Hurricane Sandy heightened the nation’s awareness of the vulnerability of coastal areas to hurricane damage. Eight U.S. cities (Miami, the New York-Newark region, New Orleans, Tampa-St. Petersburg, Boston, Philadelphia, Virginia Beach-Norfolk, and Baltimore) are among the top 20 cities in the world at risk from coastal storms, based on an estimate of potential average annual flood loss of valuable assets (e.g., buildings, transportation, utilities, personal property) (Hanson et al., 2011; Hallegatte et al., 2013). Other large cities along the East and Gulf Coasts, such as Houston, Texas, and countless smaller cities and developed areas, are also vulnerable to coastal storms. New York, New Orleans, and Miami were poorly prepared for a major storm as shown by Hurricanes Sandy (2012), Katrina (2005), and Andrew (1992). If not adequately prepared, coastal cities and developed areas are extremely vulnerable to hurricanes, which can leave many thousands of people homeless, cause extensive property damage, and result in short- and long-term economic disruptions. This chapter provides an introduction to the coastal storm-related risks (hereafter, termed “coastal risks”) faced along the U.S. East and Gulf Coasts, and discusses how those risks have changed and are continuing to change. General strategies that can reduce risk and help make communities more resilient to coastal storms are also discussed.
The United States has experienced extensive and growing loss from natural disasters. Dollar losses due to tropical storms and floods have
tripled over the past 50 years (accounting for inflation; Gall et al., 20111) and currently comprise approximately half of all natural disaster losses (Table 1-1, Figure 1-1). In addition to growth in absolute dollars, per capita natural disaster losses have also grown as have losses normalized by income, highlighting the growth in their relative economic impact (Gall et al., 2011). Appendix A provides a table of major coastal storms that have struck the United States since 1900—most of which made landfall on either the East or Gulf Coasts.
From 1980 to 2013, there were 151 weather- or climate-related natural disasters that caused a direct economic impact on the United States of greater than 1 billion dollars (in 2013 dollars).2 Tropical cyclones (including tropical storms and hurricanes) compose the single largest category, accounting for 33 of the events (or 22 percent) and 49 percent of the total damage (Table 1-1). During this period, when averaging over 5-year periods, tropical cyclone events causing billion-dollar losses increased from approximately 0.4 per year to over 1 per year, and the losses increased from approximately $1.75 billion per year to as high as $45 billion per year in the 5-year span that includes Hurricane Katrina (Figure 1-1). This increase follows a much more gradual upward trend in tropical cyclone–related economic losses extending back to at least 1900 (Pielke et al., 2008).
Causes of Increasing Disaster Losses
There are two primary reasons for the dramatic increase in natural disaster–related losses: an increase in the people and property in harm’s way and an increase in the frequency or severity of the hazard events. Pielke et al. (2008) concluded that growth in tropical cyclone–related economic losses in the United States since 1900 has been minimally influenced by changes in storm climatology and rather is primarily explained by the movement of people and accompanying wealth to areas that are at higher risk. From the east coast of Florida through the Gulf Coast, population
1Gall et al. (2011) data included direct loss estimates from the Special Hazard Events and Losses Database for the United States (SHELDUS), federal individual and public assistance and some Hazard Mitigation Grant Program spending associated with presidential disaster declarations, National Flood Insurance Program claims, and privately insured hazard claims.
2See http://www.ncdc.noaa.gov/billions/events. The total insured and uninsured direct losses considered include “physical damage to residential, commercial and government/municipal buildings, material assets within a building, time element losses (i.e., time-cost for businesses and hotel-costs for loss of living quarters), vehicles, public and private infrastructure, and agricultural assets (e.g., buildings, machinery, livestock).” The reported loss assessments do not include “losses to natural capital/assets, healthcare related losses, or values associated with loss of life” (Smith and Katz, 2013).
TABLE 1-1 Damage, Percent Damage, Frequency, and Percent Frequency by Disaster Type, 1980-2012, for All Billion-Dollar Weather- and Climate-Related Events in the United States (adjusted for inflation to 2013 dollars)
|No. of Disaster Events||No. of Deaths||Adjusted Damage (billion $)||% Damage||% Occurrence|
|Severe local storms||55||1,391||111.8||11.1||36.4|
NOTE: Damage cost totals do not include 2013 events, from which damage data are not yet available.
SOURCE: Data from http://www.ncdc.noaa.gov/billions/events.
FIGURE 1-1 Average number and costs associated with billion-dollar coastal storm events in the United States between 1980 and 2013 by 5-year time increments.
SOURCE: Data from http://www.ncdc.noaa.gov/billions/events.
has grown rapidly for at least the past 80 years. Presently, coastal counties along the entire East and Gulf Coasts, which account for 24.2 percent of the U.S. population, grew by 11.4 percent from 2000 to 2012, essentially matching population growth across the United States. However, coastal population growth has been substantially skewed toward the southeast and Gulf of Mexico coastal areas, which grew by 20.8 and 17.8 percent, respectively, during this time period (Table 1-2). These areas are the most frequently impacted by hurricanes and tropical storms (Figure 1-2) and they typically have low topographic slope, meaning they lack the most effective natural defense against coastal storm surge and wave damage—vertical elevation of the land near the water’s edge.
Natural cycles in tropical cyclone activity and observational bias in data sets before the modern satellite era (mid-1960s) make determining historical trends in either the frequency or severity of tropical cyclones during the past 100-150 years difficult (Landsea, 2007). However, despite clear increases in global mean temperature and tropical Atlantic sea surface temperature, statistically significant trends do not appear to exist in the number of Atlantic basin hurricanes or U.S. land-falling hurricanes since at least 1875 (Vecchi and Knutson, 2011; GFDL, 2013). Conversely, using six high-quality tide-gauge records (dating to 1923) from the southeastern United States, Grinsted et al. (2013) found that storm surge statistics were correlated with global temperature. This study identified a doubling of the likelihood of a Katrina-magnitude storm surge during the 20th century, which could be a significant finding because the oceanic response, represented by storm surge and waves, is usually the most destructive aspect of a hurricane.
TABLE 1-2 Population Growth in U.S. East Coast and Gulf of Mexico Coastal Counties
|Coastal County Population|
|Gulf of Mexico||14,800,000||17,500,000||17.8|
|Total 4 regions||68,200,000||75,900,000||11.4|
|East and Gulf Coast County Population as Percentage of Total U.S. Population||24.2||24.2|
SOURCE: Data from the National Ocean Economics Program, www.oceaneconomics.org.
FIGURE 1-2 Estimated average return periods (years) for all hurricanes (top) and major hurricanes (category 3 or greater, bottom) passing within 50 nautical miles (90 km) of various locations on the U.S. East and Gulf of Mexico Coasts.
SOURCE: Data from http://www.nhc.noaa.gov/climo/.
Although increases in coastal development in high hurricane hazard areas appear to have dominated the growth in coastal natural disaster–related economic losses for much of the past century, this may change in the future. Even though the total number of hurricanes is predicted to decrease in the 21st century, research suggests that climate warming may increase the intensity of hurricanes and the frequency of the strongest storms (i.e., category 4-5 hurricanes) (Bender et al., 2010; Emanuel, 2013; Knutson et al., 2013). Bender et al. (2010) estimated that in the Atlantic basin, the increase in the number of strong storms will outweigh the reduction in overall hurricane numbers yielding roughly a 30 percent increase in potential damage by 2100. Hurricanes are also projected to have higher rainfall rates than today’s hurricanes.3
In addition to changes in storm climatology, sea-level rise is raising the level of the coastal ocean relative to the land. As a result, coastal cities are increasingly exposed to flooding, and beaches and wetlands are subject to deterioration from storm surge and wave action (Titus et al., 2009; Sallenger et al., 2012). Globally, relative sea-level rise—the observed change of sea level relative to land surface at a particular point, thereby considering other factors such as subsidence—has averaged 0.12 in/yr (3.1 mm/yr) over the past two decades. However, local rates of sea-level rise vary considerably, with the largest rates in the United States in areas of the northern and western Gulf of Mexico and the mid-Atlantic (Figure 1-3) (Sallenger et al., 2012; Ezer et al., 2013). The impacts of sea-level rise over the past century can already be seen in the frequency of flooding that occurs in many low-lying areas. For example, parts of Norfolk, Virginia, that saw significant flooding only during hurricanes in the 1930s, now flood during high tides and minor storm events and therefore spend substantial amounts of time underwater (VIMS, 2013).
The most recent report from the Intergovernmental Panel on Climate Change (IPCC, 2013) predicts that climate warming will cause a mean increase in sea level by 2100 from 1.4 to 2.4 ft (44 to 74 cm). NRC (2012c) predicted an even larger increase (1.7 to 4.6 ft [51 to 140 cm]) by 2100. These increases have the potential to bring enormous damage because nearly 5 million people and 2.6 million homes in the United States are found at less than 4 ft (1.2 m) above high tide (Climate Central, 2012).
Assuming that sea level rises by only 1.6 ft (0.5 m) by the year 2100, Sweet et al. (2013) calculated that the return periods for Hurricane-Sandy–level storm surges would be reduced by a factor of approximately 4, with higher sea-level rises further reducing the intervals between major inundation events. Lin et al. (2012) considered an ensemble of tropical storm
FIGURE 1-3 Rates of relative sea-level rise (mm/yr [ft/century]) along the U.S. East and Gulf Coasts.
SOURCE: Data from http://tidesandcurrents.noaa.gov/sltrends/sltrends.shtml.
scenarios associated with climate change and found that by 2100, assuming 3.3-ft (1-m) sea-level rise, today’s 1 percent annual-chance (100-year) flooding event in the greater New York City area may increase in annual probability to 5 percent (a 20-year flood) or more. The simple addition of elevated sea levels and existing storm surge risk can be used to create approximate estimates of the spatial extent of areas that are at risk under future sea levels (Figure 1-4).
The recently released U.S. National Climate Assessment (Melillo et al., 2014) identifies the U.S. Gulf of Mexico and Atlantic Coasts as being subject to increased risk of storm surge damage and flooding due to sea-level rise combined with coastal storms. Impacts will occur to homes, critical infrastructure, cultural and historic resources, agriculture, ports, tourism, coastal resources, and coastal ecosystems. Vulnerability to these impacts is uneven due to socioeconomic disparities throughout the region.
FIGURE 1-4 Current Federal Emergency Management Agency 1 percent chance (100-year) floodplain (purple) and an approximation of the extent of flooding with the same 1 percent probability considering two sea-level rise scenarios: 11 inches (28 cm; yellow) and 31 inches (79 cm; red).
SOURCE: Adapted, with permission, from NPCC (2013).
Shifting Federal Roles
Concurrent with the growth in natural hazard economic losses, there has also been a substantial shift in the source of funds used to cover these losses in the United States. The federal government’s assistance to disaster victims is well illustrated by the large increase in the past 60 years in the number of Presidential disaster declarations that have occurred (from ap-
proximately 10 to nearly 100 per year for all weather-related disasters) and a similar relative increase (from approximately 1 to 10 per year) in coastal storm-related Presidential disaster declarations4 (Figure 1-5). There has also been a substantial increase in the percentage of severe storm-related damages covered by federal aid over this period, from 6 percent for Hurricane Diane in 1955 to more than 75 percent for Hurricane Sandy (Table 1-3). Abundant federal assistance has raised concerns of a “moral hazard” in which state and local government leaders are discouraged from investing in disaster mitigation and preparedness because they expect to be “bailed out” by the federal government (Sylves and Buzas, 2007). Federal programs supporting coastal risk management and disaster recovery are discussed in Chapter 2.
Together, the growth in coastal disaster losses associated with population redistribution, the looming implications of climate change, including sea-level rise, and the shift in the fiscal responsibility for disasters illuminate pressing challenges ahead in coastal risk management.
Full protection from coastal hazards and related damages is typically impractical at community to national scales. Even the largest levees or surge barriers could be overtopped by a large storm or suffer from structural failures. Thus, local, state, and federal governments are increasingly recognizing the importance of becoming more resilient to hazards and disasters, including coastal hazards. NRC (2012b) defines resilience as “the ability to prepare and plan for, absorb, recover from, or more successfully adapt to actual or potential adverse events.” Resilience depends on the reliability of community service systems in the face of significant disturbance or the capability to recover those services within an acceptable time period, thereby enabling a community to maintain its economic, communications, transportation, social, political, and quality-of-life functions (Tierney et al., 2001; DHS, 2007; Tierney and Bruneau, 2007). Resilience planning, therefore, focuses on the specific needs of the community served and the capacity to provide the necessary services throughout the recovery period (Corotis, 2011; NRC, 2011a).
Resilient communities are able to assess and manage risks, are generally well informed of threats, and are clear about the roles and responsibilities of individuals and organizations in the community with respect to risk (NRC, 2012b). Resilient communities take into account both pre-disaster mitigation measures and post-disaster recovery measures to determine an appropriate allocation of resources to improve resilience within budgetary
FIGURE 1-5 U.S. Presidential disaster declarations for hurricanes and coastal storms by year, 1953-2013.
SOURCE: Data from http://www.fema.gov/disasters/grid/year.
constraints. Pre-disaster mitigation can prevent property damage and some business and infrastructure impacts, but resilience can also be improved by strategies to recover more quickly (Rose et al., 2007). Actions to enhance resilience that can be implemented at the local level prior to a disaster include emergency planning drills and disaster planning for businesses (e.g., increasing inventories, identifying alternative supply-chain sources and operating locations). Other actions can be taken following a disaster, such as business relocation and conservation of critical supplies.
Understanding, managing, and reducing risk are foundations for building resilience. Risk is “the potential for hazards to cause adverse effects on
TABLE 1-3 Change in Percentage of Federal Aid Following Major Tropical Cyclones, from 1955-2012
|Disaster||Federal Aid as a Percentage of Total Damage|
|Hurricane Sandy (2012)||>75|
|Hurricane Ike (2008)||69|
|Hurricane Katrina (2005)||50|
|Hurricane Hugo (1989)||23|
|Hurricane Diane (1955)||6|
SOURCE: Michel-Kerjan (2013).
our lives; health; economic well-being; social, environmental, and cultural assets; infrastructure; and the services expected from institutions and the environment” (NRC, 2012b). In natural hazard and disaster fields, risk for a particular hazard, place, and time period is represented as the probability of a hazardous event multiplied by its consequence (Box 1-1; BSI, 2002; Gouldby and Samuels, 2005; UNISDR, 2009). Hazard refers to the physical event with the potential to result in harm (Gouldby and Samuels, 2005). Thus, flooding or overland waves caused by hurricanes or other strong coastal storms are the primary hazard—not the storm or the coastal storm surge itself. Consequence represents the impact caused by the hazard. Consequence can encompass a range of values, such as economic damage (monetary), number of people or properties affected, harm to individuals (e.g., fatalities, injuries, stress), and environmental impacts. Consequence is controlled by exposure (density of people, property, or other elements in hazard zones [UNISDR, 2009]) and vulnerability (a system’s potential to be harmed, which is a function of both the susceptibility to experience harm and the value, expressed in monetary or other terms of the people,
Components of Risk
For purposes of quantitative risk assessment, risk is represented as the probability of a hazard multiplied by the consequence:
R = H × C
in which R = risk, H = probability of the occurrence of a hazard (e.g., storm-induced flooding), and C = consequence. Consequence represents the impact and can be measured in various units including monetary damage, number of people or properties affected, harm to individuals (e.g., fatalities, injuries, stress), and environmental impacts. Consequence may be expressed as a function of the exposure (E) (the density of people, property, systems, or other elements present in hazard zones) and the vulnerability (V), which is a system’s potential to be harmed:
C = f(E, V).
Vulnerability can be defined in terms of the susceptibility to harm and the value (in monetary or other terms) of the people, property, systems, or elements present in hazard zones.
SOURCE: Data from Gouldby and Samuels (2005).
property, or other elements in the hazard zones [Box 1-1; Gouldby and Samuels, 2005]).
Coastal risk reduction focusing on the hazard is typically achieved through hard structural measures (such as construction of seawalls or levees) or nature-based approaches (such as building dunes) to reduce the wave and flood hazard probability. Risk reduction focusing on the consequence is typically achieved by an array of measures that change exposure (e.g., relocating homes and businesses away from high-hazard areas or evacuating prior to a storm event) or reduce vulnerability (e.g., elevating structures or enhancing risk awareness) (see Table 1-4). In the past decades, much more attention has been placed on strategies that reduce the probability of flooding than those that reduce exposure to storm events (i.e., the extent to which we live in harm’s way) (NRC, 2012b). To improve coastal risk management, it will be important to consider options that will address both sides of this risk-exposure equation.
Risk management is a continuous process that identifies the hazard(s) facing a community, assesses the risk from these hazards (Box 1-2), develops and implements risk reduction (mitigation) measures, reevaluates and reviews these measures, and develops and adjusts risk policies. If done well, risk management should help build capacity for communities to become more resilient to disasters (NRC, 2012b). For example, risk reduction efforts that place more value on critical infrastructure (e.g., hospitals, water and wastewater treatment facilities, power plants) than on other infrastruc-
TABLE 1-4 Risk Reduction Measures Linked to Components of Risk Reduction
|Coastal Risk Mitigation Measures||Risk Reduction|
|Probability of Hazard (Flooding, Wave Damage)||Consequence|
|Levees, sea walls||X|
|Beach nourishment and dune building||X|
|Elevating and flood-proofing structures||X|
|Flood warning and preparedness programs||X||X|
aIf flood insurance is appropriately priced, the result should communicate risk and may spur additional mitigation measures, thus reducing vulnerability in addition to transferring risk to a broader risk pool.
Evolution of Coastal Risk Assessment
Early coastal risk assessment in the United States was based on deterministic characterizations of hazards, invoking “design storms” (e.g., the standard project hurricane—the most severe storm reasonably characteristic of the project area—or the probable maximum hurricane—the most severe storm thought possible in the project area [Graham and Nunn, 1959; NOAA, 1979; Woolley and Shabman, 2008]) that were presumed to be appropriate cases for design. Within the United States, the U.S. Army Corps of Engineers (USACE) primarily used the Standard Project Hurricane to set design water levels and associated hazards in its design projects, while the Nuclear Regulatory Commission used the probable maximum hurricane for this purpose. The stipulation that the USACE design event was linked to a storm that was only “reasonably characteristic of the project” area, rather than the maximum possible storm, implied a recognition that such designs were potentially vulnerable to future storms; however, risk assessment approaches in that era considered only a few discrete failure modes and their outcomes. Although this design simplification was consistent with the approach followed in other large engineering projects at that time, it ignored the full range of storm characteristics, uncertainty in the performance of levees and other protective systems under storm loading, and the likelihood of a particular hazard result. Because of the shortcomings of deterministic risk assessment methodologies, probabilistic risk assessment has become the basis of modern risk assessment. Probabilistic risk assessment uses quantitative calculations and models to compute the probabilities that certain hazards occur, the systems’ response to those hazards, and the consequences associated with adverse outcomes of the systems’ response. Thus, its results show not only what could happen, but also how likely each outcome is to occur. Under good professional practice, uncertainty is also quantified and integrated into the decision process (NRC, 1994; IOM, 2013).
ture can help improve community resilience by allowing more rapid recovery from a disaster with less disruption to critical services. The impacts of Hurricane Sandy have led to recommendations for increased consideration of critical infrastructure in comprehensive coastal risk assessments and risk reduction planning (USACE, 2013d).
Even after risk reduction measures are taken, some risk will remain because no risk reduction measure ever provides absolute protection. The risk that remains is referred to as residual risk. In the coastal zone, residual risk exists because storms larger than those designed for may occur, or the risk reduction measures put in place have a possibility of failing to perform as designed. Communities can work collectively to determine an acceptable level of residual risk based on their risk tolerance and the benefits and costs of additional risk reduction measures (see Chapter 4).
Once calculated, risk can be used in several different decision-making approaches. A risk-standard (or “level-of-protection”) approach recommends investment in coastal risk reduction measures to drive residual risk below a specified level (such as a 1 percent annual chance of exceedance, also known as a 100-year event; Box 1-3). Congress specified the use of a 1 percent annual chance of exceedance as the design basis of the Hurricane and Storm Damage Risk Reduction System around greater New Orleans (USACE, 2013b). A benefit-cost approach determines worthy coastal risk management investments based on a comparison of benefits (measured as the value of risk reduction) to the costs of investment. Hybrid approaches are also common (see Chapter 4).
Application of either approach requires careful consideration of the long-term effects of risk reduction strategies on overall risk. Measures designed to reduce risk by decreasing the probability of the hazard, may encourage increased exposure (e.g., additional development or redevelopment) and/or increased vulnerability (e.g., higher-priced homes, risk complacency) in the hazard area and, in the long run, lead to higher risk. These risk reduction measures may thus decrease the negative consequences of small or moderate events, but increase the negative consequences of catastrophic events (Box 1-4; Hallegatte, 2012; NRC, 2013). Elevating homes in a coastal area above storm surge levels may reduce vulnerability but encourage expanded development, thereby increasing exposure to severe floods as well as other hazards such as wind or coastal erosion. Also, the expanded development may encourage investment in public infrastructure (e.g., roads, water, sewer, communications, emergency services) that are then subject to hazard damage.
Numerous designs and strategies can be used to mitigate coastal risk associated with severe storms. These include measures to reduce the hazard, such as seawalls, breakwaters, and levees; natural and nature-based features, including wetlands, natural and replenished dunes, and mangrove forests; and strategies to reduce the consequences of an event, such as land-use planning, floodproofing, and relocation (USACE, 2013a).
The primary hazards under consideration in this report are flooding and wave attack. Mitigation of coastal flooding during severe storms is largely dependent on defending against or reducing the vulnerability to storm surge. Many oceanic responses, including wind waves, swell, tides, and surge, can be classified under the general category of waves. However, throughout this report (predominantly below and in Chapter 3) waves are considered to represent only relatively short time- and spatial-scale responses to wind forcing that pass a given location in a matter of seconds
History of the 1 Percent Annual-Chance (100-Year) Flood
The concept of the “100-year” event is ever present in the probabilistic characterization of natural hazards. In recent years the trend has been to call this the “1 percent chance” event, to emphasize that the event could happen at any time.
Although considerations of annual flood hazard criteria arose in the United States by the mid-20th century (ASFPM Foundation, 2004), Executive Order 11296, signed by President Johnson in 1966, first directed federal agencies to take flood probabilities into account when making decisions in locating federally owned buildings and roads. Shortly after, in 1968, the National Flood Insurance Program (NFIP) was established to reduce future flood damages and federal disaster assistance expenditures through community-based floodplain ordinances and flood insurance (NRC, 2013). Neither the executive order nor the National Flood Insurance Act of 1968, however, defined a standard criterion for flood hazard areas.
In December 1968, a special committee of experts convened by the University of Chicago recommended that the 1 percent chance (100-year) event be considered an initial standard for the NFIP, and the Flood Insurance Administration formally established the 1 percent chance event as the regulatory standard for the NFIP in 1971 (Wright, 2000; Galloway et al., 2006). Purchase of flood insurance was required to obtain a mortgage from a federally regulated or insured lender for those living in the 1 percent chance (100-year) flood hazard area, although this requirement was later waived for properties located behind structures designed to protect against such an event. In 1972, the Federal Water Resources Council recommended that agencies use the 1 percent chance event as the baseline flood in floodplain usage decisions, although other standards were permitted when appropriate (Robinson, 2004). The 1 percent chance event was selected “because it was already being used by some agencies, and because it was thought that a flood of that magnitude and frequency represented both a reasonable probability of occurrence, a loss worth protecting against and an intermediate level that would alert planners and property owners to the effects of even greater floods” (Robinson, 2004). It did not represent an attempt to achieve optimal balancing of risks and benefits. Ultimately, it represented a compromise between decision makers and those who would be affected by its implementation, and it provided “a point of departure for adjustments that could reflect the differences that might exist in floodplains across the country and in the objectives of the States and localities that would implement the standard” (Galloway et al., 2006).
Coastal flood standards in many developed countries are far stronger (less probable) than the 1 percent chance event. For example the Netherlands and Japan use the 0.01 percent chance (10,000-year) event for some coastal works (Galloway et al., 2006), although the derivation of the flood level for such a rare event from limited duration observations introduces inherent uncertainties. Methods for dealing with those uncertainties have been (Roscoe and Diermanse, 2011) and continue to be developed. Many U.S. studies have concluded that the 1 percent chance event is inadequate as a flood risk reduction design basis for urban areas (e.g., Galloway et al., 2006; ASFPM, 2007; NRC 2009).
Flood Protection Led to Increased At-Risk Development
In 1965, after flood damages from Hurricane Betsy, Congress authorized a hurricane protection levee project along Lake Pontchartrain and vicinity, designed to protect the main urban areas of New Orleans from flooding from a Standard Project Hurricane that was described as having a likelihood of occurrence of approximately 1 in 200 years and the characteristics of a fast-moving category 3 hurricane (USACE, 1965; GAO, 2005). A feasibility study of the project found that flood protection for existing development accounted for 21 percent of the benefits, while the remaining 79 percent was associated with flood protection for new development, made possible by the enhanced levee system (GAO, 1976). In the decade after authorization of the Lake Pontchartrain project, Jefferson Parish added 47,000 housing units and Orleans Parish added 29,000 in the former low-lying wetland areas.
The development of the area east of the Industrial Canal, which contains 50 percent of New Orleans’ land area, is especially suggestive of the interaction between flood risk reduction measures and development. In 1960, before the new levee plan, eastern New Orleans consisted of a few scattered residential and commercial structures. In anticipation of construction of Interstate 10 and the extension of the city’s levee system, the city adopted a comprehensive plan in 1966 that designated the area for intensive urban development. In the 1970s, this area experienced development of 22,000 new housing units. In a retrospective assessment of the area’s development trends, the city’s 1999 land-use plan stated: “Full scale development ensued … (and) the area continued to grow from 1975 to 1985. New subdivisions were developed at a rapid pace … (and) major commercial centers developed and prospered.” (New Orleans City Planning Commission, 1999). In 2005, the entire area of urban growth that was proposed to be reasonably safe because of levee investments was flooded by Hurricane Katrina due to overtopping of design levels and structural failures at levels below the project design (IPET, 2009; Figure 1-4-1). Altogether, Hurricane Katrina caused over $148 billion in damages (in 2013 dollars) and 1,833 deaths.a
to minutes and have wave lengths measured in feet (or meters). These are commonly called wind waves or swell. Storm surge represents a much larger- and longer-scale response, sometimes inundating an area for hours, with wave lengths measured in miles (or kilometers). Storm surge is caused by the combination of winds, atmospheric pressure, the rotation of the earth, and wave-induced setup (Dean and Dalrymple, 2002). At any given time, the total coastal water level is composed of the astronomical tide plus storm surge, wave height, and freshwater input (if important).5 Due to their
FIGURE 1-4-1 Flooding of New Orleans after Hurricane Katrina.
very different time and spatial scales, storm surge and waves respond quite differently to hazard mitigation strategies and therefore these responses are discussed individually.
Measures to Reduce the Hazard—Hard Structures
Hard structural measures to address coastal storm hazards are typically static, engineered features designed to reduce wave damage and flooding, and they may also decrease shoreline erosion. Sometimes termed “gray
FIGURE 1-6 Seawall along the coast of Galveston Island, Texas.
SOURCE: Photo courtesy of Melanie Fitzpatrick, Union of Concerned Scientists.
infrastructure” or “hard engineering,” these structures include seawalls, levees and floodwalls, and surge barriers:
- Seawalls are constructed parallel to the shoreline to reduce impacts from storm surge and waves to developed lands behind the seawall. Seawalls may be vertical or curved walls (Figure 1-6) or designed as a mound built from rock or concrete blocks. The seawall reflects wave energy back to the sea, and therefore can increase erosion on the coastal side of the wall. Depending on lateral currents, seawalls may also cause increased erosion of adjacent, unprotected coastal areas.
- Levees and floodwalls are onshore engineered structures most commonly constructed along riverine floodplains that are designed “to contain, control, or divert the flow of water so as to provide protection from temporary flooding” (44 CFR § 59.1). Levees (sometimes also called dikes) are typically wide earthen embankments that are designed to control flooding over a large area up to a specific water level. Levees, however, can also be used in coastal settings, where
they may be paired with other mitigation features, such as revetments or coastal wetlands that buffer the levee against erosive wave forces. Floodwalls—typically vertical concrete walls—are usually constructed in areas where there is insufficient space for the wide footprint of an earthen levee. Floodwalls can also be constructed on top of a levee when space limits any further expansion of the levee footprint that would be required for increasing the height of the levee itself (Figure 1-7).
- Storm surge barriers are designed to block storm surges from propagating inland via rivers or other waterways (Figure 1-8). Gates in the barriers are left open to allow water to flow through under normal conditions but can be closed when storm surges are expected.
Other engineered measures, such as breakwaters (offshore rock mounds or concrete armor units), revetments (onshore armoring constructed of stone or concrete), and bulkheads (short vertical walls common in estuarine settings) are primarily intended to reduce coastal erosion but also serve to reduce wave energy that accompanies storm surge. Hard structural coastal risk reduction measures were commonly used by the USACE in the 1950s and 1960s, but their use decreased beginning in the 1970s (Figure 1-9).
Measures to Reduce the Hazard—Natural and Nature-Based Features
The presence of natural features, such as barrier islands, vegetated dunes, coastal wetlands, mangrove forests, and reefs, may reduce coastal storm hazards by attenuating wave energy and storm surge and possibly stabilizing sediment. However, the effectiveness of these features depends on the specific characteristics of the storm and the features themselves. Dunes serve as a physical barrier blocking storm surge, although their longevity depends on the adjacent beach slope, the sediment characteristics, the height and width of the dune, and the extent of dune vegetation. Coastal wetland vegetation and the land it helps retain may reduce the rate of storm surge advancement and extent (Wamsley et al., 2010). Mangroves are capable of damping incident waves, reducing wind speed within the canopy, and potentially reducing storm surge, depending on their lateral extent (see Chapter 3; McIvor et al., 2012; Zhang et al., 2012).
Nature-based coastal risk reduction strategies are designed and engineered to mimic natural features for the purpose of attenuating storm surge. The most commonly applied coastal risk reduction strategies in the United States are dune building and beach nourishment (also called beach fill), (Figure 1-10). Periodic nourishment with sand from offshore locations creates a wide beach area to absorb the energy of breaking waves, and replenished sand dunes can serve as a physical barrier, albeit a dynamic one,
FIGURE 1-7 Schematic of a floodwall paired with a levee for flood risk reduction along a canal (top) and a levee with floodwall along the London Avenue Canal in New Orleans, Louisiana (bottom).
SOURCES: USACE (http://library.water-resources.us/docs/MMDL/FLD/Feature.cfm?ID=2);http://en.wikipedia.org/wiki/File:London_West_from_Robert_E_Lee_to_DPS4_0001.jpg).
FIGURE 1-8 Fox Point hurricane barrier on the Providence River in Providence, Rhode Island, built in 1966.
SOURCE: Photo courtesy of Neil Aquino.
to reduce flooding and destructive wave energy on structures located behind the dunes. However, replenishment brings additional ecological impacts, as discussed in Chapter 3, and replenished beaches may have a different slope than natural beaches, which can alter the incident wave conditions. Other potential nature-based coastal risk reduction strategies include conservation and/or construction of wetlands and oyster reefs, which may also provide additional ecosystem services benefits.
Measures to Reduce the Consequences
Consequence reduction measures aim to reduce the exposure or vulnerability to a hazard. These approaches include elevating and floodproofing structures (and related building codes) and nonstructural strategies, such as flood warning and emergency preparedness programs, flood insurance, land-use regulations, restrictions on development in areas of severe flood hazard, and property acquisition and relocation programs (see Table 1-4). These strategies are sometimes broadly called “nonstructural” measures,
FIGURE 1-9 Percentage of total USACE coastal risk reduction expenditures (top) between hard structural measures and beach nourishment (including dune building) projects and miles of project (bottom) by decade. Recent cost data are not available, but the percentage of overall coastal risk reduction costs represented by hard structural measures has likely increased in the past decade with the post-Katrina construction of the Hurricane Storm Damage Risk Reduction System. Data used to compile the bottom figure are listed in Appendix B.
SOURCE: USACE (1996); D. Cresitello, USACE, personal communication, 2014.
FIGURE 1-10 Beach nourishment in Ocean City, New Jersey.
SOURCE: NOAA (2007).
although to minimize confusion, for the purposes of this report, the term “nonstructural” does not include floodproofing and elevation of individual structures.
Flood preparedness programs might include delineation of flood hazard areas, effective communication of risks to community residents and developers, development and communication of evacuation plans, and flood insurance for those at risk of flooding. If appropriately priced, flood insurance serves as both a risk transfer mechanism and an effective risk communication tool. Additionally, detailed and accurate forecasts and flood warning systems are essential for officials and citizens to be able to plan for and respond to a flood event, including decisions regarding evacuation (NRC, 2012b).
Flood-related impacts can also be minimized through well-enforced building codes and land-use regulations. Communities can restrict development in severe flood hazard areas and limit the construction of new public infrastructure that facilitates development (e.g., utilities, transportation). Additionally, communities can develop plans for relocating existing critical infrastructure to less risky locations, either when aging facilities require replacement or when facilities are severely damaged by coastal storms. Local
governments can require elevation and other floodproofing measures in all new construction, although new building codes will take time to produce widespread changes and the codes must be enforced if they are to be effective. Kunreuther (1996) found that one-third of the damage from Hurricane Andrew could have been avoided if the state and local building codes had been enforced. Existing structures in floodprone areas can be elevated so that the main floor is above the base flood elevation (Figure 1-11) or residents with repeated flood damage can be encouraged through economic incentives to relocate.
This study was undertaken as part of a broad 5-year effort to provide advice to the USACE on a range of scientific, engineering, and water resources planning issues through periodic reports. Prior to this current emphasis on coastal risk reduction, the NRC Committee on U.S. Army Corps of Engineers Water Resources Science, Engineering, and Planning issued two reports: National Water Resources Challenges Facing the U.S. Army Corps of Engineers (NRC, 2011b) and Corps of Engineers Water Resources Infrastructure: Deterioration, Investment, or Divestment? (NRC, 2012a). The committee was subsequently reconstituted for specific focus on reducing flood risks from coastal storm surges along the East and Gulf Coasts and was tasked to address the following questions:
- What coastal risk-reduction strategies have been used along the U.S. East and Gulf Coasts to reduce impacts of coastal flooding associated with storm surges, and what design standards or levels of protection have been used? To what extent have these many strategies and levels of protection proven effective in terms of economic return, protection of life safety, and minimizing environmental effects?
- What are the regional and national implications of expanding the extent and levels of coastal storm surge protection? Examples might include operations and maintenance costs, sediment availability, and regional-scale sediment dynamics.
- How might risk-related principles contribute to the development of design standards for coastal risk reduction projects? How might risk-related principles increase the ability of coastal regions and communities to prepare for coastal storms and surge, and adjust to changing coastal dynamics, such as prospects of sea-level rise?
- What general principles might be used to guide future investments in U.S. coastal risk reduction?
FIGURE 1-11 Example of one approach for elevating a house above the base flood elevation (BFE).
SOURCE: FEMA (2000).
The committee’s charge specifically addresses coastal storms (hurricanes, tropical storms, and extratropical storms) and associated waves, storm surge, and flooding, which in the United States primarily affect the East and Gulf Coasts. Although Hawaii, Puerto Rico, and, under rare conditions, California are also subject to such storms, the study charge is focused on the East and Gulf Coasts. However, the committee’s approach to Tasks #3 and 4 more broadly considers coastal storm surge risks throughout the nation. Other coastal hazards, such as erosion from mild or moderate storms, wind damage, or tsunami-induced flooding, were not considered in depth.
The committee’s report and its conclusions and recommendations are based on a review of relevant technical literature, briefings, and discussions at its five meetings, and the experience and knowledge of the committee members in their fields of expertise. The committee received briefings from a range of federal and nonfederal agencies and organizations involved in coastal risk management (see Acknowledgments). However, because this study was conducted as part of a 5-year USACE-sponsored effort, the committee paid particular attention to the role of and opportunities for the USACE and, more broadly, the federal government in coastal risk reduction. The project scope combined with the 13-month study period did not allow the committee to give equal attention to all federal agencies or provide detailed discussion of actions that could be taken to reduce coastal risk at state or local levels.
In some cases the availability of data limited the extent to which these questions could be addressed. For example, the limited availability of retrospective analyses of the costs and benefits of coastal risk reduction projects after a storm event prevented a thorough analysis of the economic aspects of Task 1. The committee also found that Task 2 could not be answered quantitatively, because a full discussion of regional and national implications of expanding coastal risk reduction, particularly with respect to costs, would require detailed information on current risks and possible risk reduction strategies that was not available.
Following this introduction, the statement of task is addressed in three subsequent chapters of this report:
- Chapter 2 presents the institutional landscape for coastal risk management in the United States, highlighting major programs and recent budgets, and discusses the mechanisms by which the USACE develops and implements coastal risk reduction projects.
- Chapter 3 summarizes the current state of knowledge on the effectiveness of coastal risk reduction measures based on proven performance under coastal storms. The chapter includes discussion of financial and environmental benefits, costs, associated adverse
impacts, and regional implications for sediment availability (Tasks 1 and 2).
- Chapter 4 outlines key principles to guide future investments in coastal risk reduction, including a discussion of a benefit-cost approach constrained by acceptable risk for prioritizing coastal risk measures at a regional or a national scale (Tasks 3 and 4).
- Chapter 5 offers recommendations to enhance coastal storm risk management (Task 3).