Chapter 6 – storm surge warning, mitigation and adaptation

Chapter 6

Storm Surge Warning,
Mitigation, and Adaptation
Diane P. Horn
Birkbeck College, University of London, London, UK

ABSTRACT
This chapter provides a systematic overview of measures which play a role in mitigation and adaptation to hazards associated with storm surge. Mitigation and adaptation
are interpreted in a manner comparable to the disaster risk reduction literature, referring
to the amelioration of storm surge disaster risk through the reduction of existing hazards, exposure, or vulnerability. The chapter discusses major storm surge barriers such
as the Dutch Delta Works and the Thames Barrier, and describes more recently built
barriers in St Petersburg (Russia), New Orleans, and Venice. The chapter also reviews
storm surge early warning systems, with examples from Bangladesh, the Philippines,
the United Kingdom, and the United States.

The physical causes of storm surge are well known, and models are increasingly effective at predicting the storm surge associated with particular cyclone
conditions. Despite this, we continue to see loss of life from storm surges. For
example, the surge from Typhoon Haiyan in November 2013 was not unexpected; the strength of the storm was predicted and understood and over
800,000 people were evacuated, yet it led to 6,111 deaths and over 5 million
were people displaced (Chan et al., 2013). In comparison, the 2004 Indian
Ocean tsunami and Hurricane Katrina “only” displaced 1e1.5 million each
(Ferris, 2013). The greatest loss of life related to a tropical cyclone is from
storm surge (Doocy et al., 2013). In the United States, the loss of life in the
three deadliest hurricanes (Galveston, Texas, 1900, over 8,000 deaths; Lake
Okeechobee, Florida, 1928, 2,500 deaths; Hurricane Katrina 2005, 1,833
deaths) was primarily due to storm surge (Weindl, 2012). Hurricane-surgeinduced flooding has killed more people in the United States than all other
hurricane threats combined in the twentieth and twenty-first centuries (NOAA,
2007). The extratropical Cyclone Xynthia claimed 47 lives in 2010, Europe’s
highest storm surge toll since 1962 (Kron, 2013). Why does such loss of life
continue despite better understanding of the physics of storm surge and

Coastal and Marine Hazards, Risks, and Disasters. />Copyright © 2015 Elsevier Inc. All rights reserved.

153


154

Coastal and Marine Hazards, Risks, and Disasters

improved modeling and warning systems? How can we best reduce the
negative impacts of storm surges?

6.1 MITIGATION AND ADAPTATION
The terms “mitigation” and “adaptation” are used in many disciplines, which
results in a range of interpretations. For example, the term “mitigation” can be
used to describe actions taken to reduce the likelihood of an event occurring
(e.g., reducing greenhouse gas emissions in order to reduce increases in global
temperature and thus reduce the rate of sea level rise) or actions taken to
reduce the impact if the event does occur (e.g., building flood defenses).
Within a particular discipline the meaning of such terms may be clear. There is
no interdisciplinary consensus, however, and until recently, climate scientists
and disaster risk reduction researchers had very different definitions of mitigation. In the climate change literature, mitigation refers to the reduction of
the rate of climate change via the management of its causal factors (emission
of greenhouse gases from fossil fuel combustion, agriculture, land use
changes, etc.). However, in the disaster risk reduction literature, mitigation
refers to the amelioration of disaster risk through the reduction of existing
hazards, exposure, or vulnerability. Recent Intergovernmental Panel on
Climate Change (IPCC) reports have revised their definitions to accommodate
both interpretations of mitigation. IPCC definitions now distinguish between
mitigation of climate change, which is defined as “a human intervention to

reduce the sources or enhance the sinks of greenhouse gases,” and mitigation
of disaster risk and disasters, which is defined as “the lessening of the potential
adverse impacts of physical hazards, including those that are human-induced,
through actions that reduce hazard, exposure, and vulnerability” (IPCC, 2012).
Wilby and Keenan (2012) and Cooper and Pile (2013) reviewed the
different definitions of adaptation, as summarized above, and Hallegatte
(2009) identified five categories of practical adaptation strategies. The first is
no-regret measures, which yield benefits even in the absence of increasing
hazards (although they are not cost free). The second category is reversible
strategies, which are flexible enough to reduce as much as possible the cost of
being wrong about future risks. The third category, safety margin strategies,
reduces vulnerability at low or no cost. The fourth category, soft strategies
such as land use planning and insurance, influence individual and institutional
decisions and therefore can have an effect on hard investments. The final
category is strategies to reduce decision-making time horizons. Hallegatte
(2009) also distinguished between “hard” adaptation (e.g., building sea
defenses) and “soft” adaptation such as land use planning, early warning
systems, and financial instruments such as insurance. Wilby and Keenan
(2012) distinguished between the broader enabling environment for adaptation
(information provision, institutional arrangements, and preparedness) and
specific implementing measures to reduce flood risk. They classified these


Chapter j 6

Storm Surge Warning, Mitigation, and Adaptation

155

implementing measures into three categories: defending against flood risk,

living with flood risk, and withdrawing from flood risk. Defense measures
typically involve some form of engineering to protect existing land use. Hard
engineering structures such as levees and storm surge barriers are probably the
most common; however, there is a growing interest in soft engineering
approaches, such as wetland creation. Accommodation actions include raising
buildings and roads above flood level, establishing evacuation routes and
warning systems, the creation or enhancement of stormwater system capacity,
and zoning policies aimed at preventing development in high-risk areas.
Retreat policies include those aimed at discouraging rebuilding in high-risk
areas and the reclamation or abandonment of highly flood-prone lands.
Table 6.1 summarizes the range of available approaches to coastal hazards (not
necessarily restricted to storm surge). Cooper and Pile (2013) characterized
adaptation measures as two types, those that modify the environment and those
that are aimed at changing human activities. They noted that the reduction of
harm and the realization of benefits to humans are central to all definitions of
adaptation. Wilby and Keenan (2012) argued that in the case of flood risk
management (floods in general, not storm surge explicitly), much of what is
labeled adaptation could just be described as good practice.
In the context of storm surge, mitigation can be thought of acting to reduce
the number and severity of storm surges (which would mean reducing the
number and severity of tropical, subtropical, and extratropical cyclones), while
adaptation is learning to live with the risk of storm surges. Adaptation measures are unlikely to eliminate risk, but should aim to reduce risks to levels that
are acceptable within the limits of available resources. The measures discussed
in this chapter are adaptive measures only, which may reduce the vulnerability
to storm surge rather than the storm surge itselfdnone of them can stop
cyclones from occurring.

6.2 STORM SURGE BARRIERS
The most common hard engineering structure used to protect against surgeinduced flooding is a storm surge barrier. These structures provide temporary
protection from flooding, generally for a few hours before and after high tide and

are often partly open during normal conditions to allow navigation and salt water
exchange with estuarine areas landward of the barrier (Jonkman et al., 2013). A
storm surge barrier may be only one component of a larger flood protection
scheme, which will often include other structures such as seawalls and levees.
Storm surge barriers are usually built at a position where the barrier can be
closed during times where flooding is predicted. When the barrier is not closed,
it allows free passage of water and shipping. The main disadvantages of a storm
surge barrier system are the huge construction and maintenance costs. Movable
barriers also require simultaneous investment in flood warning systems, which
provide information on when to close the barrier (Aerts et al., 2013b).


156

TABLE 6.1 Coastal Hazard Adaptation Strategies (Not Restricted to Storm Surge)

Adaptation
Option

No-Regret
Strategy

Reversible/
Flexible

Safety
Margins
Available

Hard defenses (e.g., seawalls, levees)

þ

À

þ

Defend against flood risk

Storm surge barriers

þ

À

þ

Defend against flood risk

Restore natural coastal defenses
(e.g., salt marsh, mangrove, dunes)

þ

þ

þ

Defend against flood risk

þ

þ

Live with flood risk

Type of
Response

Enhanced drainage systems

þ

À

þ

Live with flood risk

Land use planning (e.g. rezoning, setback,
compulsory purchase, restrictions on
development in flood zones)

þ

þ

þ

Live with flood risk

Improved building standards/flood-resilient
construction (buildings and infrastructure)

þ

þ

Live with flood risk

Information and warning

þþ

þ

þ

Live with flood risk

Evacuation schemes

þþ

þ

þ

Live with flood risk

Flood/storm surge shelters

þ

þ

Insurance (household to national level)

þþ

þ

Relocation and retreat

À

À

þ

Live with flood risk
þ

Live with flood risk
Withdraw from flood risk

In the classification used by Hallegatte (2009), þþ indicates options which yield benefits even without climate change, while þ indicates options that are “no-regret” only
in some cases, depending on local characteristics and À indicates that the strategy would entail significant losses in the current climate.
Source: Adapted from Hallegatte (2009) and Wilby and Keenan (2012).

Coastal and Marine Hazards, Risks, and Disasters

Temporary/demountable defenses

Soft
Strategy


Chapter j 6

Storm Surge Warning, Mitigation, and Adaptation

157

Most storm surge barriers were implemented after a flood disaster occurred
(Aerts et al., 2013b). The best known barriers are the Dutch Delta Works and
the Thames Barrier, which were both built in response to the 1953 North Sea
storm surge when 305 people died in the United Kingdom and 1835 died in the
Netherlands. The Delta Works (Figure 6.1) is the largest flood protection
project in the world and includes dams, sluices, locks, bridges, tunnels, dikes,
levees, and storm surge barriers that protect southern Holland against a
1:10,000-year storm surge (Zhong et al., 2012). The Oosterscheldekering
(Eastern Scheldt storm surge barrier) is the largest of 13 structures which make
up the Delta Works. Construction on the Delta Works began in 1958 and was
officially completed in 1997 at a cost of $7 billion (Bijker, 2002). However,
the Netherlands continues to add infrastructure to the system as needed
(Pilarczyk, 2012), with a seawall near Harlingen completed in 2010. The Delta
Committee was set up to investigate the impact of climate change and projected sea level rise in the twenty-first century. Their 2008 report (Delta Vision
Committee report, 2008) led to the Delta Programme, which addresses future
flood risk management and freshwater supplies, and the establishment of the

Delta Fund. Significant investments will need to be made after 2050; for

FIGURE 6.1 The Dutch Delta Works. />

158

Coastal and Marine Hazards, Risks, and Disasters

example, the Maeslant Barrier protecting Rotterdam is due to be replaced after
2070 to build a new closable storm surge barrier. The Delta Fund, which went
into effect at the start of 2013, sets aside money for the government
investments to works set out in the Delta Programme (Zevenbergen et al.,
2013). The central government and the water boards (regional government
bodies which levy their own taxes) have agreed to pay an equal share of the
costs of current and future flood protection measures (Verduijn et al., 2012).
They will each contribute V131 million in 2014 and V181 million annually
from 2015, with a total of V16.6 billon available from all funding
sources between 2014 and 2028 and V22 billion between 2029 and 2050
(Kabat et al., 2009). All the resources from the Delta Fund are fully allocated
until 2019, when funding for investment in new flood risk management
measures will be available (Delta Programme, 2014). The impact of the Delta
Works has been extensively studied, with research investigating the impact on
ecology (e.g., Hostens et al., 2003; Tangelder et al., 2012; Jansen et al., 2013;
van Wesenbeeck et al., 2014), geomorphology (e.g., Louters et al., 1998;
Roelvink et al., 2001; To¨nis et al., 2002; Hudson et al., 2008), and hydrology
and hydrodynamics (e.g., Jonkman et al., 2008; Augustijn et al., 2011; Zhong
et al., 2012). The references reported here represent only a small number of the
studies which have been carried out on the areas affected by the Delta Works.
Zhong et al. (2012) evaluated the impact of the Maeslant storm surge barrier
on flood frequency under rising sea levels and showed that the operation of the

Maeslant Barrier reduces flood frequency and can partly compensate for the
effect of future sea level rise, although the return periods of all water levels
will decrease as the sea level rises. The Maeslant Barrier is currently closed
when water level in Rotterdam reaches 3.0 m. According to the model of
Zhong et al. (2012), the return period of this water level in 2010 without the
barrier, 10.9 years, is increased to 2,400 years with the barrier. Under the
current closing decision water level of 3.0 m, the port of Rotterdam will be
closed once every 3.2 years in 2050 and once every 1.1 year in 2100. The
design safety level, 4.0 m, will be reached with a return period of 46,948 years
in 2010, 16,420 years in 2050, and 3,849 years in 2100.
Although the water level in the 1953 storm surge was the highest ever
recorded in London, the height of the storm surge was about 3 h before high
tide and the river level was low after a dry spell, and thus only a few locations in
east London were flooded. After the lucky escape in 1953, the UK government
appointed the Waverley Committee to study flood dangers to the city. The
Committee recommended that a structure be constructed across the Thames,
backed up with a considered approach to development in the floodplain, and 41
proposals at six sites were put forward following the report (Waverley, 1954).
However, no action was taken until the problem was passed to the Greater
London Council in 1968 with a request that a full investigation into the problem
be carried out as a matter of considerable urgency (Horner, 1979). This led to
the construction of the barrier, begun in 1976 and completed in 1984 at a cost of


Chapter j 6

Storm Surge Warning, Mitigation, and Adaptation

159

£560 million (equivalent to about £1.6 billion in 2014 prices), including the
cost of strengthened defenses both upstream and downstream of the Thames
Barrier and the associated construction of three minor barriers, two major
floodgates, and nine minor floodgates on tributaries of the Thames. The Thames
Barrier was designed to protect against the 1,000-year flood, with an allowance
for an increase in water level up to 2030. In 2010, the UK Environment Agency
(EA) published a study (TE2100) on planning for flood risk management in the
Thames Estuary, which aimed to determine the appropriate level of flood
protection needed for London and the estuary for the next 100 years (Environment Agency, 2010). The TE2100 policy recommendations were divided
into three time periods. In the first 25 years (2010e2035), the strategy is to
continue to maintain the current flood defense system including planned improvements, ensure effective floodplain management is in place, and monitor
change indicators. The strategy for the transition period (2035e2049) is to
replace and upgrade current defenses and to make the final decision on building
a new barrier or other end-of-century option, and to start planning for this. The
agreed end-of-century option is to be planned, designed, and constructed between 2050 and 2100. The extensive benefitecost analyses carried out in the
TE2100 analysis suggested that improving defense standards now is not cost
effective, as the extra benefit gained is generally not worth its costs until climate
and socioeconomic change begin to create additional risk from 2050 and toward
2070 (Penning-Rowsell et al., 2013a). No costings for the unspecified end-ofcentury option were given in the TE2100 report but other information from
the EA indicated that investment of £1.5 billion will be needed for the first
25 years, with another £1.5 billion needed for the middle 15 years, and £6e7
billion for 2050e2100 (Environment Agency, 2013). The contracts for the first
stage of TE2100 went out for bidding in September 2013. The phase 1 program
represents the capital-funded work needed to maintain tidal defenses from 2015
to 2025 and includes refurbishment of fixed assets (such as tidal walls and
embankments), active assets (including the Thames Barrier gates), and new
assets such as pumping stations. The contract for phase 1 will be awarded in
September 2014.
As of April 2014, the Thames Barrier has been closed 174 times since it
became operational in 1982, with increasingly frequent closures since 2000.

Half of these closures were to protect against tidal flooding and half to protect
against river flooding. The closure on December 6, 2013, was associated with
the biggest storm surge since 1953 and the highest tide in response to which the
Thames Barrier has closed, at an elevation of 4.1 m (Atkin, 2014). The EA
published a map showing the probable impact on central London if the barrier
had not been in place (Figure 6.2). The TE2100 report recommended that the
Thames Barrier should not be closed more than 50 times a year to reduce the
chance of it failing. The Thames Barrier was expected to reach this design limit
from 2135 onward, and the TE2100 report suggested that once this design limit
is reached, it may not be possible to close the barrier to protect against fluvial


160

Coastal and Marine Hazards, Risks, and Disasters

FIGURE 6.2 The UK Environment Agency’s simulation of the flooding which should have
resulted from the storm surge in December 2013 without the Thames Barrier. .
uk/news/nce-live-news-updates-tuesday-10-december-tidal-surge-broke-records-london-willchoke-without-crossrail-2/8656557.article.

flooding in order to maintain the reliability of the barrier against tidal flooding
(TE2100). The Thames Barrier came under unprecedented pressure in the first
months of 2014, reaching its operational safety limit of 50 closures on 4 March
of that year. This is the record for the highest number of times it has been closed
in a single season (Figure 6.3), with many of the closures to protect West
London against river flooding in the wettest winter since records began in 1766.
The EA described these frequent closures as a “blip” and is still forecasting that
a replacement is not needed until 2070 (BBC, 2014). However, the EA will
carry out an investigation of the robustness of the Thames Barrier in response to
a request from the Mayor of London and will report its findings at a Thames

Estuary Steering Group meeting in early summer 2014 (London Assembly,
2014). Even with the impetus of the loss of life in 1953, it took years to plan
and build the Delta Works and the Thames Barrier. Padron and Forsyth (2013)
estimated that the average interval between planning and constructing a major
storm surge barrier is 27 years, suggesting that perhaps the timetable for the
Thames end-of-century option should be brought forward and that something
like the Delta Fund should be set up now to plan for its financing.
Storm surge barriers in other cities are reviewed by Dircke et al. (2013) and
summarized in Table 6.2; this chapter will describe only a few more examples.
The St Petersburg Flood Prevention Facility Complex is a storm surge barrier


Chapter j 6
Storm Surge Warning, Mitigation, and Adaptation

161

FIGURE 6.3 Closures of the Thames Barrier since it was built. />

162

TABLE 6.2 Overview of Storm Surge Barriers

Type

Country

Hollandse Ijssel storm surge barrier

Lifting gate

Netherlands

4

1958

127

Oosterschelde storm surge barrier

Lifting gate

Netherlands

10

1986

5,005

Year of
Operation

Approx Cost
(Million Dollars)

Maeslant storm surge barrier

Floating sector gate

Netherlands

8

1997

709

Europoort Barrier and Hartel Barrier

Lifting gate

Netherlands

6

1997

329

Ramspol storm surge barrier

Rubber dam

Netherlands

5

2002

88

Venice storm surge barrier

Flap gate

Italy

13 estimated

2016 estimated

7.6

New Bedford storm
surge barrier

Rolling sector gate

New Bedford,
MA, USA

4

1968

72

Stamford storm surge barrier

Flap gate

Stamford, CT, USA

4

1968

19

Harvey canal flood protection
barrier, Gulf Intracoastal Waterway
West Closure Complex (GIWWCC)

Sector gate

New Orleans,
LA, USA

3

2008

1,014

Coastal and Marine Hazards, Risks, and Disasters

Barrier

Design and
Construction
Time (years)


New Orleans,
LA, USA

4

2011

1,318

Thames Barrier

Segment gate,
radial gate

United Kingdom

8

1982

2,730

Fox Point hurricane barrier

Radial gate

Providence,
RI, USA

6

1966

75

St Petersburg Barrier

Lifting gate,
floating
sector gate

Russia

Started in the
1990s, resumed
in 2002

2010

6,600

Eider Barrage

Radial gate

Germany

6

1973

113

Ems storm surge barrier

Lifting gate,
radial gate,
segment gate

Germany

5

2002

286

Marina Barrage

Crest gate

Singapore

3

2008

173

Source: Based on Dircke et al. (2013), Expanded and Updated.

Storm Surge Warning, Mitigation, and Adaptation

Sector gate

Chapter j 6

IHNC New Orleans hurricane
protection barrier

163


164

Coastal and Marine Hazards, Risks, and Disasters

system which includes 11 embankment dams, 67 hydraulic gates, a bridge and
tunnel, and two navigation channels with gates that can be closed to protect
against storm surge (Hunter, 2012). Construction on the barrier began in 1979,
halted at the breakup of the Soviet Union, and was completed in 2011 at a cost
of about $3 billion. The scheme is designed to protect against a 1,000-year
storm surge of 4.55 m with an extreme limit of 5.15 m for a 1:10,000-year
flood (Ivanov et al., 2012).
A storm surge barrier system to protect Venice, known as MOdulo Sperimentale Elettromeccanico (MOSE), is under construction. The MOSE project

consists of four tidal defense barriers at the openings in the barrier islands
where Adriatic tides enter Venice lagoon (Bajo et al., 2007). The MOSE
defenses include 21 gates in a north channel at the Lido inlet, 20 gates in the
south channel and an intermediate island linking the two barriers, a barrier
with 19 gates at the Malamocco inlet, and a barrier with 18 gates in the
Chioggia inlet (Fontini et al., 2010). Construction started in 2003 and is
expected to be completed in 2016 with an estimated cost of V5.493 million
(about $7.6 million). There has been a long debate about the optimal closing
level of the MOSE barriers: increasing the barriers’ functioning frequency
improves the degree of protection against storm surge; however, this also increases interference with harbor activities (Fontini et al., 2010). Researchers
have also debated whether the yet-to-be-completed Venice barrier will protect
the city as sea level rises (e.g., Pirazolli, 2002; Bras and Harleman, 2002;
Pirazzolli and Umgeisser, 2006; Umgiesser and Matticchio, 2006; Rinaldo
et al., 2008). The current expectation is that the MOSE barriers will protect
Venice until sea level rise is greater than 0.60 m; at that stage, new options for
the future of the Venice lagoon will have to be considered (Munaretto et al.,
2012).
Hurricane Katrina highlighted the need for improved defenses for New
Orleans and led to the construction of the Hurricane and Storm Damage Risk
Reduction System (HSDRRS) for Southeast Louisiana. The storm surge from
Katrina breached 50 floodwalls and levees and collapsed a 1,220-m section of
the Inner Harbor Navigation Canal (IHNC), a waterway that connects the
Mississippi with Lake Borgne and the Gulf of Mexico (Miller et al., 2013).
The $14.45 billion HSDRRS consists of 563 km of levees and floodwalls,
73 pumping stations, three canal closure structures with pumps, and four gated
outlets (USACE, 2010). Because evacuation is now seen as a major part of the
new emphasis on risk reduction, the HSDRRS also includes an elevated road
and bridge over a new floodwall at the entrance to the Lake Pontchartrain
Causeway Bridge (Eqecat, 2009; Reid, 2013). One component of the
HSDRSS, the 3-km IHNCdLake Borgne Storm surge barrierdwas

completed in 2013 at a cost of $1.1 billion and is designed to protect the areas
which were hit hardest by Katrina. This barrier closes off Lake Borgne, situated at the east side of New Orleans, and closes both the lake and the canals
northwest and southwest of the lake. The navigable Seabrook surge barrier


Chapter j 6

Storm Surge Warning, Mitigation, and Adaptation

165

protects the Lake Pontchartrain entrance of the IHNC. A movable gate has
been built in the canal northwest of Lake Borgne, the Gulf Intracoastal
Waterway (GIWW), making navigation possible during normal conditions.
The canal southwest of Lake Borgne, the Mississippi RivereGulf Outlet, has
been permanently closed by the storm surge barrier. The new system is
designed to withstand the 1:100-year storm event in 2057 (USACE, 2010). The
HSDRSS system was tested by Hurricane Isaac in 2012 and performed as
designed (Reid, 2013; USACE, 2013), although flooding did occur in areas
outside the system. This was attributed to an intense and long-duration surge
caused by the long duration of tropical force winds, which in some cases were
aggravated by extreme local rainfall (USACE, 2013). However, is a 1:100
protection standard adequate to protect New Orleans? Compare this to the
1:10,000 standard of the Delta Works.
Hurricane Sandy provided a demonstration of the vulnerability of New York
City to storm surge, reviving discussion of the possibility of building a protective barrier (e.g., Bowman et al., 2005; Colle et al., 2008; Aerts and Botzen,
2012; Coch, 2013; Hill, 2012, 2013). Proceedings of a conference held in 2009
under the sponsorship of the Infrastructure Group of the American Society of
Civil Engineers were reissued in 2013 in the aftermath of Sandy (Aerts et al.,
2013a). Aerts et al. (2013b) outlined three possible configurations of storm

surge barriers. The cost of these barriers is estimated at up to $15 billion,
comparable to the amount spent on the Louisiana HSDRSS system (Aerts et al.,
2013b). Note that these barriers would also protect against only the 1:100 storm
surge. In contrast, other researchers argue that the city should commit to protecting all areas to a 1:500-year standard (Tollefson, 2013). The system
envisaged by these researchers, including Bowman et al. (2005), Colle et al.
(2008), and Hill (2013), would comprise two or three barriers to protect New
York and New Jersey, including an 8-km-wide barrier approximately 6 m high
that could be opened and closed at the entrance to New York’s harbor, and a
second barrier at the entrance to Long Island Sound (Tollefson, 2012). The state
panel’s cost estimates for such a system range from $7 to $29 billion,
depending on the design (Tollefson, 2013). Wagner et al. (2014) argued that
historically, financial resources, policies, and public support have coalesced
after catastrophic events to act as catalysts for policy change; however, the
arrival of Sandy during the economic crisis restricted the availability of funding.
Structural measures can never entirely eliminate flood risk, and may not
always be an appropriate response to such risk. Storm surge barriers can
protect a large area with relatively small structures, but are not appropriate
on an open coastline. In addition, surge barriers are expensive and therefore
not economically feasible for all locations. As with all flood defense
structures, the presence of the barrier may encourage development in hazardous areas. Many coastal cities will not be able to rely on such major
infrastructure, and will need instead to rely on a mix of structural and
nonstructural measures.


166

Coastal and Marine Hazards, Risks, and Disasters

6.3 STORM SURGE WARNING
If it is not possible or desirable to completely eliminate flood risks, adaptation

can be accomplished in the form of adequate provision of emergency procedures (Cooper and Pile, 2013). Improvements in forecasting, early warning
systems, and evacuation and shelter procedures have reduced storm-surgerelated mortality (Doocy et al., 2013). Real-time observational and predictive modeling techniques make it possible to issue cyclone and related storm
surge warnings, which may trigger the implementation of emergency
procedures and provide an opportunity for the evacuation of vulnerable areas.
To be effective, early warning systems must integrate four elements: (1)
knowledge of the risks faced, (2) technical monitoring and warning service, (3)
dissemination of meaningful warnings to those at risk, and (4) public awareness and preparedness to act (UNISDR, 2009). Although comprehensive
coverage of early warning systems for storms and tropical cyclones is available, disasters such as Hurricane Katrina have highlighted inadequacies in
technologies for enabling effective and timely emergency response (Grasso
and Singh, 2009). Storm surge is generally treated as a corollary of cyclones
(both tropical and extratropical), with few early warning systems specifically
for storm surge in existence. For example, Grasso and Singh (2009) presented
detailed tables of early warning systems for different types of events, but they
did not include storm surge in these tables, with one table for floods (which
they interpreted as primarily fluvial and pluvial) and another table for severe
weather, storms, cyclones, and hurricanes. The report did show that there are
often inadequate flood warning and monitoring systems, especially in developing or least developed countries. In most of the cases that they surveyed,
Grasso and Singh (2009) found that communication systems and adequate
response plans were lacking. They argued that predictions are not useful unless
they are translated into a warning and action plan that the public can
understand.
Storm surge warning systems are most often operated at a national scale
and are usually linked to predictions of the path and landfall of tropical or
extratropical cyclones. Many countries have or are developing storm surge
prediction models and warning systems: this review has found references to at
least 28 national programs, but only a few will be described here. The Joint
Technical Commission for Oceanography and Marine Meteorology (JCOMM)
has compiled a comprehensive list of national storm surge products (JCOMM,
2014). Although that list is not exhaustive and does not explicitly address
storm surge warning systems, it gives an indication of international activity in

this area. JCOMM is currently carrying out a worldwide survey on operational
storm surge models and data, with data collection scheduled to be completed
by the end of May 2014. JCOMM has also initiated the World Meteorological
Organization (WMO) Coastal Inundation Forecasting Demonstration Project
to assist countries at risk of coastal flooding to operate and maintain a reliable


Chapter j 6

Storm Surge Warning, Mitigation, and Adaptation

167

integrated forecasting and warning system. Four national subprojects have
begun since 2011 (Bangladesh, Fiji, Dominican Republic, and Indonesia), with
two more planned (Shanghai and South Africa).
In 2007, Cyclone Sidr made landfall in southern Bangladesh, causing 3,406
deaths. Despite advance warnings being issued 5 days before landfall, only
slightly over a third of the population complied with cyclone warnings and
evacuation orders (Paul, 2012). After Sidr, the Bangladesh government
modernized the early warning systems and initiated the construction of 2,000
new cyclone shelters and as a result, in 2009, Cyclone Aila caused only
330 deaths (Haque et al., 2012; Ahamed, 2013). In Myanmar, Cyclone Nargis
caused over 146,000 deaths and more than $10 billion in damages in 2008. In
contrast, 2 years earlier a slightly stronger storm, Cyclone Mala, hit the
Myanmar coast about 150 km north of the Nargis track, but only led to 22
deaths after a well-executed evacuation effort (Fritz et al., 2011). The area hit
by Nargis lacked evacuation plans and shelters, with residents having no
experience of storm surge flooding (Brakenridge et al., 2013). In February
2011, Cyclone Yasi hit Queensland, Australia. The cyclone was 500 km wide

with 285 km/h wind speeds, yet no lives were lost as a result of advance
warning and an evacuation that was completed before the cyclone struck
(Haque et al., 2012). Early warnings disseminated 4 days before Cyclone
Phailin hit Eastern India in October 2013 allowed the evacuation of nearly
1.2 million people, with only 44 deaths. In contrast, a comparable cyclone hit
the same area in 1999, with the loss of more than 10,000 lives (Harriman,
2014).
The United Nations Development Program identified 29 developing
countries and four developed nations that are at risks from cyclones, but 42%
and 27% of cyclone deaths in the past two centuries have occurred in
Bangladesh and India, respectively (Doocy et al., 2013). (Note, however, that
although the Doocy et al. paper was published in 2013, the data analyzed in the
study was from 1980 to 2008 and does not, therefore, include any deaths from
storm surges since then, particularly Typhoon Haiyan.) Bangladesh is particularly vulnerable to storm surge because of its location, low-elevation and flat
topography, high population density, and limited coastal defense structures.
Penning-Rowsell et al. (2013b) estimated that nearly 700,000 deaths have been
caused by cyclones affecting Bangladesh since 1960. Two storms accounted
for the majority of these deaths, with more than 300,000 deaths from Bhola
Cyclone in 1970 and 138,866 from Cyclone Gorky in 1991. Economic losses
due to cyclones are not declining significantly, although the reduction in potential gross domestic product associated with major disasters has gradually
become smaller as the national economy has grown and become less dominated by agriculture (Haque et al., 2012).
The Bangladesh Cyclone Preparedness Programme began in 1972, in
response to the 1970 disaster, when no warnings were received in vulnerable
rural areas and coastal villages despite the cyclone being detected by coastal


168

Coastal and Marine Hazards, Risks, and Disasters

radar and satellite. The loss of life has decreased dramatically under this
program, with improved warning systems, infrastructure improvement to
facilitate evacuation, access to elevated cyclone shelters, the construction of
coastal embankments and more resilient buildings, and the reforestation of
mangrove forests (Haque et al., 2012; Paul, 2012). Forecasts from the Storm
Warning Centre are used to produce early warnings supported by emergency
telecommunications and cyclone preparedness activities in the coastal zone. In
particular, a large network of trained local volunteers is equipped with
megaphones and public address systems to disseminate the warning. Warning
dissemination was originally the weak link in the system and the large death
toll in 1991 has been attributed to this, and also to the fact that only two out of
every five shelters were useable because of flooding (Haque et al., 2012). Since
2007 there has been a major change in the availability of communication
technology with the widespread use of mobile phones. Colorful hot air balloons can also be used to convey cyclone warning messages in remote and
coastal areas of Bangladesh (Haque et al., 2012). However, continued challenges of illiteracy, lack of awareness, and communication problems mean that
some people do not understand or follow the evacuation warnings. Coastal
areas are not well connected by road networks, public bus service is very
limited, and few people own a car. Most coastal residents walk from their
homes to public cyclone shelters, which are mostly located within 3e5 km of
their homes; however, this can be difficult in adverse weather conditions. In
addition, many of the at-risk population are unwilling to use public shelters
because of the poor condition (e.g., lack of latrines and separate rooms for
women, no drinking water, and lighting). The shelters that are in reasonably
good condition can only accommodate about 15% of the coastal population of
Bangladesh (Paul, 2012). Despite improvements in warning systems, precyclone evacuation remains a challenge. Haque et al. (2012) recommended
that a network of small, sturdy, and safe multipurpose buildings should be
established within a 2-km walking distance of households and villages instead
of developing larger cyclone shelters.
Typhoon Haiyan (known locally as Typhoon Yolanda), which struck the
Philippines in November 2013, was the deadliest storm in the country’s history, with 6,111 deaths and 1,779 people still missing, and estimated losses of

$12.9 billion (Neussner, 2014). Haiyan is also the strongest storm yet recorded
at landfall, with sustained winds of up to 315 km/h. The storm surge was not
measured, but has been estimated at 2.3e5 m, with some estimates as high as
7e8 m (eSurge, 2014; NASA, 2014; Neussner, 2014). Although warnings
were issued and the Philippines National Disaster Risk Reduction and Management Council (NDRRMC), with local authorities, evacuated more than
800,000 people, the official death toll still exceeded 6,000 (Neussner, 2014). In
the current system, a warning is issued by Philippine Atmospheric,
Geophysical and Astronomical Services Administration (PAGASA), which
gives a warning to NDRRMC, which then releases a national statement on


Chapter j 6

Storm Surge Warning, Mitigation, and Adaptation

169

what to expect from a storm. The information goes from the national to
regional, provincial, municipal, and finally to the village level. Officials from
PAGASA said that the public lacked a clear understanding of what was meant
by a storm surge (Santos, 2013). There is no Filipino term for storm surge, and
ironically, if the warning had been described as a tsunami, people would have
had a better idea what to expect (Yates, 2013). Many residents of coastal areas
and political decision makers admitted that they were not familiar with the
term “storm surge” (Neussner, 2014). Although PAGASA accurately forecast
the typhoon’s path and strength, public warnings failed to sufficiently explain
and clarify the specific dangers, and warnings related to storm surge were
issued relatively late (Neussner, 2014). In some locations, however, such as
Balankayan, East Samar, no lives were lost because of effective warning and
evacuation drills (Yates, 2013). Postdisaster studies have suggested that up to

94% of the victims of Haiyan could have been saved if they had been properly
warned and evacuated to safe areas (Neussner, 2014). PAGASA and
NDRRMC emphasized rain, flood, and landslide warnings, but did not stress
very strongly the storm surge to come. The official storm surge map grossly
underestimated the inundation area of the storm surge (Neussner, 2014). In
addition, many of the evacuation centers were single-floor buildings without
stilts and thus inappropriate as storm surge shelters, and were flooded by the
storm surge. Even when a clear evacuation message was given by the government, a large part of the population did not leave their homes in the danger
zones. One of the main reasons given by respondents of the Gesellschaft fur
Internationale Zusammenarbeit (GIZ) survey was the fear of theft and/or
looting. Some also dismissed the evacuation order or underestimated the
height and force of the water (Neussner, 2014). In February 2014, the
Philippines Science Secretary proposed the adoption of a storm surge advisory
protocol and flood advisory system. If the storm surge warning protocol is
adopted, Storm Surge Advisory (SSA) No. 1 would indicate a storm surge
height of up to 2 m; SSA No. 2, up to 5 m; and SSA No. 3, more than 5 m.
Unlike public weather warning signals, which are updated in 24 h, a storm
surge advisory will be given 48 h in advance. The new advisory systems for
storm surges and floods are part of a disaster preparedness road map that the
President wants adopted before the onset of the rainy season in June 2014. The
storm surge disaster risk reduction efforts are geared at a reduction of exposure
in the coastal areas by implementing “no-build” zones, development of natural
barriers and construction of man-made barriers to reduce impact of hazard, and
resiliency of home, building, and other infrastructure (Ubac, 2014).
In December 2013, the United Kingdom experienced the largest storm
surge since 1953, but suffered relatively limited damage thanks to improved
defenses. About 10,000 homes were evacuated in Eastern England after
receiving coastal flood warnings from the UK Coastal Monitoring and Forecasting service (UKCMF). The UKCMF is managed by the EA in partnership
with the Met Office, the National Oceanography Centre (NOC), the Flood

170

Coastal and Marine Hazards, Risks, and Disasters

Forecasting Centre, and the Centre for Environment, Fisheries and Aquaculture Science (CEFAS). The UKCMF represents a move from forecasting alone
to forecasting and monitoring. It was originally called the Storm Tide Warning
Service and was set up after the 1953 storm. The UKCMF uses a network of
44 tide gauges and a strategic wave monitoring network hosted by CEFAS.
The surge model uses inputs from the Met Office mesoscale atmospheric
model, the WAM wave prediction model, tide-surge models developed by
Proudman Oceanographic Laboratory (part of NOC), and bathymetric data to
predict the height and timing of surge events. Surge models run four times a
day producing forecasts up to 2 days ahead, and include local models which
provide more accurate total water level forecasts (surge plus tide). Model
results are transmitted to the EA along with real-time data from the tide gauge
network. The EA operates an online flood warning service (for all floods, not
just coastal floods or storm surge) which publishes a live flood warning map
updated every 15 min, local flood information by postcode, a 3-day flood
forecast, flood warnings on Facebook, and flood warnings by phone, text or,
e-mail. The Flood Forecasting Centre also sends Flood Guidance Statements
to emergency responders.
In the United States, the National Hurricane Center (NHC) monitors
cyclone tracks and intensity in the North Atlantic and Eastern North Pacific
east of 140 W. The NHC issues a package of advisory products which are
updated every 6 h, and give information on areas of disturbed weather and
their potential for development into a tropical cyclone during the next 5 days,
which are tied to the anticipated arrival time of tropical storm force winds.
Each of the advisory categories (tropical storm watch, tropical storm warning,
hurricane watch, hurricane warning) has a clearly defined meaning which is

included in the NHC’s public advisory bulletins and includes a series of
parameters which describe the storm. Local Weather Forecast Offices issue
Hurricane Local Statements, which include detailed information on weather
conditions, evacuation decisions, recommended precautionary actions, storm
surge and tide information, and potential for other hazards (e.g., tornado, flash
flood, rip currents, beach erosion).
The NHC predicts storm surge using the Sea, Lake and Overland Surges
from Hurricanes (SLOSH) model. SLOSH estimates winds and storm surge
heights resulting from historical, hypothetical, or predicted hurricanes for a
range of input variables: atmospheric pressure, cyclone intensity, size, forward
speed, and direction of motion (hurricane track). These inputs are used to
create a model of the wind field which drives the storm surge. Predictions for a
particular location are calculated for the local bathymetry, including the
adjacent continental shelf, the local shoreline plan form, and angle of approach
of the cyclone relative to the shoreline. SLOSH does not include waves,
normal river flow and rain, or the astronomical tide (although operational runs
can be carried out at different tide levels via an initial water level anomaly).
SLOSH outputs strongly depend on the accuracy of the meteorological input,


Chapter j 6

Storm Surge Warning, Mitigation, and Adaptation

171

and meteorological uncertainty will dominate over storm surge model specifications (Lowry, 2012).
The NHC uses the SLOSH model to make advance calculations of
potential surge for 37 basins, which represent sections of coastline that are
centered on particularly susceptible features such as inlets, large coastal

centers of population, low-lying topography, and ports. These basins include
all of the US East and Gulf coasts, the Bahamas, Puerto Rico, the Virgin
Islands, and parts of Hawaii and Guam. The NHC normally updates the storm
surge data sets for five or six basins a year. These are done after a storm
changes the shape of a coastline, the high-resolution land elevation for the area
has been updated, or local emergency managers need updated surge data for
planning. When a hurricane threatens a particular basin, the NHC runs thousands of simulated hurricanes for all five storm categories, changing various
factors such as forward speeds, directions of approach, and landfall locations.
Occurrence probabilities of these hurricanes are not considered. The SLOSH
model produces three storm surge products: probabilistic surge (P-Surge),
Maximum Envelope of Water (MEOW), and Maximum of Maximums
(MOM). Storm-specific uncertainties are accounted for in the P-Surge product,
which shows the overall chances that the specified storm surge height will
occur at each individual location on the forecast map during the indicated
forecast period. The probabilities are based on errors during recent years in the
official track and intensity forecasts issued by the NHC. Variability in storm
size is also incorporated (NOAA, 2014).
Each model run for a category creates a separate MEOW, which provides a
worst-case snapshot for a particular storm category, forward speed, trajectory,
and initial tide level. All the MEOWs for a given basin are combined to form
an MOM, which is an ensemble product of maximum storm surge heights for
all hurricanes of a given category. MOMs are created by pooling all the
MEOWs for a given basin, separated by category and tide level (zero/high),
and selecting the MEOW with the greatest storm surge value for each basin
grid cell. This procedure is repeated for each storm category. In total
10 MOMs are made available for each basin: one MOM per storm category
and tide level. MOMs represent the worst-case scenario for a given category of
storm under “perfect” storm conditions (NOAA, 2014). Neither MEOWs nor
MOMs are storm specific and no single hurricane will produce the regional
flooding depicted in the MEOWs and MOMs, which are intended to capture

the worst-case high water value at a particular location for evacuation planning. Although these acronyms may seem frivolous, the idea is that they are
easier to rememberdand maybe to show that modelers have a sense of humor.
Perhaps to balance the cat names (e.g., Cat 5 storm, MEOWs), animations of
storm surge generated by SLOSH have the extension .rex, which reportedly
was the name of the developer’s dog (Williams, 2013).
The most familiar way of describing the intensity, thus destructive potential, of tropical cyclones is the SaffireSimpson Hurricane Wind Scale


172

Coastal and Marine Hazards, Risks, and Disasters

(SSWS), which is a classification of hurricanes into five categories based on a
hurricane’s sustained wind speed at a particular time. The scale provides examples of the type of damage and impact in the United States associated with
winds of indicated intensity and indicates the maximum sustained wind and
measure of the degree of damage possible in areas that experience the
maximum wind (typically the eyewall). The SSWS is not an overall indicator
of the severity of the storm, as it does not indicate the size of the wind field or
the size of the area impacted, the magnitude of the storm surge, or the amount
of rain that will fall. Earlier versions of the scale incorporated central pressure
and storm surge as components of the categories. However, storm surge is
affected by many factors other than wind intensity: the central pressure of the
cyclone, the maximum wind speed, the speed of motion of the storm, the size
of the cyclone and its wind field, the angle of approach to the coast, the width
and slope of the continental shelf, and local features of the coastline (Irish and
Resio, 2010). None of these are included in the SSWS. In 2010, after storms
such as Katrina (2005) and Ike (2008) demonstrated conclusively that estimates of storm surge based simply on wind intensity are misleading, the NHC
removed central pressure and storm surge from the scale.
Extreme storm surges such as those resulting from Katrina and Ike showed
that the SaffireSimpson categories, which were designed to be an index of the

potential intensity of wind damage, are not a good public warning scale for
storm surge (Irish and Resio, 2010; Kantha, 2013). Most people interpret the
hurricane category as an indication of the severity of the storm. Using the
SSWS category to make evacuation decisions may leave people at risk from
storm surge. For example, Katrina was a category 3 hurricane at landfall, yet
caused more damage and loss of life than either Hurricane Camille (1969) or
Hurricane Andrew (1992), both category 5 storms. A similar situation
occurred with Hurricane Ike (2008), a weak category 2 storm which caused
extensive storm surge damage on the Texas coast. Hurricane Andrew had a
landfall wind speed of 75 m/s but a radius of only 77 km and a storm surge of
2.4 m. In contrast, Katrina had a landfall speed of 56 m/s and a radius of
217 km, with a storm surge of 7.5e8.5 m. Ike had a landfall speed of 49 m/s
and a radius of 195 km, with a storm surge of 4.8e5.9 m. The relatively large
sizes of both Katrina and Ike demonstrated that the impact of a tropical
cyclone is also a function of its size (Kantha, 2013). The larger the hurricane,
the greater is the impact potential. However, because only the SSHS category
is widely disseminated, the lay public, and local officials not privy to sophisticated models and other data at the federal level, are generally unaware of
the true destructive power of the storm surge (Kantha, 2013).
The need for a change in the method of reporting storm surge hazards
became even more obvious after hurricanes Isaac and Sandy in 2012, where
despite the very accurate forecasts, many people did not respond to the
warnings from forecasters (Baker et al., 2012). At landfall, Isaac was a category 2 hurricane and Sandy was a posttropical cyclone with hurricane-force


Chapter j 6

Storm Surge Warning, Mitigation, and Adaptation

173

winds. These storm descriptions were interpreted by the general public as lowrisk categories, and the extent of the storm surge surprised many people. After
the storm, forecasters talked to focus groups and found that the public interpreted the storm’s danger in terms of its hurricane category, and storm surge
warnings did not encourage residents of threatened areas to evacuate (Lazo
and Morrow, 2013). Many residents were not fully aware of the nature of the
warnings for their area, how strong a storm Sandy was, the risk that it posed, or
how long impacts would persist. Residents who lived within one block of the
water felt that the greatest threat that the storm posed was wind rather than
water (Baker et al., 2012). Both National Weather Service (NWS) forecasters
and emergency managers reported difficulty determining inundation guidance
at specific locations. One of the biggest surprises in Sandy was the impact of
the surge and how fast it moved in; the public and National Oceanic and
Atmospheric Administration/NWS partners did not clearly understand what
storm surge was or how dangerous it could be (NOAA, 2013). A poststorm
assessment concluded that the highest priority need was for improved highresolution storm surge forecasting and communication (NOAA, 2013).
Starting in the 2014 hurricane season, the NHC is issuing potential storm
surge flooding maps as one of its products. These maps show land areas
where, based on the latest NHC forecast, storm surge could occur and the
elevation that the water could reach in those areas. The maps are created from
multiple SLOSH runs and allow for uncertainties in the track, landfall
location, intensity, and size of the cyclone. However, the maps do not account
for wave action, pluvial flooding, or flooding inside levees and overtopping
(NOAA, 2014). The potential storm surge flooding maps will be issued when
a hurricane or tropical storm watch is first issued for any portion of the Gulf
or east coast of the United States, which is approximately 48 h before the
anticipated onset of tropical storm force winds. The maps will be updated
every 6 h in association with each full advisory package, but because of the
extensive computer processing times, they will not be available until
45e60 min after the advisory is released. The new maps will be color coded,
much like the radar maps on the local news showing rain and severe weather.
The new surge warnings will not include categories; instead, the colors will

represent depths above the ground. The new maps will be used on an
experimental basis for at least 2 years while the NHC collects feedback from
emergency personnel and the public. A package of storm surge watch and
warning advisories, similar to existing hurricane products, will be rolled out
in 2015.

6.4 STORM SURGE DISASTER RISK REDUCTION
Disaster risk derives from a combination of physical hazards and the vulnerabilities of exposed elements. Hazard refers to the chance and characteristics of the hazardous phenomenon itself (e.g., flood depth, flood extent).


174

Coastal and Marine Hazards, Risks, and Disasters

Exposure refers to the presence (location) of people, livelihoods, physical
and biological systems, infrastructure, or economic, social, or cultural assets
in places that could be affected by physical events and which, thereby, are
subject to potential future harm, loss, or damage (IPCC, 2012). Vulnerability
is defined as the characteristics and circumstances of a community, system, or
asset that makes it susceptible to the damaging effects of a hazard, for
example, due to unsafe housing and living conditions (UNISDR, 2009). Both
exposure and vulnerability to storm surge are increasing. Jongman et al.
(2012) calculated that the global population exposed to 1:100-year coastal
flooding was about 271 million in 2010, with exposed assets of $13 trillion.
By 2050, Jongman et al. (2014) calculated that 345 million people will be
living in 1:100 coastal flood hazard locations, with exposed assets of
$43 trillion, an increase of approximately 200% from 2010. Dasgupta et al.
(2009) considered the potential impact of a 1:100 storm surge in 84 developing countries along with 577 of their cyclone-vulnerable coastal cities with
populations over 100,000, and compared it to a more intense impact later in
the century. Their study showed a significant asymmetry in storm surge risk,

with three cities (Manila, Alexandria, and Lagos) out of the 577 accounting
for 25% of future coastal population exposure and 10 cities accounting for
53% of the future exposure. The increased vulnerability of these cities can be
attributed to urban growth combined with physical exposure to storm surge
inundation.
The UNISDR 2013 Global Assessment Report on Disaster Risk Reduction
reported that as of December 2012, 85 countries had established multisector
national platforms for disaster risk reduction. With a few exceptions, these
institutional and legal systems have remained focused on disaster preparedness
and response rather than encouraging risk reduction. There is no quantitative
index to measure the scale of disaster caused by storm surge in different
regions. In addition, there is no easily understood scale to indicate the severity
of storm surge, with nothing comparable to the earthquake magnitude and
intensity scales or the SSWS. Few disaster risk management systems have
been able to employ land use planning to encourage disaster risk reduction
(UNISDR, 2013). As with any natural hazard, storm surge risk is a function of
the physical hazard (storm surge depth and extent), exposure (the location and
number of people and economic assets in locations that can be inundated by
storm surge), and vulnerability (susceptibility to damage and loss). Although
human vulnerability to storm surge is expected to increase in future years due
to population growth, urbanization, increased coastal settlement, poverty, and
changing weather patterns (Doocy et al., 2013), storm surge risk could be
reduced through a range of adaptive strategies. Storm surge barriers can
decrease the chance of flooding and thus reduce the physical hazard. Limiting
development in the most hazardous locations can reduce exposure. Improved
building codes, requiring flood-damaged buildings to be rebuilt, and new
developments to be constructed in a resilient fashion can reduce vulnerability,


Chapter j 6

Storm Surge Warning, Mitigation, and Adaptation

175

as can early warning systems and the establishment of effective procedures to
prepare for and recover from storm surge disasters. Insurance can help to cover
the remaining risk to individuals, can assist national governments in financing
recovery and reconstruction after a disaster, and can provide price signals to
discourage development in hazardous locations. As international efforts move
toward developing a post-2015 framework for disaster risk reduction, perhaps
progress will be made toward a unified approach to storm surge warning
similar to that which has been established for tropical cyclone and tsunami
warnings.

ACKNOWLEDGMENTS
Thanks to the Climate Change and Sea Level Rise Initiative at Old Dominion University,
who supported my research on flood insurance through a Visiting Scholarship, and where I
started thinking seriously about the issues related to adaptation.

REFERENCES
Aerts, J.C.J.H., Botzen, W.J.W., 2012. Hurricane Irene: a wake-up call for New York City? Nat.
Hazards Earth Syst. Sci. 12, 1837e1840.
Aerts, J.C.J.H., Boutzen, W.J.W., de Moel, H. (Eds.), 2013a. Cost Estimates for Flood Resilience
and Protection Strategies in New York City. Annals of the New York Academy of Sciences,
1294, pp. 1e104.
Aerts, J.C.J.C., Bowman, M., Boutzen, W.J.W., de Moel, H., 2013b. Storm surge barriers for New
York city and New Jersey. In: Aerts, J., Boutzen, W.J.W., de Moel, H. (Eds.), Cost Estimates
for Flood Resilience and Protection Strategies in New York City, Annals of the New York
Academy of Sciences, 1294, pp. 49e68.

Ahamed, M., 2013. Community based approach for reducing vulnerability to natural hazards
(cyclones, storm surges) in coastal belt of Bangladesh. Procedia Environ. Sci. 17, 361e371.
Atkin, A., April 2014. The calm after the storms. Environment Agency Blog 14. https://
environmentagency.blog.gov.uk/2014/04/14/the-calm-after-the-storms/.
Augustijn, D.C.M., van den Berg, M., de Bruine, E., Korving, H., 2011. Dynamic control of salt
intrusion in the Mark-Vliet river system, The Netherlands. Water Resour. Manage. 25,
1005e1020.
Baker, E.J., Broad, K., Czajkowski, J., Meyer, R., Orlov, B., 2012. Risk Perceptions and Preparedness Among Mid-Atlantic Coastal Residents in Advance of Hurricane Sandy. Risk
Management and Decision Processes Center, The Wharton School, University of Pennsylvania.
Working paper 2012-18.
Bajo, M., Zampato, L., Umgeisser, G., Cucco, A., Canestrelli, P., 2007. A finite element operational model for storm surge prediction in Venice. Est. Coast. Shelf Sci. 75 (1), 236e249.
BBC, March 5, 2014. Thames Barrier Breaks Closure Record. Available from .
uk/news/uk-england-london-26453484.
Bijker, W.E., 2002. The Oosterschelde storm surge barrier: a test case for Dutch water technology,
management and politics. Technol. Cult. 43 (3), 569e584.
Brakenridge, G.R., Syvitski, J.P.M., Overeem, I., Higgins, S.A., Kettner, A.J., Stewart-Moore, J.A.,
Westerhof, R., 2013. Global mapping of storm surges and the assessment of coastal vulnerability. Nat. Hazards 66, 1295e1312.


176

Coastal and Marine Hazards, Risks, and Disasters

Bras, R.L., Harleman, D.R., 2002. Obsolete? No. Necessary? Yes. The gates will save Venice. Eos
83 (10), 224e226.
Bowman, M.J., Flood, R.D., Hill, D., Wilson, R.E., 2005. Hydrologic Feasibility of Storm Surge
Barriers to Protect the Metropolitan New York e New Jersey Region. Marine Sciences
Research Center Technical Report. Stony Brook University.
Chan, E.Y.Y., Liu, S., Hung, K.C., 2013. Typhoon Haiyan and beyond. Lancet 382 (9908), 1873.
/>Coch, N.K., 2013. Hurricane flood protection in NYC e now is not the time for multiple surge

barriers. Discussion of Hill, D. 2012. The lessons of Katrina: learned and unlearned. Journal of
Coastal Research 28(2): 324-331 J. Coastal Res. 28 (5), 1313e1315.
Colle, B.A., Buonaiuto, F., Bowman, M.J., Wilson, R.E., Flood, R., Hunter, R., Mintz, A., Hill, D.,
June 2008. New York City’s vulnerability to coastal flooding: storm surge modeling of past
cyclones. Bull. Am. Meteorol. Soc., 829e841.
Cooper, J.A.G., Pile, J., 2013. The adaptation e resistance spectrum: a classification of contemporary adaptation responses to climate-related coastal change. Ocean and Coastal Manage.
Available from 10.1016/j.ocecoaman.2013.09.006.
Dasgupta, S., Laplante, B., Murray, S., Wheeler, D., 2009. Climate Change and Future Impacts of
Storm-surge Disasters in Developing Countries. CGD Working Paper 182. Center for Global
Development, Washington, DC. Available from .
Delta Committee Report., 2008. Available from />Delta Programme. Netherlands Ministry of Infrastructure and Environment and Ministry of Economic Affairs, 2014. Working on the Delta: Promising Solutions for Tasking and Ambitions.
Available from .
Dircke, P.T.M., Jongeliong, T.H.G., Jansen, P.L.M., 2013. In: Hill, D., Bowman, M.J., Khinda, J.S.
(Eds.), Storm Surge Barriers to Protect New York City: Against the Deluge; 2009 March
30e31. American Society of Civil Engineers, Reston, Virginia, pp. 65e87.
Doocy, S., Dick, S., Daniels, A., Kirsch, T.D., April 16, 2013. The human impact of tropical cyclones:
a historical review of events 1980e2009 and systematic literature review. PLOS Curr. Disasters.
Edition 1.
eSurge, 2014. Tropical Cyclone Haiyan. Available from o/stormsurge-tropical-cyclone-haiyan.
Environment Agency (UK), 2010. Thames Estuary Plan: Managing Flood Risk through London
and Thames Estuary. Crown Copyright, London. Available from />Environment Agency, 2013. Thames Estuary Phase 1 Programme e Industry Event. http://
a0768b4a8a31e106d8b0-50dc802554eb38a24458b98ff72d550b.r19.cf3.rackcdn.com/TEP1_
Industry_Day_Presentation_LIT_CODE_8772_(2)_9bae3c.pdf.
Eqecat, 2009. Technical Report: The Hurricane and Storm Damage Risk Reduction System in New
Orleans, Louisiana. Eqecat.
Ferris, E., November 26, 2013. For Typhoon Haiyan Survivors, the Recovery Is Just Beginning.
NOVA Next. Available from />Fontini, F., Umgeisser, G., Vergano, L., 2010. The role of ambiguity in the evaluation of the net
benefits of the MOSE system in the Venice lagoon. Ecol. Econ. 69, 1964e1972.
Fritz, H.M., Blount, C., Thwin, S., Thu, M.K., Chan, N., 2011. Observations and modelling of
cyclone Nargis storm surge in Myanmar. In: Wallendorf, L.A., Jones, C., Ewing, L.,

Battalio, B. (Eds.), Proceedings of Solutions to Coastal Disasters: 2011 June 26e29;
Anchorage, Alaska. American Society of Civil Engineers, Reston, Virginia, pp. 1e9.


Chapter j 6

Storm Surge Warning, Mitigation, and Adaptation

177

Grasso, V.F., Singh, A., 2009. Early Warning Systems: State-of-art Analysis and Future Directions.
Draft report. United Nations Environment Program.
Hallegatte, S., 2009. Strategies to adapt to an uncertain climate. Global Environ. Change 19,
240e247.
Harriman, L., 2014. Cyclone Phailin in India: early warning and timely action saved lives. Article
reproduced from UNEP Global Environmental Alert Service (GEAS). Environ. Dev. 9,
93e100.
Haque, U., Hashizume, M., Kolivras, K.N., Overgaard, H.J., Das, B., Yamamoto, T., 2012.
Reduced death rates from cyclones in Bangladesh: what more needs to be done? Bull. W.H.O.
90, 150e156.
Hill, D., 2012. The lessons of Katrina: learned and unlearned. J. Coastal Res. 28 (2), 324e331.
Hill, D., 2013. New York City plans to be flooded. J. Coastal Res. 29 (6A) (iii-iv).
Horner, R.W., 1979. The Thames barrier project. Geogr. J. 145 (2), 242e253.
Hostens, K., Hummel, H., Mees J., Vincx, M., Heip, C.H., 2003. The fish fauna of the Oosterschelde, a decade after completion of the engineering works. In: Hostens, K. The Demersal
Fish and Macro-invertebrate Assemblages of the Westerschelde and Oosterschelde Estuaries
(Southern Bight of the North Sea) (Ph.D. thesis). 59e77.
Hudson, P.F., Middelkoop, H., Stouthamer, E., 2008. Flood management along the lower
Mississippi and Rhine rivers (The Netherlands) and the continuum of geomorphic adjustment.
Geomorphology 101, 209e236.
Hunter, P., 2012. The St Petersburg Flood Protection Barrier: Design and Construction. CETMEF

PIANC Paris. HR Wallingford, 1e9. />IPCC, 2012. Managing the Risks of Extreme Events and Disasters to Advance Climate Change
Adaptation. A special report of Working Groups I and II of the Intergovernmental Panel on
Climate Change. Cambridge University Press, Cambridge.
Irish, J.L., Resio, D.T., 2010. A hydrodynamics-based surge scale for hurricanes. Ocean Eng. 37,
69e81.
Ivanov, S.V., Kosukhin, S.S., Kaluzhnaya, A.V., Boukhanovsky, A.V., 2012. Simulation-based collaborative decision support for surge floods prevention in St. Petersburg. J. Comput. Sci. 3, 450e455.
Jansen, O.E., Aarts, G.M., Reijnders, P.J.H., 2013. Harbour porpoises Phocoena phocoena in the
Eastern Scheldt: a resident stock or trapped by a storm surge barrier? PLoS One 8 (3), e56932.
/>JCOMM (Joint WMO/IOC Committee on Oceanography and Marine Meteorology), 2014. Guide
to Storm Surge Forecasting Geneva. WMO, Switzerland. Available from mm.
info/index.php?option¼com_content&view¼article&id¼175&Itemid¼37#04.
Jongman, B., Ward, P.J., Aerts, J.C.H., 2012. Global exposure to river and coastal flooding: long
term trends and changes. Global Environ. Change 22, 823e835.
Jonkman, S.N., Bockarjova, M., Kok, M., Bernardini, P., 2008. Integrated hydrodynamic and
economic modelling of flood damage in the Netherlands. Ecol. Econ. 66, 77e90.
Jonkman, S.N., Hillen, M.M., Nicholls, R.J., Kanning, W., van Ledden, M.M.van Ledden, 2013.
Costs of adapting coastal defences to sea-level rise - new estimates and their implications. J.
Coast.l Res. 29 (5), 1212e1226.
Jongman, B., Hochrainer-Stigler, S., Feyen, L., Aerts, C.J.H., Mechler, R., Botzen, W.J.W.,
Bouwer, L.M., Pflug, G., Rojas, R., Ward, P.J., 2014. Increasing stress on disaster-risk finance
due to large floods. Nat. Clim. Change 4, 264e268.
Kabat, P., Fresco, L.O., Stive, M.J.F., Veerman, C.P., van Alphaen, J.S.L.J.,
Parmet, B.W.A.H., Hazeleger, W., Katsman, C.A., 2009. Dutch coasts in transition.
Nat. Geosci. 2, 450e452.