Resilience to flooding: Draft building code

(1)

ICFR

International Conference on Flood Resilience: i Experiences in Asia and Europe 5-7 September 2013, Exeter, United Kingdom

RESILIENCE TO FLOODING - DRAFT BUILDING CODE

John D. Clarkson^'* Kenton Braun^, Angela Desoto-Duncan^, Graeme Forsyth"*, Dr. Ir.

J.G.de Gijt^ Dr. Nils P. Huber^, Dale Miller^, Philippe Rigo^, David Sullivan^

^ USAGE, Huntington District, Huntington WV USA; ^ PND Engineers, Inc, Ancliorage Alasl<a, USA; ^ Tetra Tech, Inc., Metairie, LA 70002 USA, '^^ TUDELFT ,Stevinweg Delft, Tfie Netherlands, BAW - Federal Watenways Engineering and Research Institute IVlannheim Area, Germany, ''University of Liege, Liege, Belgium

* CELRH-EC-DS, Huntington District, 502 Eiglitli Street, Huntington WV 25701 USA, Tel: 001-304-399-5217; E¬ mail: john.d.clarl<son ©usace.army.mil

A B S T R A C T

A significant issue associated Flood Defence Systems (FDS) is the difficulty of predicting how these structures will behave when inevitably they have been loaded beyond their designed capacity by a flood. The flood can cause these structures to fail catastrophically with loss of life and substantial damage to property. For a limited incremental investment, by including resilient features shown in this document, the FDS can dramatically lessen the chances for loss of life and property damage. While not a building code, the following provides guidance on how to improve the resilience of FDS so they will not fail catastrophically when overloaded beyond their designed capacity. Of all "lessons learnt" most important is to explicitly incorporate the consequences of failure and the possibility of being wrong in one's assumptions into the design process. Building in the flood plain will always have risk; the public should not become over confident just because a FDS is place. While it is recognized that an Integrated Water Basin perspective would include retention zones, restricted developments in flood plains, land use planning, awareness raising, flood resistant construction, drainage and water storage improvement, effective evacuation planning and other measures.

K E Y W O R D S

Climate change; catastrophic failure; draft building code, flood defence system, flood resilience.

1. INTRODUCTION

The requirement for flood resilience is to ensure the system does not fail catastrophically if the flood is higher than the design flood. It is likely that many structures associated with the FDS will be overtopped during the long service life of the structure, especially those structures that were designed to withstand and perform for events with lower return periods, (see Figure 1). If a FDS does not have resilient features and is overtopped, then the protected community is at much greater risk than if the protection had not been built, as the potential energy stored behind the wall is suddenly released when/if a breach occurs. With resilient features, a FDS can survive an overtopping flood event higher than the design flood (see Figure 2). The FDS performs as designed, with no catastrophic failure. There is time to evacuate the residents and minimal or no lives are lost. The flooding occurs gradually instead of in a sudden surge. Although for the overload event, economic losses for the community will still be significant, these resilient features will allow for the community to survive with less recovery time. A small amount of cosmetic damage to the FDS is acceptable, but the system should still be capable of withstanding the design loads after the flood event passes. It should also be capable of returning to flood protection as soon as is reasonably possible.

Design guidance which addresses structure and system resilience to extreme events that include overtopping during flooding is limited. Comprehensive "lessons learnt" and the incorporation of features to ensure resilience will help facilitate preparation of design guidance and the design of new construction or the rehabilitation of existing structures. Following Hurricane Katrina, the Final Report

of the Interagency Performance Evaluation Task Force (IPET) stated the following:

"RESILIENCE: Of the performance of the HPS, beyond the failure of the four l-wall sections,

it w a s the lack of resilience that stands out a s a major factor in the ultimate flooding and losses. If no catastrophic breaching had occurred, the flooding and losses would have

been significantly reduced, perhaps by half. Structures must be designed to withstand overtopping and to prevent catastrophic breaching. Such capability would not only have

(2)

International Conference on Flood Resilience: Experiences in Asia and Europe 5-7 September 2013, Exeter, United Kingdom

dramatically reduced the losses in New Orleans but also dramatically eased the burdens of recovery." 565 560 555 Q 550 > 545 540 530 525 1 1 J 1 1 ( -_— -• •

1

/ J -1

1

; ; 1 . ; ;

; J _{-^t—Median Flood Frequency}_{-^t—Median Flood Frequency} Wall at 150 Yr Wall at 150 Yr Wall at 500 Yr 1 1 Wall at 500 Yr T 1 1 J . ; ; ƒ i" ; ; ; ; ; 1 < -1 > _. -1 r r -J ] -' 1 1 ' : ' -100 200 300 400 Frequency in Years 500 600 700

Figure 1. Flood Frequency Curve, it is only a matter of time before most flood defence systems are overtopped. Note: some are designed to P M F with return periods as high as 10,000 years. Graphic from U S A C E .

Figure 1. Resilient structures will not fail catastrophically when actual loadings exceed the design load, note hardened splash area. Picture from U S A C E .

The need for resilience guidance was noted at the workshop, "International Flood Risk Management Approaches: From Theory to Practice", held November 30 through December 1, 2010 in Washington, D.C., http://www.nfrmp.us/IFRMA. FDS projects were stated to: "Need the best mix of protection and resilience to overtopping (one example approach in Japan is the super levee for highly populated areas)."

It is noted that a broader and more integrated flood risk management approach that strikes a balance between structural and non-structural measures is applicable for FDS in general. Of all the "lessons learnt" the most important is to explicitly incorporate the consequences of failure and the possibility of being wrong in one's assumptions into the design process. Building in the flood plain will always have risk; the public should not develop a false sense of security just because a FDS is place. Past examples have shown that some structural flood control systems may have exacerbated rather than

(3)

ICFR

reduced the extent of flooding. For instance when sediment deposits in river channels raised the height of river channels or when levees created a false sense of security, leading to excessive development in floodplains.

A fully integrated water basin perspective would also include many non-structural methods that can be developed to give rivers more room and reduce the speed of flood water. Polders are a low-lying flood plain enclosed by embankments separated from the river and are used on the Rhine River to allow for floods to be alleviated. Storing water by means of vegetation, soil, ground and wetlands, all of which are capable of retaining water, should have priority over swift water run-off. Every cubic meter of water not drained away immediately to the next body of water is a gain for the water regimen. An efficient flood risk management system needs to be complemented with integrated watershed management, retention zones, restricted developments in flood plains, land use planning, awareness raising, flood resistant construction, drainage and water storage improvement, effective evacuation planning and other measures. It is emphasized that levees are only one part of a fully functioning sustainable system.

The purpose of this report on resilience to overloading is the assembly of lessons learnt. The report includes recommendations for minimum system performance and public safety aspects of FDS which have the potential to malfunction or fail during major storm or flood events and thereby cause loss of human life, create catastrophic environmental hazards that endanger public health, disrupt lifeline services or destroy critical infrastructure needed for emergency response. This report introduces

performance requirements covering the minimum public safety aspects for FDS. The performance

requirement statement will lead the design team in how the system will perform for various loadings throughout its life. The recommendations in this document apply to the design, construction, operation, inspection and assessment of all FDS. (Dams are beyond the scope of this report.) Resilience is about designing systems that "fail gracefully" when overloaded, to avoid catastrophic failure but to allow for time to evacuate.

There is growing recognition (but little guidance) on the need for resilient systems:

"Many ecosystems have been frayed to the point where they are no longer resilient and able to withstand natural disturbances, setting the stage for 'unnatural disasters'—those made more frequent or more severe due to human actions." Janet N. Abramovitz, Averting

Unnatural Disasters, http://www.globalchange.umich.edu/gctext/lnquiries/Module%20Activities/State%20of%20the

%20World/Natural%20Disasters%202001.pdf

These issues will only become more important with growing concerns for climate change and the apprehension of high frequency flooding.

"Across Europe, the greatest natural threat in the coming years will be flooding as global warming sends more water gushing through passageways bordered by densely populated areas and overdevelopment, according to many water and engineering experts. The potential for catastrophic devastation and death is so high in so many countries that the European Union is preparing continent-wide guidelines for water management and flood control." (Molly Moore, Thie Wastiington Posf Foreign Service).

For countries in transition with limited budgets, these guidelines are paramount as the level of any protection provided will likely have lower elevations of protection and therefore more likely to be overtopped.

2. WHAT IS R E S I L I E N C E ?

Resilience is the ability of a system or a component of a system to sustain loads greater than the design load allowing the system or component to fail gradually, over some duration of time, rather than suddenly and without warning, see Figure 3. Overtopping and significant flooding are inevitable for any FDS with levels of protection below the Probable Maximum Flood (PMF). However, a complete system providing consistent levels of protection and incorporating resilient features would minimize the risk of catastrophic failure and reduced the potential of loss of life and damage to property.

(4)

(5)

ICFR

3. FUNCTIONAL P E R F O R M A N C E REQUIREMENTS

The goal of the reporl is to identify methods for ensuring resilience to the system by requiring it to survive overloads without catastrophic failure, endure only cosmetic damage, and be capable of withstanding the design loads after the flood event passes and return to normal service. This working group report introduces performance requirements covering the minimum public safety aspects for FDS. A Performance Requirement is a statement that establishes the necessary system response that minimizes risks to public safety and satisfies the purpose of the FDS. The first step is assessing how and when the structure might fail, then assess how the structure and system will perform once failure occurs. For FDS public safety is always most critical, however other design concerns need to be assessed: economic damages, reliability, maintainability, survivability and life cycle cost. The design approach and all criteria for components, units and features of the FDS should be formulated and executed in accordance with this statement. The performance requirements for all designs, evaluations, operations, inspections and assessments should be part of an integrated and cost-effective effort that will provide a high degree of confidence throughout the life cycle of the FDS. The performance requirements should focus on establishing consistent water levels, site characteristics, condition and response of structures, embankments, operating equipment and utilities, potential exposure to uncontrolled flooding, and functional consequences.

Resilience - The capability of a component, unit or system to withstand occasional small overloads that cause minimal permanent deformation, damage or cumulative degradation and then essentially recover its original state and function after the overloading event, to sustain loads greater than the design load to achieve gradual failure modes over some duration rather than sudden failure modes. From Mechanics of Materials, it is the c a p a b i l i t y of a strained body to absorb energy and recover its size and shape after deformation (elastic behaviour). Graphically, resilience is the area to the left of the resilience limit under the elastic region of a force/deformation curve as shown. See Figure 4.

Figure 4. Graphically, resilience is the area under the elastic portion of the force/deformation curve. Graphic from U S A C E .

Functional Resilience Performance Requirement. For storm events that exceed the design storm with return periods less than 500-1000 years, the system should function without sustaining damage that cannot be repaired before the next major storm. The performance of the FDS should be sufficient to prevent the interruption of all public, lifeline and business services in the protected area that have catastrophic regional or national impacts. The goal of the functional resilience performance requirement is to minimize the potential of damage to the FDS due to relatively small overloads which will allow the FDS to perform its intended function with minimal repair.

Toughness - The capability of a component, unit or system to withstand extreme

C o l l a p s e Limit

(6)

D C F R

overloads that cause extensive permanent deformation, damage or cumulative degradation but do not lead to catastrophic failure and/or uncontrolled flooding. Time to rehabilitate or replace the component, unit or system will probably be significant. From Mechanics of Materials, it is the capability of a strained body to absorb energy before rupture, but the strained body will not fully recover its size and shape after deformation (inelastic behaviour). Graphically, toughness is the entire area below the force/deformation curve as shown. See Figure 5.

Figure 5. Graphically, toughness is the area under the entire force/deformation curve. Graphic from U S A C E .

Functional Toughness Performance Requirement. For storm events that exceed the design storm with return periods greater than or equal to 500-1000 years the system is expected to perform inelastically with some permanent deformation, but the damage should not result in catastrophic failure of the system. Repair and/or replacement of damaged system components will likely be required and the time required to do so may be significant. The goal of the functional toughness performance requirement is to allow significant damage to the FDS due to large overloads while at the same time minimizing the possibility of catastrophic failure of the FDS. A system that is exposed to a large overload but that conforms to this performance requirement will minimize the risk to the public and property during those large overload events.

Note the recommendations for the design storm with return periods greater than or equal to 500-1000 years the system are site specific and very dependent on the societal value of the protected area and the resources available. A building code as recommended in the conclusions could provide bracketed parameters.

4. S Y S T E M R E S I L I E N C E

One of the key lessons learnt from Hurricane Katrina as documented in the IPET report is that "Floodwall design methods need to consider a broader spectrum of possible behaviours, and resilience to overtopping should be considered as a fundamental performance characteristic."

4.1 Minimal investment now may significantly reduce

loss of life

For a limited incremental investment, by including resilient features shown in this report, the FDS can dramatically lessen the chances for loss of life and property damage.

4.2 Accuracy of water level predictions

Design or evaluation of FDS projects should include sufficient hydraulic and hydrological information to minimize uncertainty concerning water levels and associated return periods. With the concern for Global Warming one point that needs to be made is that many of the hydrological parameters currently used in design and evaluations of

FDS need to be discussed and re-evaluated. With this concern what was believed to be a 500 yr flood may in fact be a higher frequency event. Many believe there is a correlation between recent catastrophes, failures in FDS, and Global Warming. Hydraulic parameters used in design or evaluation should provide reliability that is commensurate with project risks that provide safety to the public. The criteria selection of design parameters should be approached with a great deal of humility, that is, what if the designer is wrong, what are the consequences? To ensure this level of reliability, hydraulic and hydrological designers should have a high level of knowledge and understanding of both the anticipated storms, waves, and expected runoff as appropriate at the project site and also engineering guidance and judgment regarding selection of such parameters. Studies and modelling

Performance Evaluation of the New Orleans and Southeast Louisiana Hurricane Protection System

RnïtReprjit of the Inleragencij Perfwmance Evaluation Task Fotce

(7)

ICFR

requirements should be appropriate to attain this level of confidence. For risk based design wha\ level of confidence (i.e. what quantile in the uncertainty band) should designers use when assessing this event? The best guess (50%) estimate is too low half the time; many hydraulic designers use 90% confidence levels when assigning projections.

Also see CIRIA C635 London, 2006, Designing for exceedance in urban drainage - good practice, that develops best practices for flood conveyence in the urban environment.

4.3 Accuracy of Datum

One of tfie lessons learnt from Hurricane Katrina was to make sure the project has the correct Geospatial Vertical Datum to reduce the uncertainty of the Floodwall and Bank/Levee Elevations.

FDS protection elevations should verify the correct geodetic elevation references and account for the effects of subsidence if geological conditions suggest it is expected. For thorough discussion see EM 1110-2-6056 Standards and Procedures for Referencing Project Evaluation Grades to Nationwide Vertical Datums.

4.4Compartmentalization

Compartmentalization can be used to build redundancy for FDS by breaking a community into smaller sections such that the community has less risk with a smaller section of barrier failing. Furthermore, critical structures such as hospitals, police, and fire stations may have a separate, higher barrier built just around them to increase survivability and use for the community after an overtopping event.

4.5 Overtopping Sections

Since overtopping is an overload event for FDS, one of the key components for resilient design is to provide sufficient overtopping sections to allow for the protected area to safely flood when floods exceed the level of protection. Banks and levees should not erode and breach during flooding, even when top of the barrier levels are exceeded. The geotechnical engineer for banks and levees and the structural engineer for floodwalls should confer with the hydraulic designer to determine the need for and extent of overtopping protection such that the FDS works as a system. Usually for a river setting the use of different levee heights relative to the design water surface from reach to reach to force overtopping in a desired location. Designs using superiority (Increment of additional height added above design elevation to account for unknowns) can force initial overtopping in the least hazardous location. For a river setting continuous superiority should be provided over the length of the project to limit flow over the levee to the controlled overtopping section. Water surface profiles above the design profile need examining to apply superiority. The rate of overtopping of a levee or floodwall can be difficult to predict, however there are several good sources of design aids. For initial overtopping, the least hazardous location, like an undeveloped area, for initial inundation of the interior is preferred. It needs to be understood that no overtopping design can prevent overtopping. A good overtopping design can force overtopping in a selected reach and provides an initial cushion of water in interior areas to lessen overtopping impacts in other levee reaches. Development of a plunge pool on the protected side of the levee is a resilient feature that can reduce the erosive effects of water impacting on the protected side of the levee.

Planned overtopping sections should be considered as part of all FDS projects, but where banks and levees are composed of erodible soils or where overtopping will result in high velocity flows over the levee crest and slopes or over floodwalls, then erosion protection should be provided. For coastal levees should be provide scour protection along the entire length of the levee for the still water level plus wave heights due to the uncertainty associated with wave and storm surge overtopping rates. Flood side slope erosion and/or toe scour should also be prevented when high stream velocities are predicted during flooding.

(8)

ICFR

Figure 7. Scour hole formation by overtopping jet (from Hoffmans and Verheij 1997) Lessons Learned: Provide an erosion-resistant surface on the levee adjacent to the wall on the protective side.

Figure 8. Scour protection against overtopping shall be provided in order to prevent erosion of resisting soil on the protected side of the wall.

(9)

ICFR

4.6 Transition Sections

Transitions in FDS sinould be designed sucin that any differential deflections under load, long-term settlement, or overtopping events do not cause decreased performance of the FDS. Transitions which incorporate overflow weir sections should incorporate defensive measures to assure that flood flow will not adversely affect the stability or performance of the system and also that any incidental damage to protected structures is understood and mitigated if necessary. Transitions, connections or corners between different types of features shall include erosion protection and shall be capable of accommodating differential movements. Figure 10 shows scour at the transition from the 3-D effect of water spilling over and then flowing laterally with the wall that worsens localized erosion. Figure 11 shows a schematic of this 3-D effect.

(10)

International Conference on Flood Resilience: Experiences in Asia a n d Europe 5-7 September 2013, Exeter, United Kingdom

Flood Side

t t

Levee (dreife

m Ooriedintritio

Fioodwili

Smm Region

Protected Side

Figure 11- Schematic showing the 3D effect that happens during overtopping at transition areas, armouring is needed to provide resilience, also see Figure 10 for photograph of this 3D effect.

5. S Y S T E M OPERATIONS

5.1. General Performance Requirement.

The FDS should be designed, operated and maintained to resiliently survive the design storm event. This means that the ease of operation and maintenance of the FDS during the storm event must be primary considerations during design, selection, and layout of features. For storms that exceed the design storm, the FDS shall be designed to resiliently survive without incurring the type of damage to the system that would impact its ability to prevent catastrophic flooding or be repaired prior to the next storm. Critical maintenance, inspections and assessments shall be performed throughout the service life of the system to satisfy the design intent regarding public safety and operational adequacy.

5.2. Operation Considerations.

An acceptable state of readiness on the part of the operator must be continuously maintained and exercised. Maintenance: The report would benefit from a section on maintenance. Many levees have failed because of neglected maintenance. The FDS in their original design might have been able to withstand extreme loads but due to neglected maintenance, they are more vulnerable and breached.

Sufficient numbers of qualified and experienced personnel must be on hand to operate/install project pump stations, closure structures, interior drainage features and to patrol the project during a flood loading. The project operator should also maintain a transactional history of project operations and post-event reports capturing performance and operation lessons from each event. The project operator shall ensure that:

o Operation and maintenance manuals are in use and operating staff are knowledgeable of the content therein.

• Sufficient supplies of emergency and flood fight materials are always on hand (sand bags, shovels, plastic sheeting, etc.) Stored gates like stop-log closures need to be protected from theft.

• An addendum to the operations and maintenance manual outlining specific operations issues and lessons learnt is inserted and maintained in the record.

(11)

ICFR

5.3. Periodic Assessment/Evaluation

5.3.1. What to do with deficient, Non-Resilient Structures?

Impletment programs for the inspection and assessment of existing structures to determine resilience and recommended rehabilitation measures for deficient structures. With global warming concerns we may have larger, more intense storms, in conjunction with deteriorating levees with most that were NOT built with resilient features. Also because of the false sense of security we higher value, more vulnerable assets in the flood plain. While it is realized societal demands on funding will make it difficult repair all deficiencies ministries need to have effective prioritization program that weighs cost to consequences to achieve a tolerable risk levels. Need to have a monitoring to check datum in areas were subsidence is expected. An effective instrumentation program needs to be included to help monitor the system, especially during a storm event. Operations plan needs to assure that trained and deployable personnel are available. This would include a testing and certification programs for individuals involved in executing the work. Systems that do not rely on active decision making are preferred to an action such as deciding when to activate an overflow gate in the event of overtopping can be a highly political concern.

5.3.2. Emergency Action and Communication Plan

A well thought out plan that uses effective communications and evacuation planning can limit the loss of life. Need to identify potential risks for disruption of lifeline services and assure the flood waters do not remove emergency evacuation routes. As the population at risk is probably larger for existing levees then the plan needs to be revisited for current standards.

6. CONCLUSIONS AND RECOMMENDATIONS

Enclosed are a summary of lessons learnt and best practices for engineers that can be used in the design or evaluation of FDS to assure satisfactory performance during and after a significant flood event. Incorporating the concept of resilience to overloading into an FDS is of critical importance if the flood exceeds the top of the barrier. The features described will ensure resilience to the system by allowing it to survive this overload without catastrophic failure, endure only cosmetic damage, be capable of withstanding the design loads after the significant flood event passes, and to return to normal service. The incremental costs of these features are minor compared to the benefits of being resilient by minimizing loss of life. While failure scenarios to overloading can manifest themselves in many components of a FDS, the key resilient feature is to provide an overtopping section with armouring.

When these guidelines are followed, the FDS will become a sustainable asset. Without resilient features, a FDS will probably have to be replaced when first overloaded, such as was the case for the greater New Orleans area, where large portions of the hurricane protection system had to be replaced after the overload from Hurricane Katrina. Any FDS that is disposable after one overload, with a sudden failure mode and loss of life, is unacceptable. Sustainable systems meet the needs of the present without compromising the ability of future generations to meet their own needs. Giving explicit recognition to the linkages between sustainability, resilience and public safety is important because the sustainability and safety of engineered systems are directly related to how these systems perform under overload conditions and as they approach failure. If engineered systems are not safe and resilient, then they will be essentially disposable features and not sustainable assets.

Of all "lessons learnt" most important is to explicitly incorporate the consequences of failure and the possibility of being wrong in one's assumptions into the design process. Building in the flood plain will always have risk; the public should not develop a false sense of security just because a FDS is place. History has shown that some structural flood control may have exacerbated rather than reduced the extent of flooding. A fully integrated water basin perspective would also include many non-structural methods that can be developed to give rivers more room and reduce the speed of flood water. An efficient flood risk management systems needs to be complemented with integrated watershed management, retention zones, restricted developments in flood plains, land use planning, awareness raising, flood resistant construction, drainage and water storage improvement and other measures. It is emphasized that levees are only one part of a functioning system.

(12)

ICFR

While this report is not a building code, the recommendation is to further develop this relevant information into a comprehensive Model Flood Design Building Code specifying minimum requirements to provide resilient systems to ensure that the human population is not endangered if a system capacity is exceeded. As a "model" building code, users could simply edit for their circumstances when awarding design contracts, i.e. "remove hurricane if not in a hurricane zone". Please contact authors if interested in developing a Resilient Flood Defence Building Code.

7. REFERENCES

2009 USACE Infrastructure Conference, Proceedings from Workshop on EC 1110-2-6066 Design/Evaluation of l-Walls, Cleveland, OH, Huntington District presentation from Sullivan, Hensley, and Clarkson, July 2 1 , 2009

Inland Waterborne Transport: Connecting Countries, International Navigation Association (PIANC), The United Nations World Water Development Report 3 Water in a Changing World, March 2009

Interagency Performance Evaluation Taskforce (IPET), the report is provided in nine volumes, United States Army Corps of Engineers

International Flood Risk Management Approaches: From Theory to Practice," held November 30-December 1 in Washington, D.C. http://www.nfrmp.us/IFRMA

Rethinking Defenses Against Sea's Power - Washington Post, Washington Post Foreign Service, By

Molly Moore, Thursday, September 8, 2005.

What happened in 1953? The Big Flood in the Netherlands in retrospect, Herman Gerritsen, Royal Society Publishing, 10.1098/rsta.2005.1568 Phil. Trans. R. Soc. A 15 June 2005 vol. 363 no. 1831 1271-1291

ETL 1110-2-299 Overtopping of Flood Control Levees and Floodwalls, 22 August 1986, United States Army Corps of Engineers

CIRIA 0635 London, 2006, Designing for exceedance in urban drainage - good practice, David Balmforth MWH, Christopher Digman MWH, Richard Kellagher HR Wallingford, David Butler University of Exeter