What considerations emerge for future adaptation planning and resourcing? â Bridge (and infrastructure in general) design criteria addressing the structureâs ability to withstand flooding, â Review of riverbank design methodologies and increasing the use of riprap (stone shoreline protection), â Route logs as a resource for design engineers in identifying structures and their locations, â Design and location of infrastructure and evacu- ation strategies multimodally and on the system as a whole, â Simplification of design plans, including mini- mization of repetitive information, â Methods to reduce potential costs (e.g., with improved land use planning), and â Methods to translate uncertainty of develop- ments in the future (e.g., mobility patterns, climate change) into infrastructure planning and investments, with an eye on robustness and fitness for extreme events. â¢ Rivers are a coupled system. Measures taken in one part of the river system will affect the behavior of the system as a whole. These effects reach out over state and country borders. The same is true for the transport sys- tem (including traffic management strategies). How can the coupled effects of these two systems be determined, especially for coordination during an extreme event? â¢ How robust are the present climate modelsâ out- comes? Are their predictions sufficiently solid to justify major investments? â¢ Based on the present forecasts, flooding will be âbusiness as usualâ for some parts of Europe and the United States. How do we communicate this message and improve the preparedness of the affected communities? â¢ How can we evaluate and improve the perfor- mance of emergency measures and response? â¢ What are the most effective methods of assessing hydrological system performance (including cascades and retention) and damage? aCKnowledgments Information about the floods in Vermont (2011), Colo- rado (2013), South Carolina (2015), and Texas (2016) was reviewed and amalgamated to develop the case study. Thanks to Gina Campoli, Vermont Agency of Transportation. Information about the European floods of 2002 and 2013 (please see the bibliography) was gath- ered by Kees van Muiswinkel, Rijkswaterstaat, Nether- lands, supported by Enno Nilson, Department of Water Balance, Forecasting, and Predictions, Federal Institute of Hydrology, Germany; Heidi Kreibich, University of Potsdam, Germany; and Jens Reuber and Rita Lam- mersen, Rijkswaterstaat, Netherlands. referenCe 1. Alfieri, Lorenzo, Luc Feyen, Francesco Dottori, and Alessandra Bianchi. Ensemble Flood Risk Assessment in Europe Under High-End Climate Scenarios. Global Envi- ronmental Change, Vol. 35, 2015, pp. 199â212. bibliograPhy of PubliCations used to desCribe 2002 and 2013 Central euroPean floods Floods in Germany (June 2013). Post-flood field investigation, July 2â3, 2013. Preliminary findings by Bas Jonkman, Timo Schweckendiek, and Guy Dupuits (TU Delft); Tor- sten Heyer (TU Dresden); Joop de Bijl (STOWA); Astrid Labrujere (RWS). Das Juni-Hochwasser des Jahres 2013 in Deutschland. BfG- 1793, BELZ et al., 27.6.2013. Das Hochwasser im Juni 2013. BewÃ¤hrungsprobe fÃ¼r das Hochwasserrisikomanagement in Deutschland. Deutsches Komitee Katastrophenvorsorge e.V., UniversitÃ¤t Potsdam DKKV Schriftenreihe 53. Mitteilungen Des BfG Nr. 31: Das Hochwasserextrem des Jah- res 2013 in Deutschland: Dokumentation und Analyse, Dezember 2014. Deutscher Bundestag, Drucksache 18/2124: Antwort der Bun- desregierung auf die Kleine Anfrage der Abgeordneten BÃ¤rbel HÃ¶hn, Sven-Christian Kindler, Peter Meiwald, wei- terer Abgeordneter und der Fraktion BÃNDNIS 90/DIE GRÃNENâMittelabfluss aus dem Fluthilfefonds 2013 fÃ¼r AufbaumaÃnahmen.
76 t r a n s p o r t a t i o n r e s i l i e n c e LAWA Bund/LÃ¤nder Arbeitsgemeinschaft Wasser. Zusam- menfassende Analyse der Ergebnisse der vom Hochwasser 2013 betroffenen Flussgebietsgemeinschaften beschlossen auf der 147. LAWA-VV am 27â28 MÃ¤rz 2014 in Kiel. The Flood of June 2013 in Germany: How Much Do We Know About Its Impacts? Discussion paper under review to be published in Natural Hazard and Earth System Sciences. The Deltares and KNMI study, Implications of the KNMIâ14 Climate Scenarios for the Discharge of the Rhine and Meuse. 1220042, 2015. FLOODrisk 2016 Conference. Interpreting the Impact of Flood Forecasts by Combining Policy Analysis Studies and Flood Defence. Robert Slomp (Rijkswaterstaat); Bas Kolen (HKV); and Hilde Westera, Jaap Verweij, and Durk Riedstra (Rijkswaterstaat). annex 1 strategy for the use of main infrastruCture before, during, and after a flood (rijKswaterstaat ProjeCt grouP mego, 2016, referenCe 9) In the Netherlands, the Minister of Infrastructure and the Environment developed a strategy with respect to improving evacuation in case of major floods, mainly by marking evacuation routes, instructing the safety regions, and influencing the behavior of people by informing them about the strategy. Measures for Using Highways for Prevention or Evacuation The most cost-effective measures for using highways for prevention or evacuation are focused on administrators and citizens: â¢ Make decisions in a more timely manner; â¢ Improve the time span of forecast models; â¢ Influence the evacuation behavior and self- reliance of citizens through a well-designed and com- municated evacuation network. This beneficial influence will increase the number of people who can leave an endangered area; and â¢ As needed, add options to the current traffic man- agement plan that can strengthen the function of the main infrastructure for major evacuations. Major adjustments of infrastructure are not cost- effective, as the probability of a major flood in the Neth- erlands is low. However, regional measures on specific critical locations may be worth the investment. Recommendations to Be Discussed with the Ministry of Safety and Justice Responsible for the Crisis Organization â¢ Develop a default strategy. (In the Netherlands, this strategy is âgo upstairs and stay dry; driving away is more dangerous in many cases.â Under certain circum- stances citizens may be advised to leave an area in time if possible.) â¢ Model and simulate an evacuation over the main roads to gain insight into the available capacity. This knowl- edge is necessary to develop possible strategies for preven- tive evacuation (the current strategy has to be updated). â¢ Make the âall hazard evacuation scenario highway infrastructureâ part of the crisis and safety plans of the safety regions. â¢ Enforce unity and connection of the crisis and safety plans to ensure that the highway infrastructure is able to accommodate all the evacuation traffic. â¢ Train the people involved with the new plans.