Katrina was a unique storm. With its combination of high

Transcription

1 Katrina Policy Lessons Learned: Coping With Change is a Risky Business By Lewis E. Link, Ph.D. Hurricane Katrina was a manifestation of the increased risks we incurred but failed to adequately assess and monitor, such as the changing nature of the hazard, its protective systems, and its consequences. The author believes we must adopt risk-based, adaptive, long-term strategies to minimize future damages while creating a comprehensive, sustainable framework for national water policy. Katrina was a unique storm. With its combination of high intensity and large size, Katrina generated the largest storm surge to ever hit North America (IPET 2007a). The surge, coupled with long period waves, overtopped and overwhelmed many of the levees. It also caused floodwalls to fail, some by overtopping and some because of inadequate design. Of 350 miles of levees and floodwalls, over 200 miles sustained damage, 41 miles of which were severely damaged. There were 50 major breaches, 4 of which were because of design deficiencies. The remaining breaches were caused by forces that exceeded design criteria, but the lack of resilience significantly increased the damage and losses suffered. Without breaching, Katrina still would have created the worst flooding ever experienced in New Orleans. With breaching, Katrina created one of the worst natural disasters in the history of the nation. The U.S. Army Corps of Engineers Chief established the Interagency Performance Evaluation Task Force (IPET) to determine why the hurricane protection system in New Orleans performed as it did during Katrina. Comprised of experts from 25 universities, 23 private companies, and 10 government agencies, and using some of the most sophisticated technical tools available, the task force performed a detailed, technical examination of the performance of individual structures and components as well as a broader risk-based assessment of the reliability of the system with respect to the future. The results of this effort are available on the Corps website at WES.Army.Mil. While perhaps unparalleled in scope and depth of analysis of a hurricane-related natural disaster, the detailed findings and lessons learned from this effort only tell a part of the story. What led to this situation is a long story with multiple paths that converged in New Orleans on August 29, Unfortunately, the underlying forces that generated the Katrina disaster were working decades before the storm (Link 2006). There were many changes Lewis Link is Director of the Interagency Performance Evaluation Task Force. He is a Senior Research Engineer on the faculty of the Department of Civil and Environmental Engineering at the University of Maryland in College Park, Maryland. He formerly served as the Director of Research and Development for the U. S. Army Corps of Engineers, retiring in 2002 after 34 years of service. occurring that were creating risk. That risk, unfortunately, was not well quantified or managed. In that light it is appropriate to discuss these changes using the three major components of risk: the hazard; the system; and the consequences. The Hazard Changes were occurring in the nature of the storms threatening New Orleans, in the very nature and character of the system put in place to protect New Orleans, and in the number of people and value of property being protected. The hurricane data used to create the 1965 Standard Project Hurricane (SPH) the design criteria for the hurricane protection system developed by the National Weather Bureau (now the National Weather Service) primarily included storms prior to 1960, a period of few large storms. By 2005, the 1965 SPH, thought to represent a 1:200 to 1:300 annual chance of occurrence event, no longer equated to even a 1:100 event (IPET 2007b). During the overall planning and design of the hurricane protection system in New Orleans, the hazard, in terms of the SPH, was redefined by the National Weather Service based on an updated information base. Yet no changes were made to the criteria for the structures being designed and constructed. Some have hypothesized that this was a failure of diligence (ILIT 2006). It is more likely that changes were not made because of the inflexibility of the policies under which such projects were authorized, appropriated, planned, and constructed, and because any halt in the planning and design process could result in delaying construction of any protection at all. How we should characterize the hurricane hazard for the future is a significant policy issue. Because of the dynamic nature of climate, we have experienced at least periods with more frequent and more intense storms since Research has shown that this has not been a linear or gradual increase, but that such increases occur in cycles (Resio 2007). There was a period of high energy (counting the total energy in all storms for each year) during the 1960s (Hurricanes Camille and Betsy) followed by a window of relative lower energy. It appears that a second high energy period is occurring now (Hurricanes Rita and Katrina). Clearly, we can not use historical storms to define the future hazard. There are too few large historical storms and they do not necessarily capture the dynamic aspects of climate. When and if the next period of more intense storms will occur is an unknown. September-October

2 It is also not adequate to specify flood protection based on categories of storms. The Saffir-Simpson Hurricane Scale categories only relate to the intensity of a storm or its maximum wind speed. While this is very valuable information, especially for wind damage, it is not a complete index for surge potential. We now know that it is both the intensity and the size of the storm that defines its surge generation capability and both must be considered if we are to properly understand the threat of flooding. Perhaps we need to develop a separate classification system that better reflects the storm surge and flood hazard potential from a hurricane. From a policy perspective it makes sense to define the hazard not based on the probability of a storm, but based on the probability of experiencing water levels (surge and waves). A storm like Katrina, computed to be a 400-year event based on its surge generation potential (Resio, 2006), created historically high water levels on the East Bank of New Orleans but insignificant surge and wave levels on the West Bank. Obviously, location matters for any given storm, and it is the probability of exceeding given water levels at any location that is important in determining the level of protection provided. The Corps, the National Oceanic and Atmospheric Administration, and the Federal Emergency Management Agency (FEMA) created a team that has developed a new approach, the Joint Probability Methodology Optimal Sampling (Resio, 2006), for defining future hazard. It provides a water level exceedence relationship for many locations around an area of interest. This method has been used to define the current and future hazard for New Orleans using a spectrum of 152 hurricanes that represent storms with a wide range of sizes, intensities, and tracks. This information will support the IPET risk assessment, the Corps design of 100-year protection for New Orleans, the Corps Louisiana Coastal Protection and Restoration Study, and FEMA flood mapping. The System The original plan for protecting New Orleans was focused on life safety, not economics. It was a systems approach dimensioned to deal with the most severe storm likely for the region. What was put in place, the result of many incremental compromises to deal with Map of flood depths from Katrina in Orleans Metro sectio0n of New Orleans. Vast areas below sea level were flooded from the waters coming over and through the protective structures. Nearly 80 percent of New Orleans was flooded and 78 percent of the flood damage was in residential areas. competing priorities, had little margin for error. The levee-floodwall system along portions of the drainage canals were limited by real estate right-of-ways. Their designs, I-wall structures combined with relatively narrow levees, reflected that constraint and resulted in structures not conservative enough to deal with a failure mode that had not been considered in the design. Some extensive sections of the levees were constructed using hydraulic fill (material dredged from an adjacent channel) and capped with clay. While these levees performed well for the water levels for which they were designed, they had little resilience to overtopping. When water exceeded their elevations, they were vulnerable to breaching by scour and erosion. Resilience to breaching for water levels beyond the design criteria was simply not a part of the package, probably because of additional cost. When Katrina hit, the hurricane protection system authorized in 1965 still had yet to be completed. After 40 years, major sections remained incomplete. The funding for the project was piecemeal and the design and construction dragged out over decades. In fact, it is a rare water project that has full funding and can proceed with a coherent design and construction schedule. The incomplete areas and the transitions between types of structures and levels of completion proved to be least capable of providing protection. Fortunately, many of the least complete sections were on the west bank where Katrina had much less impact. The system was also under a constant state of subsidence, in some locations at rates that significantly reduced the elevation of the levee and floodwall crowns as well as the project s overall ability to protect. In addition, systematic errors in interpreting the local elevation datum caused some structures to be built lower than intended. Subsidence, coupled with the construction elevation error, left some structures 2 to 3 feet below their authorized elevations when Katrina struck. In some cases it was of little significance, in others it markedly reduced protection and increased vulnerability to overtopping. The lower than authorized elevations were known at least qualitatively, but as in the case of the changing hazard, structures were not augmented or modified to deal with the problem. The lack of response to the known loss in protective elevation also points to the inflexibility of the water resources project policies and processes that best fit a static world. 14 national wetlands newsletter

3 The natural environment surrounding New Orleans is also undergoing constant change. This change must be considered in any analysis of the overall protection or risk levels in New Orleans. After the canalization of the Mississippi River by the mainline levees, nutrients and sediments normally deposited in the marshes were whisked away to the Gulf. Through the natural subsidence in the region and man-made channels that allowed salt water intrusion into freshwater marshes, huge areas of natural marshes were lost. These marshes play a significant role in the overall protection of the region to storm surges. Their loss contributed to the increasing levels of surge and waves experienced by the hurricane protection system. Katrina itself caused the loss of another large area of marshes, close to 118 square miles. (IPET 2007b). While there have been rules of thumb relating to the impact of marshes on storm surge, they are far from providing an accurate means toward understanding the real role of marshes. High resolution models are just now being applied to better understand the protections offered by marshes and wetlands. While the modeling validates the value of marshes, the specific impacts can vary greatly depending on the character of the surge and geometry of the region. Of significant concern is the time for a natural marsh to recover from hurricaneinduced damage. Hurricane Rita, following close on the heels of Katrina, had less marsh to curtail it and demonstrates that we cannot assume that natural marshes will be there whenever you need them. We need considerable additional knowledge about these natural systems. How fast can they recover from large storms, how can we accelerate recovery, and perhaps, most important, how do we stop the current rates of loss? The Consequences Over the life of the hurricane protection system, the region s demographics changed significantly. As more people and property occupied former marsh lands, the harsh consequences of flooding increased dramatically. Continuous pumping of groundwater accelerated subsidence and placed many residential areas well below sea level. Residential areas suffered 78 percent of the property damage from Katrina, and many of these areas existed at the city s lowest elevations. Only 8 of 37 neighborhoods were not flooded, creating a massive Map showing estimated flooding in Orleans Metro area that would have occurred if no levee or floodwall breaching occurred. This estimate also includes full pumping capacity. This scenario results in approximately half the economic losses of the actual flooding, a testimony to the value of resilience in hurricane protection structures. disruption to the social and cultural fabric of the area and significantly impeding recovery. Seventy five percent of the fatalities were individuals over the age of 60, many of who were infirm, disabled, or poor. The people who were less likely to be able to care for themselves or self-evacuate were clearly more vulnerable to flooding. The shifting demographics of coastal regions represent a major challenge for the nation. Some estimate that increasing development in vulnerable coastal zones is the dominant cause of increases in risk (Pielke 2006), a cause significantly greater than that presented by the increasing hurricane hazard or the lack of integrity of structures. Our national water policy does not deal with this issue well. There is little leverage at the national level beyond the flood insurance program to steer development away from areas vulnerable to flooding. Land use and development authorities primarily reside at the local level where many communities rely on development to pay the bills. There is little incentive to preclude development in flood prone areas as long as the federal system permits people to move behind levees and floodwalls and be relieved of the responsibility to purchase insurance and acknowledge their risk. It is also easy to falsely assume that the impact of an event like a hurricane induced flood is primarily local. The consequence analysis for Katrina looked at regional and national impacts, demonstrating that this is not true. The cost of recovery in New Orleans will far exceed the direct losses in property. The economic cost to the nation will likely far exceed all local losses and recovery costs. The social consequences of the resettlement of flood victims will be felt in New Orleans for decades. Policy Dilemma All of these changes an increasing hazard, a system that was not a system, and increasing consequence of flooding amounted to long-term incremental increases in risk. Most knew intuitively that it was there and that it was a big problem, but since there was no comprehensive methodology available to assess and monitor that risk, it was undefined. Obviously, our current policies and practices have not dealt well with change. As a nation we are usually focused on the short term, and complacency with respect to infrequent hazards is too often a part of September-October

4 our culture. It is reflected in our incremental and often piecemeal funding of major projects, our satisfaction with achieving a 1:100 level of protection to relieve us of the economic burden of buying flood insurance even when life and safety are paramount factors, our propensity to use economic cost-benefit ratios as a primary planning criteria, our history of depending on structural solutions that too often reduce the natural protection mother nature provides, and our propensity for continuing to place people and property in harms way. Others, such as the Dutch and Japanese, seek much higher levels of protection and tend to make life cycle or long-term investments in these areas. Unfortunately, in our current policy framework, life cycle benefits are only obvious and essential when you are caught without them. We now know that New Orleans likely had protection levels well below 1:100 in August of 2005 and that Katrina was far superior to whatever protection was there (Link 2006). We lack the national will to deal with long-term hazards with long-term strategies and long-term systematic investments. The same short-term emphasis pervades local, state, and national decisions. It is easy to point to the political process and question why it is oriented to the budget and election cycles rather than the life cycle more appropriate for infrastructure and natural hazards (Link 2006). The fact is, we, the public, are still the drivers of the political decision process. If we want to change this situation, we need to change what we are telling our representatives, be it through voting, writing letters, or reaching out through other available forums. The current policies and practices are short term because we are tolerating them, and they are costing citizens and the nation an enormous tax. Some are paying for it with their life s savings or worse, their lives. We will be recovering from Katrina for a long time, in New Orleans and nationally, and the cost will be staggering compared to the investments that would have dramatically reduced the losses from this single event. Understanding Risk Alvin and Heidi Toffler s recent book, Revolutionary Wealth, highlights three deep fundamentals that underlie many of the policy challenges we face in natural hazard risk reduction. These fundamentals of time, space, and knowledge nicely capture the major issues discussed here and the challenges faced by communities, regions, and the nation with regard to hurricane and natural hazard protection. Time represents the dynamics of the hazard, the system, the consequences, and the knowledge related to the levels of risk that existed. As a nation we do not anticipate or manage change well, too often resorting to the convenience of assuming a static or status quo world. Space represents the fact that local, state, regional, and national (federal) interests were and continue to be involved in the decisionmaking and investments for New Orleans hurricane protection. These interests and perspectives are often unsynchronized and in conflict, resulting in compromises that inadvertently increase risk. Space also relates to nationwide consequences. Just as with globalization, processes that are influenced separately at different scales are complex and harder to understand. As a nation, we do not deal well with issues that have multiple scales and complex interdependencies. They do not fit into the taxonomy by which our economic, legal, social, and political processes work, inherently increasing risk. Knowledge represents the increasing body of knowledge on all aspects of hurricane protection. Knowledge of climate dynamics exposes us to the fact that the future hazard can be significantly different from that of the past. Our knowledge of storm surge and wave processes gives us the capability to more accurately understand hazards and to design damage reduction measures that can adapt to change and be effective from a systems perspective. This same knowledge can materially improve forecasting of who and what might be vulnerable from an impending storm, dramatically increasing the focus and effectiveness of emergency response activities. New capabilities such as risk and reliability assessment will allow decisionmakers at all levels to better understand the short- and long-term costs (human life, environmental impacts, economics, and social well being) of alternative approaches to reducing risk. These same methods also provide a new level of understanding the sources of risk, enabling a more focused approach toward risk reduction from a life cycle and systems perspective, not to mention the ability to better inform the public of the level of risk they are assuming given their locale. While all this new knowledge represents a significant opportunity to reduce risk, it is of little value if it is not put to use. Conclusion We must forge a new framework for national water policy that accommodates change, customizes solutions that incorporate all approaches for reducing risk, and most importantly, works toward comprehensive, sustainable long-term solutions. This will take some time and careful thought by a diverse group representing many factions and stakeholders. Some of the key aspects of this are: 1. Adopt a risk-based decision support approach for informing everyone from the people on the street to public officials 2. Create a long-term water strategy for the nation that provides integrated approaches for achieving local, regional, and national economic, security, and quality of life objectives. Hazard protection should be integrated with other water management initiatives. 3. Develop adaptive approaches to risk reduction that deal with the changing nature of the hazard, protective systems, and consequences of natural hazards. Only a systems approach that integrates the natural and man-made measures can effectively deal with significant risk reduction in the long term. 4. Increase the body of knowledge concerning the natural coastal environment to better understand the short- and long-term roles that coastal environments can play in a systems approach to risk reduction. There is a critical need for more knowledge on the regeneration of natural environments decimated by long-term activities or episodical threats like large storms. Continued on page national wetlands newsletter

5 Thomas, continued from page 32 Katrina Play Out?, Findlaw, Sept. 19, 2005, at 17 See Chu, supra note See Joseoph Treaster, Big Insurer Will Pay 640 Kartina Claims, N.Y. Times, Jan. 24, 2007, at &en=87b9f8b851669b92&ei= See Joseph Treaster, Judge Puts Katrina Settlement in Question, N.Y. Times, Jan. 27, 2007, at &en=87b9f8b851669b92&ei= See, National Flood Insurance Program: Issues Exposed by the 2005 Hurricanes: Hearing Before the H. Subcomm. on Oversight and Investigation and the H. Subcomm. on Management, Investigations and Oversight of the Committee on Homeland Security, 110th Cong., June 12, 2007 (Statement for the Record, Matt Jadacki, Deputy Inspector General for Disaster Assistance Oversight, U.S. Department of Homeland Security). 21 See Liam Pleven, As Insurers Flee Coast States Face New Threat, Wall Street J., June 7, 2007, at A See Gary Michiels, State Farm Won t Seek New Mississippi Policies, Nat l Pub. Radio, June 22, 2007, at php?storyid= See, e.g., ProtectingAmerica.org, at 24 See Edward A. Thomas, No Adverse Impact: Working Together To Prevent Harm, Nat l Wetlands Newsl., Jan.-Feb. 2007, at See, e.g., Multiple Peril Insurance Act of 2007, H.R Link, continued from page 16 In short, Katrina caught us with our risk up and our guard down. To address these issues, we must think big picture and long term. Reform of individual organizations or practices of the overall process will not solve the problem. It will require a holistic look at how the entire system of governments behaves, what drives it, and the interdependencies in the context of time, space, and knowledge. References Interagency Performance Evaluation Task Force (IPET) Performance Evaluation of the New Orleans and Southeast Louisiana Hurricane Protection System, 9 Volumes, Draft Final. IPET. 2007a. Report of the IPET, Volume I, Executive Summary and Overview, Interim Final, available at IPET. 2007b. IPET Report, Volume 7, Final, Consequences, available at IPET.WES.Army.Mil. Independent Levee Investigation Team (ILIT) Investigation of the New Orleans Flood Protection Systems in Hurricane Karina on 29 August, Link, Lewis E Stopping Another National Disaster, Viewpoint, Engineering News-Record, Dec. 11, Louisiana Coastal Protection and Restoration Preliminary Technical Report to the United States Congress (Louisiana Coastal Protection and Restoration Study), U.S. Army Corps of Engineers, Washington, DC. Pielke, Roger, Jr An Eye to the Future: Emerging Issues in Flood Risk Management, Wye River National Flood Risk Policy Summit, Queenstown, MD, Dec , 2006 Resio, Donald, T White Paper on Estimating Hurricane Inundation Probabilities, U.S. Army Corps of Engineers Engineer R&D Center, Vicksburg, MS, 29 January, Probability Methodology Optimal Sampling Toffler, Alvin and Heidi Toffler Revolutionary Wealth. Alfred N. Knopf, New York. Boesch, continued from page 5 by climate change and its attendant accelerated sea-level rise and potentially more frequent or intense tropical cyclones. This requires sober appraisal of the consequences and adaptation strategies rather than denial, on one hand, or alarmism, on the other. Integrated coastal planning founded on a sustainable coastal landscape is an enormous scientific and political challenge that will require very large financial investments, will, and determination, as well as innovative and critical contributions from the science and engineering community. Given changes in the global climate, our options are quickly becoming more narrow and less desirable with time, thereby requiring urgent action. References American Geophysical Union Hurricanes and the U.S. Gulf Coast: Science and Sustainable Rebuilding. American Geophysical Union, Washington, DC, available at America s Wetland Envisioning the Future of the Gulf Coast. America s Wetlands available at Boesch, D.F., L. Shabman, et al A New Framework for Planning the Future of Coastal Louisiana after the Hurricanes of University of Maryland Center for Environmental Science, Cambridge, MD, available at edu/la-restore/. Coastal Protection and Restoration Authority Integrated Ecosystem Restoration and Hurricane Protection: Louisiana s Comprehensive Master Plan for a Sustainable Coast. Coastal Protection and Restoration Authority, Baton Rouge, available at Day, J.W., Jr., D.F. Boesch, et al Restoration of the Mississippi Delta: Lessons from Hurricanes Katrina and Rita. Science 315: Interagency Performance Evaluation Task Force Performance Evaluation of the New Orleans and Southeast Louisiana Hurricane Protection System. MMTF , U.S. Army Corps of Engineers, Vicksburg, MS, available at McQuaid, J., and M. Schleifstein Path of Destruction: The Devastation of New Orleans and the Coming Age of Superstorms. Little, Brown, New York, NY. National Research Council Drawing Louisiana s New Map: Addressing Land Loss in Coastal Louisiana. National Academy Press, Washington, DC. U.S. Army Corps of Engineers Louisiana Coastal Protection and Restoration Preliminary Technical Report. U.S. Army Corps of Engineers, New Orleans, LA, available at U.S. Army Corps of Engineers Louisiana Coastal Protection and Restoration Plan Formulation Atlas. U.S. Army Corps of Engineers, New Orleans, LA, available at September-October