Potential for nature-based mitigation of coastal flood risks From regional to global scale assessments Potenties voor natuur-gebaseerde mitigatie van overstromingsrisico’s in kustgebieden Een regionale tot globale studie Dissertation for the degree of Doctor in Science at the University of Antwerp to be defended by Rebecca Van Coppenolle Promotor: Prof. Dr. Stijn Temmerman Faculty of Science Department of Biology Antwerp 2018
222
Embed
Potential for nature based mitigation of coastal flood risks
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Potential for nature-based mitigation of coastal flood risks
From regional to global scale assessments
Potenties voor natuur-gebaseerde mitigatie van overstromingsrisico’s in kustgebieden
Een regionale tot globale studie
Dissertation for the degree of Doctor in Science at the University of Antwerp to be defended by
Rebecca Van Coppenolle
Promotor: Prof. Dr. Stijn Temmerman
Faculty of Science Department of Biology Antwerp 2018
Figure 1.2 Mechanisms behind the formation of storm surges and their inland propagation. Adapted from http://www.nola.com/hurricane/images/scourgeofsurge.pdf
Additionally numerous factors affect the height and extent of storm surges. They
are firstly influenced by the storm characteristics, as the radius of the maximum
winds (storms of larger radius generate higher peak water levels and flood
volumes), the storm forward speed (faster forward speeds are expected to create
higher surges in combination with lower flood volumes and inversely), the storm
track (the angle and direction at which the storm approaches the coast is decisive
for its impact) and the storm wind intensity and atmospheric pressure (stronger
winds as well as lower pressures produce higher surge) (Flather, 2001; McIvor,
Spencer, et al., 2012; Rego & Li, 2009; Resio & Westerink, 2008). The height of
storm surges is also related to the timing of the storm surge relative to the tide
level, particularly in areas of large tidal variations (e.g. macro-tidal areas). As such,
a storm surge reaching the coast around high tide in a macro-tidal environment
can lead to devastating consequences while the same surge reaching the coast at
low tide may be unnoticed (Flather, 2001; McIvor, Spencer, et al., 2012; Paul,
2009). Secondly, the bio-geomorphology of the coast largely influences the impact
of the incoming storm surge, e.g. the near-shore bathymetry (surges are higher in
large shallow-water coastal areas than in narrow and steep off-shore sloping
coastal areas), the concave or convex shape of the coastline (surges become larger
in case of landward converging coastlines such as funnel-shaped embayments and
estuaries), the geometry of estuarine and deltaic channels and water bodies
(channels allow the surge to propagate inland more easily) and the friction
exerted by the land surface that is flooded by the surge (increased surface
roughness or friction will slow the surge’s inland propagation and lower its
Westerink, 2008). Moreover, storm surges are generally associated with heavy
rainfall, which may cause additional flood risks through rainfall driven runoff and
General introduction
7
the freshwater discharge in estuaries and river deltas (McIvor, Spencer, et al.,
2012; Resio & Westerink, 2008; Woodruff et al., 2013).
In the coming decades, as a response to the global climate change, the current
frequency, intensity and tracks of the tropical and extra-tropical storms (Figure
1.3) and of the associated storm surges is expected to change (Knutson et al., 2010;
Tessler et al., 2015; Vitousek et al., 2017; Webster et al., 2005; Woodruff et al.,
2013). Tropical cyclones are formed and intensified over warm sea surfaces
(temperatures higher than 26°C) (Knutson, 2014; Webster et al., 2005). As oceans
and tropical oceans become warmer (IPCC, 2013), the maximal intensity of
cyclones might increase, as the most intense cyclones seem to not develop over
cool sea surface, but over warmer seas (Knutson, 2014). This will imply a rise in
the maximal wind speed of tropical cyclones by 2 to 11 % by 2100 (Knutson, 2014;
Knutson et al., 2010) and a higher occurrence of high intensity cyclones and fewer
occurrence of low intensity cyclones (Knutson, 2014; Webster et al., 2005). The
tracks of tropical and extra-tropical cyclones are also expected to shift with for
example a strengthening of storm tracks North of the British Isles or in the eastern
Pacific (Bengtsson et al., 2006). Observations and comparisons of the last two
cyclone seasons (2016 and 2017) to the general trend highlight that warmer SST
(sea surface temperatures) are at play in the strong cyclone activity of 2017 (Lim
et al., 2018). While the cyclone activity in 2016 showed some shifts in storm
tracks, with few cyclones in the main development region and the development of
some extreme and unusual storms in the Atlantic basin (Collins & Roache, 2017).
Chapter 1
Figure 1.3 Representation of the storms’ tracks and intensities following the Saffir-Simpson scale for the period 1842 to 2017 over the world based on the International Best Track Archive for Climate Stewardship (IBTrACS) (Knapp et al., 2010) and of the different storm appellations over the world
General introduction
9
1.1.3 Coastal Population Increase and Anthropogenic Activities
The world’s coastlines are populated since the early development of the
civilizations (Maddison, 2001; McEvedy & Jones, 1978). Current trends are leading
to the densification of the coastal population and to the growing value of the
associated assets located along the coasts (Guzmán et al., 2009; Small & Nicholls,
2003). By 2060, the population density of the LECZ is expected to reach 405 to 534
inhabitants per square kilometre, which corresponds to two times the current
LECZ’s density and ten times the current world’s average (Green & Short, 2003;
Mcgranahan et al., 2006; Neumann et al., 2015; Nicholls et al., 2008). This
increasing population, mostly concentrated in highly populated cities, or
‘megacities’, is putting more and more people and assets at flooding and erosion
risks (Bernstein et al., 2007; Von Glasow et al., 2013; Hallegatte et al., 2013;
Hanson et al., 2011; de Sherbinin et al., 2007).
Additionally, the increased human pressure and anthropogenic activities are
responsible for the disturbance, degradation and loss of natural coastal processes
and ecosystems, which are associated with increasing flood and erosion risks
(Auerbach et al., 2015; Balke & Friess, 2016; Barbier, 2014; Hanson et al., 2011;
Syvitski, 2005; Syvitski et al., 2009). Some of those disturbances are related to soil
drainage for urban and agricultural development that leads to soil compaction and
lower soil permeability or to the mining and extraction of groundwater, oil and gas
from the coastal substrate that contributes to coastal land subsidence. It is also
related to the shortage of sediment supply to deltas and estuaries due to upstream
river dams and levees that contribute in some areas to an increased erosion as the
limited deposition of sediments cannot counterbalance the land subsidence
(Kirwan & Megonigal, 2013; A. Murray, 2017; Pethick & Orford, 2013; Rovere et
al., 2016; Syvitski & Saito, 2007; Tessler et al., 2015). The embankment of tidal
wetlands along deltaic or estuarine channels for agriculture, aquaculture or urban
development purposes on the other hand leads to the loss of flood water storage
areas and to a rising level of the water in channels (Pethick & Orford, 2013;
Smolders et al., 2015). And the deepening and widening of the estuarine and
deltaic channels (e.g. for industrial shipping facilities) may facilitate the inland
propagation of tides and storm surges (Loder et al., 2009; Temmerman et al.,
2012).
Chapter 1
10
1.2 Ecosystems in the coastal zone
Coastal zones are home to a large diversity of marine and coastal ecosystems
(Figure 1.4), such as mangrove forests, salt marshes, dunes, seagrass meadows,
coral reefs, oyster reefs, rocky shores... In this PhD thesis our interest is mainly
focused on tidal wetlands (i.e. mangrove forests and salt marshes) and on seagrass
meadows and coral reefs, because they are widely considered to contribute to
mitigation of the storm surges impacts (see below) and because global data are
available for these ecosystems (Figure 1.4).
1.2.1 Ecosystems Types
Salt marshes in temperate regions and mangrove forests in sub-tropical and
tropical regions, together called tidal wetlands throughout this thesis, are
predominately present in river deltas or estuaries, where the low-lying coastal
plain is large and where low wave energy allows the development of vegetation on
muddy shores (Alongi, 2009; Scott et al., 2014; Wolanski & Elliott, 2015). They are
highly influenced by regular tidal flooding and drainage; their salt tolerant
vegetation colonizes the shores between the mean sea level and the highest tides
or extreme water levels, by following a vertical plant zonation related to the
exposure time to marine water, the stress generated by waves and the salinity
(Alongi, 2009; Mcowen et al., 2017).
Salt marshes are the most widespread tidal wetlands, mostly occupying
temperate latitudes. They are also found in polar regions and at the landward side
of mangrove forests in tropical latitudes with specific plant species adapted to
extreme cold or warm temperature conditions, respectively. Salt marsh vegetation,
consisting of tall halophytic grasses, herbs and low shrubs (Mcowen et al., 2017;
Scott et al., 2014; Wolanski & Elliott, 2015), is covering an area of about 55 000
km² over the full world (Mcowen et al., 2017).
Mangrove forests are dominated by trees and shrubs and occur in
tropical and subtropical regions, where the temperatures stay relatively warm (>
16°) all year round (Giri et al., 2011; Scott et al., 2014; Spalding et al., 2010;
Wolanski & Elliott, 2015). Different types of mangroves are found based on the
hydrodynamic conditions of the environment (Giri et al., 2011; Scott et al., 2014;
Wolanski & Elliott, 2015); riverine mangroves (R-type) are found in sheltered tidal
estuaries with an upstream input of freshwaters and nutrients and tidal creeks
draining the wetlands. The tree height and density is the highest in this type of
mangroves (≈ 20 m in height). Fringing mangroves (F-type) are located along
General introduction
11
coastlines, they are protected from the sea by coral reefs or headlands, and have
lower tree height and density. Basin mangroves (B-type) are present in inland
depressions where the water flow is very limited (water stagnation), and where
the trees do not exceed 10 m in height. Lastly, dwarf mangroves, located in
isolated and very stressful environments, are often present in clumps of very small
trees (≈ 2 m) (Giri et al., 2011; Scott et al., 2014; Spalding et al., 2010; Wolanski &
Elliott, 2015). Mangrove forests are covering an area of 152 000 km² over the
world (Giri et al., 2011; Spalding et al., 2010).
Seagrass meadows or seagrass beds develop in all climate regions.
Seagrass species are permanently submerged marine and estuarine flowering
plants and show large variation in shape and size (Green & Short, 2003). They
principally develop on sandy or muddy substrates in shallow clear waters in the
form of extensive beds, isolated patches or as part of a habitat mosaic (i.e. in
proximity and with ecological links with other marine and coastal habitats
(Duarte, 1991; Green & Short, 2003; Koch et al., 2006). Changing environmental
conditions are relatively well managed by seagrasses that are highly dynamic and
can migrate to new areas in relatively short timeframes (Green & Short, 2003;
O’Brien et al., 2017; Short & Neckles, 1999). Seagrasses are observed over an area
of about 177 000 km² over the world (Green & Short, 2003).
Coral reefs are marine habitats defined by a physical structure or skeleton
that keeps growing during the coral’s life by accumulation of calcium carbonate
and that is colonized by a multitude of organisms. Present between the 30°N and
30°S latitudes, corals need a solid substrate to develop, in addition to warm (> 16-
18°C), salty and clear water (e.g. low amount of suspended sediments) (Bessell-
Browne et al., 2017; Duckworth et al., 2017; Spalding et al., 2002). Yet, once
settled, they can grow vertically to adapt to slow changes of their environmental
conditions (e.g. rising sea level, lower light availability...). Several types of coral
reefs exist based on their location from the shore (e.g. fringing reefs, barrier reefs,
atolls) and their structure (Spalding et al., 2002). Coral reefs are present over
about 285 000 km² along the world’s coastlines (Spalding et al., 2002).
Chapter 1
Figure 1.4 Worldwide distribution of the four coastal ecosystems considered in this thesis, i.e. mangrove forests (Giri et al., 2011), salt marshes (Mcowen et al., 2017), seagrass meadows (Green & Short, 2003) and coral reefs (Spalding et al., 2002)
General introduction
13
Figure 1.5 Illustration of the four above-described ecosystems; (a) salt marsh, (b) mangrove forest, (c) seagrass meadow and (d) coral reef
1.2.2 Coastal Ecosystem Functions
Each ecosystem, whether it are tidal wetlands, coral or oyster reefs, seagrass beds,
coastal dunes or rocky shores, provides a range of ecological functions and
ecosystem services. A set of those ecological functions and services, sometimes
common to multiple ecosystem types, are described here.
Nutrient cycling inside the different ecosystems generates ecological functions and
services such as nutrient and food supply, water purification (e.g. pollutant uptake,
removal of excess nutrients) or carbon sequestration by the burial of the carbon in
anoxic soil (Barbier et al., 2011; Duarte et al., 2013; Duckworth et al., 2017;
Fourqurean et al., 2012; McLeod et al., 2011; Mumby & Steneck, 2018; Tack et al.,
2007; Temmerman et al., 2004; Teuchies et al., 2013). In terms of monetary value
of those ecosystem services, the water purification function of salt marshes in the
USA is estimated to provide an equivalent to traditional waste water treatments
amounting up to 15 000 US$ per acre (0.004 km²) (Barbier et al., 2011; Breaux et
al., 1995). While the carbon sequestration capacity of the tidal wetlands at an
Chapter 1
14
overall average rate of 210 g of CO2 per square-meter per year converted to the
Carbon Emission Reduction price corresponds to a monetary valuation of about 30
US$ per hectare per year (Barbier & Barbier, 2014; Chmura, 2003; McLeod et al.,
2011).
Due to the high concentration of nutrients and food in coastal ecosystems, as well
as due to their sheltered environment, they are providing a suitable reproduction
and nursery habitat for fish, shellfish and crustaceans (Barbier et al., 2011; S. Y.
Lee et al., 2014; Millennium Ecosystem Assessment, 2005; Ondiviela et al., 2014).
Mangroves, salt marshes and seagrass meadows provide a protection against
larger predators that cannot penetrate the complex structures of vegetation, while
larger fish (e.g. tuna, sharks...) find a diverse food source in coral reefs (Barbier et
al., 2011). Coastal ecosystems play a major role in the maintenance of healthy
fisheries. For example, in Thailand the mangrove forests can contribute to up to 1
000 US$ per hectare of capitalized increased offshore fishery production (Barbier,
2007). Coral reefs in the Philippines provide fishes for local consumption and live-
fish export for up to 45 000 US$ per km² per year and 10 000 US$ per km² per
year respectively (White et al., 2000). The loss of 127 km² of seagrass in Australia
was estimated to result in a loss of fishery production of 235 000 AU$ (Barbier et
al., 2011; McArthur & Boland, 2006).
Coastal ecosystems provide a highly valuable recreational and touristic
environment due to their unique landscapes and diversity of fauna and flora.
Subsequently, they generate revenues for the local populations, as in the
Seychelles where 40 000 tourists per year visit the marine parks generating
88 000 US$ (Mathieu et al., 2003), or along the North Carolina’s beaches where
tourists spend on average 166 US$ per trip (Barbier et al., 2011; Landry & Liu,
2009).
Materials provided by coastal ecosystems (e.g. wood from mangrove forests, lime
of coral reefs, grasses and herbs from salt marshes...) can be also highly valuable
and can be cut, extracted or harvested for human use (Barbier et al., 2011; Bolund
& Hunhammar, 1999; Reddy et al., 2016). In the UK, livestock grazing in salt
marshes generates about 15 £ per hectare per year of net income (King & Lester,
1995) and the gathering of mangrove products in Thailand is estimated at some
600 US$ per hectare per year (Barbier, 2007; Gedan et al., 2011).
Moreover, coastal ecosystems are providing a protection against coastal flood and
erosion risks (see below). In the Indian Ocean, the loss of coral reefs due to the
bleaching event of 1998 is estimated to have diminished the property value by 174
US$ per hectare per year due to the loss of coastal protection offered by the former
General introduction
15
coral reefs (Barbier et al., 2011; Wilkinson et al., 1999), while the presence of salt
marshes along the U.S. Atlantic and Gulf Coasts reduced hurricane induced
damages by more than 8 000 US$ per hectare per year (Costanza et al., 2008).
However, the anthropogenic development and pressure along the coastlines
during the last centuries resulted in large losses and degradations of the different
coastal ecosystems (Almeida et al., 2014; Hoeksema, 2007; Lotze et al., 2006; Scott
et al., 2014; Valiela et al., 2009). The worldwide loss of mangrove forests, salt
marshes and seagrass meadows is estimated at 20 to 50 % (Barbier et al., 2008;
McLeod et al., 2011; Millennium Ecosystem Assessment, 2005; Spalding et al.,
1997; Valiela et al., 2001), while 75 % of the global coral reefs are under threats
from the changing environmental conditions and anthropogenic impacts (Spalding
et al., 2002; Spalding, Ruffo, et al., 2014). Although recognized as a dramatic trend,
the loss of coastal ecosystems and services for the benefit of anthropogenic
activities is still ongoing (Arkema et al., 2013; Sutton-Grier et al., 2018; Tian et al.,
2016; Valiela et al., 2009). Projections for the next 100 years estimate the future
losses at 30 to 40 % of the actual considered coastal ecosystems via land
location in the deltas and estuaries, ecosystem size...) and the bathymetry and
topography of the surrounding large-scale coastal landscape (e.g. continental shelf
slope, channels structure...) (Leonardi et al., 2018). Each of these factors can
enhance or reduce the storm surge attenuation provided by the tidal wetlands; a
summary of the existing insights from the literature is presented in Table 1.2.
In tidal wetlands, i.e. salt marshes and mangrove forests, there are two main
mechanisms behind the storm surge mitigation (Figure 1.6); (1) the so-called
within-wetland attenuation, where the surge’s energy is absorbed through friction
induced by the vegetation canopy and by the topographic variations of the bed
surface (Barbier et al., 2013; Costanza et al., 2008; Leonardi et al., 2018; Mazda et
al., 1997, 2006), and (2) the so-called along-channel attenuation, where the excess
of water brought by the surge can flow into the low-lying tidal wetlands areas
adjacent to the channels within deltas and estuaries, thereby reducing the height
of a storm surge while it is travelling inland along a channel (Kobashi & Mazda,
2005; Leonardi et al., 2018; Smolders et al., 2015; Stark et al., 2016).
Chapter 1
18
Figure 1.6 Scheme illustrating the within-wetland and along-channel attenuation potential of tidal wetlands. Adapted from Stark (2016)
Compared to tidal wetlands, there are less available studies on storm surge
mitigation by seagrass meadows and coral reefs.
Seagrass meadows significantly influence the hydrodynamic environment by
reducing the flow velocity, dissipating the wave’s energy and stabilizing the
sediments. Consequently, they diminish erosion risks, and reduce storm surges at
a magnitude expected to be comparable to salt marshes (Duarte et al., 2013;
Fonseca & Cahalan, 1992). The maximal storm surge reduction happens in shallow
water and low wave energy environments, where the canopy height accounts for
more than 15 % of the water column (Adriana Gracia et al., 2018; Ondiviela et al.,
2014).
The structural complexity of coral reefs results in hydraulic roughness and a great
frictional effect of the reef on the water column (Ferrario et al., 2014; Harris et al.,
2018; UNEP-WCMC, 2006). The storm surge reduction, by coral reefs can reach 97
% under hurricanes conditions over the whole reef, with most of the attenuation
happening over the reef flat (i.e. the shallow part of the reef that extends from the
reef crest, or the seaward edge of the reef, and the shore) (Ferrario et al., 2014;
Harris et al., 2018; Principe et al., 2012; UNEP-WCMC, 2006). Coral reefs show
comparable storm surge attenuation capacities than artificial coastal defences
(Ferrario et al., 2014; Principe et al., 2012; UNEP-WCMC, 2006).
General introduction
Table 1.1 Storm surge attenuation rates across salt marshes and mangrove forests based on in situ measurements and hydrodynamic models. Adapted and completed from McIvor et al. (2012) and Stark et al. (2015).
Location vegetation type Event Attenuation
rate (cm/km)
Reference
Southern Louisiana coastal marsh Compilation of 7 storms between 1909 and 1957
1.6 - 20 United States Army Corps of Engineers (2006)
Louisiana marsh & open water Hurricane Andrew (1992), cat. 5 4.4 - 4.9 Lovelace (1994)
Het Verdronken Land van Saeftinghe, Western Scheldt, Netherlands
marsh & channels Simulations validated with in situ measurements
0 - 25 Stark et al. 2016
Cameron Prairie, Louisiana marsh Hurricane Rita (2005), cat. 3 10.0 Wamsley et al. (2010) calculated with data from McGee et al. (2006)
Sabine, Louisiana marsh Hurricane Rita (2005), cat. 3 25.0 Wamsley et al. (2010) calculated with data from McGee et al. (2006)
Vermillion, Louisiana marsh Hurricane Rita (2005), cat. 3 4.0 Wamsley et al. (2010) calculated with data from McGee et al. (2006)
Vermillion, Louisiana marsh Hurricane Rita (2005), cat. 3 7.7 Wamsley et al. (2010) calculated with data from McGee et al. (2006)
Ten Thousand Island National Wildlife Refuge, Florida
mangrove & interior marsh
Hurricane Charley (2004), cat. 3 9.4 Krauss et al. (2009)
Everglades National Park, Florida mangrove Simulations validated with Hurricane Wilma (2005)
20 - 50 Zhang et al. (2012)
Everglades National Park, Florida no vegetation Simulations validated with Hurricane Wilma (2005)
6 - 10 Zhang et al. (2012)
Chapter 1
20
Table 1.2 Summary of the existing insights from the literature on the characteristics of storm surges, wetlands and channels that influence the attenuation rate of a storm surge (i.e. that enhance or reduce the attenuation)
Enhanced Attenuation
Reduced Attenuation
Mechanism References
Storm Surge
Level Shallow to moderate
Extreme Friction decreases with an increased water depth
Lawler et al., 2016; Resio & Westerink, 2008; Sheng et al., 2012; Wamsley et al., 2010
Duration Short (hours)
Long (days)
Long storms have more time to propagate inland and to fully flood the wetlands area
Resio & Westerink, 2008; Wamsley et al., 2010
Forward Moving Speed
Fast (e.g. 10 m/s)
Slow (e.g. 5 m/s)
Slow moving storm surges have more time to impact the coastal waters
Hu et al., 2015; Liu et al., 2013; Rego & Li, 2009; Sheng et al., 2012; Zhang et al., 2012
Wetlands
Soil Elevation High Low
Filling of the wetlands by the surge is longer and potentially still possible at the maximum surge height in high elevated wetlands soil
Loder et al., 2009; Smolders et al., 2015; Stark et al., 2016
Width Large Narrow Large wetlands are providing more storm surge water storage and for a longer time
Loder et al., 2009; Smolders et al., 2015; Stark et al., 2016
Location Upstream Downstream Upstream wetlands can store higher percentages of flood volume relative to downstream wetlands
Smolders et al., 2015
Accretion High Low Accreting wetlands keep up with sea level rise and maintain their mitigation capacities
Temmerman et al., 2012; Wamsley et al., 2010
Vegetated Wetland/Open
Water High Low
A high level of continuity in wetland vegetation cover provides higher attenuation rates
Hu et al., 2015; Loder et al., 2009; Schepers et al., 2017; Sheng et al., 2012; Zhang et al., 2012
Channels
Width Narrow Wide Wide channels concentrate and facilitate the landward propagation of the surge
Stark et al., 2016; Temmerman et al., 2012
Depth Shallow Deep Shallow wetlands channels exert more friction on the storm surge propagation
Stark et al., 2016; Temmerman et al., 2012
General introduction
21
1.4 Coastal flood defence strategies
Engineered coastal defence structures commonly consist of dikes, levees or dams.
Those structures can be found all over the world, and are especially widespread in
rich developed countries, as in The Netherlands, were most of the coastline is
delimited by dikes and embankments to prevent coastal flooding of large low-lying
lands that have been historically gained from conversion and drainage of coastal,
estuarine and deltaic wetlands (Hoeksema, 2007; Pierik et al., 2017; Wolff, 1993);
or as in the Mississippi delta or East Asian deltas (Ma et al., 2014; Temmerman &
Kirwan, 2015). These structures sometimes combine protective and recreational
functions as the ‘Promenade des Anglais’ in Nice (France) (Pranzini et al., 2015).
Whilst the knowledge and expertise on the implementation and functioning of the
engineered structures are large and well-established, their creation, construction
and maintenance is expensive (Leonardi et al., 2018; R. L. Morris et al., 2018;
Temmerman et al., 2013). Consequently, the presence of coastal protection
structures is not solely linked to the richness of the country but on the willingness
of the policy makers to invest in such structures (Nicholls et al., 2008). Some
coastlines present then low safety standards, because of non-reliable or non-
existent flood defence structures (Dasgupta et al., 2009; Mcgranahan et al., 2006;
Nicholls et al., 2008; de Sherbinin et al., 2007), that might create a false feeling of
security to the coastal communities (Sutton-Grier et al., 2015). Engineered coastal
protection structures present other disadvantages, for instance, they disturb
natural coastal processes such as sediment supply to tidal wetlands, which is an
essential process through which tidal wetlands can build up land elevation with
sea level rise (Firth et al., 2014; Sutton-Grier et al., 2015; Temmerman & Kirwan,
2015). In addition, although they are able to protect coastal communities against
the impacts of storm events during a certain period of time, they have a certain
lifetime and their strength weakens with age; furthermore engineered flood
defence structures do not have a self-adaptive capacity in response to changing
environmental conditions such as sea level rise (Firth et al., 2014; Leonardi et al.,
2018; R. L. Morris et al., 2018; Sutton-Grier et al., 2015) (Figure 1.7).
As coastal protection is a necessity and will need an enhanced efficiency and
adaptation in regards to the expected socio-economic and climatic evolutions and
associated threats, nature-based solutions for coastal flood and erosion risks
reduction are increasingly studied as an alternative and add-on to standard
engineered flood defences. The nature-based solutions rely on the ability of the
coastal ecosystems to attenuate the flood and erosions risks, while providing
additional ecological functions, or ecosystem services (e.g. carbon sequestration,
Chapter 1
22
water purification, habitat and nursery for different animals...) (Barbier et al.,
2011; McLeod et al., 2011; Sutton-Grier et al., 2018). They have the advantage to
be sustainable and self-adaptive, meaning that the need for human management of
the ecosystems is limited. Furthermore, the natural ecosystems are, under certain
conditions, prone to adapt to gradual environmental changes (e.g. sea level rise,
sea surface warming, change in sediment supply...) and cost-effective; their limited
management needs make them financially less demanding than hard engineering
structures especially in the face of climate change (Barbier et al., 2013; Rao et al.,
2013; Reguero et al., 2018; Schueler, 2017; Spalding et al., 2013)(Figure 1.7).
Figure 1.7 Traditional engineering coastal protection structures (top) versus nature-based coastal protection (bottom) and their associated impact on the environment. Blue arrows indicate an increase or decrease in intensity of storm wave, storm surge and sea level. Adapted from Temmerman et al. (2013).
General introduction
23
Over the last decade, nature-based coastal protection and mostly hybrid
structures, i.e. combinations of nature-based and hard engineering structures,
were increasingly developed at several locations around the world. It involves the
incorporation of the existing coastal ecosystems in the coastal protection planning,
but also the restoration or creation of those ecosystems. However, still little is
known on the capacities of the restored or created ecosystems for storm surge
mitigation. Although they are expected to not recreate a pristine environment
(Elliott et al., 2016; Lawrence et al., 2018), several studies suggest that they could
be able to provide ecosystems services as water quality regulation or wind waves
and storm surge mitigation, yet probably not to the same extent as natural
ecosystems (Bullock et al., 2011; Hobbs et al., 2009; Rupprecht et al., 2017;
Spalding, McIvor, et al., 2014). One of the largest examples is the creation, in the
aftermaths of the landfall of hurricanes Katrina and Rita (August and September
2005) in Louisiana, of the Coastal Protection and Restoration Authority (CPRA).
The CPRA, which for the first time integrates coastal restoration and hurricane
protection under a single clear voice for the full state, has the mandate to develop,
implement and enforce hybrid strategies for safe and sustainable coasts that will
protect the local communities, the industrial infrastructures and the natural
resources. Between 2007 and 2017, 135 projects of restoration and risks
reduction were funded and realized. They involved the restoration of 146 km² of
coastal habitat, the improvement of 454 km of levees and the construction of
about 100 km of barrier islands and berms. Following this, the five-year strategic
plan started in 2017 comprises 124 projects of ecosystem restoration and flood
risks reduction (Figure 1.8) financed for a total amount of 50 billion US$. It will
lead to the restoration of about 2 000 km² of coastal habitat, and the reduction of
the expected coastal damages by 150 billion US$ over the next 50 years (Boesch et
al., 2006; Coastal Protection and Restoration Authority of Louisiana, 2017; Day et
al., 2007). Similarly, the San Francisco Bay Joint Venture works, since 2001,
towards the restoration of wildlife and wetlands in the San Francisco Bay with the
combined objective to gain benefits for wildlife and coastal protection. To date,
some 550 km² of tidal flats, marshes and lagoons and 260 km² of seasonal
wetlands were protected, restored or enhanced (San Francisco Bay Joint Venture,
2018).
In North-western Europe, natural environments are also increasingly incorporated
in coastal planning to (re)create coastal habitats and deliver sustainable coastal
flood and erosion risks reduction to the local communities (Esteves, 2014;
Gardiner et al., 2007; Meire et al., 2014; Rupp-Armstrong & Nicholls, 2007;
SigmaPlan, 2017). In England and Wales, the Making Space for Water policy
(2005), shifted the coastal protection focus from hard engineering to hybrid
Chapter 1
24
strategies integrating the natural environment. It resulted in multiple projects of
‘managed coastal realignment’ (Figure 1.8) that consists of the landward
relocation of coastal flood defence structures in order to provide space for tidal
marshes development. The realignment will concern about 660 km of coastline by
2030 and recreate about 62 km² of intertidal areas (Esteves, 2014; French, 2006;
Pendle, 2013). Similarly, in Belgium and in The Netherlands, several projects of
‘depoldering’ are implemented. In Belgium, the SigmaPlan has the goal to improve
the protection against the flooding of the Scheldt river and to develop the valuable
nature along the Scheldt (SigmaPlan, 2017). The plan consists on the
strengthening and heightening of hard defence (especially dikes) and the
construction of controlled flooding areas in former polder areas, and some of these
controlled flooding areas will be designed as such that they can develop in tidal
flats and marshes. One of the major projects is the Hedwige-Prosper polder project
(Figure 1.8) straddling Belgium and The Netherlands, in which the tidal flats and
marshes will be restored over and area of 4.65 km² of formerly embanked land
(SigmaPlan, 2017). As in the UK, this project consists of the landward relocation of
the dikes enabling the creation of tidal marshes on formerly embanked land.
After the 2004 tsunami in South East Asia and the typhoon Haiyan in the
Philippines in 2013, the registered damages in the coastal zones highlighted lower
damages in villages located behind mangrove forests (Balke & Friess, 2016;
Dahdouh-Guebas et al., 2005; Danielsen et al., 2005), and increased the interest in
ecosystems for coastal flood and erosion risks mitigation in those regions. Several
associations and countries (i.e. Indonesia, India, Sri Lanka, Thailand and Malaysia)
collaborated to develop mangroves restoration projects (FAO, 2007; Schmitt,
2012) that allowed amongst others the restoration of 20 km² of mangrove forest
in Indonesia and the plantation of 310 000 seedlings over the Sri Lanka’s coasts
(Schmitt, 2012). Nature-based coastal protection strategies are then not just a
theoretical concept for coastal protection, they are already implemented at small
to large scales in several countries over the world (Esteves, 2014; Marois & Mitsch,
Figure 1.8 (A) Completed or funded projects of the Louisiana Coastal Master Plan for 2017. It includes 79 marsh restoration projects, 13 structural protection projects, and 32 non-structural risk reduction projects (Coastal Protection and Restoration Authority of Louisiana, 2017); (B) location and type of 54 managed realignment projects along the UK’s coasts (Esteves, 2014); (C) Map of the depoldering of the Hedwige-Prosper polder along the Scheldt (Belgium and The Netherlands) (SigmaPlan, 2017).
Chapter 1
26
1.5 Thesis Objectives and Outline
Although an increasing amount of studies accounts for the contribution of coastal
ecosystem to nature-based mitigation of coastal flood and erosion risks on local to
regional scales there are, to our knowledge, no studies that have explored the
potential of nature-based coastal risk mitigation on regional to global scales.
Global scale studies have the advantage to reach more easily the local communities
and policy-makers and are then needed to promote the integration of coastal
ecosystems in coastal flood defence strategies. As the construction and
maintenance of hard engineering structures for coastal protection is expensive
and needs extensive maintenance, large portions of the world’s coasts have a low
protection and are vulnerable to coastal flood and erosion risks. Furthermore, the
implementation of nature-based or hybrid strategies for coastal flood and erosion
risks mitigation are often made in the aftermaths of destructive storm surge
events (i.e. Caribbean coast in the USA, South East Asia or in the Philippines).
Therefore, highlighting the worldwide presence of coastal ecosystems and their
potential for coastal flood and erosion risks mitigation is a necessity to increase
their incorporation in coastal defence strategies before having to deal with the
consequences of devastating storm surge events. As such, global scale studies will
promote the need for new local scale studies and spread to the different
stakeholders and policy-makers the possibility to implement nature-based
strategies while including the local characteristics of the area. Furthermore, with
the knowledge of the benefits coastal ecosystems can provide, local communities
and policy-makers can act to terminate the practices of land reclamation for
human use that weaken the coastal zone.
Subsequently, throughout this thesis, we pursued the general aim to identify
hotspots for nature-based storm surge flood risks mitigation, at a regional and
worldwide scale. Those hotspots were determined on the one hand as the deltas
plains, worldwide coastal plains and coastal cities with high needs for storm surge
mitigation, i.e. highly populated low-lying areas, low-lying areas with valuable
assets or coastal areas frequently exposed to storm surges. And, on the other hand,
as the deltas, the worldwide coastal plains and the coastal cities that have a high
potential for nature-based strategies via the incorporation of the existing or
potentially re-created coastal ecosystems into the coastal protection planning,
mostly as add-ons to the hard engineering structures.
Firstly, we created a GIS model based on globally available datasets that defines
the storm surge mitigation function of tidal wetlands for the low-lying areas of
General introduction
27
eleven highly populated deltas (Chapter 2) and for the entire world’s coastline
(Chapter 3).
Secondly, in order to create a comprehensive global assessment of the hotspots for
nature-based flood risks mitigation, we quantified the current extent of the four
meadows and coral reefs) in front of highly populated and flood-exposed coastal
cities (Chapter 4). Subsequently, in line with existing projects of nature-based
coastal risk mitigation by tidal wetland restoration or creation that are executed at
several places around the world, we estimated the potentially available areas for
tidal wetlands restoration or creation (Chapter 5) in front of those highly
populated and flood exposed coastal cities.
We conclude this thesis by a synthesis (Chapter 6) including the main findings of
the different chapters as well as some discussions on the limitations of our
approaches, on future researches and on managerial implications.
We explored the following aspects more specifically.
In Chapter 2, we created a GIS model based on globally available data and
relatively simple assumptions, simulating how a storm surge will be routed from
the open sea towards the potential floodplain within a delta. The model allows the
identification of the surface areas and population numbers within the delta which
are flooded via a flood pathway crossing through tidal wetlands. As such the model
identifies the areas and population that receive a mitigating effect from storm
surge attenuation by the tidal wetlands, and the magnitude of this mitigating effect
is estimated by the length of the flood pathway crossing through tidal wetlands. To
take a first step towards a global assessment of the potential contribution of tidal
wetlands to storm surge risk mitigation at a quasi-global scale, we applied the
model on 11 highly populated and flood-exposed deltas having tidal wetlands and
scattered over the world.
Subsequently, in Chapter 3 we made an upscaling of our quasi-global model to a
fully global model assessing the coastal areas and population numbers benefiting
from coastal flood risks mitigation by tidal wetlands at a worldwide scale. The aim
was to identify the specific areas, or hotspots, where mangrove forests and salt
marshes can provide the highest storm surge flood risk mitigation.
The following chapters are focusing on 136 coastal cities of more than 1 million
inhabitants that are exposed to coastal flooding due to storm surges. In Chapter 4,
we developed a GIS procedure defining the most likely pathway a storm surge
Chapter 1
28
would take to propagate from the open sea towards the city centres and the area
around this likely pathway that is potentially influencing the storm surge
propagation. In the area influencing the storm surge propagation, we quantified
the extent of four coastal ecosystems that are known to mitigate storm surge
propagation, i.e. mangrove forests, salt marshes, seagrass meadows and coral
reefs. As such, we identified the hotspot cities that have large surface area of
coastal ecosystems along their storm surge pathways, and therefore have a high
potential for nature-based flood risks mitigation. Furthermore, we aimed at
identifying the social and physical parameters explaining the spatial variations in
ecosystem surface areas in front of the 136 studies cities.
In Chapter 5, we explored the possibility of tidal wetlands’ restoration or creation
in the area influencing the storm surge propagation from the open sea towards the
city centre for the 136 coastal cities as defined in Chapter 4. Based on the
topography, the tidal amplitude, the land use and the population density, we
estimated the potentially available area (km²) where tidal wetlands could be
restored or created to enhance the nature-based flood risk mitigation in front of
the cities. We identified the hotspot cities for tidal wetlands restoration or creation
and a set of social and physical parameters influencing the size of the area
potentially available for tidal wetlands creation.
Deltas
29
CHAPTER 2 Contribution of mangroves and salt marshes to nature-based mitigation of coastal flood risks in major deltas of the world
Rebecca Van Coppenolle, Christian Schwarz, Stijn Temmerman
This chapter is based on Van Coppenolle, R., Schwarz, C., & Temmerman, S. (2018). Contribution of Mangroves and Salt Marshes to Nature-Based Mitigation of Coastal Flood Risks in Major Deltas of the World. Estuaries and Coasts, 41(6), 1699–1711. https://doi.org/10.1007/s12237-018-0394-7
Chapter 2
30
Abstract
Nature-based solutions are rapidly gaining interest in the face of global change and
increasing flood risks. While assessments of flood risk mitigation by coastal
ecosystems are mainly restricted to local scales, our study assesses the
contribution of salt marshes and mangroves to nature-based storm surge
mitigation in 11 large deltas around the world. We present a relatively simple GIS
model that, based on globally available input data, provides an estimation of the
tidal wetland’s capacity of risk mitigation at a regional scale. It shows the high
potential of nature-based solutions, as tidal wetlands, to provide storm surge
mitigation to more than 80% of the flood-exposed land area for 4 of the 11 deltas
and to more than 70% of the flood-exposed population for 3 deltas. The magnitude
of the nature-based mitigation, estimated as the length of the storm surge pathway
crossing through tidal wetlands, was found to be significantly correlated to the
total wetland area within a delta. This highlights the importance of conserving
extensive continuous tidal wetlands as a nature-based approach to mitigate flood
risks. Our analysis further reveals that deltas with limited historical wetland
reclamation and therefore large remaining wetlands, such as the Mississippi, Niger
and part of the Ganges-Brahmaputra deltas, benefit from investing in the
conservation of their vast wetlands, while deltas with extensive historical wetland
reclamation, such as the Yangtze and Rhine deltas, may improve the sustainability
of flood protection programs by combining existing hard engineering with new
nature-based solutions through restoration of former wetlands.
Deltas
31
2.1 Introduction
Global climate change induces acceleration of sea level rise and is expected to
increase the intensity of storm surges, and as such is threatening coastal and
deltaic areas worldwide (Bengtsson et al., 2006; Hallegatte et al., 2013; Hinkel et
al., 2014; Woodruff et al., 2013). Storm surges originating from severe storms,
such as tropical cyclones, propagate from the sea towards the land with surge
heights that can reach several meters above mean sea level, causing densely
populated low-lying areas in river deltas to be particularly vulnerable to storm
surge flood risks; (Day et al., 2007; Tessler et al., 2015). Additionally, the globally
averaged population density in the Low Elevation Coastal Zone (LECZ, i.e. less than
10 meters above mean sea level) is expected to grow from 241 people/km² in
2015 (i.e. five times the world’s average) to 405 to 534 people/km² by 2060
(McGranahan et al., 2007; Neumann et al., 2015; Small & Nicholls, 2003).
This increase of both coastal population density and risk probability of coastal
flooding events calls for the development of sustainable coastal management
strategies. Apart from traditional hard engineered flood defence structures, such
as dams or dikes, nature-based or ecosystem-based coastal flood defence is
increasingly proposed as an alternative or addition to traditional hard
engineering, and relies on the conservation and in certain cases restoration of
coastal and deltaic ecosystems (Cheong et al., 2013; Duarte et al., 2013;
Temmerman et al., 2013). Here we focus on salt marshes and mangrove forests,
which we collectively call tidal wetlands throughout this paper. Among their
valuable ecosystem services, tidal wetlands have the capacity to attenuate waves,
reduce shoreline erosion and inland storm surge propagation, and to sustain
themselves with sea level rise by allochthonous sediment accretion (Gedan et al.,
2011; Kirwan et al., 2016; Shepard et al., 2011; Temmerman & Kirwan, 2015). As
such, nature-based flood risk mitigation is increasingly regarded as a polyvalent,
self-adaptive and sustainable strategy (Temmerman et al., 2013).
Observational and hydrodynamic modelling studies have demonstrated the value
of tidal wetlands for storm surge mitigation due to the resistance exerted by the
wetland vegetation and topography on incoming storm surges, implying a
landward attenuation in storm surge height further referred to as storm surge
attenuation or reduction (Barbier et al., 2013; Costanza et al., 2008; Krauss et al.,
2009; Stark et al., 2015; Wamsley et al., 2010; Zhang et al., 2012). This attenuation
is quantified as the rate of vertical reduction in storm surge height per horizontal
inland distance over the delta plain (expressed in cm/km). It depends on various
factors as the flow resistance provided by the coastal geomorphology and its
Chapter 2
32
vegetation or land cover type and on the specific properties of the storm surge,
such as its height and duration (Loder et al., 2009; McIvor, Spencer, et al., 2012;
Stark et al., 2015; Wamsley et al., 2010). Observed rates of storm surge attenuation
over tidal wetlands range from a couple of cm/km to 25 cm/km (Krauss et al.,
2009; Stark et al., 2015; Wamsley et al., 2010), with maximum rates of up to 50
cm/km reported from a hydrodynamic modelling study in Florida’s mangroves
(Zhang et al., 2012). Although such local to regional observational and modelling
studies play an important role in advancing our understanding of the role of tidal
wetlands in storm surge risk mitigation, there are no upscaling studies yet that
have explored the potential contribution of tidal wetlands to storm surge risk
mitigation on a quasi-global scale.
Our study aims to take a first step towards such a global assessment of the
contribution of tidal wetlands to nature-based storm surge risk mitigation, by
selecting 11 of the most populated deltas around the world. We present results
from a GIS model based on globally available data and on relatively simple
assumptions to define the storm surge mitigation function of tidal wetlands for
low-lying delta areas and populations.
2.2 Material and Method
2.2.1 Study areas
For the selection of the studied deltas, the world’s deltas were ranked according to
their total population size as reported by Ericson et al. (2006). Starting from this
list, the selection of the deltas was firstly based on the presence of tidal wetlands,
then on the highest population, and lastly on their global distribution, so that at
least one delta per continent (North America, South America, Europe, Africa, Asia
and Australia) was selected (Table 2.1).
Some deltas could not be included in the study due to the lack of data regarding
the distribution of salt marshes as it is the case for the Pearl River delta in China.
Deltas
33
Table 2.1 Main characteristics and geographical distribution of the deltas selected for the study. A more extensive description of the deltas can be found in the Supplementary Information.
Country Tidal
wetland type
Population (Ericson et al., 2006)
Delta area as delineated
in the study (km²)
Tidal wetlands within the
delta (km²) Ganges-
Brahmaputra delta India/
Bangladesh Mangrove 111,000,000 78,453.9 6,431.6
Yangtze delta China Salt Marsh 42,100,000 61,251.9 121.1
Amazon delta Brazil Mangrove 2,930,000 42,028.3 1,676.0
Rhine delta The
Netherlands Salt Marsh 1,940,000 9,139.9 40.0
Mississippi delta USA Salt Marsh 1,790,000 36,894.3 6,014.3
Mahakam delta Indonesia Mangrove 706,000 2,425.1 503.5
Burdekin delta Australia Mangrove 5,800 1,441.4 129.7
2.2.2 Datasets
The following datasets were used.
The land elevation is defined with the NASA Shuttle Radar Topography
Mission (SRTM) Global 3 arc second V003 dataset (NASA JPL, 2013). The SRTM
dataset is so far the best-known Digital Elevation Model (DEM) available at a
global scale (Rodriguez et al., 2006; Sun et al., 2003). It is found to be more
accurate in areas with small slopes, such as deltas, yet there can be errors due to
reflection of the radar signal on vegetation canopies with an absolute vertical
error up to 16 m (http://www2.jpl.nasa.gov/srtm/datafinaldescriptions.html)
(Rodriguez et al., 2006; Sun et al., 2003).
The tidal wetlands distribution is based on different types of datasets. The
representation of mangroves is based on the Global distribution of mangroves from
the United States Geologic Survey (USGS, www.unep-wcmc.org) (Giri et al., 2011).
The distribution of salt marshes is a compilation of different country wide or
continent wide datasets as the European Commission program Corine Land Cover
of 2006 for Europe, the Geohabitats for Australia (Heap et al., 2001), the
Classification of Wetlands and Deepwater habitats for the United States (Federal
Chapter 2
34
Geographic Data Committee, 2013) and the Chinese wetlands mapping from Niu et
al. (2009).
The storm surge height is taken from the DIVA database (Hinkel et al., 2014;
Vafeidis et al., 2005). It corresponds to the storm surge water level above mean
sea level and is calculated by model simulations based on tidal levels, barometric
pressures, wind speeds and sea bed slopes for return periods of 10, 100 and 1000
years. The DIVA database uses the coastline of the Digital Chart of the World
(DCW, Environmental Systems Research Institute, ESRI, 2002) divided in
segments based on administrative and environmental parameters (Vafeidis et al.,
2005). To avoid inconsistencies due to differences in scale between the datasets
(e.g. tidal wetlands on the seaward side of the DIVA coastline), the DIVA coastline
was not directly used as the source of the flooding. Alternatively a ‘flood source’,
i.e. the coastline along the seaward delta front, is interpolated from a convex hull
based on the land area (the latter is defined more below). The storm surge heights
of the different segments stored in the DIVA coastline are then transferred to the
flood source with a shortest Euclidean distance algorithm (for further information
see the Supplementary Information).
The population distribution originates from the LandScan 2013 Global
Population Database (Bright et al., 2013). It represents the population over a 30
arc second grid resolution and integrates the diurnal movements and collective
travelling behaviour of the world population, i.e. the so-called “ambient
population”, averaged over 24 hours (Bright et al., 2013; Dobson et al., 2000). The
dataset of 30 arc second resolution was resampled to a resolution of 3 arc second
(to match the resolution of the SRTM land elevation dataset) based on the
guidelines of the LandScan documentation (Bright et al., 2013; UT BATTELLE
LLC.).
The extent of the world countries is the representation of the country boundaries
as they exist in January 2015 and is available through the ESRI platform (ESRI,
DeLorme Publishing Company, Inc., 2015). Due to the fact that the borders of the
different global datasets do not perfectly overlap (Lichter et al., 2011), the most
seaward extent of the emerging land was defined by merging the extent of the
world countries dataset and the tidal wetlands datasets, and this land extent is
further referred to as the land area.
The delta areas were delineated in accordance with spatial delineation of the
deltas in other studies (e.g. Coleman & Huh, 2004; Syvitski et al., 2009).
Deltas
35
The global datasets used present some limitations in regards to local data
accuracy and local data artefacts. Such limitations may include vegetation artefacts
in the SRTM dataset or the moderate resolution of the tidal wetlands datasets that
can omit some of the smaller wetlands areas for example. In addition, the use of a
global scale storm surge dataset based on the static calculation of the storm surge
height can introduce limitations related for example to the angle at which the
storm is making landfall that is not accounted for, or to the exact storm surge
height at one specific spot of the delta as the surge heights are defined for
segments that may not account for very local variations of the coastal area.
The resolution of all raster layers was converted to a 3 arc second grid based on
the World Geodetic System 1984 ellipsoid, which corresponds to the resolution of
the SRTM data grid.
2.2.3 Model description
The model simulates how a storm surge flood wave, entering a delta system from
the seaward delta front, would be routed in a landward direction through the
channels and over the potential floodplain of the delta, assuming that no artificial
flood protecting structures like dikes or dams would be present. The non-inclusion
of the protective structures as dikes or dams in the model refer to what would
happen if the protecting structure would fail. In this case of failure the model
highlights which coastal areas and coastal populations could still benefit from
storm surge mitigation thanks to the existing coastal ecosystems. As such, the salt
marshes and mangrove forests that existed before the construction of the
protecting structures are not accounted for in the study. It is based on a GIS
procedure developed in ArcGIS (10.3.1), and is similar to previously published
procedures that assess the coastal areas and number of people vulnerable to
storm surge flooding on regional to global scales (Arkema et al., 2013; Dasgupta et
al., 2011). The model does not simulate the full complexity of hydrodynamic
processes involved in flood propagation and therefore is not able to calculate
accurate flood depths and absolute values of reduction in flood depth behind tidal
wetlands. Nevertheless, it has the major advantage to be globally applicable to
make a relative comparison between delta systems around the world.
The input data of the model are globally available GIS data presented above in
combination with the storm surge attenuation rates derived from the range of
values found in the literature (Table 2.2) for the three land cover types considered
in this study (Table 2.3). The open water and channel areas are attributed a very
low attenuation value of 0.1 cm/km (Table 2.3) assuming that the flood height
Chapter 2
36
attenuation by friction over these areas is small. The remaining land area within
the delta is considered as a unique land cover type and is based on the literature
values of hydrodynamic modelling of storm surge propagation over coastal low
land areas (Table 2.2, (Zhang et al., 2012)), and on the assumption of a lower
attenuation rate over human-developed land (typically dominated by agricultural
land in deltas) than over natural wetlands. A conservative option was taken and
assumes that the remaining land areas have an average attenuation rate of 6
cm/km (Table 2.3). The attenuation rates for the tidal wetlands relate then to this
conservative rate over the remaining land. Additionally, the higher vegetation
canopy of the mangroves is expected to exert more friction and to result in higher
storm surge height attenuation (10 cm/km) than the lower vegetation of salt
marshes (8 cm/km) (Table 2.3). It is important to note, that there is much
uncertainty about which precise values to use for these attenuation rates as they
will vary with factors like the different land use. Currently little is known on how a
storm surge will be attenuated over an urban land use for example. For this reason
a sensitivity analysis was performed using a likely range of input values for the
attenuation rates for these land cover types (see Table 2.3), showing that the
model is relatively insensitive to it (see next section below).
Deltas
37
Table 2.2 Rates of storm surge height attenuation across various tidal wetland types based on observations during storm surges over tidal wetlands areas or on models calibrated by observations. Adapted and completed from McIvor et al. (2012) and Stark et al. (2015).
Location vegetation
type Event
Attenuation rate
(cm/km) Reference
Southern Louisiana coastal marsh Compilation of 7 storms between 1909 and 1957
1.6 - 20 United States Army Corps of Engineers (2006)
Louisiana marsh & open water
Hurricane Andrew (1992), cat. 5 4.4 - 4.9 Lovelace (1994)
Cameron Prairie, Louisiana marsh Hurricane Rita (2005), cat. 3 10.0 (Wamsley et al. 2010 calculated with data from McGee et al. 2006)
Sabine, Louisiana Marsh Hurricane Rita (2005), cat. 3 25.0 (Wamsley et al. 2010 calculated with data from McGee et al. 2006)
Vermillion, Louisiana Marsh Hurricane Rita (2005), cat. 3 4.0 (Wamsley et al. 2010 calculated with data from McGee et al. 2006)
Vermillion, Louisiana Marsh Hurricane Rita (2005), cat. 3 7.7 (Wamsley et al. 2010 calculated with data from McGee et al. 2006)
Ten Thousand Island National Wildlife Refuge, Florida
mangrove and interior marsh
Hurricane Charley (2004), cat. 3 9.4 (Krauss et al., 2009)
Everglades National Park, Florida Mangrove Simulations validated with Hurricane Wilma (2005)
20 - 50 (Zhang et al., 2012)
Everglades National Park, Florida no vegetation Simulations validated with Hurricane Wilma (2005)
6 - 10 (Zhang et al., 2012)
Chapter 2
38
Table 2.3 Storm surge attenuation rates attributed to the three land cover types considered in the study and used for the sensitivity analysis
Land cover type Attenuation rate
(cm/km) Attenuation rates for the
sensitivity analysis (cm/km)
Open water and Channels 0.1 Mangrove Salt Marsh
10.0 8.0
5, 10, 15, 20, 30, 40, 50
Remaining land 6.0 4, 6, 8, 10, 15, 20
The model works as follows. A cost distance algorithm is applied over the delta to
define the route that the storm surge follows during its landward propagation, i.e.
the storm surge flood pathway. The cost distance algorithm defines the flood
pathway between the flood source and every location (every pixel) of the delta
system as the route where the traveling cost of the storm surge is the lowest based
on the distance travelled and on the friction of the different land covers, or
attenuation rates (Table 2.3). It then allocates the cost of traveling, i.e. the
attenuation experienced by the storm surge, to every location. Subsequently,
pixels where the resulting storm surge height is higher than the land elevation are
considered at risk of flooding. The different steps of the model are presented for
the example of the Ganges-Brahmaputra delta in India and Bangladesh (Figure
2.1).
The model produces two main outputs for storm surge return periods of 10, 100
and 1000 years:
(1) It identifies the areas within the delta, and the number of people living in
those areas, that would be flooded via flood pathways crossing tidal
wetlands. We assume that areas flooded via those routes would benefit
more from nature-based attenuation of the storm surge, as compared to
areas flooded via pathways that do not cross tidal wetlands. In order to
select the pixels that are flooded or not flooded via tidal wetlands, two
scenarios were compared. The first scenario represents the existing extent
of tidal wetlands while the second scenario represents a situation where
the tidal wetlands would be replaced by the remaining land cover type and
its corresponding average attenuation rate (Table 2.3). Finally, all pixels
with a higher storm surge attenuation for scenario 1 as compared to
scenario 2 are identified as pixels having a storm surge pathway crossing
tidal wetlands.
Deltas
39
(2) It identifies the length of the flood route crossing tidal wetlands as a proxy
for the magnitude of nature-based storm surge flood risk mitigation. We
assume that the longer the flood wave travels through tidal wetlands
before it reaches inhabited land, the more that flood wave will be
attenuated. This was done by dividing the difference in storm surge levels
between the two scenarios by the difference in attenuation rates of the
tidal wetlands and remaining land (Table 2.3).
Figure 2.1 Maps of the Ganges-Brahmaputra delta illustrating the steps of the model. (a)Input data: topography of the delta (meters above mean sea level), area of the mangrove forest and location of the source of the storm surge. (b) Estimated flood-prone areas for a 1 in 100 year storm surge accounting for different storm surge attenuation rates over mangroves, open water and land area (Table 2.3), i.e. scenario 1. (c) Flood-prone areas with a flood pathway passing through the mangrove forest. (d) Mangrove forest length along the flood pathway for every pixel classified into the five distance classes
Further, we introduced a measure for the relative magnitude of nature-based flood
risk mitigation, which gives one value for the entire delta based on the length of
the tidal wetlands along the flood pathways. This length is classified into distance
classes i, which are classes of length of tidal wetlands along the flood pathway
(Table 2.4). The relative magnitude is calculated then in terms of land area (Mland)
and population (Mpop) benefiting from nature-based flood risk mitigation from the
following formula:
Chapter 2
40
𝑀 = ∑𝑁𝑖
𝑁∗ 𝑊𝑖
(Equation 1)
Where M is the relative magnitude of the nature-based flood risk mitigation, Ni is
the number of pixels or inhabitants in the distance class i, N is the total amount of
pixels or inhabitants having a storm surge pathway crossing tidal wetlands and W
is the weight of the distance class i (Table 2.4). The weight values follow a linear
function based on the mean of each length class.
Table 2.4 Value of the weight (W) of each class of length of tidal wetlands along the flood pathway
Distance class (i)
Classes of tidal wetlands length along the flood pathway (m)
but these deltas do not represent the largest land areas buffered by tidal wetlands
(Figure 2.3). In terms of relative percentages, the Ganges-Brahmaputra, the
Irrawaddy, the Amazon and the Rhine deltas, have about 15 to 20 % of their flood-
exposed land area buffered by tidal wetlands. This percentage rises up to about 40
to 60 % for the Mekong, Burdekin and Yangtze deltas, while the other deltas
present percentages of more than 80 % of the flood-exposed land area benefiting
from flood risk mitigation by tidal wetlands.
Chapter 2
Figure 2.2 Land surface area (left/blue) and number of people (right/red) benefiting from flood risk mitigation by tidal wetlands. The map represents the absolute land area (km²) or population (number of people) through the colour of the symbols, while the size represents the percentage of land area or population buffered by wetlands relative to flood-exposed land area or population
Deltas
43
Figure 2.3 (a) Relation between the land area benefiting from flood risk mitigation (km²) and the tidal wetlands surface area (km²) (Pearson’s r = - 0.073; p = 0.83), (b) Relation between the population benefiting from flood risk mitigation (inhabitants) and the tidal wetlands surface area (km²) (Pearson’s r = - 0.17, p = 0.61), for the 11 deltas studied
A similar pattern is found for the population buffered by tidal wetlands. In
absolute numbers, the deltas with the highest number of inhabitants buffered by
tidal wetlands are the Yangtze (5 922 009 inhabitants) and Mekong (4 602 641
inhabitants) delta, while the Burdekin delta is by far the delta with the lowest
number of people (169 people) buffered by tidal wetlands. Also for the total
population buffered by wetlands within a delta, there is no significant correlation
to the total tidal wetland area within that delta (Pearson’s r = -0.17, p = 0.61). In
relative percentages, the Ganges-Brahmaputra, Irrawaddy, Amazon and Rhine
deltas show the lowest percentages of flood-exposed population benefiting from
flood risk mitigation by tidal wetlands (less than 20 %). The percentages of the
Mekong, Burdekin, Yangtze and Mississippi deltas are of 25.4, 46.8, 49.1, and 58.1
% respectively, and the other deltas have more than 70 % of their flood-exposed
population buffered by tidal wetlands, rising up to 98 % for the Mahakam delta.
2.3.2 Relative magnitude of the nature-based storm surge flood risk
mitigation
The magnitude of the nature-based storm surge flood risk mitigation defined via
the length of the flood pathway passing through tidal wetlands shows a large
variability among deltas, with mean distances inside tidal wetlands ranging from
234 m to more than 2 km.
The comparison of the relative magnitude of nature-based flood risk mitigation is
based on the magnitude M ( (Equation 1) in terms of land surface area (Mland) and
Chapter 2
44
delta population (Mpop) (Table 2.5). The magnitude is higher when a larger
proportion of land area or population is buffered by a wider length of tidal
wetlands.
Table 2.5 Value of the relative magnitude of nature-based flood risk mitigation in term of land area (Mland) and population (Mpop) for every delta
Delta
Ranking Ranking
Mland Mland Mpop Mpop
Ganges-Brahmaputra delta 6.26 3 1.53 9
Yangtze delta 1.07 11 1.02 11
Mekong delta 2.70 8 2.41 5
Chao Phraya delta 3.10 7 1.89 6
Irrawaddy delta 3.62 6 1.62 8
Niger delta 8.09 2 5.31 1
Amazon delta 4.47 4 2.95 4
Rhine delta 1.37 10 1.12 10
Mississippi delta 9.26 1 1.86 7
Mahakam delta 2.53 9 3.07 3
Burdekin delta 4.37 5 4.42 2
The relative magnitudes for land area range from 1.07 to 9.26 and for the delta
population from 1.02 to 5.31. Except for the Mahakam and Burdekin deltas, all the
deltas present higher magnitudes in terms of land area than in terms of
population. However, the degree of deviation between Mland and Mpop varies among
the deltas (Table 2.5). The Mississippi, Ganges-Brahmaputra and Niger deltas
present the higher differences, while the Chao Phraya, Amazon and Irrawaddy
deltas have smaller differences and the Yangtze, Burdekin, Rhine, Mekong and
Mahakam present the lowest differences between Mland and Mpop (Figure 2.4).
Deltas
Figure 2.4 Relative magnitude of nature-based storm surge flood risk mitigation for every delta in terms of land area (left/blue) and of population (right/red)
Chapter 2
46
The comparison of the tidal wetlands area (km²) and the magnitude of flood risk
mitigation for the land area (Figure 2.5) shows a significant positive correlation
(Pearson’s r = 0.89, p = 0.0003). For an increasing tidal wetland area the
magnitude of flood risk mitigation for the land increases. The correlation between
the tidal wetlands area (km²) and the relative magnitude of flood risk mitigation
for the population is non-significant (Pearson’s r = 0.17, p = 0.61).
Figure 2.5 (a) Relation between the tidal wetlands surface area (km²) and the relative magnitude of nature-based flood risk mitigation for the terrestrial land area (Mland) (Pearson’s r = 0.88; p = 0.0003). (b) Relation between the tidal wetlands surface area (km²) and the relative magnitude of nature-based flood risk mitigation for the population (Mpop) (Pearson’s r = 0.17; p = 0.61)
2.4 Discussion
Although nature-based coastal risk mitigation is rapidly gaining interest in the face
of global change (Cheong et al., 2013; Giosan et al., 2014; Spalding, McIvor, et al.,
2014; Sutton-Grier et al., 2015; Temmerman et al., 2013), current insights into the
role of tidal wetlands for storm surge risk mitigation are mainly based on local or
regional scale assessments (Costanza et al., 2008; Das & Vincent, 2009; Krauss et
al., 2009; Stark et al., 2015; Wamsley et al., 2010; Zhang et al., 2012), and
methodologies for intermediate to global scale assessments are scarce. Our study
contributes to fill this gap by developing a GIS model assessing the impact of the
presence of tidal wetlands along the storm surge flood pathway in deltas around
the world. We applied the model on 11 deltas, selected based on their population
density, the presence of tidal wetlands and a worldwide distribution. Our results
indicate that tidal wetlands provide storm surge mitigation to large percentages of
the flood-exposed land area (> 80 %) for 4 of the 11 studied deltas, and to large
percentages of the flood-exposed population (> 70 %) for 3 of the deltas. The land
area and population buffered by tidal wetlands within a delta were not found to be
Deltas
47
correlated to the total wetland area in the delta, suggesting that more complex
factors are at play, such as the spatial distribution of population within the delta,
the delta geomorphology or the location of the tidal wetlands compared to the
location of the population. The magnitude of the nature-based storm surge risk
mitigation, estimated as the length of the storm surge pathway crossing tidal
wetlands, was found to be significantly correlated to the tidal wetlands area within
a delta. The latter finding highlights the importance of the conservation and, where
possible, restoration of extensive continuous tidal wetland areas as a nature-based
approach to mitigate storm surge flood risks in deltas.
Our quasi-global modelling approach differs from the existing complex
hydrodynamic models applied in local to regional studies, which incorporate
physical mechanisms of storm surge generation on the open sea, landward surge
propagation and attenuation of peak surge levels by friction exerted by the
landforms and vegetation of tidal wetlands and other coastal land use types
(Haddad et al., 2016; Liu et al., 2013; Marsooli et al., 2016; Resio & Westerink,
2008; Smolders et al., 2015; Stark et al., 2016; Wamsley et al., 2010; Zhang et al.,
2012). Such hydrodynamic modelling approaches are data-demanding and
computationally expensive, enabling their application on local to regional scales,
but excluding their feasibility for application to many deltas worldwide. In
contrast, our modelling approach is simple, computationally much less intensive,
and based on input datasets that are globally available. As a first step, we
demonstrated its applicability for 11 deltas, yet, the same approach is applicable to
much more deltaic or non-deltaic areas around the world. The release of new
global datasets is highly interesting, and such updated datasets should, when
possible, be used in the future applications of our modelling approach. For
example, there is the recently published global distribution of saltmarshes by
McOwen et al. (2017) and the global dataset on storm surge levels by Muis et al.
(2016).
The results presented highlight that our relatively simple method based on
globally available data of generally lower resolution than local data can provide an
estimation of the tidal wetland’s capacity of risk mitigation at a regional scale.
Nevertheless, despite the advantage of its intermediate to global applicability, our
model, unlike hydrodynamic models, does not include parameters such as the
complex geomorphology of the delta, the storm surge characteristics (such as
storm intensity, duration, direction, forward moving speed) or the wetland
characteristics (such as vegetation and geomorphic properties), making it unable
to accurately predict the flooding extent, depth or duration as a result of a specific
storm surge event.
Chapter 2
48
Hence, a direct comparison between hydrodynamic models and our GIS model is
not very informative as the methods represent different conditions in their
simulations. Hydrodynamic models are setup for specific storm surge
characteristics, including specific wind velocity fields, storm track, duration etc.,
with typical output being among others the reduction of peak storm surge height
due to the presence of tidal wetlands (Wamsley et al., 2010; Zhang et al., 2012).
Whereas our GIS model is setup for ‘statistical’ storm surge height of a given
return period, neglecting the specific storm surge characteristics mentioned
above, and aims to identify the land area and population that is exposed to flood
risks via flood pathways crossing tidal wetlands and that will benefit from reduced
flood risks due to the presence of the tidal wetlands. Nevertheless, a number of
qualitative conclusions derived from both modelling approaches are comparable,
as discussed below.
The analysis of the sensitivity of the model output to the variation of the input
values for attenuation rates for the land area (1 664.20 km² ± 12.80 %) and for the
tidal wetlands (i.e. mangrove forests) (1 812.60 km² ± 1.2 %) reveals that the
model is robust in regards to those parameters. The reason why the model output
(i.e. the land area flooded via flood pathways crossing through tidal wetlands) is
relatively insensitive to the wetland attenuation rates that are applied can be
ascribed to two points. First, a main point explaining this low variation of the land
area benefiting from storm surge mitigation by tidal wetlands is the continuity or
connectivity of the tidal wetlands. A tidal wetland area dissected by channels,
embayment and land areas will introduce a non-linear parameter in the reduction
of the surge, as it will not only be related to the attenuation rate of the wetlands,
but also to the pathway the flood can follow. Because a higher attenuation rate is
applied for the wetlands and land areas compared to the water areas, the flood
will preferably propagate over areas of lower friction, such as the channels and
water bodies. As a result, the area benefiting from flood risk mitigation by tidal
wetlands is relatively insensitive to the range of attenuation rates that were
applied in our sensitivity analysis (5 to 50 cm/km for the wetlands and 4 to 20
cm/km for the land areas), as channels in the delta plain will remain the main
pathways of flood propagation.
The second point is related to the different land use classes. In the current design
of the model, the land area is divided into two classes, the tidal wetlands and the
other land area, whilst, the division of the land area into an increased number of
classes (e.g. forests, urban areas, agricultural fields...) is expected to modify and
refine the pathway of the storm surge in the delta area.
Deltas
49
Our analysis shows that there is no correlation between the surface area of the
delta’s tidal wetlands and the land area benefiting from storm surge mitigation.
This is corroborating the fact that even small wetlands can provide flood wave
attenuation to large areas (Gedan et al., 2011). The comparison of the Ganges-
Brahmaputra and the Chao Phraya deltas further illustrates this. The Ganges-
Brahmaputra delta has a total surface area of tidal wetlands of 6 432 km² that
provides flood risk mitigation to 1 821 km² of the delta area. In comparison, the
Chao Phraya delta has 174 km² of tidal wetlands and 1 666 km² of land area
benefiting from nature-based flood risk mitigation. The non-correlation between
the wetlands surface area and the land area benefiting from storm surge
mitigation can also be related to the effect of the tidal wetlands location in the
delta and along the channels that is known to influence their capacity to mitigate
storm surges (Smolders et al., 2015; Stark et al., 2015). Following the previous
example of the Ganges-Brahmaputra and Chao Phraya deltas, the effect of the
location of the tidal wetlands in the delta can be observed. Tidal wetlands of the
Chao Phraya delta are more scattered and present along the main channels of the
delta (see maps of the deltas in the Supplementary Information), which implies
that they influence the propagation of the flood wave for a large part of the delta (1
666 km²). In contrast, tidal wetlands of the Ganges-Brahmaputra delta are
clustered, leaving some of the main channels, such as the ones running to Kolkata
or Dhaka for example, exempt of tidal wetlands. Hence, the large tidal wetland
area within the Ganges-Brahmaputra delta (i.e. the Sundarbans mangrove forest)
provides nature-based flood risk mitigation to only a certain part (1 821 km²) of
this large deltaic area.
Nevertheless, the analysis shows a significant, positive correlation between the
surface area of tidal wetlands and the magnitude of nature-based storm surge risk
mitigation (Pearson’s r = 0.88, p-value = 0.0003). This implies that a delta with a
large and/or continuous surface area occupied by tidal wetlands will benefit from
a higher magnitude of flood risk mitigation due to a longer width of tidal wetlands
crossed by the flood pathway. This finding relates to several hydrodynamic studies
that have identified the importance of wetland continuity for effective attenuation
of storm surges (Loder et al., 2009; McIvor et al., 2012; Phan et al., 2015; Zhang et
al., 2012).
In addition, studies have pointed out that tidal wetlands dissected by deep and
wide channels provide less storm surge attenuation than wetlands with narrow
and shallow channels (Loder et al., 2009; Stark et al., 2016). On the large scale of a
delta area, our results show similar effects of the dissection of the tidal wetlands
by deltaic channels. A qualitative observation of the studied deltas shows that the
Chapter 2
50
deltas having the lower relative magnitude of nature-based risk mitigation are also
the deltas having large channels dissecting the wetlands, as for example the Rhine
or the Yangtze deltas.
When a delta has a high value of land area benefiting from flood risk mitigation,
like the 86 % of the Chao Phraya delta or the 95 % of the Mahakam delta, this does
not necessarily imply a high magnitude of storm surge flood risk mitigation. The
Mland of the Chao Phraya and Mahakam deltas are 3.10 and 2.53 respectively,
almost three times lower than the magnitude of the Mississippi or the Niger deltas.
This means that although the Chao Phraya and Mahakam deltas have a high land
area benefiting from flood risk mitigation by tidal wetlands, the magnitude of the
mitigation is rather small compared to the magnitude estimated for the Mississippi
or Niger delta for which the flood pathway is crossing in general a longer width of
tidal wetlands.
Neither the absolute number of people buffered by tidal wetlands nor the relative
magnitude of nature-based storm surge mitigation in terms of population is
correlated to the total tidal wetlands area in the delta, implying that other
parameters must be considered. This may be attributed to factors such as the
spatial population distribution and density, or the historical settlement of the
population and their spatial relation to tidal wetlands. The Niger, Burdekin and
Mahakam deltas have the population benefitting from the highest magnitude of
storm surge risk mitigation in terms of population, with values of 5.31, 4.42 and
3.07 respectively. The tidal wetlands in those deltas are located in between the
population and the sea and are present over the full extent of the coastline. They
differ in population distribution, as the population of the Niger delta is distributed
over the delta plain, while the population of the Mahakam and Burdekin deltas is
concentrated in cities. However, the presence of tidal wetlands along the coastline
and along either side of the channels is influencing the propagation of the flood
pathway that crosses the tidal wetlands before reaching the population located
behind. Those deltas are in contrast to other deltas, such as the Yangtze and Rhine
deltas, where the channels are wider, with smaller tidal wetlands located only in
some regions of the delta; or to the Ganges-Brahmaputra delta, where the tidal
wetlands occupy a large portion of the delta, but a large part of the population is
not located directly behind the wetlands resulting in a lower magnitude in term of
population.
Ranking the deltas according to the relative magnitude of storm surge mitigation
in terms of land area, demonstrates that the deltas with the highest ranking are
deltas where large tidal wetlands exist and where historic wetland reclamation
Deltas
51
and conversion into human land use has been limited, such as in the Mississippi,
Niger and Ganges-Brahmaputra deltas (Table 2.5). In contrast, the deltas with the
lowest ranking have experienced large scale historical wetland reclamation, such
as in the Yangtze and Rhine delta. Hence, the difference in historical land use
management of the deltas induces differences in the future management. Deltas
with limited wetland reclamation and large remaining wetlands, such as the
Mississippi, Niger and Ganges-Brahmaputra delta would benefit from investing in
the conservation of their tidal wetlands as a nature-based strategy to mitigate
storm surge flood risks, while deltas with extensive historical wetland
reclamation, such as the Yangtze and Rhine deltas, should not only rely on hard
engineering of flood defence structures but also invest in restoration of formerly
reclaimed wetlands where possible.
2.5 Acknowledgements
The author would like to thank the different data providers and Dr. Chen Wang for
her help in the gathering of the Chinese wetlands data. This work was funded by
the University of Antwerp.
Chapter 2
52
Supplementary Information
Deltas
Ganges-Brahmaputra delta, India and Bangladesh
Figure SI 2.1 Left: Topography and location of the tidal wetlands in the Ganges-Brahmaputra delta in India and Bangladesh. Right: Ambient population in the delta
Yangtze delta, China
Figure SI 2.2 Left: Topography and location of the tidal wetlands in the Yangtze delta in China. Right: Ambient population in the delta
Deltas
53
Mekong delta, Vietnam
Figure SI 2.3 Left: Topography and location of the tidal wetlands in the Mekong delta in Vietnam. Right: Ambient population in the delta
Chao Phraya delta, Thailand
Figure SI 2.4 Left: Topography and location of the tidal wetlands in the Chao Praya delta in Thailand. Right: Ambient population in the delta
Chapter 2
54
Irrawaddy delta, Myanmar
Figure SI 2.5 Left: Topography and location of the tidal wetlands in the Irrawaddy delta in Myanmar. Right: Ambient population in the delta
Niger delta, Nigeria
Figure SI 2.6 Left: Topography and location of the tidal wetlands in the Niger delta in Nigeria. Right: Ambient population in the delta
Deltas
55
Amazon delta, Brazil
Figure SI 2.7 Left: Topography and location of the tidal wetlands in the Amazon delta in Brazil. Right: Ambient population in the delta
Rhine delta, The Netherlands
Figure SI 2.8 Left: Topography and location of the tidal wetlands in the Rhine delta in The Netherlands. Right: Ambient population in the delta
Chapter 2
56
Mississippi delta, United States of America
Figure SI 2.9 Left: Topography and location of the tidal wetlands in the Mississippi delta in the USA Right: Ambient population in the delta
Mahakam delta, Indonesia
Figure SI 2.10 Left: Topography and location of the tidal wetlands in the Mahakam delta in Indonesia. Right: Ambient population in the delta
Deltas
57
Burdekin delta, Australia
Figure SI 2.11 Left: Topography and location of the tidal wetlands in the Burdekin delta in Australia. Right: Ambient population in the delta
Chapter 2
58
DIVA dataset
The DIVA coastline is based on the coastline of the Digital Chart of the World
(DCW, Environmental Systems Research Institute, ESRI, 2002) and stores the
values of storm surge height for every segment of the coastline and for several
return periods.
In the study, the origin of the storm surge was defined as a flood source (red
line in Figure SI 2.12) and the value of the storm surge heights was transferred to
this flood source via a Euclidean distance algorithm. The Euclidean distance
algorithm delineated for every segment of the DIVA coastline its area of influence
(thin black lines in Figure SI 2.12). Then, the flood source was segmented and the
value of every segment corresponds to the value of the area of influence in which
the segment is located.
Figure SI 2.12 Representation of the DIVA coastline (blue), the flood source (red) and the areas of influence of every segment of the DIVA coastline (black lines).
Deltas
59
Results for the 1 in 10 and 1 in 1000 year storm surges
Table SI 2.1 Results of the analysis for the three storm surge return periods, 1 in 10, 100 and 1000 year. The population corresponds to the number of people benefiting from flood risk mitigation by tidal wetlands. The percentage is calculated relative to the total number of people flooded in the case of the scenario 1, i.e. with the presence of the tidal wetlands. The surface corresponds to the square kilometres of land area (excluding the tidal wetlands area itself) benefitting from flood risks mitigation by the tidal wetlands. The percentage of surface area is calculated relative to the land area (excluding the tidal wetlands area itself) flooded in the case of the scenario 1.
Delta Country Population (number of people) Population (relative %)
1 in 10 year
1 in 100 year
1 in 1000 year
1 in 10 year
1 in 100 year
1 in 1000 year
Ganges-Brahmaputra
India and Bangladesh
1,114,765 1,322,735 1,541,264 16.3 16.8 15.4
Yangtze China 4,438,443 5,922,009 7,322,102 49.1 49.1 49.0
Amazon Brazil 1,832.7 2,109.4 2,373.4 17.4 17.6 18.1
Rhine The Netherlands 986.5 1,043.6 1,093.2 12.7 12.8 12.9
Mississippi USA 3,467.6 4,334.5 5,051.6 84.7 82.8 80.2
Mahakam Indonesia 375.6 525.2 553.7 94.8 94.3 94.7
Burdekin Australia 65.9 70.7 89.0 55.7 53.4 49.9
Chapter 2
60
Sensitivity analysis
In order to compare the flood-exposed areas predicted by the GIS model and the
areas that can be indeed flooded during flood events, the flooding extend of two
cyclones were selected, the cyclone Sidr of November 2007 for the Ganges-
Brahmaputra and the cyclone Nargis of May 2008 for the Irrawaddy delta. The
result of this comparison shows that about 50 % of the areas that were flooded
during those cyclone events are flood-exposed areas predicted by the model
(Figure SI 2.13). The model is not designated to predict the areas that are flooded
during specific events, as those areas depends on factors such as the specific
cyclone properties like the direction of the cyclone or its speed of forward
propagation, the failure of flood protecting structures like dikes and dams present
in both deltas, or the rainfall driven fluvial floods. Nevertheless, it is able to
designate as flood-exposed areas those areas that experienced storm surge
flooding from specific cyclone events.
Figure SI 2.13 Comparison of the delta’s flood-exposed areas predicted by the GIS model and the areas flooded during previous cyclone events for (a) the Ganges-Brahmaputra delta with the cyclone Sidr (15/11/2007) and (b) the Irrawaddy delta with the cyclone Nargis (02/05/2008).
Global hotspots
61
CHAPTER 3 Identifying global hotspots for nature-based mitigation of coastal flood risks
Rebecca Van Coppenolle and Stijn Temmerman
Chapter 3
62
Abstract
Low-lying coastal zones are increasingly exposed to flood risks due to global
change including sea level rise, increasing storm intensity and growing coastal
population densities. Nature-based risk mitigation, by conservation or restoration
of ecosystems, such as tidal wetlands (salt marshes and mangroves) that have the
natural capacity to mitigate storm surge related flood risks, has been
demonstrated by local to regional-scale studies, but yet, we currently lack global-
scale assessments of where hotspots are located of large flood-exposed coastal
areas and populations that can receive nature-based risk mitigation from tidal
wetlands. Here we present the results of a global-scale GIS model assessing the
worldwide contribution of tidal wetlands to coastal flood risks mitigation. It
identifies the inland surface areas and population numbers receiving storm surge
mitigation by mangrove forests and salt marshes, and it quantifies the distance
travelled by a storm surge through tidal wetlands as a measure of the magnitude
of storm surge mitigation. Results show that on a worldwide scale, about 30 % of
the flood-exposed low-lying coastal plain benefit from nature-based storm surge
mitigation by tidal wetlands, with the largest areas located in deltas, estuaries and
lagoons (e.g. Mississippi delta, Elbe estuary, Venice Lagoon). About 40 % of the
global flood-exposed coastal population receives nature-based storm surge
mitigation. The majority of that population (80 %) is located in five countries, i.e.
China, Vietnam, the Netherlands, India and Indonesia. Areas more exposed to
extreme storm surges (Eastern America, Caribbean Sea, Eastern Asia) include
hotspot areas where storm surges are travelling through wider tidal wetlands
generating higher mitigation, as for example in the Mississippi delta, Chesapeake
bay, Ganges-Brahmaputra delta or Yangtze delta. Our global scale assessment aims
to increase general awareness on the high capacity of nature-based flood risk
mitigation, and to inspire further local scale analyses in support of the wider
application of nature-based risk mitigation as a sustainable strategy to mitigate
increasing coastal flood risks around the world.
Global hotspots
63
3.1 Introduction
Coastal areas are increasingly exposed to flood and erosion risks due to sea level
rise, increasing intensity of storms and cyclones (Hallegatte et al., 2013; Hinkel et
al., 2014; de Sherbinin et al., 2007; Vitousek et al., 2017), and subsidence by
human actions such as reduction of sediment supply by river dams or conversion
and drainage of coastal wetland ecosystems into human land use (Adam, 2002;
Auerbach et al., 2015; Balke & Friess, 2016; Gedan et al., 2011; Kirwan &
2008; Syvitski et al., 2009; Tessler et al., 2015; Thampanya et al., 2006).
In parallel, the coastal population will continue to grow, reaching globally
averaged densities of 405 to 534 people/km² by 2060 (or ten times the current
world’s average) (Guzmán et al., 2009; Kron, 2013; Mcgranahan et al., 2006;
Neumann et al., 2015), with more and more people concentrated in large coastal
cities (Von Glasow et al., 2013; Sengupta et al., 2018; United Nations, 2012),
increasing the number of people and assets exposed to coastal flood risks (Hanson
et al., 2011; Small & Nicholls, 2003).
The standard strategy for coastal protection is the construction of hard
engineering structures such as dams or dikes that protect the low-lying coastal
areas from the coastal flood and erosion risks (Adriana Gracia et al., 2018;
Pranzini, 2018; Rangel-Buitrago et al., 2018). However, those structures are more
and more challenged for their negative consequences on the natural environment
(disturbance of natural habitats, disturbance of sediment supply, accelerated
erosion at the bottom...) and the practical and financial difficulties to maintain
them in the face of projected climate and socio-economic changes. While nature-
based solutions, or combined hybrid solutions, are more and more regarded as a
sustainable, self-sufficient and cost-effective strategy to mitigate coastal flood and
erosion hazards (Adriana Gracia et al., 2018; Griggs, 2005; Temmerman et al.,
2013). Nature-based solutions are based on the conservation, restoration or
creation of coastal ecosystems, in particular mangrove forests and salt marshes
(further referred to as tidal wetlands), for their capacity to reduce the inland
propagation of storm surges, to reduce wind waves and shoreline erosion, and to
adapt to sea level rise by sedimentation (Costanza et al., 2008; Krauss et al., 2014;
McIvor, Möller, et al., 2012; McIvor, Spencer, et al., 2012; Shepard et al., 2011). In
the last decades, projects of nature-based coastal protection were developed in
several coastal areas around the world, as along the Mississippi delta plain
(Boesch et al., 2006; Coastal Wetlands Planning Protection and Restoration Act
(CWPPRA), n.d.; Day et al., 2007) or along the UK, Belgian and Dutch coastal plains
Chapter 3
64
and estuaries (R. A. Garbutt et al., 2006; Gardiner et al., 2007; Rupp-Armstrong &
Nicholls, 2007; SigmaPlan, 2017).
Tidal wetlands are increasingly recognized as having the capacity to attenuate
wind waves and storm surges, acting as a buffer in between the sea and the low-
lying coastal areas (Barbier et al., 2013; Costanza et al., 2008; Gedan et al., 2011;
Krauss et al., 2009; Mazda et al., 2006; Narayan et al., 2016, 2017; Wamsley et al.,
2010; Zhang et al., 2012). The mechanisms of storm surge reduction rely on the
friction exerted by the tidal wetlands’ geomorphology and vegetation on the water
column during the landward propagation of the surge (Barbier et al., 2013;
Costanza et al., 2008; Leonardi et al., 2018; Smolders et al., 2015; Stark et al.,
2016). Often expressed as a rate of storm surge height reduction per unit of
distance travelled through the tidal wetlands, the attenuation rates derived from
observations range from a couple of centimetres to 25 cm/km for salt marshes
(Krauss et al., 2009; Stark et al., 2015; Wamsley et al., 2010; Zhang et al., 2012),
and up to 50 cm/km for mangrove forests as reported by the hydrodynamic
modelling study of Zhang et al. (2012).
Existing studies on storm surge risk mitigation by tidal wetlands, discussed above,
mostly focus on local to regional scales and are mostly concentrated on specific
locations in the USA and to a lesser extent in Europe (Arkema et al., 2013; Das &
Vincent, 2009; Krauss et al., 2009; McGee et al., 2006; Stark et al., 2015), while
studies elsewhere in the world are much scarcer. As a consequence, until now
there is poor insight in the global scale possibilities for nature-based storm surge
mitigation by mangroves and salt marshes. From the upscaling of the GIS model
presented in Van Coppenolle et al. (2018) and based on globally available data, we
aim to identify the global hotspots of large flood-exposed coastal areas and
populations that can receive nature-based flood risk mitigation, defined as any
reduction of the storm surge due to its travelling through the existing mangrove
and salt marsh ecosystems. Such a global assessment should increase awareness
on the possibilities for nature-based risk mitigation and should stimulate further
developments in nature-based mitigation policies as a strategy against increasing
coastal flood risks.
Global hotspots
65
3.2 Methods
3.2.1 Datasets
The following datasets were used.
The values for the topography and the bathymetry are coming from the
General Bathymetric Chart of the Oceans (GEBCO) (British Oceanographic Data
Center, 2017) that represents a gridded bathymetry of the oceans coupled with the
NASA Shuttle Radar Topography Mission (NASA SRTM, NASA JPL, 2013) digital
elevation model of the continents. Both datasets have a resolution of 30 arc-
second. The SRTM dataset is found to be the best known global digital elevation
model (Rodriguez et al., 2006; Sun et al., 2003).
The worldwide distribution of the tidal wetlands was determined by the
Global distribution of Mangroves (Giri et al., 2011) and the Global distribution of
Saltmarshes (Mcowen et al., 2017) (USGS, www.unep-wcmc.org). The coastlines
delimiting the land and sea environment were defined by the combination of the
country boundaries as they exist in January 2015 (ESRI, DeLorme Publishing
Company, Inc., 2015), and the mangrove forests and salt marshes extent, as the
different dataset do not perfectly overlap (Lichter et al., 2011).
The storm surge heights for a 1 in 100 year return period accounted for
are coming from two datasets. The first one is the DINAS-COAST Extreme Sea
Level dataset from the Dynamic Interactive Vulnerability Assessment (DIVA)
database (Hinkel et al., 2014; Vafeidis et al., 2005). It corresponds to the storm
surge water level above mean sea level and is calculated by model simulations
based on tidal levels, barometric pressures, wind speeds and sea bed slopes for
return periods of 10, 100 and 1000 years. These storm surge water levels are
given for coastline segments, that correspond to the coastline of the Digital Chart
of the World (DCW, Environmental Systems Research Institute, ESRI, 2002)
divided in segments based on administrative and environmental parameters
(Vafeidis et al., 2005). The second storm surge height dataset is the Global Tide
and Surge Reanalysis (GTSR) (Muis, Verlaan, Winsemius, et al., 2016) that is also
based on the coastline segments of the Digital Chart of the World, as the DIVA data.
The GTSR corresponds to the near-coast global reanalysis of storm surges over the
period 1979-2014, with large validation of the results against observations. The
main difference between the two datasets is the static (DIVA) or dynamic (GTSR)
way of calculating the sea level extremes (Muis, Verlaan, Nicholls, et al., 2016). It
results in differences in the extremes, with an overestimation of the extremes by
Chapter 3
66
the DIVA data (mean bias of -0.19 m and mean absolute error of 0.23 m) and an
underestimation of the extremes by the GTSR dataset (mean bias of 0.55 m and
mean absolute error of 0.64 m) (Muis et al. 2016a; Muis et al. 2016b).
The population distribution originates from the LandScan 2013 Global
Population Database (Bright et al. 2013). It represents the population over a 30 arc
second grid resolution and integrates the diurnal movements and collective
travelling behaviour of the world population, i.e. the so-called “ambient
population”, averaged over 24 hours (Bright et al., 2013; Dobson et al., 2000).
The distribution of the historical tracks of the cyclones is coming from the
Global Cyclone Hazard and Frequency Distribution that is a compilation of more
than 1 600 storm tracks over the period of January 1980 to December 2000; the
wind speeds around the tracks have been modelled using the Holland’s model
(1997). The value of each cell corresponds to a decile ranking, a higher ranking
implies a greater frequency of the hazard relative to the other cells (Center for
Hazards and Risk Research - CHRR - Columbia University, Center for International
Earth Science Information Network - CIESIN - Columbia University, International
Bank for Reconstruction and Development - The World Bank, 2005; Dilley et al.,
2005)
Because there are local differences in the exact position of the coastlines between
the different datasets, e.g. tidal wetlands and land areas can locally appear on the
seaward side of the storm surge coastline segments and that the goal of the
analysis is to account for all the tidal wetlands and land areas that can influence
the landward propagation of the storm surge, the coastline segments of the storm
surge datasets were not directly used as the source of the flooding. Alternatively a
‘flood source line’ was defined via the creation of a buffer of 15 km around the
original storm surge coastline segments. Only the offshore limit of this buffer was
kept and defined as the ‘flood source line’. As such, the flood source line
corresponds to a simplified coastline 15 km offshore of the original storm surge
datasets coastline segments, to assure that all tidal wetlands and land areas that
influence the propagation of the storm surge are located on the landward side of
this flood source line. The storm surge heights of the different segments stored in
the DIVA and GTSR datasets are transferred to the flood source line segments of
various lengths (average length of 59.04 ± 87.03 km) with a shortest Euclidean
distance algorithm (for further information see Chapter 2).
The model has a resolution of 30 arc second in accordance with the original
resolution of the bathymetry (GEBCO), topography (SRTM) and population
(LandScan) datasets. In such, the other datasets, i.e. the countries boundaries, the
Global hotspots
67
salt marshes and the mangrove forests areas, were transformed to raster datasets
of 30 arc second resolution. This manipulation generated the loss of the smallest
salt marshes and mangrove forests areas (< 1 km²), however, due to the global
character of the model, those losses are unavoidable. Hence our model accounts
only for storm surge mitigation by wetlands patches larger than 1 km². We argue
that this is acceptable, as storm surge attenuation rates are up to 25 cm/km in salt
marshes and 50 cm/km in mangroves (e.g. Zhang et al. 2012; Mcivor et al. 2012;
Krauss et al. 2009; Wamsley et al. 2010; Lovelace 1994), hence less than 1 km
wide wetlands provide a relatively low degree of storm surge height reduction and
are not considered here.
The global datasets used present some limitations in regards to local data accuracy
and local data artefacts. Such limitations may include vegetation artefacts in the
elevation dataset that over-estimates the land elevation by one to several meters
(Rodriguez et al., 2006; Sun et al., 2003), or the moderate resolution of the tidal
wetlands datasets that can locally result in over- or under-estimations of the
surface area of the tidal wetlands (Giri et al., 2011; Mcowen et al., 2017). The
variable lengths of the DIVA coastline segment involve that some very local
characteristics of the coastal plain or some possible local increase or decrease in
storm surge height due to the geomorphology of the coast may not be accounted
for in the model.
3.2.2 Model
The model corresponds to the GIS procedure described in Van Coppenolle et al.
(2018), but applied on a worldwide scale, while in Van Coppenolle et al. (2018) it
was tested for 11 large deltas around the world. The model was developed in
ArcGIS (10.3.1) and Python (2.7). The model is similar to previously published
procedures that assess the coastal areas and number of people vulnerable to
storm surge flooding on regional to global scales (Arkema et al., 2013; Dasgupta et
al., 2011).
The model simulates how a storm surge flood wave would be routed from the
above-described flood source to the coastal plain, i.e. the land area below 10 m of
elevation (corresponding to the Low Elevation Coastal Zone and to the maximal
storm surge height of both datasets). The model resolution did not allow
accounting for flood protecting structures like dikes or dams, and hence it
evaluates the flood risks in case existing flood protecting structures would fail.
Four land covers are considered, i.e. (1) open water and channel areas, (2) salt
marshes, (3) mangrove forests and (4) remaining land area. For each of them, a
Chapter 3
68
storm surge attenuation rate, i.e. reduction of the surge height with distance the
surge has travelled over these land cover types (in cm/km), was defined based on
literature (Table 3.1) (for details, we refer to Van Coppenolle et al. 2018). The
storm surge propagation pathways were defined by the cost distance algorithms
that account for both the distance travelled and the friction generated by the land
covers. As such, every pixel of the coastal plain is associated with a given travelling
cost that is subsequently used in the cost distance algorithm, and corresponds to
the attenuation rate (Table 3.1) the storm surge will undergo by its propagation
from the flood source towards the pixel. The model is run for two scenarios: the
first scenario considers the current extent of the tidal wetlands, while, for the
second scenario, the tidal wetlands are considered as remaining land area. Finally,
the comparison of the two scenarios is used to determine the coastal areas and
populations that will be flooded via a storm surge flood wave that travelled
through tidal wetlands, namely, the areas where the storm surge attenuation is
higher in the case of the first scenario including wetlands.
Table 3.1 Attenuation rates attributed to the land covers considered in this study, in accordance with the model approach presented in Van Coppenolle et al. (2018).
Substrate Attenuation rate
(cm/km)
Tidal wetlands Mangrove Salt Marsh
10.0 8.0
Remaining land area 6.0 Open water and Channels 0.1
For every pixel located on the coastal floodplains, the length of the flood pathway
crossing through tidal wetlands was defined as an indication of the magnitude of
the storm surge mitigation by the tidal wetlands, as propagation through longer
distances of tidal wetlands are expected to generate a larger reduction of the
storm surge height. The distance travelled through tidal wetlands was calculated
by dividing the difference in storm surge height reduction between the two
scenarios by the difference of attenuation rates between the tidal wetland type,
either mangrove or salt marsh, and the remaining land (i.e. 10-6 cm/km, or 4
cm/km for mangrove forests and 2 cm/km for salt marshes). In the situation were
the two types of tidal wetlands were present, the value used for the division
corresponds to the averaged value, i.e. 3 cm/km.
From a comparative perspective, the simulations were made with the two storm
surge height datasets, the DIVA data and the GTSR data. The main discussion is
Global hotspots
69
based on the results of the DIVA data (as it shows the smallest error on predicted
extreme sea levels; see above and Muis et al (2016a,b)), while a comparison of the
results of the two datasets is presented in the Results section.
The model does not simulate the full complexity of atmospheric and
hydrodynamic processes involved in flood propagation and therefore is not able to
calculate accurate flood depths and absolute values of reduction in flood depth
behind tidal wetlands during specific storm surge events. Instead it calculates the
surface area and population numbers flooded via flood pathways crossing through
tidal wetlands, and it calculates the distance or length of the flood pathway
crossing through tidal wetlands, which are variables that are not dependent on
complex atmospheric and hydrodynamic processes during specific storm surge
events. As such, it has the major advantage to be globally applicable to compare
coastlines and coastal plains around the world.
For representation purposes, the model output will be presented on the coastline
segments defined by the Digital Chart of the World. A section of the coastal plain is
associated to each coastline segment via Euclidean distance. For each segment the
model output exists of (1) a value for the total surface area within the associated
coastal plain that is flooded via pathways crossing through tidal wetlands – further
called “area benefiting from storm surge mitigation”; (2) a value for the total
population number within the associated coastal plain that is flooded via pathways
crossing through tidal wetlands – further called “population benefiting from storm
surge mitigation”; (3) mean distance travelled by a storm surge through tidal
wetlands.
3.3 Results
3.3.1 Coastal Plain Areas Benefiting from Storm Surge Mitigation
The results show that for a 1-in-100 year storm surge event, without accounting
for any flood protecting structures and without the existing tidal wetlands
(scenario 2), 439 525 km² of the world’s coastal plain is exposed to storm surge
flood risks. However, when accounting for the currently existing tidal wetlands
(scenario 1), 135 911 km² (i.e. 31 % of the previous number) of the world’s coastal
plain benefits from a reduction in storm surge height as the storm surge pathway
passes through tidal wetlands (see Supplementary Information Figures SI 3.1, 3.2
and 3.5).
The locations having the largest coastal plain area benefiting from storm surge
mitigation by tidal wetlands (> 1 000 km² of the coastal plain associated to one
Chapter 3
70
segment, see Method section, segments are highlighted in red in Figure 3.1), are
mainly found in or close to deltas, estuaries and lagoons. Those hotspots include
the Northern part of the Yangtze delta, in front of the city of Yancheng (China),with
3 112.3 km² benefiting from storm surge mitigation associated to one segment of
the delta; the Northern part of the Elbe estuary (Germany), with one segment
having 2 215.2 km² of coastal plain benefiting from storm surge mitigation; the
Mississippi delta (USA) that has a segment with 2 050.4 km² benefiting from storm
surge mitigation; and the Laguna de Terminos in the Bahia de Campeche (Mexico)
that presents a segment with a surface area of 1 999.4 km² benefiting from storm
surge mitigation by tidal wetlands.
Not all coastline segments have the same length, and therefore, in order to
standardize the results, we also plotted what we call the standardized surface area
benefiting from storm surge mitigation, i.e. the absolute surface area divided by
the length of the associated coastline segment (km²/km) (Figure 3.2). The results
show that hotspots for storm surge mitigation by tidal wetlands per unit of
shoreline length also correspond to bays, lagoons, deltas and estuaries, yet, the
locations show some divergences with the absolute surface area benefiting from
mitigation. The hotspots are mainly located along the Northern European coasts
(France, Belgium, Netherlands, Germany and United Kingdom) and along the East
Asian coasts.
Global hotspots
Figure 3.1 Absolute surface area benefiting from a storm surge pathway crossing through tidal wetlands represented on the associated coastal segment (km²), with circles highlighting the segments for which the coastal plain area benefiting from storm surge mitigation is greater than 1 000 km².
Chapter 3
Figure 3.2 Standardized area, i.e. surface area per unit of shoreline length (km²/km), benefiting from a storm surge pathway crossing through tidal wetlands represented on the associated coastal segment, with circles highlighting the segments for which the surface area benefiting from storm surge mitigation is greater than 15 km² per 1 km of shoreline.
Global hotspots
73
3.3.2 Magnitude of Storm Surge Mitigation
Not only the surface area benefiting from a storm surge pathway crossing through
tidal wetlands, but particularly the distance travelled by the storm surge through
the tidal wetlands is a very relevant parameter determining the magnitude to
which wetlands can contribute to nature-based mitigation of storm surge flood
risks. Therefore the distance travelled by the storm surge through tidal wetlands
was defined for every pixel of the coastal plain and averaged over the areas
associated to each segment (Figure 3.3). The locations having the longest distance
(> 5 km, highlighted in red in Figure 3.3) travelled by the storm surge through
tidal wetlands are considered as having the highest degree of storm surge
mitigation. They are mainly found in areas where large tidal wetlands exist, again
in large deltas, such as in the Guayas delta in Ecuador, where parts of the coastal
plain benefit of storm surge mitigation by more than 12 km travelled through the
mangrove forests. The Kolyma delta in north-eastern Siberia presents large areas
of tidal wetlands (> 1 000 km²) that influence the storm surge propagation, as
some parts of the coastal plain can benefit from distances of more than 10 km of
salt marshes crossed by a storm surge. The coastal plains in the Mississippi delta
(USA), Chesapeake bay (USA), Ganges-Brahmaputra delta (India and Bangladesh)
or Tidung estuary in the North Kalimantan region of Borneo (Indonesia) also
benefit from more than 5 km travelled by a storm surge through tidal wetlands.
The comparison of the areas with a long distance of tidal wetlands along the storm
surge flood pathway (> 5 km) with the areas that are exposed to cyclone
conditions is presented on Figure 3.3. The value of the global cyclone hazard
distribution represents the frequency of the hazard relative to the other areas. Six
hotspots having a coastal plain area benefiting from storm surge mitigation by a
long distance (> 5 km) of tidal wetlands are also areas where the likelihood of
being exposed to cyclone hazards is greater than elsewhere (indicated in red
colours in Figure 3.3). In such, those areas whilst exposed to higher frequency of
cyclones could benefit from higher storm surge mitigation by the tidal wetlands.
When up-scaled to the country level, the results show that on the 114 countries
having tidal wetlands in their coastal plain, 101 benefit from flood risks mitigation
by the tidal wetlands (See Supplementary Information). The thirteen countries
that have tidal wetlands but no coastal area buffered by tidal wetlands are mainly
countries where the coastal plain is rapidly gaining in altitude, as in Equatorial
Guinea, where the 142 km² of mangrove forests are bordered by land areas with
an altitude rapidly reaching 5 m, while the 1 in 100 years storm surge is less than
2 meters high.
Chapter 3
Figure 3.3 Mean distance (m) travelled through tidal wetlands by a 1-in-100 years storm surge during its landward propagation. The circles highlight the segments for which the mean distance travelled through tidal wetlands by a storm surge is longer than 5 km, while red colours indicate hotspots were the long distance of wetlands coincides with high exposure to cyclones.
Global hotspots
75
3.3.3 Coastal Population Benefiting from Storm Surge Mitigation
Globally around 73.5 million people are exposed to coastal flood risks from a 1-in-
100 year storm surge in the case of scenario 2 (no protecting structures and no
tidal wetlands). When considering the current tidal wetlands (scenario 1), around
29.4 million people (i.e. 40 % of the previous number) benefit from nature-based
storm surge mitigation (See Supplementary Information Figures SI 3.3, 3.4 and
3.6). Hotspot areas (i.e. coastline segments with > 10 000 people benefiting from
storm surge risk mitigation) are predominantly located in large deltaic and coastal
lowland areas in Asia and Europe, such as in the Ganges-Brahmaputra delta (India
and Bangladesh), the Mekong delta (Vietnam), the Pearl and Red river deltas
(China), the Rhine-Meuse-Scheldt delta (Belgium and Netherlands) and the
Humber estuary (UK) (Figure 3.4). At a country level, the highest number of
people benefiting from storm surge mitigation by tidal wetlands is found in China
with 9.5 million people, followed by Vietnam (9.3 million people), India (2.8
million people), The Netherlands (1.7 million people) and Indonesia (1.2 million
people) (See Supplementary Information Figure SI 3.3). Those five countries
together make up for 83 % of the global population benefiting from nature-based
flood risks mitigation.
As for the coastal plain area, the population benefiting from storm surge
mitigation was also standardized by calculation per unit of shoreline segment
length (Figure 3.4). The hotspots where the population benefiting from storm
surge mitigation per 1 km of shoreline is the largest partly correspond to the
hotspots of the absolute number of people benefiting from storm surge mitigation
associated to one coastal segment, i.e. the Humber estuary (UK), the Weser estuary
(Germany) and the Rhine-Meuse-Scheldt delta (The Netherlands and Belgium) in
Europe, and most of the deltas and bays highlighted in Asia, at the exception of the
Ganges-Brahmaputra, Irrawaddy and Yangtze delta.
Chapter 3
Figure 3.4 Absolute number of people benefiting from a storm surge pathway crossing tidal wetlands represented on the associated coastal segment, with circles highlighting the segments for which the population benefiting from storm surge mitigation is higher than 100 000 people.
Global hotspots
Figure 3.5 Standardized population, i.e. people per unit of shoreline length (number of people/km), benefiting from a storm surge pathway crossing tidal wetlands, with circles highlighting the segments for which the population benefiting from storm surge mitigation is greater than 10 000 people per 1 km of shoreline.
Chapter 3
78
3.3.4 Comparison of the Storm Surge Height Datasets
In general, the GTSR sea level extremes have been found to underestimate storm
surge heights in comparison to the DIVA storm surge heights (See Supplementary
Information Figure SI 3.5) (Muis, Verlaan, Nicholls, et al., 2016; Muis, Verlaan,
Winsemius, et al., 2016). Whilst the correlation coefficient between the modelled
and observed sea level extremes is 0.70 for the DIVA database and 0.84 for the
GTSR database, the tropical cyclones in the GTSR data are under-represented, with
a bias that is larger in the tropical regions (Muis, Verlaan, Nicholls, et al., 2016).
Therefore, for the purpose of our analysis, we plotted the results from the DIVA
values in the main text of this paper, the results for the GTSR values are presented
in Supplementary Information.
Results from the simulations based on the GTSR dataset (Supplementary
Information Figures SI 3.8 and 3.9) show that for the 281 750 km² of coastal plain
exposed to 1-in-100 year storm surge flood risks, 80 307 km² are benefiting from a
storm surge pathway crossing tidal wetlands. Compared to the DIVA results
(Figure 3.1; resulting in 439 525 km² exposed to 1-in-100 year storm surges), the
GTSR dataset results in a smaller area exposed to coastal flood risks,
corresponding to an under-representation of the exposed areas by 36 %, which is
concordant with the difference observed between both datasets by Muis et al.
(2016). In regards to the hotspots of coastal plain areas benefiting from storm
surge mitigation (i.e. areas > 1 000 km²), fewer locations are identified based on
the GTSR dataset, yet, all GTSR hotspots are also identified as hotspots based on
the DIVA data.
The comparison of the mean distance travelled by the storm surge through tidal
wetlands before reaching the coastal plain also shows a fewer number of hotspots
(distance of more than 5 km) resulting from the GTSR data as compared to the
DIVA data. Most of the GTSR hotspots (Supplementary Information Figure SI 3.10)
are corresponding to the DIVA hotspots, while the Everglades in Florida and the
low-lying areas close to the Dee River in the United Kingdom are GTSR specific. In
the Everglades, the difference is relatively small between the two results, 500 m,
while for the Dee River, the GTSR storm surge height of 5.86 m (3.61 m for the
DIVA) results in a larger area identified as benefitting from nature-based storm
surge mitigation.
In terms of population that benefits from storm surge mitigation by the tidal
wetlands, over the 38.3 million people that are globally exposed to coastal flood
risks, 13.5 million can benefit from storm surge mitigation by the tidal wetlands,
Global hotspots
79
based on the GTSR data (Supplementary Information Figures SI 3.11 and 3.12). As
for the DIVA results, the highest absolute numbers of people benefiting from storm
surge mitigation are located along the Belgian-Dutch-German and East-Asian
coastlines.
The comparison of the results shows that the differences in the incoming storm
surge height between the two datasets is influencing the coastal areas and
populations benefiting from storm surge mitigation by the tidal wetlands.
Nonetheless, the general trends observed throughout the results are similar.
3.4 Discussion and Conclusion
In the face of global climate change and the associated increasing risks of coastal
flooding from more severe storm surges and expected sea level rise (Bengtsson et
al., 2006; Hallegatte et al., 2013; Hinkel et al., 2014; IPCC, 2013; Knutson et al.,
2010; Webster et al., 2005; Woodruff et al., 2013), the conservation of tidal
wetlands can contribute to the mitigation of coastal flood risks by their ability to
attenuate storm surges, reduce the impact of waves and shoreline erosion, and
accumulate sediments in balance with sea level rise (Kirwan et al., 2016; Krauss et
al., 2014; Lovelock et al., 2015; Sandi et al., 2018). As such, nature-based risk
mitigation can reduce the threats to flood-exposed coastal areas and populations
(Cheong et al., 2013; Costanza et al., 2008; Duarte et al., 2013; Sutton-Grier et al.,
2015; Temmerman et al., 2013). Current assessments on the role of tidal wetlands
for coastal flood risk mitigation are based on in situ observations (Das & Vincent,
2009; Krauss et al., 2009; McGee et al., 2006; Stark et al., 2015) and/or on
modelling studies (Arkema et al., 2013; Stark et al., 2016; Zhang et al., 2012) at
local to regional scales. Such site-specific studies have substantially advanced our
understanding of the mechanisms determining the rate of storm surge mitigation
by salt marshes and mangroves (i.e. how much the peak water level is reduced per
distance that the storm surge has travelled through salt marshes or mangroves).
However, we currently lack a global scale assessment of the possibilities for
nature-based mitigation of coastal flood risks. That is, the location of global
hotspots of large flood-exposed coastal areas and populations that can receive
nature-based risk mitigation from existing salt marsh and mangrove ecosystems.
Our study is to our knowledge the first worldwide assessment identifying such
global hotspots and quantifying the coastal area and number of people that can
benefit from nature-based coastal flood risk mitigation.
The presented model allows a global-scale assessment of the locations where tidal
wetlands are expected to play a role in the mitigation of storm surge flood risks,
Chapter 3
80
but it does not accurately define the intensity of the storm surge mitigation nor the
flooding depth, extent or duration resulting from a specific storm surge event.
Furthermore, our model analysis does not account for the socio-economic
limitations that may hamper the applicability of nature-based strategies. To
precisely evaluate the attenuation of a specific storm surge due to the presence of
a specific tidal wetland area, a hydrodynamic modelling approach would be
needed based on high resolution input data regarding the storm characteristics
(duration, intensity, track, wind velocity field...), the wetlands’ vegetation and
geomorphology (vegetation type, density and continuity, soil surface
topography...) as well as the geomorphology of the surrounding coastal area (off-
shore bathymetry, shoreline shape, flood protection structures...) (Leonardi et al.,
2018; Marsooli et al., 2016; Resio & Westerink, 2008; Smolders et al., 2015; Stark
et al., 2016; Temmerman et al., 2012, 2013). However, the high computational
demand of such a hydrodynamic modelling approach does not enable a worldwide
assessment, while the more simple approach of our model and the use of globally
available datasets enable its worldwide applicability with a much lower
computational demand than hydrodynamic models (Van Coppenolle et al., 2018).
The results show that about one third of the global flood-exposed coastal plains
(31 % for the DIVA data and 28.5 % for the GTSR data) and almost 40 % of the
global flood-exposed population (40 % for the DIVA data and 34.5 % for the GTSR
data) experience nature-based storm surge mitigation by existing mangrove
forests or salt marshes. The coastal plains with the largest absolute surfaces
benefiting from storm surge mitigation are mainly found in deltas (e.g. Mississippi
2017). In other hotspots, policy and decision-makers are only beginning to
account for the coastal protection value of tidal wetlands and nature-based
strategies are starting to be implemented along with classical hard engineering, as
in the Ganges-Brahmaputra delta, the Yangtze delta or the Mekong delta (Käkönen,
2008; Seavitt, 2013; Ysebaert et al., 2017). Nevertheless, the application of nature-
based strategies, and then the conservation and maintenance of the coastal
ecosystems, is highly related to the human land claim in coastal zones. As such, in
certain areas over the world, the implementation of those nature-based strategies
would be less realistic, like along the China’s coasts, where despite the local
implementation of nature-based strategies, large natural coastal areas are still
expected to be turned into human land use (Ma et al., 2014; Meng et al., 2017;
Wang et al., 2014).
Creating sustainable and cost-effective coastal protection strategies to adapt to the
increasing coastal flood risks would require policy makers to account for the
presence of tidal wetlands, and coastal ecosystems in general, in the design and
development of coastal protection structures and renounce to the practice of
large-scale tidal wetlands reclamation (Duke et al., 2007; McLeod et al., 2011;
Chapter 3
82
Pendleton et al., 2012; Spalding et al., 1997; Valiela et al., 2001, 2009). Indeed, salt
marshes and mangroves have been converted on large scales to human land use
such as for agriculture, aquaculture, industry and urbanization, and this is still
actively going on in several places around the world, especially in fast developing
regions such as in Southeast Asia (An et al., 2007; Jiang et al., 2015; Tian et al.,
2016; Wang et al., 2014). Apart from the fact that such large-scale tidal wetland
reclamation implies the loss of biodiversity and valuable ecosystem services, we
argue that wetland reclamation should be planned carefully and avoided as much
as possible in order to maximize the flood risk mitigation function of remaining
tidal wetlands. Additionally, where possible restoration or creation of tidal
wetlands should be considered to enhance the nature-based adaptation capacity of
the coastal communities exposed to coastal flood risks. Our global-scale
assessment aims to increase the awareness on the value of nature-based coastal
protection strategies and stimulate further site-specific studies as a first step
towards a more worldwide implementation of nature-based mitigation policies as
a strategy against increasing coastal flood risks.
Global hotspots
Supplementary Information
Results at the country level and percentages
Figure SI 3.1 Surface area (km²) of the flood-exposed coastal plain that benefits from storm surge pathway passing through tidal wetlands at a country level
Chapter 3
Figure SI 3.2 Percentage (%) of the flood-exposed coastal plain benefiting from a storm surge pathway passing through tidal wetlands at a country level
Global hotspots
Figure SI 3.3 Number of people in the flood-exposed coastal plain that benefits from storm surge pathway passing through tidal wetlands at a country level
Chapter 3
Figure SI 3.4 Percentage (%) of the flood-exposed population benefiting from a storm surge pathway passing through tidal wetlands at a country level
Global hotspots
Figure SI 3.5 Percentage (%) of the flood-exposed coastal plain associated to each segment benefiting from storm surge mitigation by tidal wetlands
Chapter 3
Figure SI 3.6 Percentage (%) of the flood-exposed population associated to each segment benefiting from storm surge mitigation by tidal wetlands
Global hotspots
89
Results for the DIVA dataset
Table SI 3.1 Values of the surface area and type of tidal wetlands as well as of the different variables related to storm surge risk mitigation based on a 1-in-100 year storm surge levels from the DIVA dataset at a country level for the countries having tidal wetlands along their coastline. The percentages of the fifth and seventh columns correspond to the percentage of the coastal plain or population benefiting from flood risk mitigation by tidal wetlands relative to the total flood-exposed coastal plain area or population.
Figure SI 3.7 Maps representing the differences between the DINAS-COAST Extreme Sea Levels (DCESL, called DIVA here) and the Global Tide and Surge Reanalysis (GTSR) extremes for a return period of 100 years: (a) the sea level height for the DCESL/DIVA data relative to mean sea level; (b) the sea level height for the GTSR data relative to mean sea level; (c) the difference between DCELS/DIVA and GTSR; (d) whether the DCELS/DIVA extremes are within the 5 and 95 % confidence bounds of the fitted Gumbel distribution of the GTSR extremes. From (Muis, Verlaan, Nicholls, et al., 2016)
Chapter 3
94
Results for the GTSR dataset
Table SI 3.2 Values of the surface area and type of tidal wetlands as well as of the different variables related to storm surge risk mitigation based on a 1-in-100 year storm surge levels from the GTSR dataset at a country level for the countries having tidal wetlands along their coastline. The percentages of the fifth and seventh columns correspond to the percentage of the coastal plain or population benefiting from flood risk mitigation by tidal wetlands relative to the flood-exposed coastal plain area or population.
Figure SI 3.8 GTSR results of the absolute surface area benefiting from a storm surge pathway crossing through tidal wetlands represented on the associated coastal segment (km²), with circles highlighting the segments for which the coastal plain area benefiting from storm surge mitigation is greater than 1 000 km².
Global hotspots
Figure SI 3.9 GTSR results of the standardized area, i.e. surface area per unit of shoreline length (km²/km), benefiting from a storm surge pathway crossing through tidal wetlands represented on the associated coastal segment, with circles highlighting the segments for which the surface area benefiting from storm surge mitigation is greater than 15 km² per 1 km of shoreline.
Chapter 3
Figure SI 3.10 GTSR results of the mean distance (m) travelled through tidal wetlands by the storm surge during its landward propagation. The circles highlight the segments for which the mean distance travelled through tidal wetlands by the storm surge is longer than 5 km, in red are the hotspots were the long distance of wetlands coincides with high exposure to cyclones.
Global hotspots
Figure SI 3.11 GTSR results of the absolute number of people benefiting from a storm surge pathway crossing tidal wetlands represented on the associated coastal segment, with circles highlighting the segments for which the population benefiting from storm surge mitigation is higher than 100 000 people.
Chapter 3
Figure SI 3.12 GTSR results of the standardized population, i.e. people per unit of shoreline length (number of people/km), benefiting from a storm surge pathway crossing tidal wetlands, with circles highlighting the segments for which the population benefiting from storm surge mitigation is greater than 10 000 people per 1 km of shoreline.
Cities
103
CHAPTER 4 Potential for nature-based flood risk mitigation in coastal cities around the world
Rebecca Van Coppenolle and Stijn Temmerman
Based on the paper submitted to Earth’s Future in July 2018
Chapter 4
104
Abstract
Nature-based risk mitigation is increasingly proposed as a strategy to cope with
global changes that increase flood risks in coastal areas. However, loss of coastal
ecosystems reduces their mitigating effect on coastal flood risks in many places
around the world. Here we identify global urban hotspots exposed to storm surge
flood risks, where conservation of existing coastal ecosystems can contribute to
nature-based risk mitigation. We present a global procedure identifying the most
likely pathways followed by storm surges from the open sea towards 136 cities
around the world, and quantifying the extent of mangrove forests, salt marshes,
seagrass meadows and/or coral reefs along these storm surge pathways. Cities
that combine large flood-exposed populations (> 400 000 people exposed to 1-in-
100 years storm events) and large potential for nature-based risk mitigation (>
200 km² of coastal ecosystems) are located in large river deltas and estuaries, such
as Khulna (Ganges-Brahmaputra delta in Bangladesh), Guayaquil (Guayas delta in
Ecuador), Ho Chi Minh City (Mekong delta in Vietnam) and New Orleans
(Mississippi delta in USA). Here conservation of mangroves and salt marshes plays
a key role. Cities with large populations and/or assets at risk, but few ecosystems,
are either located directly adjacent to coastlines, or where former wetlands have
been reclaimed, especially in European and Asian cities. Overall, an encouraging
75 % of the studied cities benefit from present ecosystems. Hence our study calls
for conservation and (re-)creation of coastal ecosystems as a sustainable strategy
for nature-based mitigation of increasing coastal flood risks.
Cities
105
4.1 Introduction
Coastal populations are exposed to increasing natural hazards due to socio-
economic and climatic changes (Barbier, 2014; Syvitski et al., 2009). Global climate
change causes accelerating sea level rise and increasing frequency of high intensity
storms, and as such increases the risks of coastal flooding and erosion by storm
surges and wind waves (Bengtsson et al., 2006; Knutson et al., 2010; Webster et
al., 2005; Woodruff et al., 2013). In addition, socio-economic developments are
leading to an increasing coastal population density and increasing value of assets,
in particular in coastal cities (Green & Short, 2003; Hallegatte et al., 2013;
Mcgranahan et al., 2006; Neumann et al., 2015; Nicholls et al., 2008; de Sherbinin
et al., 2007). Human activities in the coastal zone, such as the conversion of natural
wetlands into agricultural, industrial or urban areas, may interfere with local flows
of water and sediments, may cause land subsidence, and as such further aggravate
coastal flood and erosion risks (Adam, 2002; Auerbach et al., 2015; Kirwan &
Megonigal, 2013; Pethick & Orford, 2013; Syvitski et al., 2009). All together these
socio-economic and climatic changes highlight the need for sustainable protection
of coastal societies, and especially densely populated coastal cities, against
increasing flood and erosion risks.
The building of protective structures, such as dikes, levees and dams, is commonly
considered the standard solution to protect against coastal flood and erosion risks.
Although such coastal protection structures are implemented in several coastal
zones around the world, a high share of coastal cities remain highly vulnerable to
coastal flooding due to flood defences with relatively low safety standards (e.g.
protection against a 1 in 100 year event instead of 1 in 1 000 year event), or non-
reliable or absent coastal protection structures (Dasgupta et al., 2009; Green &
Short, 2003; Mcgranahan et al., 2006; Nicholls et al., 2008; de Sherbinin et al.,
2007). While the implementation of coastal protection is necessary, it is not solely
linked to the richness of the country, but the decision to build effective defences
against coastal flooding also strongly depends on the agenda of policy makers
(Nicholls et al., 2008).
Nature-based solutions for coastal flood and erosion risk reduction rely on the
natural ability of coastal ecosystems to mitigate flood and erosion risks (Cheong et
al., 2013; Gedan et al., 2011; Sutton-Grier et al., 2015; Temmerman et al., 2013),
while at the same time providing additional valuable ecosystem services as carbon
sequestration, contribution to fisheries production and water quality regulation
(Barbier et al., 2011; McLeod et al., 2011). The flood and erosion risk mitigation
capacities of natural coastal ecosystems were demonstrated in several studies,
Chapter 4
106
mainly for mangrove forests, salt marshes, seagrass meadows and coral reefs
(Guannel et al., 2016; Koch et al., 2009; Narayan et al., 2016). These coastal
ecosystems provide a barrier against landward propagation of wind waves
(Fonseca & Cahalan, 1992; Gedan et al., 2011; McIvor et al., 2012) and storm
surges (Ferrario et al., 2014; Guannel et al., 2016; McIvor, Spencer, et al., 2012;
Möller et al., 2014), and therefore reduce the risks of shoreline erosion and
flooding. Additionally, they are able to adapt to the rising sea level by natural
processes of sediment accretion (Buddemeier & Smith, 1988; Lovelock et al., 2015;
J. T. Morris et al., 2002), although this adaptation ability depends on local
conditions such as the rate of relative sea level rise, the tidal range, the availability
of sediments and nutrients (Alongi, 2008; Kirwan et al., 2010, 2016; Simas et al.,
2001). The magnitude of the flood risk mitigation, measured as the amount of
wave or storm surge height reduction per distance travelled through the
ecosystem, is driven by a multitude of parameters such as the phenological and
morphological traits of the species present in the coastal ecosystems, the
geomorphology of the ecosystem and of the wider surrounding coastal area, and
the characteristics of the wind waves or storm surges (Gedan et al., 2011; Koch et
al., 2009; Mazda et al., 2006). As a consequence, a unique quantitative value of the
rate of wind wave or storm surge height reduction by the different ecosystems
cannot be defined. Nevertheless, ranges of values of wave height and storm surge
height reduction are known from local studies.
Tropical mangrove forests have been shown to mitigate wind waves, storm surges
and to a limited extent small tsunamis (Alongi, 2008; Krauss et al., 2009; Mazda et
al., 2006; McIvor, Möller, et al., 2012; McIvor, Spencer, et al., 2012; Zhang et al.,
2012). Recorded values show that 500 m of mangrove forest can reduce small
wind waves (< 70 cm high) by 50 to 99 % (McIvor, Möller, et al., 2012), while
storm surge height reduction ranges from several centimetres to 50 cm per
kilometre travelled by the storm surge through the mangrove forests (Gedan et al.,
2011; McIvor, Spencer, et al., 2012; Zhang et al., 2012). The lower vegetation of
salt marshes is generally providing less friction than mangrove trees.
Nevertheless, salt marshes can reduce wind waves up to 80 % over several tens of
meters (Moller et al., 1999; Ysebaert et al., 2011) and can lower storm wind waves
by 60 % over 40 m (Möller et al., 2014). Salt marshes can also attenuate the height
of storm surges at rates of 1.7 to 25 cm per kilometre travelled through the marsh
(Leonardi et al., 2018; Shepard et al., 2011; Stark et al., 2015; Wamsley et al.,
2010). Seagrass meadows are expected to reduce the flood wave energy with a
magnitude comparable to salt marshes under wind wave and storm conditions
(Duarte et al., 2013; Fonseca & Cahalan, 1992), with the maximum reduction in
shallow water and low wave energy environments (Fonseca & Cahalan, 1992;
Cities
107
Ondiviela et al., 2014). Coral reefs can provide wave energy attenuation similar to
artificial defences (Ferrario et al., 2014), by reducing in average the wave energy
by 97 %, with most of the reduction provided by the reef crest (Ferrario et al.,
2014; Principe et al., 2012; UNEP-WCMC, 2006). While most of the studies on
wave attenuation by coastal ecosystems are focusing on moderate wind waves, the
evidence of wave mitigation under storm conditions is increasing (Guannel et al.,
2016; McIvor, Spencer, et al., 2012; Narayan et al., 2016; Stark et al., 2015).
As the flood and erosion defence function of natural ecosystems becomes
increasingly demonstrated, nature-based coastal defence is starting to be
implemented in a growing number of coastal areas, often as an add-on to
traditional engineered defence structures (Barbier et al., 2008; Costanza et al.,
2008; Gedan et al., 2011; McGranahan et al., 2007; Spalding, Ruffo, et al., 2014;
Temmerman et al., 2013). The conservation and/or restoration or creation of
ecosystem buffers between the sea and populations at risk, is often a more cost-
efficient strategy than only classic engineering solutions, as coastal ecosystems are
self-adaptive to sea level rise by sediment accretion (Alongi, 2008; Kirwan et al.,
2016; Simas et al., 2001) and hence need less maintenance than engineered
defence structures. Some observations by the National Park Service in the US
based on real projects show that the installation of engineering infrastructures
(e.g. seawalls, dikes, breakwaters...) cost usually from 6 500 to 9 800 $ per linear
meter, while nature-based projects usually have a cost of installation ranging from
0 (ecosystems already present) to 6 600 $ per meter (Beavers et al., 2016; Sutton-
Grier et al., 2018). The nature-based projects are also cost-efficient in terms of
maintenance and reparation, with observations of 0 to 328 $ per meter, in
comparison to the maintenance and repair costs of the engineering infrastructures
ranging from 0 to 1 710 $ per meter (Beavers et al., 2016). Furthermore, natural
ecosystems provide additional societal benefits through ecosystem services. This
makes nature-based solutions particularly relevant for areas with lower financial
resources. Yet, existing coastal flood protection programs that include nature-
based approaches are still relatively scarce on a global scale. Examples include the
large-scale restoration of marshes in the Mississippi deltaic plain with the
objective to reduce landward propagation of storm surges, in combination with
engineered flood defences (Coastal Wetlands Planning Protection and Restoration
Act (CWPPRA), n.d.; Day et al., 2007), or the projects of ‘managed coastal
realignment’, which is the landward relocation of flood defence structures to
accommodate space for coastal ecosystem development, and which is applied for
example in the UK and elsewhere in Europe (R. A. Garbutt et al., 2006; Gardiner et
analyst-toolbox/how-the-cost-distance-tools-work.htm) were used to determine
the most likely pathway followed by a storm surge from the open sea to the city
centre. For the cost distance algorithms, the land areas were given a cost of 1000,
while the water bodies were given a cost of 1. As such the probable flood pathways
between the open sea and the city centre were defined for every city. A few
examples are presented in the Supplementary Information.
The third step defined the likely area influencing the propagation of the
storm surge. It assumes that a buffer area of 20 km around the probable storm
surge pathway is including both the potential variations in the storm surge’s
pathways and the area in which the presence of ecosystems (mangroves, salt
marshes, seagrasses, and coral reefs) can have an effect on the reduction of wave
and surge heights, and of erosion risks. With this procedure the buffer also extends
20 km offshore on the ‘open sea’ (see examples in the Supplementary Information)
and as such can also include the offshore ecosystem types (seagrasses and coral
reefs). These offshore ecosystems also contribute to wave, surge and erosion risk
reduction for the urban populations at risk, while far offshore ecosystems located
outside this 20 km buffer are not considered relevant anymore for risk reduction
in the city. Different values of buffer areas were tested (10, 20 and 30 km) to
evaluate their impact on the surface area of coastal ecosystems per city. The
surface area of coastal ecosystems inside the three buffer sizes (10, 20 and 30 km)
is presented in the Supplementary Information.
Chapter 4
110
The comparison of the coastal ecosystems surface area in the three buffer
size shows that for the cities that have no coastal ecosystems in the 10 km buffer
(i.e. 39 cities) four cities, (i.e. Xiamen, Yantai, Rangoon and Rotterdam) have a
presence of coastal ecosystems in their 20 km buffer area, and 6 more cities have a
presence of coastal ecosystems in the 30 km buffer (i.e. Qingdao, Athens, Kolkata,
Nagoya, Tokyo and Benghazi). For the cities having coastal ecosystems in the 10
km buffer area (i.e. 97 cities), 60 % have an increase of coastal ecosystems surface
area smaller than 60 km² in the 30 km buffer area, the other 40 % of those cities
have a higher increase in surface area. However, the difference in surface areas of
coastal ecosystems for the different cities and for the different buffer size remains
relative. As such, the comparison of the coastal ecosystems area between the cities
remains similar in the case of the use of the three buffer sizes.
The fourth step of the analysis was to determine the surface area of the
four coastal ecosystems accounted for inside the area influencing the propagation
of the storm surge. The spatial distribution of the different coastal ecosystems is
based on global scale datasets: the Global distribution of Mangroves (Giri et al.,
2011), the Global distribution of Saltmarshes (Mcowen et al., 2017), the World Atlas
of Seagrasses (Green & Short, 2003) and the World Atlas of Coral Reefs (Spalding et
al., 2002) from the United Nations Environmental Program – World Conservation
Monitoring Centre (www.unep-wcmc.org).
The second part of the study defined the surface area inside the 20 km buffer
around the probable flood pathway that probably consisted of coastal wetlands in
historic time and that was reclaimed for human land use throughout history. This
estimation of historically reclaimed coastal wetlands was used then to test if it can
explain the current distribution of coastal wetlands in between cities and the open
sea. As such, this part of the analysis is focussing specifically on two of the
considered ecosystem types, mangroves and salt marshes, that were in many
places around the world converted into human land use such as agricultural fields,
aquaculture ponds, industrial or urban areas - called here generally ‘land
reclamation’. This estimation of the surface area of land reclamation was made by
selecting for every city the land area that sits below the mean high tide, and
therefore is expected to be intertidal, but is currently not covered by mangroves or
marshes. The absence of mangroves and marshes in those areas is interpreted as a
historical conversion of former wetlands (mangroves, marshes) into human land
use.
The mean high tide is obtained from mean sea level augmented by the tidal
amplitude. The datasets used in this step of the analysis are the NASA Shuttle
Cities
111
Radar Topography Mission v3.0 at a resolution of 1 arc-second and the Finite
Element Solution (2012) – Global Tide from AVISO that contains the information
on the tidal amplitude. The Principal Lunar semi-diurnal component (M2) was
used to define the average tidal amplitude in front of every city. Due to data
limitations, the city of Helsinki (Finland) could not be included in this step of the
analysis.
The third part of our analysis aims to explore which parameters can explain the
presence and the size of the coastal ecosystems in front of the studied cities. For
this, two statistical analyses (logistic and linear regressions) were performed on a
set of social and physical parameters (Table 4.1). Except for the country’s GDP per
capita (http://databank.worldbank.org/data/home.aspx), the parameters are
obtained through the above-mentioned datasets and analysis in ArcGIS (10.3.1).
Except where mentioned, all the values are extracted for the probable area
influencing the propagation of the storm surge (i.e. the 20 km buffer around the
probable flood pathway).
Table 4.1 Physical and Social parameters extracted from the data for each city
Physical Parameters Social Parameters Type Unit Type unit
Latitude of the city Degree Short-distance population density (within 20 km around the flood pathway)
Inhabitants/km²
Distance between the sea and city
km Intermediate-distance population density (within 50 km around the flood pathway)
Inhabitants/km²
Coastline length km Long-distance population density (within 100 km around the flood pathway)
Inhabitants/km²
Area below mean high tide km² Country’s GDP per capita Constant 2005 US$
Shallow water area (> -100 m)
km²
Deep water area (< -100 m)
km²
Through logistic regression including all the studied cities and focussing on one
ecosystem at the time, each of the social and physical parameters (i.e. candidate
explanatory variables) was tested for association with the presence of the
ecosystem type (dependent variable). To do so, the cities were split in two groups
according to the presence or absence of the considered coastal habitat. The results
of the logistic regression consists of odds ratios (and their corresponding p-value)
that quantify the increase in odds of finding a coastal ecosystem for a specific
increase in the value of the explanatory variable.
Chapter 4
112
The influence of the candidate explanatory variables (social and physical
parameters) on the surface area of the coastal ecosystems (dependent variable)
was tested through linear regression. In this analysis, only the cities in which the
coastal ecosystem under study is present were retained. One assumption of a
linear regression model is the normality of the residuals. As deviations from
normality were observed in the data, the model was refitted and the dependent
variable used is the natural logarithm of the surface area of the ecosystem. As
such, the residuals showed a normal distribution. The results obtained from the
linear regression correspond to the regression coefficients of the slope, the
standard error of the coefficient and the associated p-value.
Finally the last part of the study focussed on a comparison of the surface area of
the four coastal ecosystems, the population at risk and the assets at risks in order
to identify the cities with both large population and/or assets at risks of flooding
and large coastal ecosystem areas. The population and assets at risk of coastal
flooding in every city were taken from the results of Nicholls et al. (2007) and
correspond to the population and assets exposed to coastal flooding due to storm
surge and high wind damages for a 1 in 100 years storm surge, without any
consideration of coastal defences or adaptations.
4.3 Results and Discussion
4.3.1 Coastal Ecosystems Areas
From the 136 cities studied, 101 cities have at least one natural coastal ecosystem
along their likely flood pathway, while 35 cities have a complete absence of coastal
ecosystems. The majority of cities (52) have only one ecosystem type along their
likely flood pathway, 39 cities benefit from the presence of two coastal ecosystem
types, while seven and three cities have respectively three and four ecosystem
types along their likely flood pathway (see Figure 4.1). The more frequent
combinations of ecosystem types are salt marshes and seagrass meadows, in front
of North American, European and Australian cities, while the combination of
mangrove forests and coral reefs is present in front of tropical region’s cities.
In total, the most represented coastal ecosystem is seagrass meadows with 5 195
km² followed by 4 890 km² of mangrove forests and 1 974 km² of salt marshes.
Coral reefs are much less present with a total of 282 km², partially due to their
more offshore location in comparison to the three other ecosystems.
The total surface area of natural coastal ecosystems varies greatly between the
different cities, with the highest surfaces being the 2 006 km² of mangrove forest
Cities
113
in front of Khulna in the Ganges-Brahmaputra delta in Bangladesh (but note that
there are no coastal ecosystems in front of Kolkata or Dhaka in the same delta) and
the 1 138 km² of seagrass meadow (1 022 km²) and mangrove forest (116 km²) in
front of Conakry in Guinea.
Based on the number of cities and the total surface area of coastal ecosystems per
continents, Africa is the continent with the largest averaged surface area of coastal
ecosystems per city (149.0 km²), while Europe is the continent with the lowest
averaged surface area of coastal ecosystems per city (5.7 km²) (Figure 4.2). The
surface of coastal ecosystems in front of cities in Africa, Asia and to a lower extent
South America shows a large variation, with a combination of cities having no or
small coastal ecosystems and cities having a high surface area occupied by coastal
ecosystems along their likely flood pathway, as Khulna in Bangladesh (2 006 km²),
Conakry in Guinea (1 138 km²) or Guayaquil in Ecuador (741 km²). Oceania is the
only continent where all the cities benefit from the presence of coastal ecosystems
along their probable flood pathway.
Chapter 4
Figure 4.1 Representation of the surface area (km²) occupied by coastal ecosystems (size of the circle) and of the different ecosystem types present (colours) within the 20 km buffer zones along the likely storm surge flood pathways for each of the 136 cities with highest flood-exposed populations in 2005.
Cities
115
Figure 4.2 Comparison of the surface area of coastal ecosystems (km²) in front of the studied cities per continent. The width of the boxes corresponds to the square root of the number of cities per continent. The cities of Khulna in Asia and Conakry in Africa are not represented due to their exceptionally high values of 2 006 km² and 1 138 km² respectively
4.3.2 Historical Influence on the Tidal Wetlands Distribution
The surface area of tidal wetlands, as mangroves and salt marshes, in front of the
studied cities is influenced by several parameters. It was tested if an explanation
could be found in the historical societal development of the continents. The
earliest highly populated continents are Asia and Europe (Maddison, 2001;
McEvedy & Jones, 1978; World Population History, n.d.), which was accompanied
by an increasing demand for food production and agricultural land. This was
particularly the case in or near coastal zones, where demand for food production,
urban and industrial development resulted in reclamation of marshes and
mangroves into human land use (Almeida et al., 2014; Hoeksema, 2007; Lotze et
al., 2006; Scott et al., 2014; Tian et al., 2016; Valiela, 2006).
Figure 4.3 represents for each city the surface area within the 20 km buffer zone
around the likely flood pathways where the land elevation is below the mean high
tide but no tidal wetlands exist presently. This area is considered here as an
estimation of the surface area of land that was reclaimed over time by conversion
of formerly existing tidal wetlands into human land use. The observation of the
estimated reclaimed land area per continent shows that Oceania, Africa and South
Chapter 4
116
America are the continents where there was the least reclamation, in contrast to
Europe, Asia and Northern America. The city with the largest extent of estimated
reclaimed land area is Hamburg in Germany with 1 391 km² of reclaimed
intertidal land area in its likely area for storm surge propagation, i.e. the Elbe
estuary. This number is in accordance with previous assessments, showing that
the original intertidal flood plain area decreased enormously by intertidal wetland
reclamation over the past centuries (Hamburg Port Authority, 2006; Hansen,
2015; Reise, 2005) (Supplementary Information Figure SI 4.2 ). Hamburg is
followed by Guangzhou, along the Pearl River delta in China, with an estimated
reclaimed intertidal area of 1 119 km², corresponding with reported estimations
of original wetland loss reaching 50 to 60 % over the delta region (Li & Lee, 1997;
Tian et al., 2016). Ho Chi Minh City in Vietnam has an estimated reclaimed
intertidal area of 785 km². For Rotterdam in the Netherlands this number is of 782
km², resulting from historical marsh embankment and drainage into agricultural
land – so-called “polders” – since the Middle Ages, and more recent construction of
the harbour of Rotterdam (de Haas et al., 2018; Pierik et al., 2017; Ysebaert et al.,
2016). The observations made here are consistent with other studies assessing the
land reclamation over the world (Airoldi & Beck, 2007; Ganong, 1903; Hatvany,
2003; Hoeksema, 2007; Murray et al., 2014).
Our analysis indicates that the lower surface area of coastal ecosystems in front of
European cities can be partly attributed to the long history of conversion of coastal
wetlands into human land use in Europe (often so-called polders). Throughout
Europe, evidence of coastal wetland reclamation for agriculture can be traced back
to the Middle Ages (Hoeksema, 2007; Scott et al., 2014), while in China, the
thirteenth century was already a time of high human influence in the Yangtze and
Yellow river deltas (Scott et al., 2014). In Asia, most of the cities are still fronted by
large areas of coastal ecosystems, mainly mangroves and salt marshes,
nonetheless, the estimation of reclaimed land in front of Guangzhou, Ho Chi Minh
City, Tianjin, Kolkata or Khulna are among the highest in the world, with
estimations ranging from 245 km² for Khulna to 1 119 km² of reclaimed land for
Guangzhou. Indeed these Asian cities are located in the world’s largest deltas,
where on the one hand large surface areas of tidal wetlands have been reclaimed
into human land use, but on the other hand also still relatively large wetlands
remain (Auerbach et al., 2015; Wang et al., 2014). An extreme example in this
respect, is Khulna, located in the Ganges-Brahmaputra delta, with a large
estimated reclaimed land surface area (245 km²) and at the same time the largest
remaining mangrove surface area (2 006 km²; in the Sundarbans) along its likely
flood pathway (Auerbach et al., 2015).
Cities
117
The growth of the population in North and South America and Africa appeared
later (from the end of the 18th century) with the industrialization and the end of
both the colonization and the slave trade. Yet, recent observations of coastal
ecosystems over the world are highlighting the dramatic trend of current large
ecosystem losses (Ma et al., 2014; Scott et al., 2014; Spalding et al., 2010; World
Population History, n.d.). The rise in worldwide commercial relationships
concentrated the increasing population to settle in or near the coasts and major
river mouths, in order to benefit from the water way connections with the rest of
the world, which at the same time increased the human pressure on the coastal
ecosystems (de Sherbinin et al., 2007). During the nineteenth and twentieth
centuries, all the continents knew a peak in land reclamation for human activities,
resulting in a major loss of coastal wetlands. In addition, the pollution,
acidification, warming and rising of the oceans among others are threatening
seagrass meadows and coral reefs (Green & Short, 2003; Spalding, Ruffo, et al.,
2014). It is estimated that 75 % of the world’s coral reefs are under threat
(Spalding, Ruffo, et al., 2014), while about one third of mangrove forests and
seagrass meadows disappeared and half of the world’s salt marshes were
destroyed or degraded over the last 30 years (Barbier et al., 2008; Millennium
Ecosystem Assessment, 2005; Valiela et al., 2009).
Chapter 4
Figure 4.3 Estimation of the surface area (km²) of tidal wetlands (mangroves and marshes) reclaimed within the likely area for storm surge propagation for every city.
Cities
119
4.3.3 Parameters Influencing Coastal Ecosystems Presence and Area
The long term historical evolution of human settlement and activities on the
continents in addition to the recent history of land reclamation and environmental
disturbance is only a part of the explanation behind the variation of the surface
area of coastal ecosystems along the cities’ likely flood pathway. In order to
identify the social and physical environmental variables that may explain the
variations in the presence or absence, or the surface area of the different coastal
ecosystems, logistic (presence/absence) and linear (surface area) regressions
were performed, both as simple and multiple regressions. The results of the simple
logistic and simple linear regressions are presented in the Supplementary
Information, while the results of the multiple logistic and linear regressions are
presented and discussed here after.
The multiple logistic regressions test the influence of each explanatory variable on
the presence or absence of the coastal ecosystem (i.e. dependent variable)
assuming all the rest remains constant. The odd ratio gives the factor of change in
odds to find an ecosystem for an increase in one unit of the explanatory variable
assuming that all the other explanatory variables remain constant. The Table 4.2
present the results of the multiple logistic regressions, i.e. the explanatory
variables significantly (p-value < 0.05) influencing the presence or absence of each
coastal ecosystem.
Table 4.2 Odds Ratio resulting from the multiple logistic regressions testing the influence of the explanatory variables (Table 4.1) on the odds of finding coastal habitats in front of the coastal cities for a significance of 95 % (p-value < 0.05), the ‘X’ corresponds to non-significant relations
Dependent variables
Explanatory variables
Unit of increase
Odd Ratio
Mangrove Salt Marsh Seagrass Coral Reef
Physical Parameters
Latitude 1 ° 0.961 X X 0.971
Distance between the sea and the city 1 km
0.954 X X X
Coastline length 1.006 X X X
Area below mean high tide 1 km²
X X 0.985 X
Shallow Water area (depth > -100 m) X X 1.003 X
Deep Water area (depth < -100 m) X X 1.005 1.007
Social Parameters
GDP Per Capita of the Country 1 US$ 0.999 1.0001 1.0001 X
Short-distance population density 1 inhabitant
/km²
X 0.9997 X X
Intermediate-distance population density X 1.001 X X
Long-distance population density X X X X
Chapter 4
120
The results highlight that the presence or absence of the four coastal ecosystems is
related to a set of explanatory variables. For example, the presence of a mangrove
forest is positively correlated to an increase of the coastline length assuming that
all the other parameters remain constant, while the increase of the latitude or the
GDP per capita each decrease the odds to find a mangrove forest, assuming that
when one variable changes, all the others remain constant. The negative relation
between the latitude and the presence of a mangrove forest is related to the
intrinsic characteristics of mangroves, as they develop in tropical regions that are
in low latitudes.
The condition that all the other parameters remain constant is obviously not
realistic in practice. However, it allows highlighting the ‘pure’ effect of each
parameter on the presence or absence of a coastal ecosystem.
Secondly, the multiple linear regressions focus on the influence of the tested
explanatory variables on the surface area of the four considered coastal
ecosystems. As for the multiple logistic regressions, the influence of each
explanatory variable is considered assuming all the other variables remain
constant.
Table 4.3 Regression coefficients resulting from the multiple linear regressions testing the influence of the explanatory variables (Table 4.1) on the size of the coastal habitats in front of the coastal cities for a significance of 95 % (p-value < 0.05), the ‘X’ corresponds to non-significant relations
Dependent variables
Explanatory variables
Unit of increase
Multiple Regression coefficients
Mangrove Salt Marsh Seagrass Coral Reef
Physical Parameters
Latitude 1 ° X X X X
Distance between the sea and the city 1 km
0.156 X X X
Coastline length 1.992 X X X
Area below mean high tide
1 km²
X X X X
Shallow Water area (depth > -100 m) 0.887 1.040 1.112 X
Deep Water area (depth < -100 m) X X X 1.021
Social Parameters
GDP Per Capita of the Country 1 US$ 0.998 X X X
Short-distance population density 1 inhabitant
/km²
X X 1.018 X
Intermediate-distance population density X 0.974 0.960 X
Long-distance population density 0.935 1.083 X X
Cities
121
The results show, for example, that the surface area of seagrasses is increasing for
an increase of the shallow water area, assuming all the other variables remain
constant. This seems to be related to the fact that seagrasses are growing in
shallow water areas, and as such an increase of their favourable environment lead
to an increasing possibility to have larger seagrass bed areas.
For the multiple logistic and linear regressions, the results highlight statistically
significant relations between the explanatory variables and the presence/absence
or surface area of the four coastal ecosystems. Nonetheless, those results must be
interpreted carefully, as the natural mechanisms behind the presence and size of a
coastal ecosystem are not accounted for. For example, according to the results, the
surface area of salt marshes is increasing for an increase of the population density
in the long-distance environment of the city (100 km). While this is statistically
significant, it seems unlikely in regards to the pressure the populations puts on the
coastal ecosystems.
The different regression analyses performed can then statistically back-up some
expected relations, as the disappearance of the mangrove forests and coral reefs as
the latitude increases, or the increase of the seagrass bed areas for an increase in
shallow water area. Yet, some relations must be interpreted carefully and should
be explored further to confirm or reject the statistical significance. Parameters as
the bio-geomorphological characteristics of the region (e.g. soil constitution,
granulometry), the water chemistry (e.g. salinity, pH, and temperature), or the
waves and currents conditions (e.g. orientation, energy...) are also expected to play
an important role in the establishment and development of coastal ecosystems and
differ from one city to another. As a consequence, some locations are not suitable
for the development of the considered coastal ecosystem types, such as the
Western coast of South America due to its steep topography and bathymetry (but
large deltas also exist there such as the Guayas delta in Ecuador) (Scott et al.,
2014).
4.3.1 Hotspots of Populations and Assets at Risk and Large Ecosystems
Finally, we compared our results of the geographical repartition of the coastal
ecosystems in front of every city to the people and assets exposed to coastal
flooding as defined by Nicholls et al. (2007), in order to highlight the share of the
world’s urban population and assets at risk that can benefit from the presence of
coastal ecosystems. Figure 4.4 shows on the one hand that some of the cities with
the largest populations at risk (> 412 000 people) can benefit from the presence of
large surface areas of coastal ecosystems (> 322.09 km²), such as Khulna
Chapter 4
122
(Bangladesh), Guayaquil (Ecuador), Shanghai and Guangzhou (China), Ho Chi Minh
City (Vietnam) and New Orleans (USA). All of these cities are located in large river
delta plains, where large mangroves or marsh areas exist in portions of the delta
plain between the city and the open ocean. On the other hand, a high number of
cities with a large population at risk of flooding either benefit from a limited
surface area of coastal wetlands (e.g. Boston, Rotterdam or Lagos) or are deprived
from any coastal ecosystem (e.g. Tokyo and other Japanese cities, Dhaka,
Amsterdam or Montreal). These are mostly cities either directly located along the
open sea (e.g. most Japanese cities) or where coastal ecosystems between the city
and sea have been historically reclaimed and turned into human land use (e.g.
Amsterdam, Rotterdam, Boston).
By looking at the ranking of the cities based on their population at risk, assets at
risk and surface area of coastal ecosystems, the top twenty cities with the largest
populations exposed to coastal flooding contain 71 % of the population and 69 %
of the assets at risk against only 15 % of the surface area of coastal ecosystems
that could participate to risk mitigation (Figure 4.5; all percentages are relative to
the total values for all 136 cities considered in this study). A similar result is found
for the top twenty cities with the most assets exposed to coastal flooding. In
contrast, the top twenty cities with the highest surface area of coastal ecosystems
contain 77 % of the coastal ecosystems found in front of the 136 studied cities, but
only about 30 % of both the population and assets at risk are found in those
twenty cities (Figure 4.5).
Several cities combine high values of population and assets at risk of flooding (12
cities with > 696 000 people), among which eight benefit from the presence of
coastal ecosystems, but only four have large surface areas of coastal ecosystems,
Miami (237 km²), New Orleans (241 km²), Guangzhou (304 km²) and Shanghai
(397 km²). In addition, Ho Chi Minh City is characterized by a large population at
risk (1 931 000 people) and a large surface area of coastal ecosystems (373 km²),
while Hong Kong combines a high value of assets at risk (36 billion USD) and a
large surface area of coastal ecosystems (349 km²) along its likely flood pathway.
In total, 6 of the 28 world’s cities having the highest values of population and/or
assets at risk benefit from the presence of large coastal ecosystem areas. For more
information of those top 20 cities, see Supplementary Information.
Cities
Figure 4.4 Representation, for each of the 136 most populated world’s coastal cities in 2005, of the surface area (km²) occupied by coastal ecosystems (size of the circle) and of population at risk of coastal flood damages based on the study of Nicholls et al. (2007) (colours)
Chapter 4
124
Figure 4.5 Representation of the population and assets at risk as well as the coastal ecosystem area for the top 20 cities with (1) the largest population at risk, (2) the highest amount of assets at risk and (3) the largest surface area of coastal ecosystems
4.4 Conclusion
The results show that on the 136 studied cities that have the largest flood-prone
population in the world, 75 % can benefit from existing coastal ecosystems, while
35 cities have no coastal ecosystems and 73 cities have less than 100 km². The
results highlight then that several densely populated and industrialized cities, i.e.
Rotterdam, London, Shanghai or Boston, can benefit from the presence of different
coastal ecosystems and that those ecosystems could be included in sustainable and
efficient coastal protection strategies. Our analysis implies that the parameters
influencing the presence of the coastal ecosystems on the one hand and their
surface area on the other hand are a combination of historical, social, and physical
factors. Several expected relations between the presence and extent of the
different coastal ecosystems could be backed up by the logistic and linear
regressions, while further research would be needed to fully understand the
influence of the studied and other parameters in the occurrence and extent of the
four types of coastal ecosystems in front of the world’s coastal cities.
Subsequently, future coastal protection strategies should account for the presence
or absence, and extent of coastal ecosystems in front of the city. For example, a city
Cities
125
like Khulna, with 441 000 people exposed to flood risks, but located far inland
(more or less 100 km from the open sea) and with large deltaic mangrove forests
(2 006 km²) between the city and sea, can largely benefit from nature-based flood
risk mitigation. The conservation or even restoration of formerly reclaimed coastal
ecosystems for those cities located in large river delta plains, can contribute to the
mitigation of storm surge flood risks, while providing additional valuable
ecosystem services. In contrast, while in the same delta, the cities of Kolkata
(almost 2 million people at risk) and Dhaka (844 000 people at risk) do not benefit
from the coastal flood and erosion risks mitigation offered by the mangrove forest
of the Ganges-Brahmaputra delta (i.e. the Sundarbans). Or a city like New York,
with 1 540 000 flood-exposed people and an estimated 23 km² of salt marshes
along its likely storm surge flood pathway, will benefit only marginally from
nature-based flood risk mitigation, as the available buffer area between the city
and open sea is small or non-existent. Those cities must primarily rely on investing
in hard engineering structures to protect their populations, and where possible
can benefit from the creation or restoration of coastal ecosystems as add-on to
hard structures, in order to obtain sustainable and cost-effective flood protection
solutions.
4.5 Acknowledgments
The author would like to thank Chris McOwen for sharing the salt marsh dataset
and Erik Fransen for the help on statistical analysis. The data used are listed in the
method and references. The research was funded by the University of Antwerp.
Chapter 4
126
Supplementary Information
Probable flood pathway
The probable flood pathway, i.e. the path preferably followed by the storm surge
between the open sea and the city, accounting for the friction exerted by the land
area and the water bodies, is presented here after for a selection of cities.
Figure SI 4.1 Representation of the probable flood pathways as calculated in the analysis for seven cities located at the end of long and small river channels (Guangzhou, Hamburg and Shenzhen), in bays (New Orleans), or at the coast (Hong Kong, Los Angeles, San Diego).
The comparison of the coastal ecosystems surface area in the three buffer size
(Table SI 4.1) shows that for the cities that have no coastal ecosystems in the 10
km buffer (i.e. 39 cities) four cities, (i.e. Xiamen, Yantai, Rangoon and Rotterdam)
have a presence of coastal ecosystems in their 20 km buffer area, and 6 more cities
have a presence of coastal ecosystems in the 30 km buffer (i.e. Qingdao, Athens,
Kolkata, Nagoya, Tokyo and Benghazi). For the cities having coastal ecosystems in
the 10 km buffer area (i.e. 97 cities), 60 % have an increase of coastal ecosystems
surface area smaller than 60 km² in the 30 km buffer area, the other 40 % of those
cities have a higher increase in surface area. However, the difference in surface
areas of coastal ecosystems for the different cities and for the different buffer size
remains relative. As such, the comparison of the coastal ecosystems area between
the cities remains similar in the case of the use of the three buffer sizes.
Cities
127
Buffer sizes
Table SI 4.1 The different surface area of coastal ecosystems (km²) are presented for the three buffer size tested around the probable flood pathways for every city
City Country Continent
Coastal Ecosystems Surface Area (km²) for each buffer size
10 km 20 km 30 km
Algiers Algeria Africa 1.4 5.3 8.9 Luanda Angola Africa 99.8 260.7 464.5 Buenos Aires Argentina South America 0 0 0 Adelaide Australia Oceania 61.0 148.1 233.2 Brisbane Australia Oceania 44.1 82.0 268.3 Perth Australia Oceania 65.8 161.6 286.5 Melbourne Australia Oceania 0.2 22.0 44.6 Sydney Australia Oceania 12.7 22.7 26.3 Khulna Bangladesh Asia 963.2 2006.0 3005.0 Dhaka Bangladesh Asia 0 0 0 Chittagong Bangladesh Asia 1.0 1.3 1.6 Grande Vitoria Brazil South America 19.8 20.0 20.0 Santos Brazil South America 31.7 70.1 79.1 Maceio Brazil South America 9.8 16.6 22.4 Natal Brazil South America 17.9 22.9 25.4 Belem Brazil South America 9.6 36.4 49.9 Porto Alegre Brazil South America 0 0 0 Fortaleza Brazil South America 10.3 15.0 53.8 Salvador Brazil South America 0.9 8.7 37.9 Recife Brazil South America 4.5 15.2 28.7 Rio de Janeiro Brazil South America 0.04 13.3 48.7 Douala Cameroon Africa 238.9 482.6 838.4 Montreal Canada North America 0 0 0 Vancouver Canada North America 0.5 2.5 3.3 Xiamen China Asia 0 55.8 82.0 Yantai China Asia 0 3.0 3.0 Zhanjiang China Asia 60.9 208.1 521.5 Dalian China Asia 0 0 0 Qingdao China Asia 0 0 3.0 Wenzhou China Asia 92.1 102.4 123.0 Ningbo China Asia 0.0 0.3 7.8 Shenzhen China Asia 9.0 69.0 277.5 Fuzhou China Asia 96.3 99.2 110.2 Guangzhou China Asia 180.1 304.0 491.9
Chapter 4
128
City Country Continent
Coastal Ecosystems Surface Area (km²) for each buffer size
10 km 20 km 30 km
Tianjin China Asia 5.7 9.3 9.8 Hangzhou China Asia 177.1 231.9 249.8 Shanghai China Asia 116.2 396.6 682.0 Barranquilla Colombia South America 60.4 98.6 145.5 Havana Cuba North America 3.7 11.0 22.2 Copenhagen Denmark Europe 5.0 23.4 39.6 Santo Domingo Dominican Republic North America 24.4 79.5 143.4 Guayaquil Ecuador South America 377.4 741.2 866.7 Alexandria Egypt Africa 0 0 0 Helsinki Finland Europe 1.0 3.4 3.4 Marseille France Europe 7.7 12.6 22.9 Hamburg Germany Europe 4.1 6.0 12.3 Accra Ghana Africa 154.3 426.6 571.7 Athens Greece Europe 0 0 7.7 Conakry Guinea Africa 303.1 1138.2 2454.4 Port-au-Prince Haiti North America 5.2 9.9 30.7 Hong Kong Hong Kong S.A.R. Asia 69.5 348.8 838.7 Vishakhapatnam India Asia 0 0 0 Kochi India Asia 0 0 0 Chennai India Asia 0 0 0 Surat India Asia 3.0 13.7 35.5 Mumbai India Asia 8.1 47.1 96.1 Kolkata India Asia 0 0 0.2 Palembang Indonesia Asia 112.4 263.0 388.3 Ujung Pandang Indonesia Asia 10.8 33.7 94.6 Surabaya Indonesia Asia 5.4 17.6 33.9 Jakarta Indonesia Asia 0.8 3.2 6.3 Dublin Ireland Europe 2.9 4.2 5.5 Tel Aviv-Yafo Israel Asia 0.8 2.7 2.7 Naples Italy Europe 0.9 2.2 12.5 Abidjan Ivory Coast Africa 0 0 0 Fukuoka Japan Asia 0 0 0 Nagoya Japan Asia 0 0 0.7 Hiroshima Japan Asia 0.1 1.5 2.3 Sapporo Japan Asia 0 0 0 Osaka Japan Asia 0 0 0 Tokyo Japan Asia 0 0 1.0 Kuwait Kuwait Asia 186.8 621.7 885.4
Cities
129
City Country Continent
Coastal Ecosystems Surface Area (km²) for each buffer size
10 km 20 km 30 km
Beirut Lebanon Asia 0 0 0 Benghazi Libya Africa 0 0 0 Tripoli Libya Africa 0 0 6.6 Kuala Lumpur Malaysia Asia 124.5 188.3 191.7 Rabat Morocco Africa 0 0 0 Casablanca Morocco Africa 0 0 0 Maputo Mozambique Africa 0.7 13.0 25.3 Rangoon Myanmar Asia 0 0.6 0.9 Rotterdam Netherlands Europe 0 5.7 7.3 Amsterdam Netherlands Europe 0 0 0 Auckland New Zealand Oceania 4.5 17.9 24.5 Lagos Nigeria Africa 9.2 49.7 87.0 Nampo North Korea Asia 0 0 0 Karachi Pakistan Asia 10.8 13.4 28.6 Panama City Panama North America 1.2 12.6 19.6 Lima Peru South America 0.2 2.9 4.5 Davao Philippines Asia 85.2 325.2 624.7 Manila Philippines Asia 174.0 522.3 958.3 Porto Portugal Europe 0 0 0 Lisbon Portugal Europe 1.8 9.7 23.7 San Juan Puerto Rico South America 17.1 34.8 55.9 St. Petersburg Russia Europe 0 0 0 Jeddah Saudi Arabia Asia 37.8 131.1 286.2 Dakar Senegal Africa 68.1 213.7 351.6 Singapore Singapore Asia 11.2 138.3 502.3 Mogadishu Somalia Africa 4.5 10.5 17.3 Durban South Africa Africa 0 0 0 Cape Town South Africa Africa 0 0 0 Ulsan South Korea Asia 0 0 0 Incheon South Korea Asia 0 0 0 Busan South Korea Asia 0 0 0 Barcelona Spain Europe 0.4 5.5 8.0 Stockholm Sweden Europe 0.1 0.3 0.6 Taipei Taiwan Asia 24.1 25.6 26.5 Dar-es-Salaam Tanzania Africa 23.3 53.1 77.5 Bangkok Thailand Asia 18.4 32.8 50.0 Lome Togo Africa 108.8 177.7 178.3 Izmir Turkey Europe 3.1 14.8 43.9
Chapter 4
130
City Country Continent
Coastal Ecosystems Surface Area (km²) for each buffer size
10 km 20 km 30 km
Istanbul Turkey Europe 0 0 0 Odessa Ukraine Europe 0 0 0 Dubai United Arab Emirates Asia 0.5 6.0 10.4 Glasgow United Kingdom Europe 0 0 0 London United Kingdom Europe 11.2 21.0 34.1 Washington, D.C. United States of America North America 37.8 47.9 51.8 Providence United States of America North America 4.4 7.6 10.7 Virginia Beach United States of America North America 2.0 8.6 22.6 Baltimore United States of America North America 1.1 4.8 15.1 San Jose United States of America North America 141.3 183.7 189.0 Portland United States of America North America 12.4 19.9 21.1 Seattle United States of America North America 0.1 0.1 0.2 San Diego United States of America North America 1.2 4.4 11.7 New Orleans United States of America North America 121.9 241.4 546.7 Boston United States of America North America 3.6 12.4 16.2 Tampa United States of America North America 9.2 56.3 96.7 Philadelphia United States of America North America 28.4 54.4 109.8 San Francisco United States of America North America 126.4 148.2 154.6 Houston United States of America North America 5.8 6.7 24.5 Miami United States of America North America 77.0 236.9 458.6 Los Angeles United States of America North America 0.1 2.6 3.5 New York United States of America North America 13.6 23.0 36.7 Montevideo Uruguay South America 15.5 18.5 21.6 Maracaibo Venezuela South America 4.0 6.1 72.3 Haiphong Vietnam Asia 31.9 55.7 69.9 Ho Chi Minh City Vietnam Asia 180.9 372.7 486.7
Cities
131
Elbe estuary embankments
Comparison of the embankments of low-lying lands along the Elbe estuary over
the last centuries (Hansen, 2015) and of the areas below mean high tide as defined
by our study along the Elbe estuary.
Figure SI 4.2 (A) Overview of the extent and period of construction of the embanked areas along the Elbe estuary adapted from Hansen (2015) and (B) representation of the area located below mean high tide in the likely pathway of storm surge propagation towards Hamburg according to our analysis
Chapter 4
132
Simple logistic and linear regressions
Table SI 4.2 shows the significant odds ratio resulting from the simple logistic
regression, i.e. the measure of the association between the explanatory variable
(social or physical parameter) and the dependent variable (presence of coastal
ecosystem). An odd ratio (OR) greater than 1 means that for an increase of one
unit in the explanatory variable, the odds of finding the ecosystem is increasing by
the OR to the power of 1.
Table SI 4.2 Odds Ratio resulting from the logistic regression testing the influence of the explanatory variable (Table 4.1) on the odds of finding coastal habitats in front of the coastal cities for a significance of 95 % (p-value < 0.05), the ‘X’ corresponds to non-significant relations
Dependent variables
Explanatory variables
Unit of increase
Odds Ratio (p-value < 0.05)
Mangrove Salt Marsh Seagrass Coral Reef
Physical Parameters
Latitude 1 ° 0.854 1.079 X 0.918
Distance between the sea and the city 1 km
X X X 0.887
Coastline length X X X 0.996
Area below mean high tide 1 km²
X X 0.991 X
Shallow Water area (depth > -100 m) 1.002 X X X
Deep Water area (depth < -100 m) X X X 1.006
Social Parameters
GDP Per Capita of the Country 1 US$ 0.999 1.0001 1.00004 X
Short-distance population density 1
inhabitants/km²
X 0.999 X X
Intermediate-distance population density X X X X
Long-distance population density X X X X
Cities
133
The Table SI 4.3 represent the results of the simple linear regressions determining
the parameters influencing the surface area of the coastal ecosystems in front of
cities. The regression coefficients in Table SI 4.3 are significant (p-value < 0.05),
they correspond to the factor by which the surface area of the ecosystem will be
augmented for an increase of one unit of the explanatory variable, as presented in
the Table SI 4.3.
Table SI 4.3 Regression coefficients resulting from the linear regression testing the influence of the explanatory variable (Table 4.1) on the size of the coastal habitats in front of the coastal cities for a significance of 95 % (p-value < 0.05), the ‘X’ corresponds to non-significant relations
Dependent variables
Explanatory variables
Unit of increase
Regression coefficients (p-value < 0.05)
Mangrove Salt Marsh Seagrass Coral Reef
Physical Parameters
Latitude 1 ° X 0.941 0.903 X
Distance between the sea and the city 1 km
1.040 1.018 X X
Coastline length 1.003 1.004 X X
Area below mean high tide
1 km²
X X X 0.987
Shallow Water area (depth > -100 m) X X 1.003 X
Deep Water area (depth < -100 m) X X X X
Social Parameters
GDP Per Capita of the Country 1 US$ X 0.999 0.999 X
Short-distance population density 1
inhabitants/km²
X X 1.0003 0.999
Intermediate-distance population density X 1.001 X 0.999
Long-distance population density X 1.002 X X
Chapter 4
Top 20 cities
Table SI 4.4 Top 20 world port cities ranked by population exposure to flood and wind damages and the corresponding surface and type of natural coastal ecosystems along their probable flood pathway
Rank City Country Population
2005 (000)
Population at risk (000)
Assets at risk
(US$bil)
Coastal ecosystem
(km²)
Coastal Ecosystem (type)
1 Mumbai India 18,196 2,787 46 47.1 Mangrove forest, Coral reef 2 Guangzhou China 8,425 2,718 84 304.0 Mangrove forest, Salt marsh 3 Shanghai China 14,503 2,353 73 396.5 Salt marsh
4 Miami USA 5,434 2,003 416 236.9 Mangrove forest, Salt marsh, Seagrass meadow, Coral reef
5 Ho Chi Minh City Vietnam 5,065 1,931 27 372.7 Mangrove forest 6 Kolkata India 14,277 1,929 32 0 7 New York USA 18,718 1,540 320 23.0 Salt marsh 8 Osaka Japan 11,268 1,373 216 0 9 Alexandria Egypt 3,770 1,330 28 0
10 New Orleans USA 1,010 1,124 234 241.4 Salt marsh 11 Tokyo Japan 35,197 1,110 174 0 12 Tianjin China 7,040 956 30 9.3 Salt marsh 13 Bangkok Thailand 6,593 907 39 32.8 Mangrove forest 14 Dhaka Bangladesh 12,430 844 8 0 15 Amsterdam Netherlands 1,147 839 128 0 16 Haiphong Vietnam 1,873 794 11 55.7 Mangrove forest 17 Rotterdam Netherlands 1,101 752 115 5.7 Salt marsh 18 Shenzhen China 7,233 701 22 69.0 Mangrove forest, Salt marsh 19 Nagoya Japan 3,179 696 109 0 20 Abidjan Ivory Coast 3,577 519 4 0
Cities
Table SI 4.5 Top 20 world cities ranked by assets exposure to flood and wind damages and the corresponding surface and type of natural coastal ecosystems along their probable flood pathway
Rank City Country Population
2005 (000)
Population at risk (000)
Assets at risk
(US$bil)
Coastal ecosystem
(km²)
Coastal Ecosystem (type)
1 Miami USA 5,434 2,003 416 236.9 Mangrove forest, Salt marsh, Seagrass meadow, Coral reef
2 New York USA 18,718 1,540 320 23.0 Salt marsh
3 New Orleans USA 1,010 1,124 234 241.4 Salt marsh
4 Osaka Japan 11,268 1,373 216 0
5 Tokyo Japan 35,197 1,110 174 0
6 Amsterdam Netherlands 1,147 839 128 0
7 Rotterdam Netherlands 1,101 752 115 5.7 Salt marsh
8 Nagoya Japan 3,179 696 109 0
9 Tampa USA 2,252 415 86 56.3 Mangrove forest, Salt marsh, Seagrass meadow
10 Virginia Beach USA 1,460 407 85 8.6 Salt marsh, Seagrass meadow
11 Guangzhou China 8,425 2,718 84 304.0 Mangrove forest, Salt marsh
12 Boston USA 4,361 370 77 12.4 Salt marsh, Seagrass meadow
13 Shanghai China 14,503 2,353 73 396.6 Salt marsh
20 Hong Kong Hong Kong SAR 7,041 223 36 348.8 Mangrove forest, Seagrass meadow
Chapter 4
Table SI 4.6 Top 20 world cities ranked by their natural coastal ecosystem surface along their probable flood pathway and the corresponding population and assets exposed to flood and wind damages
These examples of medium to large scale tidal wetland creation programs for
mitigation of coastal flood and erosion risks exemplify the potential of nature-
based mitigation programs on local to regional scales. A global scale analysis of the
potential of tidal wetland creation for coastal risk mitigation is lacking so far, but
could contribute to promote the more widespread implementation of nature-
based flood risk mitigation programs into policy on coastal zone management at
several places around the world. Our study aims to provide a global estimation of
the land surface areas where tidal wetlands, i.e. salt marshes and mangrove
forests, could be restored or created in front of the world’s most flood-exposed
Chapter 5
142
coastal cities, and to explore which factors influence the geographical variation in
the area available for tidal wetlands creation.
5.2 Method
The coastal cities considered in the analysis correspond to 135 cities studied by
Nicholls et al. (2008) that have a population of more than 1 million people in 2005
(UN, 2005) and are exposed to coastal flood damages generated by storm surges
and high winds without any consideration of coastal defences or adaptations. The
city of Helsinki in Finland could not be included due to data availability.
5.2.1 Data
The bathymetry is coming from the General Bathymetric Chart of the Oceans
(GEBCO) (British Oceanographic Data Center, 2017) and represents a gridded
bathymetry of the oceans at a 30 arc second resolution combined with the land
topography defined by the NASA Shuttle Radar Topography Mission (SRTM).
The NASA Shuttle Radar Topography Mission (SRTM) Global 3 arc second
V003 dataset (NASA JPL., 2013) is used as digital elevation model (DEM), as it is
the best known DEM available at global scale (Rodriguez et al., 2006; Sun et al.,
2003).
Information on the tidal amplitude in front of every city is derived from the
Finite Element Solution (2012) – Global Tide from AVISO. The Principal Lunar
semi-diurnal component (M2) was used to define the averaged tidal amplitude in
front of every city.
The location of the urban areas was determined from the Global Land Cover
by National Mapping Organizations dataset that classifies the status of the world’s
land cover into 20 categories (see below) based on the Land Cover Classification
System (LCCS) developed by the Food and Agriculture Organization of the United
Nations (Tateishi et al., 2014).
The population distribution originates from the LandScan 2013 Global
Population Database (Bright et al., 2013). It represents the population over a 30
arc second grid resolution and integrates the diurnal movements and collective
travelling behaviour of the population, i.e. the so-called “ambient population”,
averaged over 24 hours (Bright et al., 2013; Dobson et al., 2000). The dataset was
adapted to deliver values of population density following the guidelines of the
LandScan documentation (Bright et al., 2013; UT BATTELLE LLC., n.d.).
Wetlands creation
143
The location of the tidal wetlands, as salt marsh and mangrove forest, was
determined based on the Global distribution of Mangroves (Giri et al., 2011) and
the Global distribution of Saltmarshes (Mcowen et al., 2017) from the United Nation
Environmental Program – World Conservation Monitoring Centre (www.unep-
wcmc.org).
The study aims to identify areas that meet a selection of conditions for mangrove
and salt marsh development within the likely area of storm surge propagation in
front of the world’s most populated coastal cities, in order to contribute to the
nature-based mitigation of coastal flood risks. The likely area of storm surge
propagation in front of the cities is delineated using the procedure presented in
Chapter 4. The four conditions retained to identify the suitable locations for tidal
wetland creation are (1) elevation below mean high tide, (2) absence of existing
tidal wetlands, (3) location outside the urbanized area and (4) a population
density lower than 50 inhabitants per square kilometre. The logical steps are
presented in Figure 5.1.
The value of 50 inhabitants per square kilometre is an arbitrary value based on the
literature (Mcgranahan et al., 2006; Neumann et al., 2015). Scenarios with
different population threshold values (20, 35 and 50 inhabitants/km²) were
explored and are presented in Supplementary Information. The comparison of the
results based on the three population density thresholds is showing a general
increase of the average surface area available for tidal wetlands development of
8.99 ± 34.04 km² between the lowest and highest thresholds, while keeping the
relative difference between the cities very similar. It is important to note that the
considered threshold remains high and represents a theoretical exploration of the
potential to restore tidal wetlands. In practice, the displacement of a small number
of inhabitants may already generate so much societal-political resistance that tidal
wetland creation or restoration may be impossible to realize.
Figure 5.1 Representation of the logical steps for the selection of the pixels suitable for tidal wetlands restoration or creation
Chapter 5
144
The areas available for tidal wetlands creation were compared to the current land
cover based on the Global Land Cover by National Mapping Organization dataset.
For each city, the different land covers were defined based on five categories,
namely (1) the vegetated areas grouping the land cover classes of broadleaf and
needle leaf evergreen and deciduous forest, mixed forests, tree open areas, shrub,
herbaceous, herbaceous with sparse trees or shrub, sparse vegetation and
wetlands areas; (3) the cropland and paddy fields areas corresponding to the
cropland, cropland with other vegetation mosaic and paddy fields areas; (4) the
bare land corresponding to the bare consolidated or unconsolidated land, and (5)
the water areas corresponding to the water bodies (i.e. pounds or lakes and areas
of coastal water considered inland following the delineation of the country’s
limits).
Due to the global scale of the different datasets, the vegetated areas include areas
defined as wetlands and mangroves in the Global Land Cover dataset. This is
because the use of different global-scale datasets implies limitations in local data
accuracy or artefacts and the overlap of features. In this analysis, the extent of salt
marshes and mangrove forests is based on the Global distribution of saltmarshes
(Mcowen et al., 2017) and the Global distribution of Mangroves (Giri et al., 2011)
that are considered as more accurate than the Global Land Cover dataset.
A linear regression was performed to test the influence of social and physical
parameters (Table 5.1) (explanatory variables) on the geographical variations of
the size of the suitable area for tidal wetlands restoration or creation (dependent
variable). The parameters were extracted through ArcGIS (10.3.1), while the value
of the country’s GDP per capita was recovered from the World Bank data base
(http://databank.worldbank.org/data/home.aspx). More information on the
method can be found in Chapter 4. The regression was performed on the logarithm
value of surface area available for tidal wetlands creation augmented by one to
account for both the normality of the residuals and for the cities having no
available area for tidal wetlands creation. The results obtained from the linear
regression are the regression coefficients (slope), their standard error and the
associated p-value.
Wetlands creation
145
Table 5.1 Physical and Social parameters extracted from the data for each city, for the method, see Chapter 4
Physical Parameters Social Parameters Type Unit Type unit
Latitude of the city Degree Short-distance population density (within 20 km around the flood pathway)
Inhabitants/km²
Distance between the sea and city
km Intermediate-distance population density (within 50 km around the flood pathway)
Inhabitants/km²
Coastline length km Long-distance population density (within 100 km around the flood pathway)
Inhabitants/km²
Area below mean high tide km² Country’s GDP per capita Constant 2005 US$
Shallow water area (> -100 m)
km²
Deep water area (< -100 m)
km²
5.3 Results
5.3.1 Area Below Mean High Tide for Tidal Wetlands Restoration
The zone that sits below mean high tide was defined for every studied city and
divided into three categories, (1) the existing tidal wetlands, (2) the area
potentially available for tidal wetland restoration or creation (i.e. non-urban area
with less than 50 inhabitants/km²) and (3) the area not available for tidal wetland
creation (i.e. urban areas or with more than 50 inhabitants/km²) (Figure 5.2).
The cities with the largest areas below mean high tide are Hamburg in Germany (1
396 km²), Guangzhou in China (1 128 km²), Ho Chi Minh City in Vietnam (860
km²), Rotterdam in The Netherlands (784 km²) and Guayaquil in Ecuador (562
km²). At continental scale, the largest zones below mean high tide are found in
European cities (174 ± 355 km²; i.e. average ± standard deviation) where the
existing tidal wetlands are scarce (maximum of 12 km² in front of London in the
UK). Asia and North America also present large areas below mean high tide with
averages and standard deviations of 128 ± 201 km² and 80 ± 96 km² respectively.
In general the areas located below mean high tide are smaller in South America
(67 ± 130 km²), Africa (44 ± 106 km²) and Oceania (21 ± 11 km²).
The cities that have the highest potentially available area for tidal wetlands
creation are Hamburg in Germany (881 km²), Guayaquil in Ecuador (399 km²) and
Tianjin in China (233 km²), while eleven cities have no available space for tidal
wetlands creation (Figure 5.3). Over the 135 studied cities, the available area for
tidal wetlands is 34 ± 90 km² (average ± standard deviation), with large variations
Chapter 5
146
between the continents (Figure 5.4). Oceania and Africa present the lowest
available surface area for tidal wetlands development with 6 ± 3 km² and 12 ± 22
km² respectively, followed by Asia with 28 ± 41 km². Values for North and South
America are slightly higher with 40 ± 56 km² and 41 ± 94 km², respectively, while
Europe has the largest averaged surface area available for tidal wetlands
development with 73 ± 200 km².
Wetlands creation
Figure 5.2 Surface area located below mean high tide in front of every city (size of the symbol), categorized to surface areas occupied by (1) the existing tidal wetlands, (2) the area potentially available for tidal wetlands creation (i.e. non-urban areas with less than 50 inhabitants/km²) and (3) the area not available for tidal wetlands creation (i.e. urban areas or with more than 50 inhabitants / km²) (colour of the symbol)
Chapter 5
Figure 5.3 Surface area potentially available for tidal wetlands creation (size of symbols) and the current land use in those areas (colours of symbols)
Wetlands creation
149
Figure 5.4 Comparison of the available areas for tidal wetlands creation (km²) in front of the 135 studied cities, categorized per continent. The width of the boxes corresponds to the square root of the number of cities per continent. The city of Hamburg in Europe is not represented due to its high value of 888.8 km².
When looking at the current land use types within the areas that are identified as
potentially suitable for tidal wetlands restoration or creation, the most dominant
land use types are cropland and paddy fields, mostly in European, Asian, South and
North American cities (Figure 5.5), in combination with vegetated areas (1 672
km² and 1 621 km² respectively, or 36 % and 35 % of the total potentially
available area). The cities having the largest cropland and paddy fields areas
below mean high tide are Hamburg in Germany (624 km² of cropland, 71 % of the
potentially available surface area for tidal wetlands restoration or creation),
Guayaquil in Ecuador (70 km² of croplands and 41 km² of paddy field, 28 %),
Tianjin in China (67 km² of cropland, 30 %) or Rotterdam in The Netherlands (66
km² of cropland, 52 %) (see Supplementary Information). Croplands are mainly
found in North America and Europe, while paddy fields are mainly found in Asia,
and occasionally on large surfaces in other continents such as in Guayaquil
(Ecuador). For most of the cities, part of the potentially available area for tidal
wetlands restoration or creation is currently defined as water bodies by the Land
Cover Dataset; the largest areas are found in Asian and North American cities
(Figure 5.5), as in Tianjin (China) with 126 km² and San Jose (USA) with 110 km².
Bare land represents a really small share of the potentially available area for tidal
wetlands restoration or creation for all the continents, with less than 1 km² on
average (Figure 5.5).
Chapter 5
150
Figure 5.5 Comparison of the land use types in the potentially available areas for tidal wetlands creation in front of the 135 studied cities, categorized per continent. The width of the boxes corresponds to the square root of the number of cities per continent. Note that Y-axes have different scales.
5.3.2 Social and Physical Parameters Influencing the Potentially Available
Area for Tidal Wetlands Creation
A multiple linear regression was performed on several social and physical factors
susceptible to influence the potentially available area for tidal wetlands creation.
The significant regression coefficients (p-value < 0.05) are presented in Table 5.2.
Wetlands creation
151
Table 5.2 Regression coefficients resulting from the mutliple linear regression testing the influence of the social and physical factors (Table 5.1) on the geographical variation of the size of the tidal wetlands in front of the coastal cities for a significance of 95 % (p-value < 0.05). The ‘X’ corresponds to a non-significant regression coefficient.
Dependent variable
Explanatory variables
Unit of increase
Regression coefficients (p-value < 0.05)
Physical Parameters
Latitude 1 ° X
Distance between the open sea and the city 1 km
X
Coastline Length 1.002
Area Below Mean High Tide
1 km²
1.004
Shallow Water Area (depth > -100 m) X
Deep Water Area (depth < -100 m) X
Social Parameters
GDP Per Capita of the Country 1 US$ X
Short-distance population density 1 inhabitants
/km²
0.999
Intermediate-distance population density X
Long-distance population density X
The results can be interpreted as follows. One unit of increase (see Table 5.2) of
the explanatory variable is generating a multiplication of the surface area
potentially available for tidal wetlands creation by a factor corresponding to the
regression coefficient, assuming all the other variables remain constant. The
available area for tidal wetlands creation is positively linked to two variables, as
the coastline length and the area below mean high tide, while it is negatively
linked to the population density in the close vicinity of the city. It is important to
note that the hypothesis that all variables remain constant is highly improbable in
nature, as such the results must be interpreted carefully, as they mostly highlight
the ‘pure’ effect of each explanatory variable on the potentially available are for
tidal wetlands restoration or creation.
5.4 Discussion
Nature-based mitigation of coastal flood risks, by conserving, restoring or creating
coastal ecosystems that are known to attenuate the impacts of sea level rise, storm
surges, wind waves and shoreline erosion, is increasingly proposed as a
sustainable, cost-efficient strategy to mitigate and adapt to increasing coastal flood
risks (R. L. Morris et al., 2018; Narayan et al., 2016; Vuik et al., 2016; van
Wesenbeeck et al., 2014). Although this concept is increasingly adopted in
scientific literature, so far there are no global-scale studies that explored the
Chapter 5
152
potential of tidal wetland restoration or creation in front of flood-exposed coastal
cities, while such an analysis may contribute to increase global interest in the
implementation of nature-based risk mitigation programs into policy on coastal
zone management. We presented here a methodology for such a global-scale
analysis, which identified the potentially suitable and available areas for tidal
wetlands creation within the likely area of storm surge propagation from the sea
towards the 135 studied cities. Large variations in this available area between the
cities mainly reflect the differences in geomorphological setting, population
settlement and land use history within the likely area of storm surge propagation.
Note that the areas identified as potentially available for tidal wetlands restoration
or creation are theoretical areas based on parameters that do not include the
socio-economic limitations of the implementation of nature-based strategies. In
practice, socio-economic and political-governance factors highly influence the
possibility and the success of tidal wetland restoration or creation (Darwiche-
Criado et al., 2017; Hartman & Cleveland, 2018; Perillo et al., 2009). Although, such
socio-economic and political-governance factors are relatively poorly studied, the
public support and acceptance of tidal wetland restoration or creation is at least as
important as the financial and ecological considerations (Hartman & Cleveland,
2018; Perillo et al., 2009; Suman, 2019). In countries where the government’s
regulation capacities are weak for example, the conservation and restoration of
natural areas is often highly difficult; the natural areas are freely accessible by the
public with little monitoring of the different activities taking place. In general, the
local communities are willing to support the development of such nature-based
strategies, but not at their expense (e.g. livelihood reduction, land loss...), which
can seriously hamper the possibilities of tidal wetland restoration or creation
(Perillo et al., 2009; Suman, 2019).
In the studied factors, the geomorphic setting is a first factor that may explain the
differences in potential areas for tidal wetland creation. As shown by the logistic
regression analysis, the potentially available area for tidal wetlands creation is
increasing if the area below mean high tide increases, all other things remaining
constant, or with an increasing coastline length, all other things remaining
constant. This relates to the fact that cities located along deltaic or estuarine
channels or adjacent to bays and lagoons (i.e. having longer coastlines within their
zone of likely storm surge propagation), have greater low-lying zones and
subsequently larger potentially suitable areas for salt marshes and mangrove
forests development (Grobicki et al., 2016; Mcowen et al., 2017; Pennings &
Bertness, 2000; Scott et al., 2014; Spalding et al., 1997; Wolanski & Elliott, 2015).
This can be illustrated by the city of Hamburg (Germany) located adjacent to the
Wetlands creation
153
Elbe estuary at 110 km from the estuary mouth, for which our analysis identified
that the zone influencing the propagation of a storm surge includes a record area
below mean high tide of 1 396 km² and a potentially available area for tidal
wetlands development of 881 km² (see Supplementary Information Figure SI 5.1).
Another, tropical example, is Guayaquil (Ecuador), located alongside the main
river channel within the large Guayas river delta at 60 km from the open sea, for
which our analysis indicates an area below mean high tide of 562 km² and an area
potentially suitable for tidal wetlands development of 399 km².
Besides geomorphology, the potentially available area for tidal wetlands
restoration or creation is influenced by the population density in the short-
distance environment around the city (Table 5.2). It highlights the fact that the
population settlement in the low-lying zone can hamper the creation of new tidal
wetlands. This can be illustrated by Guangzhou in China, for instance, where, on
the 1 128 km² located below mean high tide (a huge area due to the location of
Guangzhou in the large Pearl river delta), only 8 km² are at present occupied by
existing tidal wetlands and 97 km² could be potentially available for new tidal
wetlands creation, while 91 % of the area is occupied by densely populated urban
zones (Figure SI 5.2). Indeed the city of Guangzhou is part of a huge agglomeration
occupying most of the delta. The situation is similar for a number of cities as
Nagoya, Kolkata or Shanghai (Figure SI 5.3), located especially in Asia (China,
Japan, India), where several of such large deltaic agglomerations developed with
little open space left for tidal wetlands restoration or creation (Figure 5.3).
The geomorphic and population factors together suggest that in the future, the
reduction of area between the open sea and the city, by marine transgression
through sea level rise and shoreline erosion, combined with an increasing
population in the coastal zone, i.e. coastal squeeze, may limit the potential
development of new tidal wetlands (Pontee, 2013; Rupp-Armstrong & Nicholls,
2007).
Land use history also plays a role. Over the world, the human influence on the
coastal areas and particularly around coastal cities led to the degradation,
destruction and conversion of hundreds of square kilometres of tidal wetlands,
leaving cities with few or no remaining tidal wetlands (Airoldi & Beck, 2007;
Alongi, 2008; Duke et al., 2007; He & Zhang, 2001). This is the case for many
European cities in an estuarine or deltaic setting, which present the largest areas
of croplands located below mean high tide level (Figure 5.3 & Figure 5.5).
Examples include Hamburg in Germany, London in the UK, or Rotterdam and
Amsterdam in The Netherlands, where the embankment and drainage of coastal,
Chapter 5
154
estuarine and deltaic wetlands into so-called “polders”, mostly for agriculture
purposes, dates back to the Middle Ages and even earlier (Airoldi & Beck, 2007;
Hansen, 2015; Hatvany, 2003; Hoeksema, 2007; Reise, 2005). As such, for instance
almost 2 500 km² of salt marshes were reclaimed along the Elbe estuary between
Hamburg and the sea (Figure SI 5.1) (de Haas et al., 2018; Hamburg Port
Identified as a currently large “polder” (881 km²), especially consisting of cropland
(624 km²), our analysis identified this area as potentially available for tidal
wetland creation. In China, from the 1960s, the embankment of mangrove areas
into rice fields, aquaculture ponds or areas for industrial and urban development,
resulted in the loss of nearly 60 % of the Chinese mangroves; at a local scale, a city
like Hong Kong lost 85 % of its original mangrove forests (Li & Lee, 1997; Meng et
al., 2017). In other (sub-)tropical areas, historical land use changes from
mangroves into human land use, often aquaculture ponds, is widespread (Chen et
al., 2017; Deb & Ferreira, 2015; Meng et al., 2017; Scott et al., 2014; Valiela et al.,
2001; Zhu et al., 2016). In Guayaquil (Ecuador), for instance, the potentially
available area for tidal wetlands restoration coincides with present-day
aquaculture ponds that were created over the past decades in former mangrove
areas in the Guayas river delta (Delgado, 2013; Parés-Ramos et al., 2013).
Over the last decades, restoration of salt marshes and mangrove forests is
observed at several places around the world, with among others the restoration of
marshland in the Mississippi deltaic plain (Coastal Wetlands Planning Protection
and Restoration Act (CWPPRA), n.d.; Day et al., 2007), the restoration of tidal
marshes in the Rhine-Meuse-Scheldt delta (Eertman et al., 2002; Oosterlee et al.,
2018; Ysebaert et al., 2016), the projects of tidal wetlands restoration in the
Yellow river and along the Chinese coasts (An et al., 2007; Cui et al., 2009; Jiang et
al., 2015), or the reforestation of mangrove forest in front of Ho Chi Minh City in
Vietnam that was destroyed during the war by chemical spraying (Hong, 2001;
Marchand, 2008). Between 1978 and 2000, the efforts of reforestation in the Can
Gio region (Ho Chi Minh City, Vietnam) resulted in the restoration of around 200
km² of healthy and diverse mangrove forest (Hong, 2001).
The effectiveness of the creation or restoration of tidal wetlands strongly depends
on the current land use both in terms of the success of the tidal wetlands
development and of the future increase in coastal protection (Lewis & Brown,
2014; Q. Zhao et al., 2016). Therefore, locations where we identified possible areas
for tidal wetland creation may necessitate different measures and may experience
different rates of success, depending on their present land use type (Figure 5.3).
Firstly, the restoration or creation of tidal wetlands necessitates the presence of
Wetlands creation
155
several hydro-geomorphic conditions. For example, establishment of tidal wetland
vegetation may be limited when tidal flooding is too excessive and soil drainage
during ebb is poor. Consequently tidal wetland vegetation is usually only able to
grow in the upper portion of the intertidal zone, where soil and topographic
conditions allow good drainage. As such, the establishment of tidal wetland
vegetation where land use currently consists of water bodies (Figure 5.3) can be
very difficult as the water bodies should be drained or elevated to create a tidal
regime allowing the development of the wetland’s vegetation (Haltiner et al.,
1997). Areas that are currently used as agricultural or paddy fields for food
production (Figure 5.3), and that are currently protected from tidal flooding by
structures such as dikes, dams or levees, may be converted to tidal wetlands by
introducing a tidal regime, but also here, care should be taken that the elevation,
tidal inundation regime and drainage is suitable to allow successful wetland
vegetation establishment (Beauchard et al., 2011; Maris et al., 2007). Additionally,
those agricultural areas may be polluted with fertilizer, leading to high
concentrations of nitrate and phosphorus for example. When these agricultural
areas are restored in tidal wetlands, the release of those nutrients during tidal
cycles can lead to severe problems, such as on the water quality of the estuarine or
coastal system (Ardón et al., 2017; Shoemaker et al., 2017). Similarly, the pollution
in industrial or urbanized soils also implies limitations to the restoration or
creation of tidal wetlands that have to be accounted for.
Secondly, the effectiveness of tidal wetland creation for nature-based storm surge
mitigation also depends on the present land use type. Tidal wetlands reduce the
height of storm surges due to their bed roughness and the friction exerted by their
vegetation on the water column; the latter is dependent on amongst others the
vegetation density, height and stiffness (Shepard et al., 2011; Sutton-Grier et al.,
2015; Wamsley et al., 2009). For tidal wetlands, salt marsh vegetation (consisting
of grasses, herbs, and low shrubs) exerts less friction than mangrove forests, yet
they generate more friction on propagating storm surges than agricultural fields
(i.e. croplands or paddy fields) or bare soil surfaces (Mattocks & Forbes, 2008;
Passeri et al., 2018; Wamsley et al., 2009). The conversion of agricultural fields and
bare soil surfaces to tidal wetlands will then increase the friction on landward
propagating storm surges and hence will increase the attenuation rate of storm
surges. On the other hand, forested areas have a friction comparable to mangrove
forests (Mattocks & Forbes, 2008), making their conversion less interesting in
terms of storm surge attenuation. Nonetheless, the propagation of storm surges
through freshwater plants implies a salinity intrusion that is not well managed by
a number of freshwater species (Carter et al., 2018; Middleton, 2016; Stanturf et
al., 2007). Thus, keeping freshwater vegetation in order to protect the coastal
Chapter 5
156
population and areas from storm surges may be inefficient, as the resilience of the
freshwater vegetation to salinity intrusion is uncertain. Subsequently, in areas
with a high intensity and frequency of storm surges, the conversion of freshwater
vegetation to saltwater vegetation (i.e. tidal wetlands) might be seen as a valuable
nature-based strategy to increase the resilience of the vegetation to storm surges
(Middleton, 2016).
Local knowledge on how and where to restore and recreate tidal wetlands is
growing and highlights this unique character of each local setting and the
importance of understanding amongst others the different hydrodynamic,
geomorphic and ecological characteristics of the specific area that influence the
success of wetland creation (Balke & Friess, 2016; Elliott et al., 2016; Oosterlee et
al., 2018; Simenstad et al., 2006). Depending on the situation, the restoration or
creation of tidal wetlands may necessitate active re-conversion of the area by
restoring the natural hydrodynamic and subsurface hydrological flow patterns,
reshaping the topography of the area, restoring the sediment supply or planting
the appropriate vegetation for example, as in old aquaculture ponds, agriculture
fields or in more urbanized areas (R. A. Garbutt et al., 2006; Lawrence et al., 2018;
S.-M. Lee et al., 2012; Lewis & Brown, 2014; Spalding, McIvor, et al., 2014), while in
other places, wetland vegetation may spontaneously re-colonize the area without
much intervention once the appropriate hydrodynamic and bio-geomorphic
conditions are set (Eertman et al., 2002; Pethick, 2002). Although restoration
projects can be successful, the restored area will often not recreate a pristine
environment in terms of plant diversity, topography or hydrology (Bullock et al.,
2011; Elliott et al., 2016; Hobbs et al., 2009; Lawrence et al., 2018; Spalding,
McIvor, et al., 2014; Yepsen et al., 2014). However, restoration or creation of tidal
wetlands can be successful in a large variety of environments, and are expected to
be able to deliver ecosystem services such as water quality regulation, carbon
sequestration and protection against wind waves and storm surges (Adam, 2019;
Bullock et al., 2011; Hobbs et al., 2009; Spalding, McIvor, et al., 2014).
5.5 Conclusion
There is a pressing need for adaptation of the coastal zone to increasing threats
due to climate change (increased frequency of storm surges, sea level rise...) and
due to socio-economic changes (increasing coastal population density, coastal
megacities...). The development of nature-based and hybrid protection structures
for mitigation of coastal flood and erosion risks is increasingly regarded as a
sustainable and cost-efficient strategy over the long term.
Wetlands creation
157
Our study reveals that on the 135 studied cities, 60 % (8 332 km²) of the area
below mean high tide is urbanized or densely populated and 34 % (4 624 km²,
distributed over 124 cities) is potentially available for tidal wetlands restoration
or creation. Key factors influencing this potentially available space are the size of
the low-lying zone in front of the city (distance between the open sea and the city,
area below mean high tide...) as well as the population density in the close
surrounding of the city. The land use in the potentially available area for tidal
wetlands restoration or creation is mainly composed of croplands, paddy fields,
water bodies and vegetated areas, and influences the effectiveness of tidal wetland
creation for nature-based flood risk mitigation.
The analysis, which is based on global-scale datasets, is providing first estimations
regarding the globally available areas for tidal wetlands restoration or creation.
The development of specific successful restoration projects should necessitate
further local- to regional-scale analyses including a combination of scientific,
socio-economic, policy and management approaches. Local studies based on more
high-resolution datasets are needed to identify the potentially available area for
tidal wetlands restoration or creation at specific locations. This chapter has to be
regarded as a starting point to promote global awareness of the possibility to
restore or create tidal wetlands as nature-based risk mitigation in front of flood-
exposed coastal cities around the world.
Chapter 5
158
Supplementary Information
Population thresholds
Table SI 5.1 Values of the surface area available for tidal wetlands creation or restoration based on the three population threshold tested (20, 35 and 50 inhabitants/ km²)
Cities Continent Area for tidal wetlands development (km²)
< 20 inhab/km² < 35 inhab/km² < 50 inhab/km²
Abidjan Africa 15.0 16.8 18.1
Accra Africa 6.9 6.9 6.9
Adelaide Oceania 10.0 10.0 10.3
Alexandria Africa 86.8 96.3 101.4
Algiers Africa 0.0 0.0 0.0
Amsterdam Europe 71.1 97.4 123.8
Athens Europe 2.7 2.7 3.0
Auckland Oceania 5.0 6.5 6.5
Baltimore North America 3.8 4.9 5.5
Banghazi Africa 19.6 21.1 21.1
Bangkok Asia 86.3 89.5 95.8
Barcelona Europe 2.9 3.1 4.3
Barranquilla South America 105.8 113.8 118.4
Beirut Asia 0.0 0.0 0.0
Belem South America 19.7 25.9 27.8
Boston North America 4.8 5.3 6.1
Brisbane Oceania 2.1 5.5 6.1
Buenos Aires South America 0.4 0.7 1.4
Busan Asia 22.6 25.5 27.3
Cape Town Africa 0.0 0.0 0.0
Casablanca Africa 0.0 0.0 0.0
Chennai Asia 4.9 4.9 4.9
Chittagong Asia 3.9 4.2 4.2
Conakry Africa 15.1 15.7 15.7
Dakar Africa 0.0 0.0 0.0
Dalian Asia 6.0 7.0 8.0
Dar-es-Salaam Africa 2.5 2.7 2.7
Davao Asia 6.2 6.4 7.1
Dhaka Asia 97.3 102.4 107.2
Douala Africa 4.3 4.4 4.4
Dubai Asia 22.9 29.1 31.2
Dublin Europe 1.2 1.2 1.9
Wetlands creation
159
Cities Continent Area for tidal wetlands development (km²)
< 20 inhab/km² < 35 inhab/km² < 50 inhab/km²
Durban Africa 0.0 0.0 0.0
Fortaleza South America 3.6 3.8 3.8
Fukuoka Asia 2.2 2.7 3.3
Fuzhou Asia 14.0 15.5 19.5
Glasgow Europe 3.4 4.5 5.1
Guangzhou Asia 46.9 70.9 97.4
Guayaquil South America 323.9 372.5 399.4
Haiphong Asia 49.5 65.0 74.7
Hamburg Europe 505.3 747.6 880.8
Hangzhou Asia 2.7 4.2 7.4
Havana North America 0.0 0.0 0.0
Hiroshima Asia 3.2 3.4 3.8
Ho Chi Minh City Asia 49.6 68.8 98.1
Hong Kong Asia 14.5 17.4 17.6
Houston North America 16.3 17.5 19.2
Incheon Asia 17.4 20.8 21.9
Istanbul Europe 1.0 1.3 2.8
Izmir Europe 8.4 10.4 11.1
Jakarta Asia 5.6 9.0 9.0
Jeddah Asia 3.3 3.7 4.5
Karachi Asia 4.0 4.0 4.0
Copenhagen Europe 23.0 29.3 29.8
Khulna Asia 50.6 63.3 73.2
Kochi Asia 4.7 6.8 8.8
Kolkata Asia 15.1 20.2 23.7
Kuala Lumpur Asia 12.3 15.9 17.5
Kuwait Asia 2.7 3.7 3.7
Lagos Africa 13.7 14.2 15.6
Lima South America 1.6 2.0 2.0
Lisbon Europe 9.8 10.3 11.9
Lomé Africa 10.5 12.2 13.9
London Europe 21.8 35.5 48.4
Los Angeles North America 3.9 4.6 6.3
Luanda Africa 4.1 4.2 4.5
Maceio South America 7.0 8.0 8.3
Manila Asia 5.6 5.8 8.6
Maputo Africa 10.1 10.6 10.6
Maracaibo South America 8.5 9.1 9.5
Chapter 5
160
Cities Continent Area for tidal wetlands development (km²)
< 20 inhab/km² < 35 inhab/km² < 50 inhab/km²
Marseille Europe 3.4 3.9 3.9
Melbourne Oceania 2.2 2.6 2.7
Miami North America 2.2 2.9 2.9
Mogadishu Africa 4.4 4.4 4.4
Montevideo South America 9.4 10.1 10.8
Montreal North America 132.7 140.1 144.0
Mumbai Asia 12.8 16.2 20.7
Nagoya Asia 6.5 7.6 9.2
Nampo Asia 11.7 14.9 16.5
Naples Europe 0.8 0.9 0.9
Natal South America 8.8 13.8 14.8
New Orleans North America 107.5 112.9 116.0
New York North America 13.3 17.1 18.2
Ningbo Asia 5.6 8.2 8.2
Odessa Europe 49.9 50.5 50.6
Osaka Asia 0.0 0.0 0.0
Palembang Asia 38.0 39.9 45.9
Panama City North America 6.9 6.9 6.9
Perth Oceania 2.9 3.2 3.2
Philadelphia North America 71.3 77.6 80.7
Port-au-Prince North America 2.3 2.3 2.3
Portland North America 165.6 177.8 184.2
Porto Europe 0.0 1.3 1.3
Porto Alegre South America 20.9 23.4 25.3
Providence North America 2.9 3.1 3.7
Qingdao Asia 11.9 12.3 13.2
Rabat Africa 0.0 0.0 3.1
Rangoon Asia 4.8 10.3 33.6
Recife South America 1.1 1.1 1.1
Rio de Janeiro South America 3.1 3.8 3.8
Rotterdam Europe 66.4 94.0 126.6
Salvador South America 0.6 1.0 1.0
San Diego North America 2.4 3.0 3.0
San Francisco North America 8.5 11.9 13.3
San Jose North America 142.4 152.6 159.0
San Juan South America 8.7 9.4 9.9
Santo Domingo North America 2.1 2.1 2.1
Santos South America 3.4 4.2 4.5
Wetlands creation
161
Cities Continent Area for tidal wetlands development (km²)
< 20 inhab/km² < 35 inhab/km² < 50 inhab/km²
Sapporo Asia 10.4 13.1 18.2
Seattle North America 49.4 51.5 53.5
Shanghai Asia 6.9 8.3 9.7
Shenzhen Asia 6.9 12.1 13.7
Singapore Asia 25.2 27.5 28.5
Stockholm Europe 10.7 11.9 12.6
St Petersburg Europe 0.2 0.4 0.9
Surabaya Asia 30.6 44.1 46.2
Surat Asia 50.4 53.3 53.3
Sydney Oceania 2.2 4.1 4.4
Taipei Asia 2.1 2.6 2.6
Tampa North America 7.7 8.2 8.2
Tel Aviv-Yafo Asia 0.0 0.0 0.0
Tianjin Asia 174.1 211.6 233.4
Tokyo Asia 2.1 2.2 2.3
Tripoli Africa 0.0 0.0 0.0
Ujung-Pandang Asia 9.7 13.6 15.1
Ulsan Asia 1.0 2.0 2.3
Vancouver North America 1.4 5.1 8.2
Virginia Beach North America 3.8 4.3 4.3
Vishakhapatnam Asia 0.0 0.2 0.8
Grande Vitoria South America 45.1 49.9 55.7
Washington D.C. North America 48.5 51.7 54.7
Wenzhou Asia 5.6 5.8 8.4
Xiamen Asia 9.3 9.4 11.0
Yantai Asia 0.0 0.0 0.0
Zhanjiang Asia 60.1 66.3 73.9
Chapter 5
162
Land uses
Table SI 5.2 Surface areas (km²) of the land use types within the potentially available area for tidal wetlands creation or restoration in front of the 135 coastal cities, with the distinction between croplands and paddy fields.
City Continent Vegetated
Areas (km²)
Croplands (km²)
Paddy Fields (km²)
Bare Areas (km²)
Water Bodies (km²)
Abidjan Africa 10.7 0.9 6.6
Accra Africa 4.4 2.6
Adelaide Oceania 4.1 0.7
5.5
Alexandria Africa 43.9 33.7 23.8
Algiers Africa
Amsterdam Europe 78.5 35.2 10.0
Athens Europe 1.2 0.6
1.2
Auckland Oceania 4.2 2.3
Baltimore North America 2.4 0.1
2.9
Banghazi Africa 13.6 2.2 5.3
Bangkok Asia 13.2 21.4 5.8
55.5
Barcelona Europe 2.1 1.1 1.1
Barranquilla South America 93.1 8.0
17.2
Beirut Asia
Belem South America 24.7
3.1
Boston North America 1.3 4.8
Brisbane Oceania 3.1 1.1
1.9
Buenos Aires South America 0.3 1.1
Busan Asia 2.4 14.1 7.6
3.2
Cape Town Africa
Casablanca Africa
Chennai Asia 0.8 0.7 3.3
Chittagong Asia 0.9 1.8 0.3
1.3
Conakry Africa 11.1 0.7 3.9
Dakar Africa
Dalian Asia 3.5 0.2 4.3
Dar-es-Salaam Africa 0.8
1.9
Davao Asia 2.3 0.2 4.6
Dhaka Asia 14.1 8.7 34.8 0.8 48.9
Douala Africa 3.2 1.2
Dubai Asia 12.5
18.3 0.4
Dublin Europe 0.4 0.5 0.9
Durban Africa
Wetlands creation
163
City Continent Vegetated
Areas (km²)
Croplands (km²)
Paddy Fields (km²)
Bare Areas (km²)
Water Bodies (km²)
Fukuoka Asia 0.4 0.3 0.5
2.0
Fuzhou Asia 3.9 2.3 1.9 11.5
Glasgow Europe 3.2 0.5
1.4
Grande Vitoria South America 25.2 30.4 0.1
Guangzhou Asia 24.3 28.0 16.1 1.5 27.5
Guayaquil South America 210.4 70.4 41.3 77.3
Haiphong Asia 15.7 9.2 13.0
36.8
Hamburg Europe 251.5 623.9 5.5
Hangzhou Asia 0.3 2.0 1.9
3.2
Havana North America
Hiroshima Asia 1.2
2.6
Ho Chi Minh City Asia 35.6 14.7 27.2 20.6
Hong Kong Asia 8.5
9.1
Houston North America 8.5 3.0 0.7 0.6 6.4
Incheon Asia 13.9 3.6 1.7 0.3 2.5
Istanbul Europe 0.7 1.3 0.8
Izmir Europe 2.6 7.5 0.7
0.4
Jakarta Asia 0.8 0.8 7.3
Jeddah Asia 0.3
1.6 2.7
Karachi Asia 0.9 0.8 1.0 0.7 0.7
Copenhagen Europe 8.4 19.9
1.5
Khulna Asia 18.7 6.4 7.7 40.4
Kochi Asia 1.5
7.3
Kolkata Asia 1.5 4.6 2.5 15.1
Kuala Lumpur Asia 7.1 6.1 0.4 0.2 3.8
Kuwait Asia 0.6 0.5 1.5 1.2
Lagos Africa 11.6 0.7
1.5 1.8
Lima South America 0.9 1.2
Lisbon Europe 5.3 1.9
4.7
Lome Africa 11.9 2.0 0.1
London Europe 12.8 26.4
9.2
Los Angeles North America 1.9 0.2 4.2
Luanda Africa 1.0
3.5
Maceio South America 4.4 2.1 1.8
Manila Asia 0.3 0.3
8.0
Maputo Africa 4.6 6.0
Maracaibo South America 4.1
0.8 0.8 3.8
Chapter 5
164
City Continent Vegetated
Areas (km²)
Croplands (km²)
Paddy Fields (km²)
Bare Areas (km²)
Water Bodies (km²)
Marseille Europe 1.9 0.5 1.6
Melbourne Oceania 0.1 0.2
2.4
Miami North America 1.3 1.6
Mogadishu Africa 3.1 0.8
0.5
Montevideo South America 9.1 1.4 0.2
Montreal North America 29.3 54.2
60.5
Mumbai Asia 8.4 5.4 1.4 5.5
Nagoya Asia 1.1 2.8 2.6
2.6
Nampo Asia 0.7 3.8 0.8 0.1 11.1
Naples Europe
0.7
0.3
Natal South America 5.9 8.2 0.4 0.4
New Orleans North America 72.2 10.2 1.3
32.3
New York North America 7.2 0.4 10.7
Ningbo Asia 4.4
0.7
3.0
Odessa Europe 5.9 3.4 41.3
Osaka Asia
Palembang Asia 24.3 16.5 2.0 3.0
Panama City North America 2.1 0.7 0.5
3.7
Perth Oceania 1.3 1.9
Philadelphia North America 33.1 23.3 7.5 0.2 16.6
Port-au-Prince North America 0.3 1.1 0.3 0.6
Portland North America 81.4 52.9 9.6 0.2 40.2
Porto Europe 0.3 1.0
Porto Alegre South America 15.3 0.7 3.8
5.4
Providence North America 1.7 2.0
Qingdao Asia 6.1
7.1
Rabat Africa 0.7 2.4
Rangoon Asia 0.5 23.8 5.0 0.5 3.9
Recife South America 0.2 0.9
Rio de Janeiro South America 1.9
1.9
Rotterdam Europe 52.2 66.4 8.1
Salvador South America 0.3
0.7
San Diego North America 2.4 0.1 0.5
San Francisco North America 5.1
0.5 7.7
San Jose North America 42.8 4.3 1.9 110.0
San Juan South America 6.2
3.8
Santo Domingo North America 0.6 0.3 1.2
Wetlands creation
165
City Continent Vegetated
Areas (km²)
Croplands (km²)
Paddy Fields (km²)
Bare Areas (km²)
Water Bodies (km²)
Santos South America 3.5
1.0
Sapporo Asia 4.4 7.7 3.2 2.8
Seattle North America 3.8
0.4 49.2
Shanghai Asia 2.5 2.9 1.2 3.1
Shenzhen Asia 8.7
2.0 3.1
Singapore Asia 4.8 11.5 12.3
St. Petersburg Europe 0.4
0.5
Stockholm Europe 5.5 7.0
Surabaya Asia 5.2 3.4 8.1
29.5
Surat Asia 6.3 40.1 4.5 0.4 2.1
Sydney Oceania 0.9
3.5
Taipei Asia 0.2 2.4
Tampa North America 2.9 0.8 0.8
3.9
Tel Aviv-Yafo Asia
Tianjin Asia 35.8 67.2 2.0 2.0 126.3
Tokyo Asia 0.7 1.6
Tripoli Africa
Ujung Pandang Asia 1.7 0.8 12.6
Ulsan Asia 0.6 0.6 0.2
0.8
Vancouver North America 5.9 1.1 1.2
Virginia Beach North America 2.6
1.8
Vishakhapatnam Asia 0.3 0.5
Washington D.C. North America 26.3 8.0 0.4
20.1
Wenzhou Asia 3.8 1.1 0.7 2.8
Xiamen Asia 3.0 1.1
0.4 6.5
Yantai Asia
Zhanjiang Asia 5.9 10.7 5.7 51.6
Chapter 5
166
Simple linear regression
A linear regression was performed on several social and physical factors
susceptible to influence the potentially available area for tidal wetlands creation.
The significant regression coefficients (p-value < 0.05) are presented in Table 5.2.
Table SI 5.3 Regression coefficients resulting from the linear regression testing the influence of the social and physical factors (Table 5.1) on the geographical variation of the size of the tidal wetlands in front of the coastal cities for a significance of 95 % (p-value < 0.05). The ‘X’ corresponds to a non-significant regression coefficient.
Dependent variable
Explanatory variables
Unit of increase
Regression coefficients (p-value < 0.05)
Physical Parameters
Latitude 1 ° X
Distance between the open sea and the city 1 km
1.018
Coastline Length 1.003
Area Below Mean High Tide
1 km²
1.005
Shallow Water Area (depth > -100 m) X
Deep Water Area (depth < -100 m) X
Social Parameters
GDP Per Capita of the Country 1 US$ X
Short-distance population density 1 inhabitants
/km²
0.999
Intermediate-distance population density X
Long-distance population density X
The results can be interpreted as follow; one unit of increase (see Table SI 5.3) of
the explanatory variable is generating a multiplication of the surface area
potentially available for tidal wetlands creation by a factor corresponding to the
regression coefficient.
Wetlands creation
167
Examples of land reclamation
Historical land reclamation from the mouth of the Elbe estuary to the city of
Hamburg in Germany and comparison of the estimated land reclamation from our
analysis.
Figure SI 5.1 (A) Overview of the extent and period of construction of the embanked areas along the Elbe estuary adapted from Hansen (2015) and (B) representation of the urban, below mean high tide and potentially available for tidal wetlands creation areas in the likely pathway of storm surge propagation towards Hamburg according to our analysis
Chapter 5
168
Figure SI 5.2 (A) Representation of the urban expansion in the Pearl River delta and the city of Guangzhou via infrared-enhanced satellite images for the years 1979 and 2013. The red areas correspond to the delta vegetation, the blue areas to the water bodies and the grey areas to the urbanized land. From (H. Zhao et al., 2010) (B) representation of the areas urbanized, below mean high tide and potentially available for tidal wetlands creation in the likely pathway of storm surge propagation of the cities of Guangzhou, Shenzhen and Hong Kong according to our analysis.
Wetlands creation
169
Figure SI 5.3 (A) Representation of the land uses in for the city of Shanghai and the adjacent Yangtze river delta adapted from Haas et al. (2015) (B) representation of the areas urbanized, below mean high tide and potentially available for tidal wetlands creation in the likely pathway of storm surge propagation of the city of Shanghai according to our analysis.
171
CHAPTER 6 Synthesis
Chapter 6
172
The growing coastal populations and assets are increasingly put under threats of
coastal flooding and erosion risks (Von Glasow et al., 2013; Hallegatte et al., 2013;
Neumann et al., 2015) due to the effects induced by global climate change on sea
level rise and increasing storm activity (Bengtsson et al., 2006; Bernstein et al.,
2007; Webster et al., 2005). It is therefore necessary to set up sustainable and
effective coastal protection programs to mitigate the devastating consequences
that such hazards can have (Sutton-Grier et al., 2015; Temmerman et al., 2013).
One strategy that is gaining wide interest is the development of nature-based
solutions, consisting on the conservation, restoration or creation of coastal
ecosystems that have the abilities to mitigate flood risks associated with storm
surges, attenuate erosion risks from waves, and adapt to sea level rise by sediment
accumulation, and in addition provide other valuable ecosystem functions and
services (e.g. water purification, nursery for fishes and crustaceous, carbon
sequestration...) (Barbier et al., 2011; Gedan et al., 2011; Leonardi et al., 2018).
Nature-based strategies are then considered as sustainable, self-adaptive and cost-
effective solutions that can be established alone or often in combination with hard
engineering solutions (ecosystems in front of dikes), i.e. so-called hybrid
approaches (Buhl-Mortensen et al., 2017; Duarte et al., 2013; Temmerman et al.,
2013).
So far, the assessment of the contribution of coastal ecosystems to nature-based
storm surge flood risk mitigation is mainly based on local to regional scale studies
(e.g. Arkema et al., 2013; Das & Vincent, 2009; Krauss et al., 2009; McGee et al.,
2006; Stark et al., 2015). In this thesis we pursued the aim to upscale those
assessments from regional to global scales. With, for objective, on the one hand to
highlight the widespread presence of coastal ecosystems and their capacity to
attenuate storm surges as well as the possibility, under certain conditions, to
restore or create them for coastal flood and erosion risks mitigation and on the
other hand to compare the presence of the coastal ecosystems with the coastal
areas densely populated and exposed to storm surge flood risks. Furthermore,
with such global scale studies, we aimed to increase the awareness of the local
communities and policy-makers in the possibility to implement, at global scale,
nature-based strategies alone or in combination with hard engineering structures
for coastal flood and erosion risks mitigation. In such, we developed procedures to
identify the coastal plain areas and populations exposed to coastal flood risks via
flood pathways crossing through tidal wetlands, and hence that benefit from storm
surge flood risks mitigation provided by the tidal wetlands (Chapters 2 & 3); we
quantified the existing extent of different coastal ecosystems (salt marshes,
mangroves, seagrasses and coral reefs) in front of highly populated and flood-
Synthesis
173
exposed coastal cities (Chapter 4); and we estimated the potentially available
space for tidal wetlands restoration or creation in front of those cities (Chapter 5).
6.1 Nature-based mitigation of coastal flood risks
The contribution of tidal wetlands to storm surge mitigation was defined at the
delta scale (Chapter 2) and at the worldwide scale (Chapter 3), while the
potential for nature-based coastal flood risks mitigation was estimated at the city
scale (Chapter 4 & 5).
At the delta and worldwide scales, the results showed high variability in the
amount of coastal protection offered by the tidal wetlands against storm surge
flood risks, and suggests, as explored by Gedan et al. (2011), that even the smallest
tidal wetlands can provide a certain level of storm surge mitigation. The analysis
of the tidal wetlands’ protection to the delta coastal plains (Chapter 2) showed
that the three deltas with the largest percentage of flood-exposed surface area (>
80 %) and population (> 70 %) benefiting from storm surge mitigation are the
Mahakam (Indonesia), Niger (Nigeria) and Chao Phraya (Thailand) deltas. While,
at the global scale (Chapter 3), we found that about 30 % of the flood-exposed
coastal plain and 40 % of the flood-exposed population benefit from storm surge
mitigation by tidal wetlands.
Based on Chapter 2, we could observe that, on the one hand, scattered tidal
wetlands located along the main channels within a delta (e.g. Chao Phraya delta)
are contributing to a larger land area benefiting from storm surge mitigation, than
clustered tidal wetlands (e.g. Ganges-Brahmaputra delta) and on the other hand,
that tidal wetlands dissected by numerous and wide channels provide storm surge
mitigation to a lesser coastal plain area (e.g. Yangtze or Rhine deltas) than tidal
wetlands that are not so densely dissected by wide channels (e.g. Mississippi delta
or Niger delta). Similarly, the magnitude of storm surge flood risks mitigation, as
estimated in Chapter 2 by the length of the storm surge pathway that is crossing
through tidal wetlands, is the highest when large and continuous tidal wetlands
are present. This finding corroborates previous studies highlighting that the
continuity of the wetlands is a key factor in the storm surge mitigation capacity of
tidal wetlands (Loder et al., 2009; Phan et al., 2015; Smolders et al., 2015; Stark et
al., 2016; Zhang et al., 2012).
Low-lying coastal plains found in deltas, estuaries, bays and lagoons are the most
favourable environment for the development of extensive and continuous tidal
wetlands (Leonardi et al., 2018). In such, at the worldwide scale (Chapter 3), the
largest relative coastal plains (> 15 km² per kilometre of shoreline) benefiting
Chapter 6
174
from storm surge mitigation by tidal wetlands are in or close to deltas, estuaries,
bays and lagoons, and mainly located along the Northern European and Eastern
Asian coasts, e.g. Aiguillon bay (France), Rhine-Meuse-Scheldt delta (Belgium and
urban, agricultural, and industrial land use types, etc.
Additionally, the need of global (or quasi-global) scale datasets of coastal
ecosystems implied a limitation of the coastal ecosystems considered in our
analyses. Therefore we focused on salt marshes and mangrove forests in Chapters
2 and 3, as the rate of storm surge height attenuation is best studied in literature
for salt marshes and mangrove forests, while largely unquantified for other coastal
ecosystem types. In Chapters 4 and 5, we estimated the spatial extent of four
coastal ecosystem types in front of flood-exposed coastal cities, namely salt
marshes, mangroves forest, seagrass beds and coral reefs. Other ecosystems such
as dunes, kelp beds, oyster beds, and others that may contribute to coastal risk
mitigation, even though they are known to provide a certain level of nature-based
coastal protection, couldn’t be incorporated in the analysis due to the lack of global
data on the worldwide spatial distribution of these ecosystem types (Beck et al.,
2017; Hanley et al., 2014; Hoggart et al., 2015; Narayan et al., 2016; Reguero et al.,
2014). In such, accounting for those coastal ecosystems in the GIS model of
Chapters 2 and 3 would modify the results. By highlighting, in addition to the
areas benefiting from the influence of tidal wetlands, the areas benefiting from the
presence of the other coastal and marine ecosystems as well as the areas
benefiting from the presence of multiple ecosystems that together provide higher
nature-based coastal flood risks mitigation (Guannel et al., 2016). Including more
coastal ecosystems in the cities’ analyses in Chapters 4 and 5 could lead to an
increase of the hotspots for nature-based coastal flood and erosion risks
mitigation.
The combination of improved knowledge on the values of storm surge attenuation
rates with an increased number of coastal ecosystems and datasets of higher
resolution (e.g. better delineation of the channels and wetlands, increased
information on the ecosystems vegetation density, height or structure...) would, in
the case of our GIS model, allow a more accurate prediction of both the landward
propagation of storm surges and the flood-exposed coastal plains and populations
benefiting from storm surge mitigation by coastal ecosystems.
Nevertheless, our global GIS model assessing the contribution of tidal wetlands to
coastal flood risks mitigation (Chapters 2 and 3), despite being of lower
resolution, for a selected number of coastal ecosystems and not simulating the full
Chapter 6
178
complexity of the hydrodynamic processes involved in storm surge propagation,
provides insights on the location and magnitude of storm surge mitigation by tidal
wetlands from regional to global scale. Its major advantages are to be globally
applicable, based on global or regional datasets and computationally much less
demanding than hydrodynamic models. Such quasi-global and global scale
analyses, by opposition to more local studies, allow the comparison of the
worldwide coastlines and selected deltas. Moreover, it defines and highlights
hotspots for nature-based storm surge flood risks mitigation on a global scale,
even for countries or world areas where such projects are not considered.
The GIS based procedures of Chapters 4 and 5 are on the other hand giving
estimations of the potential for nature-based coastal flood and erosion risks
mitigation in front of coastal cities corroborating the literature (e.g. Airoldi & Beck,
2007; Haas et al., 2015; Hansen, 2015; Zhao et al., 2010). The insights allow the
comparison of the different cities over the world in terms of existing coastal
ecosystems for nature-based storm surge flood risk mitigation and in terms of
potentially available space for tidal wetlands development to enhance the nature-
based storm surge mitigation. However, the estimates for tidal wetlands
restoration or creation are theoretical, which implies to account for other factors
such as the current land use or the socio-economic situation to fully determine the
possibility to restore or create wetlands in the highlighted areas. The current land
use involves parameters such as the soil elevation and hydrodynamic regime, the
soil pollution due to human use and the consequence of the release of those
pollutants, or the soil properties like the soil compaction that influences the
groundwater flow for example. Whilst the socio-economic situation may hamper
the restoration or creation of ecosystems due to the non-support or the
disapproval of such nature-based projects by the local communities that may be
willing to implement nature-based strategies, but not at their expenses (e.g. loss of
land, livelihood...) (Goeldner-Gianella, 2008; Temmerman et al., 2013). In addition,
the knowledge on ecosystem restoration is growing, and mostly highlights the
unique character of each project (Balke & Friess, 2016; Elliott et al., 2016).
Consequently, each project of wetland restoration or creation needs to rely on site-
specific, in depth scientific studies on the capacity of the wetlands to develop in
the delineated areas, depending on the hydrodynamic features, the
geomorphology, the sediment types, the tidal regime..., but also on an inclusive
analysis of the socio-economic situation of the area.
Additionally, the effects of global climate change on the coastal ecosystems for the
next centuries are not accounted for in the analyses. It is complex to define the
ecological effects of climate change as they are linked to different climate drivers,
Synthesis
179
the biotic and environmental conditions as well as to the anthropogenic activities
(Nicholls et al., 2018; Saunders et al., 2014). However, several studies highlight
that wetlands could suffer from sea level rise, with estimates, including
geomorphological and socio-economic feedbacks, giving up to 30 % of wetlands
loss by 2100, other studies estimate the loss to up to 90 % (Schuerch et al., 2018).
With as major factors for the resilience of the wetlands the sediment accretion for
vertical and lateral expansion and the accommodation space, to avoid the coastal
squeeze (Schuerch et al., 2018). Seagrasses on the other hand if exposed to the
changing climate conditions will have higher rates of photosynthesis, carbon
sequestration and growth, the latter, could lead to a long-term vulnerability of the
plants to storm conditions as their growth can modify their biomechanical
properties (De los Santos et al., 2017; Nicholls et al., 2018; O’Brien et al., 2017).
The coral reefs are expected to suffer from the warming and acidification of the
ocean (decreasing pH), yet it is not clear what would be their resilience to those
changes (Hoegh-Guldberg et al., 2007; Van Hooidonk et al., 2016; Nicholls et al.,
2018). Overall, global change over the next centuries is then expected to influence
the different coastal ecosystems and the changes in the environmental conditions
should be taken into account when planning for nature-based or hybrid coastal
protection strategies.
Overall, our global scale analyses emphasize the presence of coastal ecosystems
over the world’s coastlines and the possibility to account for them in coastal
planning strategies. And, the results presented in this thesis are then to be seen as
a step towards a better consideration of the benefits and value of the coastal
ecosystems for coastal flood and erosion risks reduction that could reach the
public and policy-makers.
Future research on the contribution of coastal ecosystems for coastal flood and
erosion risks mitigation should have two main objectives. Firstly, in order to
widen our understanding of nature-based coastal flood and erosion risks
mitigation, we should learn more about the mechanisms and processes behind the
attenuation of storm surges by coastal ecosystems while accounting for the
influence of the ecosystem structure and local landscape characteristics. This can
be done via in-situ monitoring of storm surge propagation over the coastal plain,
as it was done for several studies (e.g. Krauss et al., 2009; Stark et al., 2015) or via
the development of hydrodynamic models based on the insights resulting from the
previously mentioned field observations (e.g. Stark et al., 2016; Zhang et al., 2012).
Although an increasing number of such local-scale observational and modelling
studies is being published, the case studies are to be broadened to include a
Chapter 6
180
maximum of coastal ecosystem types, coastal bio-geomorphic settings, and storm
surge conditions.
Secondly, the definition, based on our GIS models and procedures, of hotspots for
nature-based coastal flood and erosion risks mitigation at the delta, city or global
scale provides inputs on where nature-based strategies could be applied.
Subsequently, it determines where local to regional studies on the implementation
of nature-based or hybrid coastal protection structures are needed. Those studies
should rely on local to regional high resolution datasets and account for the
landscape settings (including topography, bathymetry, geomorphology, slope of
the continental shelf...), the local storm surge characteristics (e.g. forward moving
speed, intensity, duration, direction...), the coastal land uses and their
characteristics (e.g. coastal ecosystems, agricultural fields, urban and populated
areas...) and the socio-economic situation. From there, the actual benefits of
coastal ecosystems for coastal protection would be refined for each considered
location and, where possible, coastal protection strategies could be implemented
to minimize the vulnerability of the coastal areas and communities to coastal flood
and erosion risks.
6.3 Implications for coastal zone management
The increasing threats exerted on coastal zones and associated populations and
assets, makes it imperative to create sustainable, cost-effective and efficient long-
term strategies to mitigate coastal flood risks (Adriana Gracia et al., 2018; Ma et al.,
2014). Nature-based strategies are increasingly proposed and implemented as
part of coastal risk reduction programs, for example along Northern European
coasts (Gardiner et al., 2007; SigmaPlan, 2017; Ysebaert et al., 2016) and in the UK
(Esteves, 2014; Pendle, 2013; Pethick, 2002), or in major coastal areas in the USA
(Boesch et al., 2006; Coastal Wetlands Planning Protection and Restoration Act
(CWPPRA), 1990; Day et al., 2007; Esteves, 2014; RESTORE, 2017). Nonetheless,
the historical trend of coastal ecosystem degradation and loss, and the
construction of hard coastal engineering structures with little attention for
adverse effects on the surrounding natural environment, are still problematic
(Hoeksema, 2007; Lotze et al., 2006; Valiela et al., 2009). Throughout this thesis,
we pursued the general aim to highlight the importance of conservation and
restoration/creation of coastal ecosystems and tidal wetlands as integrated
elements within programs for coastal flood risk mitigation. Whilst our
assessments are at regional to global scales, and consequently do not include
specific local conditions, they highlight that conservation and restoration/creation
Synthesis
181
of coastal ecosystems is beneficial to enhance the mitigation of storm surge and
erosion risks for coastal communities at several locations around the globe.
Here, in the scope of enhancing the coastal protection from storm surge and
erosion risks by accounting for nature-based strategies relying on four types of
coastal ecosystems (i.e. salt marshes, mangrove forests, seagrass meadows and
coral reefs), the focus was put on the environments where those ecosystems either
are present or can be developed. That does not imply, however, that coastal zones
not suitable for the establishment of the considered coastal ecosystems should not
rely on other and better suited ecosystems (e.g. dunes, oyster reefs...) to enhance
their resilience against coastal flood and erosion risks (Barbier et al., 2011; Cheong
et al., 2013; Grabowski et al., 2012). Sand dunes for example can mitigate storm
surges as they form geomorphic barriers against waves and storm surges, while
their above and below-ground vegetation increases resistance to the storm surge.
The above-ground vegetation will exert friction on the water column, in a similar
way as in tidal wetlands, while the below-ground vegetation provides sediment
stability and erosion resistance (Sigren et al., 2018; Silva et al., 2016).
Consequently, sand dunes are important elements of nature-based coastal
protection strategies such as in The Netherlands, in France or along the coastline
in Texas (USA) (Hanley et al., 2014; Rozé & Lemauviel, 2004; Sigren et al., 2014).
In the locations suitable for the development of the considered coastal ecosystems
(i.e. salt marshes, mangrove forests, seagrass beds and coral reefs), management
strategies should account for their presence or absence, but also to the protected
status of the area, such as international nature conservation legislations, e.g.
Ramsar or Natura2000 or national legislations. Along coastlines that were heavily
altered by human use over past and recent history (e.g. North European coasts),
the existing coastal ecosystems are often scarce or inexistent, limiting the
immediate application of nature-based coastal flood and erosion risks mitigation
(Chapters 4 and 5). For those coasts, which often rely on hard engineering
structures for coastal protection (e.g. dikes, dams...), ecosystems restoration or
creation should be stimulated as add-on to the existing engineered flood defence
structures, because vegetated ecosystems in front of these structures can
contribute significantly to reduced wave and storm surge impacts on the flood
defence structures, and hence can reduce the risk of flood disasters by failure of
these structures (Voortman et al., 2003; Vuik et al., 2016, 2018; van Wesenbeeck
et al., 2014). In those coasts, it is in addition crucial to include and gain the support
of the local communities in the development of such nature-based strategies. On
the other hand, many coastlines have large remaining ecosystems (e.g. Asian or
North and South American coasts) valuable for nature-based risks mitigation
Chapter 6
182
(Chapters 4 and 5), but the current widespread practice of ecosystem conversion
for human land use (e.g. aquaculture ponds, industrial areas...) makes them highly
vulnerable. Management practices should then invest in reducing the conversion
of coastal ecosystems to human land use, and in careful spatial planning of where
wetland conversion should be especially avoided in order to sustain their value for
nature-based storm surge risk mitigation. Nature-based strategies, although able
to reduce the risks of flooding, may not be sufficient on their own to fully protect
the low-lying coastal populations and associated assets. As such, hybrid strategies,
or the combination of nature-based strategies with other coastal protection
strategies, e.g. engineering structures, will often be of highest interest for coastal
protection.
As local communities are often unaware of the benefits and functions ecosystems
can provide, they may be reluctant to the development of nature-based strategies
(Esteves, 2014; French, 2006; Ledoux et al., 2004; The World Bank, 2017). A first
reason to this unwillingness of nature-based strategies is their skepticism about
the effectiveness of ecosystems for coastal flood and erosion risks mitigation,
while hard engineering structures make people feel safe (Esteves, 2014). Secondly,
the conversion of reclaimed land areas by previous generations (such as the
polders in Belgium and the Netherlands) into natural ecosystems is seen as a loss
of hard work and economic inputs (Goeldner-Gianella, 2008; Temmerman et al.,
2013). The integration of nature-based strategies into coastal planning
necessitates then to spread the scientific knowledge on the mechanisms and
processes behind the coastal flood and erosion mitigation capacities of the
ecosystems (Spalding, McIvor, et al., 2014; Sutton-Grier et al., 2015) as well as to
highlight the economic value of the ecosystem services such as coastal protection,
water filtration, habitat for fishes and crustaceous... (Barbier, 2015b; Dewsbury et
al., 2016; Himes-Cornell et al., 2018; Pascal et al., 2016) towards the local
communities, stakeholders, decision- and policy-makers. Both will then serve as
arguments to increase the support of the stakeholders and policy-makers for the
implementation of nature-based or hybrid strategies for coastal planning (De
Groot et al., 2013; Hill, 2015; Ledoux et al., 2004; Sutton-Grier et al., 2015).
Yet, if the coastal protection and other services provided by coastal ecosystems
along with their cost-efficiency and self-maintenance characteristics are
convincing arguments to maintain or create ecosystems, pilot studies on the
feasibility of nature-based strategies are necessary. The knowledge on the creation
or conservation of coastal ecosystems is growing and highlights the need to
understand the unique character of each coastal zone in regards to the
hydrodynamic, geomorphologic, ecological and socio-economic processes (Balke &
Synthesis
183
Friess, 2016; Simenstad et al., 2006). Successful nature-based strategies will then
rely on the capacity of the coastal managers to address the challenges created by
the different environmental and socio-economic situations. For example by
accounting for the needed environmental factors for ecosystems development (e.g.
favourable drainage, sediment supply...), by creating efficient population
settlement planning (e.g. to avoid coastal squeeze) and preventing the degradation
or modification of the environment that would affect the growth of the coastal
ecosystems (Lewis & Brown, 2014; The World Bank, 2017). In the case of
restoration projects, an extra vigilance is to be observed in defining the suitable
area for the establishment of the considered ecosystem, regarding the type of land
use for example; the conversion of agriculture or aquaculture fields into
ecosystems is often more feasible than converting urban or industrial areas.
Via the global analyses presented in this thesis, we demonstrated that despite the
various worldwide coastal environments, multiple coastal areas can already
benefit from nature-based storm surge mitigation and that this nature-based
mitigation could increase with the restoration or creation of coastal ecosystems.
Yet, only the combination of inclusive policy, long term management practices and
in depth scientific studies will lead to efficient nature-based coastal flood risk
mitigation strategies.
References
185
References
Adam, P. (2002). Saltmarshes in a time of change. Environmental Conservation, 29(01), 39–61. https://doi.org/10.1017/S0376892902000048
Adam, P. (2019). Salt Marsh Restoration. In Coastal Wetlands (pp. 817–861). Elsevier. https://doi.org/10.1016/B978-0-444-63893-9.00023-X
Adriana Gracia, C., Rangel-Buitrago, N., Oakley, J. A., & Williams, A. T. (2018). Use of ecosystems in coastal erosion management. Ocean and Coastal Management, 156, 277–289. https://doi.org/10.1016/j.ocecoaman.2017.07.009
Agardy, T., Alder, J., Dayton, P., Curran, S., Kitchingman, A., Wilson, M., et al. (2005). Coastal Systems. In Ecosystems and Human Well-being: Current Status and Trends, Volume 1 (pp. 513–550). Retrieved from http://www.millenniumassessment.org/documents/document.288.aspx.pdf
Airoldi, L., & Beck, M. (2007). Loss, Status and Trends for Coastal Marine Habitats of Europe. In R. N. Gibson, R. J. A. Atkinson, & J. D. M. Gordon (Eds.), Oceanography and Marine Biology An Annual Review, Volume 45 (1st Editio, pp. 345–405). https://doi.org/10.1201/9781420050943.ch7
Almeida, D., Neto, C., Esteves, L. S., & Costa, J. C. (2014). The impacts of land-use changes on the recovery of saltmarshes in Portugal. Ocean and Coastal Management, 92, 40–49. https://doi.org/10.1016/j.ocecoaman.2014.02.008
Alongi, D. M. (2008). Mangrove forests: Resilience, protection from tsunamis, and responses to global climate change. Estuarine, Coastal and Shelf Science, 76(1), 1–13. https://doi.org/10.1016/j.ecss.2007.08.024
Alongi, D. M. (2009). The Energetics of Mangrove Forests. Dordrecht: Springer Netherlands. https://doi.org/10.1007/978-1-4020-4271-3
An, S., Li, H., Guan, B., Zhou, C., Wang, Z., Deng, Z., et al. (2007). China’s natural wetlands: Past problems, current status, and future challenges. Ambio, 36(4), 335–342. https://doi.org/10.1579/0044-7447(2007)36[335:CNWPPC]2.0.CO;2
Ardón, M., Helton, A. M., Scheuerell, M. D., & Bernhardt, E. S. (2017). Fertilizer legacies meet saltwater incursion: challenges and constraints for coastal plain wetland restoration. Elem Sci Anth, 5(0), 41. https://doi.org/10.1525/elementa.236
Arkema, K. K., Guannel, G., Verutes, G., Wood, S. a., Guerry, A., Ruckelshaus, M., et al. (2013). Coastal habitats shield people and property from sea-level rise and storms. Nature Climate Change, 3(10), 913–918. https://doi.org/10.1038/nclimate1944
Auerbach, L. W., Goodbred Jr, S. L., Mondal, D. R., Wilson, C. a., Ahmed, K. R., Roy, K., et al. (2015). Flood risk of natural and embanked landscapes on the Ganges–Brahmaputra tidal delta plain. Nature Climate Change, 5(2), 153–157. https://doi.org/10.1038/nclimate2472
Balke, T., & Friess, D. A. (2016). Geomorphic knowledge for mangrove restoration: A pan-tropical categorization. Earth Surface Processes and Landforms, 41(2), 231–239. https://doi.org/10.1002/esp.3841
References
186
Barbier, E. B. (2007). Valuing ecosystem services as productive inputs. In Economic Policy (Vol. 22, pp. 177–229). https://doi.org/10.1111/j.1468-0327.2007.00174.x
Barbier, E. B. (2014). A global strategy for protecting vulnerable coastal populations. Science, 345(6202), 1250–1251.
Barbier, E. B. (2015a). Climate change impacts on rural poverty in low-elevation coastal zones. Estuarine, Coastal and Shelf Science, 165, A1–A13. https://doi.org/10.1016/j.ecss.2015.05.035
Barbier, E. B. (2015b). Valuing the storm protection service of estuarine and coastal ecosystems. Ecosystem Services, 11, 32–38. https://doi.org/10.1016/j.ecoser.2014.06.010
Barbier, E. B., & Barbier, B. E. B. (2014). A global strategy for protecting vulnerable coastal populations. Science, 345(6202), 1250–1251.
Barbier, E. B., Koch, E. W., Silliman, B. R., Hacker, S. D., Wolanski, E., Primavera, J., et al. (2008). Coastal ecosystem-based management with nonlinear in ecological functions and values. Science, 319(5861), 321–3. https://doi.org/10.1126/science.1150349
Barbier, E. B., Hacker, S. D., Kennedy, C., Koch, E. W., Stier, A. C., & Silliman, B. R. (2011). The value of estuarine and coastal ecosystem services. Ecological Monographs. https://doi.org/10.1890/10-1510.1
Barbier, E. B., Georgiou, I. Y., Enchelmeyer, B., & Reed, D. J. (2013). The Value of Wetlands in Protecting Southeast Louisiana from Hurricane Storm Surges. PLoS ONE, 8(3), 1–6. https://doi.org/10.1371/journal.pone.0058715
Beauchard, O., Jacobs, S., Cox, T. J. S., Maris, T., Vrebos, D., Van Braeckel, A., & Meire, P. (2011). A new technique for tidal habitat restoration: Evaluation of its hydrological potentials. Ecological Engineering, 37(11), 1849–1858. https://doi.org/10.1016/j.ecoleng.2011.06.010
Beavers, R. L., Babson, A. L., & Schupp, C. A. (Eds.). (2016). Coastal Adaptation Strategies Handbook (NPS 999/13). Washington, DC: National Park Service.
Beck, M. W., Losada, I. J., Reguero, B. G., Mendendez, P., & Burke, L. (2017). The global flood protection savings provided by coral reefs. Nature Communications, (2018). https://doi.org/10.1038/s41467-018-04568-z
Bengtsson, L., Hodges, K. I., & Roeckner, E. (2006). Storm Tracks and Climate Change. Journal of Climate, 19(15), 3518–3543. https://doi.org/10.1175/JCLI3815.1
Bernstein, L., Bosch, P., Canziani, O., Chen, Z., Christ, R., Davidson, O., et al. (2007). Climate Change 2007 : An Assessment of the Intergovernmental Panel on Climate Change. Change, 446(November), 12–17. https://doi.org/10.1256/004316502320517344
Bessell-Browne, P., Negri, A. P., Fisher, R., Clode, P. L., Duckworth, A., & Jones, R. (2017). Impacts of turbidity on corals: The relative importance of light limitation and suspended sediments. Marine Pollution Bulletin, 117(1–2), 161–170. https://doi.org/10.1016/j.marpolbul.2017.01.050
Blankespoor, B., Dasgupta, S., & Laplante, B. (2014). Sea-Level Rise and Coastal Wetlands. Ambio, 996–1005. https://doi.org/10.1007/s13280-014-0500-4
Boesch, D. F., Shabman, L., Antle, L. G., John W. Day, J., Dean, R. G., Galloway, G. E., et al. (2006). A New Framework for Planning the Future of Coastal Louisiana after the Hurricanes of 2005. Smithsonian.
References
187
Bolund, P., & Hunhammar, S. (1999). Ecosystem services in urban areas. Ecological Economics, 29(2), 293–301. https://doi.org/10.1016/S0921-8009(99)00013-0
Brandon, C. M., Woodruff, J. D., Orton, P. M., & Donnelly, J. P. (2016). Evidence for elevated coastal vulnerability following large-scale historical oyster bed harvesting. Earth Surface Processes and Landforms, 41(8), 1136–1143. https://doi.org/10.1002/esp.3931
Breaux, A., Farber, S., & Day, J. (1995). Using natural coastal wetlands systems for wastewater treatment: An economic benefit analysis. Journal of Environmental Management, 44(3), 285–291. https://doi.org/10.1006/jema.1995.0046
Bright, E. A., Coleman, P. R., Rose, A. N., & Urban, M. L. (2013). LandScan 2013. Oak Ridge, TN: Oak Ridge National Laboratory SE - July 1, 2012. Retrieved from http://www.ornl.gov/landscan
British Oceanographic Data Center. (2017). GEBCO 30 arc-second grid. Retrieved December 6, 2017, from https://www.gebco.net/data_and_products/gridded_bathymetry_data/gebco_30_second_grid/
Brown, B. E., & Dunne, R. P. (1988). The Environmental Impact of Coral Mining on Coral Reefs in the Maldives. Environmental Conservation, 15(2), 159–165. https://doi.org/10.1017/S0376892900028976
Buddemeier, R. W., & Smith, S. V. (1988). Coral Reefs predictions and suggestions for long-term research. Coral Reefs, 7, 51–56.
Buhl-Mortensen, L., Galparsoro, I., Vega Fernández, T., Johnson, K., D’Anna, G., Badalamenti, F., et al. (2017). Maritime ecosystem-based management in practice: Lessons learned from the application of a generic spatial planning framework in Europe. Marine Policy, 75, 174–186. https://doi.org/10.1016/j.marpol.2016.01.024
Bullock, J. M., Aronson, J., Newton, A. C., Pywell, R. F., & Rey-benayas, J. M. (2011). Restoration of ecosystem services and biodiversity : conflicts and opportunities, 26(10), 541–549. https://doi.org/10.1016/j.tree.2011.06.011
Carter, G. A., Otvos, E. G., Anderson, C. P., Funderburk, W. R., & Lucas, K. L. (2018). Catastrophic storm impact and gradual recovery on the Mississippi-Alabama barrier islands, 2005–2010: Changes in vegetated and total land area, and relationships of post-storm ecological communities with surface elevation. Geomorphology. https://doi.org/10.1016/j.geomorph.2018.08.020
Center for Hazards and Risk Research - CHRR - Columbia University, Center for International Earth Science Information Network - CIESIN - Columbia University, International Bank for Reconstruction and Development - The World Bank, and U. N. E. P. G. R. I. D. G.-U.-G. (2005). Global Cyclone Hazard Frequency and Distribution. Retrieved June 22, 2018, from http://dx.doi.org/10.7927/H4CZ353K
Chen, W., Wang, D., Huang, Y., Chen, L., Zhang, L., Wei, X., et al. (2017). Monitoring and analysis of coastal reclamation from 1995–2015 in Tianjin Binhai New Area, China. Scientific Reports, 7(1), 3850. https://doi.org/10.1038/s41598-017-04155-0
References
188
Cheong, S.-M. M., Silliman, B., Wong, P. P., van Wesenbeeck, B., Kim, C.-K. K., & Guannel, G. (2013). Coastal adaptation with ecological engineering. Nature Climate Change, 3(9), 787–791. https://doi.org/10.1038/nclimate1854
Chesapeake Bay Program. (2000). Chesapeake 2000. Manage, 1–13.
Chmura, G. L. (2003). Global carbon sequestration in tidal, saline wetland soils. Global Biogeochemical Cycles, 17(4). https://doi.org/10.1029/2002GB001917
Church, J. a., White, N. J., Konikow, L. F., Domingues, C. M., Graham Cogley, J., Rignot, E., et al. (2011). Revisiting the Earth’s sea-level and energy budgets from 1961 to 2008. Geophysical Research Letters, 38, 8. https://doi.org/10.1002/grl.50752
Coastal Protection and Restoration Authority of Louisiana. (2017). Louisiana’s Comprehensive Master Plan for a Sustainable Coast.
Coastal Wetlands Planning Protection and Restoration Act (CWPPRA). (n.d.). The Mississippi River Delta Basin. Retrieved August 17, 2017, from https://www.lacoast.gov/new/About/Basin_data/mr/Default.aspx
Coleman, J. M., & Huh, O. K. (2004). Major world deltas: A perspective from space. Retrieved from http://128.138.136.5/science/groups/wessman/projects/wdn/papers/colemanManuscript.pdf
Collins, J. M., & Roache, D. R. (2017). The 2016 North Atlantic hurricane season: A season of extremes. Geophysical Research Letters, 44(10), 5071–5077. https://doi.org/10.1002/2017GL073390
Van Coppenolle, R., Schwarz, C., & Temmerman, S. (2018). Contribution of Mangroves and Salt Marshes to Nature-Based Mitigation of Coastal Flood Risks in Major Deltas of the World. Estuaries and Coasts, 41(6), 1699–1711. https://doi.org/10.1007/s12237-018-0394-7
Costanza, R., Pérez-Maqueo, O., Martinez, M. L., Sutton, P., Anderson, S. J., & Mulder, K. (2008). The value of coastal wetlands for hurricane protection. Ambio, 37(4), 241–248. https://doi.org/10.1579/0044-7447(2008)37[241:tvocwf]2.0.co;2
Cui, B., Yang, Q., Yang, Z., & Zhang, K. (2009). Evaluating the ecological performance of wetland restoration in the Yellow River Delta, China. Ecological Engineering, 35(7), 1090–1103. https://doi.org/10.1016/j.ecoleng.2009.03.022
Dahdouh-Guebas, F., Jayatissa, L. P., Di Nitto, D., Bosire, J. O., Lo Seen, D., & Koedam, N. (2005). How effective were mangroves as a defence against the recent tsunami? Current Biology, 15(14), 1337–1338. https://doi.org/10.1016/j.cub.2005.07.025
Danielsen, F., Sørensen, M. K., Olwig, M. F., Selvam, V., Parish, F., Burgess, N. D., et al. (2005). The Asian Tsunami : A Protective Role for Coastal Vegetation. Science, 310(October), 643.
Darwiche-Criado, N., Sorando, R., Eismann, S. G., & Comín, F. A. (2017). Comparing Two Multi-Criteria Methods for Prioritizing Wetland Restoration and Creation Sites Based on Ecological, Biophysical and Socio-Economic Factors. Water Resources Management, 31(4), 1227–1241. https://doi.org/10.1007/s11269-017-1572-2
Das, S., & Vincent, J. R. (2009). Mangroves protected villages and reduced death toll during Indian super cyclone. Proceedings of the National Academy of Sciences of the United States of America, 106(18), 7357–7360. https://doi.org/10.1073/pnas.0810440106
Dasgupta, S., Laplante, B., Meisner, C., Wheeler, D., & Yan, J. (2009). The impact of sea level rise on developing countries: A comparative analysis. Climatic Change, 93(3–4), 379–388. https://doi.org/10.1007/s10584-008-9499-5
References
189
Dasgupta, S., Laplante, B., Murray, S., & Wheeler, D. (2011). Exposure of developing countries to sea-level rise and storm surges. Climatic Change, 106(4), 567–579. https://doi.org/10.1007/s10584-010-9959-6
Day, J. W., Boesch, D. F., Clairain, E. J., Kemp, G. P., Laska, S. B., Mitsch, W. J., et al. (2007). Restoration of the Mississippi Delta: lessons from Hurricanes Katrina and Rita. Science (New York, N.Y.), 315(5819), 1679–1684. https://doi.org/10.1126/science.1137030
Deb, M., & Ferreira, C. M. (2015). Potential impacts of the Sunderban mangrove degradation on future coastal flooding in Bangladesh. Journal of Hydro-Environment Research, 17, 30–46. https://doi.org/10.1016/j.jher.2016.11.005
Delgado, A. (2013). Guayaquil. Cities, 31, 515–532. https://doi.org/10.1016/j.cities.2011.11.001
Dewsbury, B. M., Bhat, M., & Fourqurean, J. W. (2016). A review of seagrass economic valuations: Gaps and progress in valuation approaches. Ecosystem Services, 18, 68–77. https://doi.org/10.1016/j.ecoser.2016.02.010
Dilley, M., Chen, R. S., Deichmann, U., Lerner-Lam, A. L., Arnold, M., Agwe, J., et al. (2005). Natural Disaster Hotspots A Global Risk Analysis. Earth Science. https://doi.org/10.1080/01944360902967228
Dobson, J. E., Bright, E. A., Coleman, P. R., Durfee, R. C., & Worley, B. A. (2000). LandScan: A global population database for estimating populations at risk. Photogrammetric Engineering and Remote Sensing, 66(7), 849–857. Retrieved from http://apps.webofknowledge.com.libezp.lib.lsu.edu/full_record.do?product=UA&search_mode=GeneralSearch&qid=3&SID=1D8vegIjNxsjNKm4nMs&page=1&doc=3
Duarte, C. M. (1991). Seagrass depth limits. Aquatic Botany, 40(4), 363–377. https://doi.org/10.1016/0304-3770(91)90081-F
Duarte, C. M., Losada, I. J., Hendriks, I. E., Mazarrasa, I., & Marbà, N. (2013). The role of coastal plant communities for climate change mitigation and adaptation. Nature Climate Change, 3(11), 961–968. https://doi.org/10.1038/nclimate1970
Duckworth, A., Giofre, N., & Jones, R. (2017). Coral morphology and sedimentation. Marine Pollution Bulletin, 125(1–2), 289–300. https://doi.org/10.1016/j.marpolbul.2017.08.036
Duke, N. C., Meynecke, J.-O., Dittmann, S., Ellison, A. M., Anger, K., Berger, U., et al. (2007). A World Without Mangroves? Science, 317(July), 41–43. https://doi.org/10.1126/science.317.5834.41b
Eertman, R. H. M., Kornman, B. A., Stikvoort, E., & Verbeek, H. (2002). Restoration of the sieperda tidal marsh in the Scheldt estuary, The Netherlands. Restoration Ecology, 10(3), 438–449. https://doi.org/10.1046/j.1526-100X.2002.01034.x
Elliott, M., Mander, L., Mazik, K., Simenstad, C., Valesini, F., Whitfield, A., & Wolanski, E. (2016). Ecoengineering with Ecohydrology: Successes and failures in estuarine restoration. Estuarine, Coastal and Shelf Science, 176, 12–35. https://doi.org/10.1016/j.ecss.2016.04.003
Ericson, J., Vorosmarty, C., Dingman, S., Ward, L., & Meybeck, M. (2006). Effective sea-level rise and deltas: Causes of change and human dimension implications. Global and Planetary Change, 50(1–2), 63–82. https://doi.org/10.1016/j.gloplacha.2005.07.004
References
190
Esteves, L. S. (2014). Managed Realignment : A Viable Long-Term Coastal Management Strategy? SpringerBriefs in Environmental Science. Dordrecht: Springer Netherlands. https://doi.org/10.1007/978-94-017-9029-1
FAO. (2007). Coastal protection in the aftermath of the Indian Ocean Tsunami: What role for forest and trees?
Feagin, R., Martinez, M. L., Mendoza-Gonzalez, G., & Costanza, R. (2010). Salt marsh zonal migration and ecosystem service change in response to global sea level rise: A case study from an urban region. Ecology and Society, 15(4). https://doi.org/14
Feagin, R. A., Mukherjee, N., Shanker, K., Baird, A. H., Cinner, J., Kerr, A. M., et al. (2010). Shelter from the storm? Use and misuse of coastal vegetation bioshields for managing natural disasters. Conservation Letters, 3(1), 1–11. https://doi.org/10.1111/j.1755-263X.2009.00087.x
Federal Geographic Data Committee. (2013). Classification of wetlands and deepwater habitats of the United States. FGDC-STD-004-2013. Second Edition, (August), 79. https://doi.org/FWS/OBS-79/31
Ferrario, F., Beck, M. W., Storlazzi, C. D., Micheli, F., Shepard, C. C., & Airoldi, L. (2014). The effectiveness of coral reefs for coastal hazard risk reduction and adaptation. Nature Communications, 5(May), 3794. https://doi.org/10.1038/ncomms4794
Firth, L. B., Thompson, R. C., Bohn, K., Abbiati, M., Airoldi, L., Bouma, T. J., et al. (2014). Between a rock and a hard place: Environmental and engineering considerations when designing coastal defence structures. Coastal Engineering, 87, 122–135. https://doi.org/10.1016/j.coastaleng.2013.10.015
Flather, R. A. (2001). Storm surges. Encyclopedia of Ocean Sciences, 2882–2892. https://doi.org/10.1016/B978-012374473-9.00124-7
Fonseca, M. S., & Cahalan, J. A. (1992). A preliminary evaluation of wave attenuation by four species of seagrass. Estuarine, Coastal and Shelf Science, 35(6), 565–576. https://doi.org/10.1016/S0272-7714(05)80039-3
Food and Agriculture Organization (FAO) of the United Nations. (2007). The World’s Mangroves 1980-2005. (FAO, Ed.). Rome. Retrieved from https://books.google.be/books?id=tLdlpOiuSmEC&pg=PA84&lpg=PA84&dq=978-92-5-105856-5&source=bl&ots=0yQSkC15h5&sig=4b8DAqZkt7MdXGHZ-J3q_7wrTeA&hl=fr&sa=X&ved=0ahUKEwiizqiL24naAhVM1RQKHbcYCfMQ6AEIQTAE#v=onepage&q=978-92-5-105856-5&f=false
Fourqurean, J. W., Duarte, C. M., Kennedy, H., Marbà, N., Holmer, M., Mateo, M. A., et al. (2012). Seagrass ecosystems as a globally significant carbon stock. Nature Geoscience, 5(7), 505–509. https://doi.org/10.1038/ngeo1477
French, P. W. (2006). Managed realignment - The developing story of a comparatively new approach to soft engineering. Estuarine, Coastal and Shelf Science, 67(3), 409–423. https://doi.org/10.1016/j.ecss.2005.11.035
Frihy, O., Fanos, A., Khafagy, A., & Aesha, K. (1996). Human impacts on the coastal zone of Hurghada, northern Red Sea, Egypt. Geo-Marine Letters, 1993(December 1993), 324–329.
Ganong, W. F. (1903). The Vegetation of the Bay of Fundy Salt and Diked Marshes: An Ecological Study. Botanical Gazette, 36(3), 161–186. https://doi.org/10.1086/676943
References
191
Garbutt, A., de Groot, A., Smit, C., & Pétillon, J. (2017). European salt marshes: ecology and conservation in a changing world. Journal of Coastal Conservation, 21(3), 405–408. https://doi.org/10.1007/s11852-017-0524-6
Garbutt, R. A., Reading, C. J., Wolters, M., Gray, A. J., & Rothery, P. (2006). Monitoring the development of intertidal habitats on former agricultural land after the managed realignment of coastal defences at Tollesbury, Essex, UK. Marine Pollution Bulletin, 53(1), 155–164. https://doi.org/10.1016/j.marpolbul.2005.09.015
Gardiner, S., Hanson, S., Nicholls, R. J., Zhang, Z., Jude, S., Jones, A., et al. (2007). The Habitats Directive, coastal habitats and climate change - case studies from the south coast of the U.K. In Proceedings of the International Conference on Coastal Management (pp. 1–10). Retrieved from http://eprints.soton.ac.uk/53142/
Gedan, K. B., Kirwan, M. L., Wolanski, E., Barbier, E. B., & Silliman, B. R. (2011). The present and future role of coastal wetland vegetation in protecting shorelines: Answering recent challenges to the paradigm. Climatic Change, 106(1), 7–29. https://doi.org/10.1007/s10584-010-0003-7
Gilman, E. L., Ellison, J., Duke, N. C., & Field, C. (2008). Threats to mangroves from climate change and adaptation options: A review. Aquatic Botany, 89(2), 237–250. https://doi.org/10.1016/j.aquabot.2007.12.009
Giosan, L., Syvitski, J. P. M., Constantinescu, S., & Day, J. (2014). Protect the world’s deltas. Nature, 516, 31–33.
Giri, C., Ochieng, E., Tieszen, L. L. L., Zhu, Z., Singh, A., Loveland, T., et al. (2011). Status and distribution of mangrove forests of the world using earth observation satellite data. Global Ecology and Biogeography, 20(2011), 154–159. https://doi.org/10.1111/j.1466-8238.2010.00584.x
Von Glasow, R., Jickells, T. D., Baklanov, A., Carmichael, G. R., Church, T. M., Gallardo, L., et al. (2013). Megacities and large urban agglomerations in the coastal zone: Interactions between atmosphere, land, and marine ecosystems. Ambio, 42(1), 13–28. https://doi.org/10.1007/s13280-012-0343-9
Goeldner-Gianella, L. (2008). Perceptions and attitudes towards de-polderisation in Europe : a comparison of five opinion surveys. Journal of Coastal R, 23(5), 1218–1230.
Grabowski, J. H., Brumbaugh, R. D., Conrad, R. F., Keeler, A. G., Opaluch, J. J., Peterson, C. H., et al. (2012). Economic Valuation of Ecosystem Services Provided by Oyster Reefs. BioScience, 62(10), 900–909. https://doi.org/10.1525/bio.2012.62.10.10
Green, E. P., & Short, F. (2003). World atlas of seagrasses. Prepared by the UIMEP World Conservation Monitoring Centre. University of California Press, Berkeley, USA. (Vol. 47). https://doi.org/10.1515/BOT.2004.029
Griggs, G. B. (2005). The Impacts of Coastal Armoring. Shore & Beach, 73(13), 13–22.
Grobicki, A., Chalmers, C., Jennings, E., Jones, T., & Peck, D. (2016). An Introduction to the RAMSAR Convention on Wetlands. Ramsar Convention Secretariat (7th ed. (p). Gland, Switzerland. Retrieved from https://www.mendeley.com/viewer/?fileId=658dea6f-a739-d995-65b0-e9d2a5a9f4c6&documentId=158ad1b6-0317-3653-ac6d-fb0139511534
References
192
De Groot, R. S., Blignaut, J., Van Der Ploeg, S., Aronson, J., Elmqvist, T., & Farley, J. (2013). Benefits of Investing in Ecosystem Restoration. Conservation Biology, 27(6), 1286–1293. https://doi.org/10.1111/cobi.12158
Guannel, G., Arkema, K., Ruggiero, P., & Verutes, G. (2016). The power of three: Coral reefs, seagrasses and mangroves protect coastal regions and increase their resilience. PLoS ONE, 11(7), 1–22. https://doi.org/10.1371/journal.pone.0158094
Guzmán, J. M., Martine, G., McGranahan, G., Schensul, D., & Tacoli, C. (2009). Population Dynamics and Climate Change. (J. M. Guzmán, G. Martine, G. McGranahan, D. Schensul, & C. Tacoli, Eds.).
Haas, J., Furberg, D., & Ban, Y. (2015). Satellite monitoring of urbanization and environmental impacts—A comparison of Stockholm and Shanghai. International Journal of Applied Earth Observation and Geoinformation, 38, 138–149. https://doi.org/10.1016/j.jag.2014.12.008
de Haas, T., Pierik, H. J., van der Spek, A. J. F., Cohen, K. M., van Maanen, B., & Kleinhans, M. G. (2018). Holocene evolution of tidal systems in The Netherlands: Effects of rivers, coastal boundary conditions, eco-engineering species, inherited relief and human interference. Earth-Science Reviews, 177(October 2017), 139–163. https://doi.org/10.1016/j.earscirev.2017.10.006
Haddad, J., Lawler, S., & Ferreira, C. M. (2016). Assessing the relevance of wetlands for storm surge protection: a coupled hydrodynamic and geospatial framework. Natural Hazards, 80(2), 839–861. https://doi.org/10.1007/s11069-015-2000-7
Hallegatte, S., Green, C., Nicholls, R. J., & Corfee-Morlot, J. (2013). Future flood losses in major coastal cities. Nature Climate Change, 3(9), 802–806. https://doi.org/10.1038/nclimate1979
Haltiner, J. B. ., Boyer, K. E., Williams, G. D., Callaway, J. C., & Zedler, J. (1997). Influence of physical processes on the design, functioning and evolution of restored tidal wetlands in California (USA). Wetlands Ecology and Management, 4(2), 73–91. https://doi.org/10.1007/BF01876230
Hamburg Port Authority. (2006). Concept for a sustainable development of the Tidal Elbe River as an artery of the metropolitan region Hamburg and beyond, 20.
Hanley, M. E., Hoggart, S. P. G., Simmonds, D. J., Bichot, A., Colangelo, M. A., Bozzeda, F., et al. (2014). Shifting sands? Coastal protection by sand banks, beaches and dunes. Coastal Engineering, 87, 136–146. https://doi.org/10.1016/j.coastaleng.2013.10.020
Hansen, K. (2015). Ecosystem functions of tidal marsh soils of the Elbe estuary. University of Hamburg.
Hanson, S., Nicholls, R., Ranger, N., Hallegatte, S., Corfee-Morlot, J., Herweijer, C., et al. (2011). A global ranking of port cities with high exposure to climate extremes. Climatic Change, 104(1), 89–111. https://doi.org/10.1007/s10584-010-9977-4
Harris, D. L., Rovere, A., Casella, E., Power, H., Canavesio, R., Collin, A., et al. (2018). Coral reef structural complexity provides important coastal protection from waves under rising sea levels. Science Advances, 4(2), 1–8. https://doi.org/10.1126/sciadv.aao4350
Hartman, B. D., & Cleveland, D. A. (2018). The socioeconomic factors that facilitate or constrain restoration management: Watershed rehabilitation and wet meadow (bofedal) restoration in the Bolivian Andes. Journal of Environmental Management, 209, 93–104. https://doi.org/10.1016/j.jenvman.2017.12.025
References
193
Hatvany, M. G. (2003). Marshlands: Four centuries of environmental change in the shores of the St. Lawrence. Sainte-Foy: Les Presses de l’Université Laval. Retrieved from https://books.google.be/books?hl=fr&lr=&id=-Hk9eDMLFFkC&oi=fnd&pg=PR17&dq=Marshlands:+Four+centuries+of+environmental+change++in++the++shores++of++the++St.++Lawrence.&ots=g8KIwhVDaK&sig=NPJNqp9LCNN_io912s7jJpcCTB8#v=onepage&q&f=false
He, Y., & Zhang, M. (2001). Study on wetland loss and its reasons in China. Chinese Geographical Science, 11(3), 241–245. https://doi.org/10.1007/s13398-014-0173-7.2
Heap, A., Bryce, S., Ryan, D., Radke, L., Smith, C., Smith, R., et al. (2001). Australian estuaries & coastal waterways: a geoscience perspective for improved and integrated resource management. Australian Geological Survey Organisation. Retrieved from http://www.ga.gov.au/metadata-gateway/metadata/record/gcat_36203
Hill, K. (2015). Coastal infrastructure: A typology for the next century of adaptation to sea-level rise. Frontiers in Ecology and the Environment, 13(9), 468–476. https://doi.org/10.1890/150088
Himes-Cornell, A., Pendleton, L., & Atiyah, P. (2018). Valuing ecosystem services from blue forests: A systematic review of the valuation of salt marshes, sea grass beds and mangrove forests. Ecosystem Services, 30, 36–48. https://doi.org/10.1016/j.ecoser.2018.01.006
Hinkel, J., Lincke, D., Vafeidis, A. T., Perrette, M., Nicholls, R. J., Tol, R. S. J., et al. (2014). Coastal flood damage and adaptation costs under 21st century sea-level rise. Proceedings of the National Academy of Sciences of the United States of America, 66(9), 3292–7. https://doi.org/10.1002/elsc.201300165
Hobbs, R. J., Higgs, E., & Harris, J. A. (2009). Novel ecosystems : implications for conservation and restoration, (August), 599–605. https://doi.org/10.1016/j.tree.2009.05.012
Hoegh-Guldberg, O., Mumby, P. J., Hooten, A. J., Steneck, R. S., Greenfield, P., Gomez, E., et al. (2007). Coral Reefs Under Rapid Climate Change and Ocean Acidification. Science, 318(5857), 1737–1742. https://doi.org/10.1126/science.1152509
Hoeksema, R. J. (2007). Three stages in the history of land reclamation in the Netherlands. Irrigation and Drainage, 56(S1), S113–S126. https://doi.org/10.1002/ird.340
Hoggart, S., Hawkins, S. J., Bohn, K., Airoldi, L., van Belzen, J., Bichot, A., et al. (2015). Ecological Approaches to Coastal Risk Mitigation. Coastal Risk Management in a Changing Climate. https://doi.org/10.1016/B978-0-12-397310-8.00004-X
Hong, P. N. (2001). Reforestation of mangroves after severe impacts of herbicides during the Vietnam war: the case of Can Gio. FAO Unasylva, 207(52), 57–60. Retrieved from http://www.fao.org/docrep/004/y2795f/y2795f11.htm#o
Van Hooidonk, R., Maynard, J., Tamelander, J., Gove, J., Ahmadia, G., Raymundo, L., et al. (2016). Local-scale projections of coral reef futures and implications of the Paris Agreement. Scientific Reports, 6(December), 1–8. https://doi.org/10.1038/srep39666
Hu, K., Chen, Q., & Wang, H. (2015). A numerical study of vegetation impact on reducing storm surge by wetlands in a semi-enclosed estuary. Coastal Engineering, 95, 66–76. https://doi.org/10.1016/j.coastaleng.2014.09.008
References
194
Huguet, J. R., Bertin, X., & Arnaud, G. (2017). Managed realignment to mitigate storm-induced flooding: A case study in La Faute-sur-mer, France. Coastal Engineering, (December 2016), 1–9. https://doi.org/10.1016/j.coastaleng.2017.08.010
IPCC. (2007). Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. In S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, K. B. Averyt, et al. (Eds.), Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, United Kingdom and New York, NY, USA: Cambridge University Press. Retrieved from https://www.ipcc.unibe.ch/publications/wg1-ar4/faq/docs/AR4WG1_FAQ-Brochure_LoRes.pdf
IPCC. (2013). Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press (Vol. AR5). Cambridge, United Kingdom and New York, NY, USA.
Jiang, T. ting, Pan, J. fen, Pu, X. M., Wang, B., & Pan, J. J. (2015). Current status of coastal wetlands in China: Degradation, restoration, and future management. Estuarine, Coastal and Shelf Science, 164, 265–275. https://doi.org/10.1016/j.ecss.2015.07.046
Käkönen, M. (2008). Mekong Delta at the crossroads: more control or adaptation? Ambio, 37(3), 205–212. https://doi.org/10.1579/0044-7447(2008)37[205:MDATCM]2.0.CO;2
King, S. E., & Lester, J. N. (1995). The value of salt marsh as a sea defence. Marine Pollution Bulletin, 30(3), 180–189. https://doi.org/10.1016/0025-326X(94)00173-7
Kirwan, M. L., & Megonigal, J. P. (2013). Tidal wetland stability in the face of human impacts and sea-level rise. Nature, 504(7478), 53–60. https://doi.org/10.1038/nature12856
Kirwan, M. L., Guntenspergen, G. R., D’Alpaos, A., Morris, J. T., Mudd, S. M., & Temmerman, S. (2010). Limits on the adaptability of coastal marshes to rising sea level. Geophysical Research Letters, 37(23), 1–5. https://doi.org/10.1029/2010GL045489
Kirwan, M. L., Temmerman, S., Skeehan, E. E., Guntenspergen, G. R., & Faghe, S. (2016). Overestimation of marsh vulnerability to sea level rise. Nature Climate Change, 6(3), 253–260. https://doi.org/10.1038/nclimate2909
Knapp, K. R., Kruk, M. C., Levinson, D. H., Diamond, H. J., & Neumann, C. J. (2010). The International Best Track Archive for Climate Stewardship (IBTrACS). Bulletin of the American Meteorological Society, 91(3), 363–376. https://doi.org/10.1175/2009BAMS2755.1
Knutson, T. R. (2014). Tropical Cyclones and Hurricanes: Tropical Cyclones and Climate Change. Encyclopedia of Atmospheric Sciences: Second Edition (Second Edi, Vol. 6). Elsevier. https://doi.org/10.1016/B978-0-12-382225-3.00508-9
Knutson, T. R., McBride, J. L., Chan, J., Emanuel, K., Holland, G., Landsea, C., et al. (2010). Tropical cyclones and climate change. Nature Geoscience, 3(February), 157–163. https://doi.org/10.1038/ngeo779
Kobashi, D., & Mazda, Y. (2005). Tidal flow in riverine-type mangroves. Wetlands Ecology and Management, 13(6), 615–619. https://doi.org/10.1007/s11273-004-3481-4
References
195
Koch, E. W., Ackerman, J. D., Verduin, J., & Kuelen, M. Van. (2006). Fluid Dynamics in Seagrass Ecology—from Molecules to Ecosystems. Seagrasses: Biology, Ecology and Conservation, 193–225. Retrieved from http://link.springer.com/chapter/10.1007/978-1-4020-2983-7_8%5Cnhttp://books.google.com/books?id=quMJfqMSFGAC&pg=PR7&lpg=PR7&dq=Fluid+dynamics+in+seagrass+ecology:+from+molecules+to+ecosystems&source=bl&ots=-ZROD5r85p&sig=TW4jGM91lqAXzUJ5H2G2h8harZY&hl=en&s
Koch, E. W., Barbier, E. B., Silliman, B. R., Reed, D. J., Perillo, G. M. E., Hacker, S. D., et al. (2009). Non-linearity in ecosystem services: Temporal and spatial variability in coastal protection. Frontiers in Ecology and the Environment, 7(1), 29–37. https://doi.org/10.1890/080126
Komar, P. D. (1998). Wave erosion of a massive artificial coastal landslide. Earth Surface Processes and Landforms, 23(5), 415–428. https://doi.org/10.1002/(SICI)1096-9837(199805)23:5<415::AID-ESP855>3.0.CO;2-T
Krauss, K. W., Doyle, T. J. T. W., Doyle, T. J. T. W., Swarzenski, C. M., From, A. S., Day, R. H., & Conner, W. H. (2009). Water level observations in mangrove swamps during two hurricanes in Florida. Wetlands, 29(1), 142–149. https://doi.org/10.1672/07-232.1
Krauss, K. W., McKee, K. L., Lovelock, C. E., Cahoon, D. R., Saintilan, N., Reef, R., & Chen, L. (2014). How mangrove forests adjust to rising sea level. The New Phytologist, 202(1), 19–34. https://doi.org/10.1111/nph.12605
Kron, W. (2013). Coasts: The high-risk areas of the world. Natural Hazards, 66(3), 1363–1382. https://doi.org/10.1007/s11069-012-0215-4
Landry, C. E., & Liu, H. (2009). A semi-parametric estimator for revealed and stated preference data-An application to recreational beach visitation. Journal of Environmental Economics and Management, 57(2), 205–218. https://doi.org/10.1016/j.jeem.2008.05.002
Lawler, S., Haddad, J., & Ferreira, C. M. (2016). Sensitivity considerations and the impact of spatial scaling for storm surge modeling in wetlands of the Mid-Atlantic region. Ocean and Coastal Management, 134, 226–238. https://doi.org/10.1016/j.ocecoaman.2016.10.008
Lawrence, P. J., Smith, G. R., Sullivan, M. J. P., & Mossman, H. L. (2018). Restored saltmarshes lack the topographic diversity found in natural habitat. Ecological Engineering, 115(February), 58–66. https://doi.org/10.1016/j.ecoleng.2018.02.007
Ledoux, L., Cornell, S., O’Riordan, T., Harvey, R., & Banyard, L. (2004). Towards sustainable flood and coastal management: Identifying drivers of, and obstacles to, managed realignment. Working Paper - Centre for Social and Economic Research on the Global Environment, 22(1), 1–32. https://doi.org/10.1016/j.landusepol.2004.03.001
Lee, S.-M., Cho, Y.-C., & Lee, C.-S. (2012). Feasibility of seed bank for restoration of salt marsh: a case study around the Gwangyang Bay, southern Korea. Journal of Ecology and Field Biology, 35(2), 123–129. https://doi.org/10.5141/JEFB.2012.016
Lee, S. Y., Primavera, J. H., Dahdouh-Guebas, F., Mckee, K., Bosire, J. O., Cannicci, S., et al. (2014). Ecological role and services of tropical mangrove ecosystems: A reassessment. Global Ecology and Biogeography, 23(7), 726–743. https://doi.org/10.1111/geb.12155
References
196
Leonardi, N., Carnacina, I., Donatelli, C., Ganju, N. K., Plater, A. J., Schuerch, M., & Temmerman, S. (2018). Dynamic interactions between coastal storms and salt marshes: A review. Geomorphology, 301, 92–107. https://doi.org/10.1016/j.geomorph.2017.11.001
Lewis, R. R., & Brown, B. (2014). Ecological Mangrove Rehabilitation A field manual for practitioners, 275.
Li, M. S., & Lee, S. Y. (1997). Mangroves of China: A brief review. Forest Ecology and Management, 96(3), 241–259. https://doi.org/10.1016/S0378-1127(97)00054-6
Lichter, M., Vafeidis, A. T., Nicholls, R. J., & Kaiser, G. (2011). Exploring data-related uncertainties in analyses of land area and population in the “Low-Elevation Coastal Zone” (LECZ). Journal of Coastal Research, 27(4), 757–768. https://doi.org/10.2112/JCOASTRES-D-10-00072.1
Lim, Y.-K., Schubert, S. D., Kovach, R., Molod, A. M., & Pawson, S. (2018). The Roles of Climate Change and Climate Variability in the 2017 Atlantic Hurricane Season Global Modeling & Assimilation O ffice, 2018. Retrieved from https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20180002568.pdf
Liu, H., Zhang, K., Li, Y., & Xie, L. (2013). Numerical study of the sensitivity of mangroves in reducing storm surge and flooding to hurricane characteristics in southern Florida. Continental Shelf Research, 64, 51–65. https://doi.org/10.1016/j.csr.2013.05.015
Loder, N. M., Irish, J. L., Cialone, M. a., & Wamsley, T. V. (2009). Sensitivity of hurricane surge to morphological parameters of coastal wetlands. Estuarine, Coastal and Shelf Science, 84(4), 625–636. https://doi.org/10.1016/j.ecss.2009.07.036
De los Santos, C. B., Godbold, J. A., & Solan, M. (2017). Short-term growth and biomechanical responses of the temperate seagrass Cymodocea nodosa to CO2enrichment. Marine Ecology Progress Series, 572, 91–102. https://doi.org/10.3354/meps12153
Lotze, H. K., Lenihan, H. S., Bourque, B. J., Bradbury, R. H., Richard G. Cooke, Kay, M. C., et al. (2006). Depletion, Degradation, and Recovery Potential of Estuaries and Coastal Seas. Science, 312(5781), 1806–1809. https://doi.org/10.1126/science.1128035
Lovelace, J. K. (1994). Storm-Tide Elevations Produced By Hurricane Andrew Along the Louisiana Coast, August 25-27,1992. U.S. Geological Survey Open-File Report 94-371. Prepared in Cooperation with the Federal Emergency Management Agency.
Lovelock, C. E., Cahoon, D. R., Friess, D. A., Guntenspergen, G. R., Krauss, K. W., Reef, R., et al. (2015). The vulnerability of Indo-Pacific mangrove forests to sea-level rise. Nature. https://doi.org/10.1038/nature15538
Ma, Z., Melville, D. S., Liu, J., Chen, Y., Yang, H., Ren, W., et al. (2014). ECOSYSTEMS MANAGEMENT Rethinking China’s new great wall. Science, 346(6212), 912–914. https://doi.org/10.1126/science.1257258
Maddison, A. (2001). The World Economy: A Millennial Perspective. Oecd, 384. https://doi.org/10.1787/9789264189980-en
Marchand, M. (2008). Mangrove restoration in Vietnam, 34.
Maris, T., Cox, T., Temmerman, S., De Vleeschauwer, P., Van Damme, S., De Mulder, T., et al. (2007). Tuning the tide: Creating ecological conditions for tidal marsh development in a flood control area. In Hydrobiologia (Vol. 588, pp. 31–43). https://doi.org/10.1007/s10750-007-0650-5
References
197
Marois, D. E., & Mitsch, W. J. (2015). Coastal protection from tsunamis and cyclones provided by mangrove wetlands – a review. International Journal of Biodiversity Science, Ecosystem Services & Management, 11(June 2015), 1–13. https://doi.org/10.1080/21513732.2014.997292
Marsooli, R., Orton, P. M., Georgas, N., & Blumberg, A. F. (2016). Three-dimensional hydrodynamic modeling of coastal flood mitigation by wetlands. Coastal Engineering, 111, 83–94. https://doi.org/10.1016/j.coastaleng.2016.01.012
Masters, J. (2012). A Detailed View of the Storm Surge: Comparing Katrina to Camille. Retrieved from https://www.wunderground.com/hurricane/surge_details.asp
Mathieu, L. F., Langford, I. H., & Kenyon, W. (2003). Valuing marine parks in a developing country: A case study of the Seychelles. Environment and Development Economics, 8(2), 373–390. https://doi.org/10.1017/S1355770X03000196
Mattocks, C., & Forbes, C. (2008). A real-time, event-triggered storm surge forecasting system for the state of North Carolina. Ocean Modelling, 25(3–4), 95–119. https://doi.org/10.1016/j.ocemod.2008.06.008
Mazda, Y., Wolanski, E., King, B., Sase, A., Ohtsuka, D., & Magi, M. (1997). Drag force due to vegetation in mangrove swamps. Mangroves and Salt Marshes, 1, 193–199. https://doi.org/10.1023/A:1009949411068
Mazda, Y., Magi, M., Ikeda, Y., Kurokawa, T., & Asano, T. (2006). Wave reduction in a mangrove forest dominated by Sonneratia sp. Wetlands Ecology and Management, 14(4), 365–378. https://doi.org/10.1007/s11273-005-5388-0
McArthur, L. C., & Boland, J. W. (2006). The economic contribution of seagrass to secondary production in South Australia. Ecological Modelling, 196(1–2). https://doi.org/10.1016/j.ecolmodel.2006.02.030
McEvedy, C., & Jones, R. (1978). Atlas of World Population History. New York: NY:Facts on File.
McGee, B. D., Goree, B. B., Tollett, R. W., Woodward, B. K., & Kress, W. H. (2006). Hurrican Rita Surge Data, Southwestern Louisiana and Souteasthern Texas, September to November 2005. U.S. Geological Survey Data Series 220. Retrieved from http://pubs.usgs.gov/ds/2006/220/index.htm
Mcgranahan, G., Balk, D., & Anderson, B. (2006). Low coastal zone settlements, (59), 23–26.
McGranahan, G., Balk, D., & Anderson, B. (2007). The rising tide: assessing the risks of climate change and human settlements in low elevation coastal zones. Environment and Urbanization, 19(1), 17–37. https://doi.org/10.1177/0956247807076960
McIvor, A. L., Möller, I., Spencer, T., & Spalding, M. (2012). Reduction of Wind and Swell Waves by Mangroves. Natural Coastal Protection Series, 1–27. https://doi.org/ISSN 2050-7941.
McIvor, A. L., Spencer, T., Möller, I., & Spalding, M. (2012). Storm Surge Reduction by Mangroves. Natural Coastal Protection Series, 35. Retrieved from http://www.naturalcoastalprotection.org/documents/storm-surge-reduction-by-mangroves
McIvor, A. L., Spencer, T., & Möller, I. (2013). The response of mangrove soil surface elevation to sea level rise. Natural Coastal Protection Series, 1–59.
References
198
McLeod, E., Chmura, G. L., Bouillon, S., Salm, R., Björk, M., Duarte, C. M., et al. (2011). A blueprint for blue carbon: Toward an improved understanding of the role of vegetated coastal habitats in sequestering CO2. Frontiers in Ecology and the Environment, 9(10), 552–560. https://doi.org/10.1890/110004
Mcowen, C., Weatherdon, L., Bochove, J.-W., Sullivan, E., Blyth, S., Zockler, C., et al. (2017). A global map of saltmarshes. Biodiversity Data Journal, 5, e11764. https://doi.org/10.3897/BDJ.5.e11764
Meire, P., Dauwe, W., Maris, T., Peeters, P., Coen, L., Deschamps, M., et al. (2014). Sigma Plan Proves Efficiency. ECSA Bulletin, 62, 19–23.
Meng, W., Hu, B., He, M., Liu, B., Mo, X., Li, H., et al. (2017). Temporal-spatial variations and driving factors analysis of coastal reclamation in China. Estuarine, Coastal and Shelf Science, 191, 39–49. https://doi.org/10.1016/j.ecss.2017.04.008
MFF Pakistan. (2016). A Handbook on Pakistan’s Coastal and Marine Resources. Pakistan: MFF Pakistan.
Middleton, B. A. (2016). Differences in impacts of Hurricane Sandy on freshwater swamps on the Delmarva Peninsula, Mid-Atlantic Coast, USA. Ecological Engineering, 87, 62–70. https://doi.org/10.1016/j.ecoleng.2015.11.035
Millennium Ecosystem Assessment. (2005). ECOSYSTEMS AND HUMAN WELL-BEING: WETLANDS AND WATER Synthesis. (World Resources Institute, Ed.). Washington, DC.
Moller, I., Spencer, T., French, Leggett, D., & Dixon, M. (1999). Wave transformation over salt marshes: A field and numerical modelling study from north Norfolk, England, 411–426. https://doi.org/10.1006/ecss.1999.0509
Möller, I., Kudella, M., Rupprecht, F., Spencer, T., Paul, M., van Wesenbeeck, B. K., et al. (2014). Wave attenuation over coastal salt marshes under storm surge conditions. Nature Geoscience, 7(10), 727–731. https://doi.org/10.1038/ngeo2251
Morris, J. T., Sundareshwar, P. V., Nietch, C. T., Kjerfve, B., & Cahoon, D. R. (2002). Responses of Coastal Wetlands to Rising Sea Level. Ecology, 83(10), 2869. https://doi.org/10.2307/3072022
Morris, R. L., Konlechner, T. M., Ghisalberti, M., & Swearer, S. E. (2018). From grey to green: Efficacy of eco-engineering solutions for nature-based coastal defence. Global Change Biology, 24(5), 1827–1842. https://doi.org/10.1111/gcb.14063
Muis, S., Verlaan, M., Winsemius, H. C., Aerts, J. C. J. H., & Ward, P. J. (2016). A global reanalysis of storm surges and extreme sea levels. Nature Communications, 7(May), 11969. https://doi.org/10.1038/ncomms11969
Muis, S., Verlaan, M., Nicholls, R. J., Brown, S., Hinkel, J., Lincke, D., et al. (2016). An intercomparison of two datasets of extreme sea levels and resulting global flood risk. Earth’s Future. https://doi.org/10.1002/eft2.197
Mumby, P. J., & Steneck, R. S. (2018). Paradigm Lost: Dynamic Nutrients and Missing Detritus on Coral Reefs. BioScience, 68(7), 487–495. https://doi.org/10.1093/biosci/biy055
Murray, A. (2017). Natural Flood Management. Adopting ecosystem approaches to managing flood risk, (February). Retrieved from https://www.foe.ie/download/pdf/natural_flood_management_a_study_for_friends_of_the_earth_february_2017.pdf
References
199
Murray, N. J., Clemens, R. S., Phinn, S. R., Possingham, H. P., & Fuller, R. A. (2014). Tracking the rapid loss of tidal wetlands in the Yellow Sea. Frontiers in Ecology and the Environment, 12(5), 267–272. https://doi.org/10.1890/130260
Narayan, S., Beck, M. W., Reguero, B. G., Losada, I. J., Van Wesenbeeck, B., Pontee, N., et al. (2016). The effectiveness, costs and coastal protection benefits of natural and nature-based defences. PLoS ONE, 11(5), 1–17. https://doi.org/10.1371/journal.pone.0154735
Narayan, S., Beck, M. W., Wilson, P., Thomas, C. J., Guerrero, A., Shepard, C. C., et al. (2017). The Value of Coastal Wetlands for Flood Damage Reduction in the Northeastern USA. Scientific Reports, 7(1), 9463. https://doi.org/10.1038/s41598-017-09269-z
Neumann, B., Vafeidis, A. T., Zimmermann, J., & Nicholls, R. J. (2015). Future Coastal Population Growth and Exposure to Sea-Level Rise and Coastal Flooding - A Global Assessment. PLoS ONE, 10(3), e0118571. https://doi.org/10.1371/journal.pone.0118571
Newman, W. S. (1982). Isostatic adjustment. In Beaches and Coastal Geology (pp. 497–498). Boston, MA: Springer US. https://doi.org/10.1007/0-387-30843-1_244
Nicholls, R. J., & Cazenave, A. (2010). Sea Level Rise and Its Impact on Coastal Zones. Science, 328(2010), 1517–1520. https://doi.org/10.1126/science.1185782
Nicholls, R. J., & Small, C. (2002). Improved estimates of coastal population and exposure to hazards released. Eos, 8(2), 301–305. https://doi.org/10.1029/2002EO000216
Nicholls, R. J., Hanson, S., Herweijer, C., Patmore, N., Hallegatte, S., Corfee-Morlot, J., et al. (2007). Ranking of The World’s Cities Most Exposed to Coastal Flooding Today Executive Summary. Organisation for Economic Cooperation and Development.
Nicholls, R. J., Hanson, S., Herweijer, C., Patmore, N., Hallegatte, S., Corfee-Morlot, J., et al. (2008). Ranking port cities with high exposure and vulnerability to climate extremes: exposure estimates. OECD Environment Working Papers (Vol. 1). https://doi.org/10.1787/011766488208
Nicholls, R. J., Brown, S., Goodwin, P., Wahl, T., Lowe, J., Solan, M., et al. (2018). Stabilization of global temperature at 1.5°C and 2.0°C: implications for coastal areas. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 376(2119), 20160448. https://doi.org/10.1098/rsta.2016.0448
Niu, Z. G., Gong, P., Cheng, X., Guo, J. H., Wang, L., Huang, H. B., et al. (2009). Geographical characteristics of China’s wetlands derived from remotely sensed data. Science in China, Series D: Earth Sciences, 52(6), 723–738. https://doi.org/10.1007/s11430-009-0075-2
O’Brien, K. R., Waycott, M., Maxwell, P., Kendrick, G. A., Udy, J. W., Ferguson, A. J. P., et al. (2017). Seagrass ecosystem trajectory depends on the relative timescales of resistance, recovery and disturbance. Marine Pollution Bulletin, (September), 0–1. https://doi.org/10.1016/j.marpolbul.2017.09.006
Ondiviela, B., Losada, I. J., Lara, J. L., Maza, M., Galván, C., Bouma, T. J., & van Belzen, J. (2014). The role of seagrasses in coastal protection in a changing climate. Coastal Engineering, 87, 158–168. https://doi.org/10.1016/j.coastaleng.2013.11.005
References
200
Oosterlee, L., Cox, T. J. S., Vandenbruwaene, W., Maris, T., Temmerman, S., & Meire, P. (2018). Tidal Marsh Restoration Design Affects Feedbacks Between Inundation and Elevation Change. Estuaries and Coasts, 41(3), 613–625. https://doi.org/10.1007/s12237-017-0314-2
Parés-Ramos, I., Álvarez-Berríos, N., & Aide, T. (2013). Mapping Urbanization Dynamics in Major Cities of Colombia, Ecuador, Per, and Bolivia Using Night-Time Satellite Imagery. Land, 2(1), 37–59. https://doi.org/10.3390/land2010037
Pascal, N., Allenbach, M., Brathwaite, A., Burke, L., Le Port, G., & Clua, E. (2016). Economic valuation of coral reef ecosystem service of coastal protection: A pragmatic approach. Ecosystem Services, 21, 72–80. https://doi.org/10.1016/j.ecoser.2016.07.005
Passeri, D. L., Long, J. W., Plant, N. G., Bilskie, M. V., & Hagen, S. C. (2018). The influence of bed friction variability due to land cover on storm-driven barrier island morphodynamics. Coastal Engineering, 132(February 2017), 82–94. https://doi.org/10.1016/j.coastaleng.2017.11.005
Paul, B. K. (2009). Why relatively fewer people died? The case of Bangladesh’s cyclone sidr. Natural Hazards, 50(2), 289–304. https://doi.org/10.1007/s11069-008-9340-5
Pendle, M. (2013). Estuarine and coastal managed realignment sites in England selected case studies.
Pendleton, L., Donato, D. C., Murray, B. C., Crooks, S., Jenkins, W. A., Sifleet, S., et al. (2012). Estimating Global “Blue Carbon” Emissions from Conversion and Degradation of Vegetated Coastal Ecosystems. PLoS ONE, 7(9). https://doi.org/10.1371/journal.pone.0043542
Pennings, S. C., & Bertness, M. D. (2000). Salt Marsh Communities. In Marine Community Ecology (pp. 289–316). https://doi.org/1605352284
Perillo, G. M. E., Wolanski, E., Cahoon, D. R., & Brinson, M. M. (Eds.). (2009). Coastal wetlands : an integrated ecosystem approach (First edit). https://doi.org/10.1016/j.annfar.2004.08.007
Pethick, J. (2002). Estuarine and tidal wetland restoration in the United Kingdom: Policy versus practice. Restoration Ecology, 10(3), 431–437. https://doi.org/10.1046/j.1526-100X.2002.01033.x
Pethick, J., & Orford, J. D. (2013). Rapid rise in effective sea-level in southwest Bangladesh: Its causes and contemporary rates. Global and Planetary Change, 111, 237–245. https://doi.org/10.1016/j.gloplacha.2013.09.019
Phan, L. K., van Thiel de Vries, J. S. M., Stive, M. J. F., & de Vries, Jaap S M van Thiel, Stive, M. J. F. (2015). Coastal Mangrove Squeeze in the Mekong Delta. Journal of Coastal Research, 31(2), 233–243. https://doi.org/10.2112/JCOASTRES-D-14-00049.1
Pierik, H. J., Cohen, K. M., Vos, P. C., van der Spek, A. J. F., & Stouthamer, E. (2017). Late Holocene coastal-plain evolution of the Netherlands: the role of natural preconditions in human-induced sea ingressions. Proceedings of the Geologists’ Association, 128(2), 180–197. https://doi.org/10.1016/j.pgeola.2016.12.002
Pontee, N. (2013). Defining coastal squeeze: A discussion. Ocean and Coastal Management, 84, 1–4. https://doi.org/10.1016/j.ocecoaman.2013.07.010
Pranzini, E. (2018). Shore protection in Italy: From hard to soft engineering … and back. Ocean and Coastal Management, 156, 43–57. https://doi.org/10.1016/j.ocecoaman.2017.04.018
References
201
Pranzini, E., Wetzel, L., & Williams, A. T. (2015). Aspects of coastal erosion and protection in Europe. Journal of Coastal Conservation, 19(4), 445–459. https://doi.org/10.1007/s11852-015-0399-3
Principe, P. P., Bradley, P., Yee, S. H., Fisher, W. S., Johnson, E. D., Allen, P., & Campbell, D. E. (2012). Quantifying Coral Reef Ecosystem Services, EPA/600/R-(January), 158.
Rangel-Buitrago, N., Williams, A. T., & Anfuso, G. (2018). Hard protection structures as a principal coastal erosion management strategy along the Caribbean coast of Colombia. A chronicle of pitfalls. Ocean and Coastal Management, 156, 58–75. https://doi.org/10.1016/j.ocecoaman.2017.04.006
Rao, N. S., Carruthers, T. J. B., Anderson, P., Sivo, L., Saxby, T., Durbin, T., et al. (2013). An economic analysis of ecosystem-based adaptation and engineering options for climate change adaptation in Lami Town, Republic of the Fiji Islands. A technical report by the Secretariat of the Pacific Regional Environment Programme. Retrieved from http://ian.umces.edu/pdfs/ian_report_392.pdf
Reddy, S. M. W., Guannel, G., Griffin, R., Faries, J., Boucher, T., Thompson, M., et al. (2016). Evaluating the role of coastal habitats and sea-level rise in hurricane risk mitigation: An ecological economic assessment method and application to a business decision. Integrated Environmental Assessment and Management, 12(2), 328–344. https://doi.org/10.1002/ieam.1678
Rego, J. L., & Li, C. (2009). On the importance of the forward speed of hurricanes in storm surge forecasting: A numerical study. Geophysical Research Letters, 36(7), 1–5. https://doi.org/10.1029/2008GL036953
Reguero, B. G., Bresch, D. N., Beck, M. W., Calil, J., & Meliane, I. (2014). Coastal Risks, Naure-Based Defenses and the Economics of Adaptation: An Application in the Gulf of Mexico, USA. Coastal Engineering, 1–15. https://doi.org/http://dx.doi.org/10.9753/icce.v34.management.25
Reguero, B. G., Beck, M. W., Bresch, D. N., Calil, J., & Meliane, I. (2018). Comparing the cost effectiveness of nature- based and coastal adaptation : A case study from the Gulf Coast of the United States, 1–24. https://doi.org/10.17605/OSF.IO/D6R5U.
Reise, K. (2005). Coast of change: Habitat loss and transformations in the Wadden Sea. Helgoland Marine Research, 59(1), 9–21. https://doi.org/10.1007/s10152-004-0202-6
Resio, D. T., & Westerink, J. J. (2008). Modeling the physics of storm surges. Physics Today, 61(9), 33–38. https://doi.org/10.1063/1.2982120
RESTORE. (2017). Restoring the Mississippi River Delta. Retrieved from http://mississippiriverdelta.org/files/2017/11/MRD_PriorityProjectReport.pdf
Rodriguez, E., Morris, C. C., & Belz, J. J. (2006). A global assessment of the SRTM performance. Photogrammetric Engineering and Remote Sensing, 72(3), 249–260. https://doi.org/10.14358/PERS.72.3.249
Rovere, A., Stocchi, P., & Vacchi, M. (2016). Eustatic and Relative Sea Level Changes. Current Climate Change Reports, 2(4), 221–231. https://doi.org/10.1007/s40641-016-0045-7
Rozé, F., & Lemauviel, S. (2004). Sand dune restoration in North Brittany, France: A 10-year monitoring study. Restoration Ecology, 12(1), 29–35. https://doi.org/10.1111/j.1061-2971.2004.00264.x
References
202
Rupp-Armstrong, S., & Nicholls, R. J. (2007). Coastal and Estuarine Retreat: A Comparison of the Application of Managed Realignment in England and Germany. Journal of Coastal Research, 236, 1418–1430. https://doi.org/10.2112/04-0426.1
Rupprecht, F., Möller, I., Paul, M., Kudella, M., Spencer, T., van Wesenbeeck, B. K., et al. (2017). Vegetation-wave interactions in salt marshes under storm surge conditions. Ecological Engineering, 100, 301–315. https://doi.org/10.1016/j.ecoleng.2016.12.030
San Francisco Bay Joint Venture. (2018). San Francisco Bay Joint Venture A partnership working to protect wetlands for the benefit of wildlife and people in the Bay Area. Retrieved June 27, 2018, from http://www.sfbayjv.org/about-goals.php
Sandi, S., Rodriguez, J., Saintilan, N., Riccardi, G., & Saco, P. (2018). Rising tides, rising gates: the complex ecogeomorphic response of coastal wetlands to sea-level rise and human interventions. Advances in Water Resources, 114, 135–148. https://doi.org/10.1016/j.advwatres.2018.02.006
Sasmito, S. D., Murdiyarso, D., Friess, D. A., & Kurnianto, S. (2016). Can mangroves keep pace with contemporary sea level rise? A global data review. Wetlands Ecology and Management, 24(2), 263–278. https://doi.org/10.1007/s11273-015-9466-7
Saunders, M. I., Leon, J. X., Callaghan, D. P., Roelfsema, C. M., Hamylton, S., Brown, C. J., et al. (2014). Interdependency of tropical marine ecosystems in response to climate change. Nature Climate Change, 4(8), 724–729. https://doi.org/10.1038/nclimate2274
Schepers, L., Kirwan, M., Guntenspergen, G., & Temmerman, S. (2017). Spatio-temporal development of vegetation die-off in a submerging coastal marsh. Limnology and Oceanography, 62(1), 137–150. https://doi.org/10.1002/lno.10381
Schmitt, K. (2012). Mangrove planting, community participation and integrated management in Soc Trang Province, Viet Nam. Sharing Lessons on Mangrove Restoration. Retrieved from http://www.mangrovesforthefuture.org/assets/Repository/Documents/Call-for-Action-and-Proceedings-from-2012-Colloquium-Mamallapuram-India.pdf
Schueler, K. (2017). Nature-based solutions to enhance coastal resilience. Retrieved from https://publications.iadb.org/bitstream/handle/11319/8526/Nature_based_solutions_to_enhance_coastal_resilience.pdf?sequence=1&isAllowed=y
Schuerch, M., Spencer, T., Temmerman, S., Kirwan, M. L., Wolff, C., Lincke, D., et al. (2018). Future response of global coastal wetlands to sea-level rise. Nature, 561(7722), 231–234. https://doi.org/10.1038/s41586-018-0476-5
Scott, D. B., Frail-Gauthier, J., & Mudie, P. J. (2014). Coastal Wetlands of the World. (Cambirdge University Press, Ed.). https://doi.org/10.1017/CBO9781107296916
Seavitt, C. (2013). Yangtze River Delta Project. Retrieved August 21, 2018, from https://scenariojournal.com/article/yangtze-river-delta-project/
Sengupta, D., Chen, R., & Meadows, M. E. (2018). Building beyond land: An overview of coastal land reclamation in 16 global megacities. Applied Geography, 90(May 2017), 229–238. https://doi.org/10.1016/j.apgeog.2017.12.015
References
203
Sheng, Y. P., Lapetina, A., & Ma, G. (2012). The reduction of storm surge by vegetation canopies: Three-dimensional simulations. Geophysical Research Letters, 39(20), 1–5. https://doi.org/10.1029/2012GL053577
Shepard, C. C., Crain, C. M., & Beck, M. W. (2011). The protective role of coastal marshes: A systematic review and meta-analysis. PLoS ONE, 6(11). https://doi.org/10.1371/journal.pone.0027374
de Sherbinin, A., Schiller, A., & Pulsipher, A. (2007). The vulnerability of global cities to climate hazards. Environment and Urbanization, 19(1), 39–64. https://doi.org/10.1177/0956247807076725
Shoemaker, C. M., Ervin, G. N., & DiOrio, E. W. (2017). Interplay of water quality and vegetation in restored wetland plant assemblages from an agricultural landscape. Ecological Engineering, 108(May), 255–262. https://doi.org/10.1016/j.ecoleng.2017.08.034
Short, F. T., & Neckles, H. A. (1999). The effects of global climate change on seagrasses. Aquatic Botany, 63(3–4), 169–196. https://doi.org/10.1016/S0304-3770(98)00117-X
SigmaPlan. (2017). Over het SigmaPlan. Retrieved August 17, 2017, from http://sigmaplan.be/nl/over-het-sigmaplan/
Sigren, J. M., Figlus, J., Armitage, A. R., Barone, D. A., McKenna, K. K., Farrell, S. C., et al. (2014). Coastal sand dunes and dune vegetation: restoration, erosion, and storm protection. Shore & Beach, 82(4), 5.
Sigren, J. M., Figlus, J., Highfield, W., Feagin, R. A., & Armitage, A. R. (2018). The Effects of Coastal Dune Volume and Vegetation on Storm-Induced Property Damage: Analysis from Hurricane Ike. Journal of Coastal Research, 341(1), 164–173. https://doi.org/10.2112/JCOASTRES-D-16-00169.1
Silva, R., Martínez, M. L., Odériz, I., Mendoza, E., & Feagin, R. A. (2016). Response of vegetated dune-beach systems to storm conditions. Coastal Engineering, 109, 53–62. https://doi.org/10.1016/j.coastaleng.2015.12.007
Simas, T., Nunes, J. ., & Ferreira, J. . (2001). Effects of global climate change on coastal salt marshes. Ecological Modelling, 139(1), 1–15. https://doi.org/10.1016/S0304-3800(01)00226-5
Simenstad, C., Reed, D., & Ford, M. (2006). When is restoration not? Incorporating landscape-scale processes to restore self-sustaining ecosystems in coastal wetland restoration. Ecological Engineering, 26(1), 27–39. https://doi.org/10.1016/j.ecoleng.2005.09.007
Small, C., & Nicholls, R. J. (2003). A Global Analysis of Human Settlement in Coastal Zones. Journal of Coastal Reserach, 19(3), 584–599.
Smolders, S., Plancke, Y., Ides, S., Meire, P., & Temmerman, S. (2015). Role of intertidal wetlands for tidal and storm tide attenuation along a confined estuary: a model study. Natural Hazards and Earth System Sciences Discussions, 3(5), 3181–3224. https://doi.org/10.5194/nhessd-3-3181-2015
Spalding, M. D., Blasco, E., & Field, C. D. (1997). World Mangrove Atlas. The International Society for Mangrove Ecosystems. https://doi.org/10.1017/S0266467498300528
Spalding, M. D., Ravilious, C., & Green, E. P. (2002). World Atlas of Coral Reefs (Prepared a, Vol. 44). University of California Press, Berkeley, USA. https://doi.org/10.1016/S0025-326X(01)00310-1
References
204
Spalding, M. D., Kainuma, M., & Collins, L. (2010). World Atlas of Mangroves. (Earthscan, Ed.). London: Published with ISME, ITTO and project partners FAO, UNESCO-MAB, UNEP-WCMC and UNU-INWEH. Retrieved from https://books.google.be/books?id=wzSCkulW9SQC
Spalding, M. D., McIvor, A. L., Beck, M. W., Koch, E. W., Möller, I., Reed, D. J., et al. (2013). Coastal ecosystems: A critical element of risk reduction. Conservation Letters, 7(3), 293–301. https://doi.org/10.1111/conl.12074
Spalding, M. D., McIvor, A. L., Tonneijck, F., Tol, S., & van Eijk, P. (2014). Mangroves for coastal defence. Guidelines for coastal managers & policy makers.
Spalding, M. D., Ruffo, S., Lacambra, C., Meliane, I., Hale, L. Z., Shepard, C. C., & Beck, M. W. (2014). The role of ecosystems in coastal protection: Adapting to climate change and coastal hazards. Ocean and Coastal Management, 90, 50–57. https://doi.org/10.1016/j.ocecoaman.2013.09.007
Stanturf, J. A., Goodrick, S. L., & Outcalt, K. W. (2007). Disturbance and coastal forests: A strategic approach to forest management in hurricane impact zones. Forest Ecology and Management, 250(1–2), 119–135. https://doi.org/10.1016/j.foreco.2007.03.015
Stark, J. (2016). Effects of intertidal ecosystems on estuarine hydrodynamics and flood wave attenuation : a multi-scale study. https://doi.org/10.13140/RG.2.2.24404.19840
Stark, J., Van Oyen, T., Meire, P., & Temmerman, S. (2015). Observations of tidal and storm surge attenuation in a large tidal marsh. Limnology and Oceanography, n/a-n/a. https://doi.org/10.1002/lno.10104
Stark, J., Plancke, Y., Ides, S., Meire, P., & Temmerman, S. (2016). Coastal flood protection by a combined nature-based and engineering approach: Modeling the effects of marsh geometry and surrounding dikes. Estuarine, Coastal and Shelf Science, 175, 34–45. https://doi.org/10.1016/j.ecss.2016.03.027
Storlazzi, C. D., Elias, E., Field, M. E., & Presto, M. K. (2011). Numerical modeling of the impact of sea-level rise on fringing coral reef hydrodynamics and sediment transport. Coral Reefs, 30(SUPPL. 1), 83–96. https://doi.org/10.1007/s00338-011-0723-9
Suman, D. O. (2019). Mangrove Management. In Coastal Wetlands (pp. 1055–1079). Elsevier. https://doi.org/10.1016/B978-0-444-63893-9.00031-9
Sun, G., Ranson, K. J., Kharuk, V. I., & Kovacs, K. (2003). Validation of surface height from shuttle radar topography mission using shuttle laser altimeter. Remote Sensing of Environment, 88(4), 401–411. https://doi.org/10.1016/j.rse.2003.09.001
Sutton-Grier, A. E., Wowk, K., & Bamford, H. (2015). Future of our coasts: The potential for natural and hybrid infrastructure to enhance the resilience of our coastal communities, economies and ecosystems. Environmental Science and Policy, 51, 137–148. https://doi.org/10.1016/j.envsci.2015.04.006
Sutton-Grier, A. E., Gittman, R. K., Arkema, K. K., Bennett, R. O., Benoit, J., Blitch, S., et al. (2018). Investing in natural and nature-based infrastructure: Building better along our coasts. Sustainability (Switzerland), 10(2), 1–11. https://doi.org/10.3390/su10020523
Syvitski, J. P. M. (2005). Impact of Humans on the Flux of Terrestrial Sediment to the Global Coastal Ocean. Science, 308(5720), 376–380. https://doi.org/10.1126/science.1109454
References
205
Syvitski, J. P. M. (2008). Deltas at risk. Sustainability Science, 3(1), 23–32. https://doi.org/10.1007/s11625-008-0043-3
Syvitski, J. P. M., & Saito, Y. (2007). Morphodynamics of deltas under the influence of humans. Global and Planetary Change, 57(3–4), 261–282. https://doi.org/10.1016/j.gloplacha.2006.12.001
Syvitski, J. P. M., Kettner, A. J., Overeem, I. I., Hutton, E. W. H., Hannon, M. T., Brakenridge, G. R., et al. (2009). Sinking deltas due to human activites. Nature Geoscience, 2(September), 6–11. https://doi.org/10.1038/ngeo629
Tack, F. M. G., De Pauw, N., Du Laing, G., & Rousseau, D. (2007). Contaminants in natural and constructed wetlands: Pollutant dynamics and control. Science of the Total Environment, 380(1–3), 1–2. https://doi.org/10.1016/j.scitotenv.2007.02.018
Tateishi, R., Hoan, N. T., Kobayashi, T., Alsaaideh, B., Tana, G., & Phong, D. X. (2014). Production of Global Land Cover Data – GLCNMO2008. Journal of Geography and Geology, 6(3). https://doi.org/10.5539/jgg.v6n3p99
Temmerman, S., & Kirwan, M. L. (2015). Building land with a rising sea. Science. https://doi.org/10.1126/science.aac8312
Temmerman, S., Govers, G., Wartel, S., & Meire, P. (2004). Modelling estuarine variations in tidal marsh sedimentation: Response to changing sea level and suspended sediment concentrations. Marine Geology, 212(1–4), 1–19. https://doi.org/10.1016/j.margeo.2004.10.021
Temmerman, S., De Vries, M. B., & Bouma, T. J. (2012). Coastal marsh die-off and reduced attenuation of coastal floods: A model analysis. Global and Planetary Change, 92–93, 267–274. https://doi.org/10.1016/j.gloplacha.2012.06.001
Temmerman, S., Meire, P., Bouma, T. J., Herman, P. M. J., Ysebaert, T., & De Vriend, H. J. (2013). Ecosystem-based coastal defence in the face of global change. Nature, 504(7478), 79–83. https://doi.org/10.1038/nature12859
Tessler, Z. D., Vörösmarty, C. J., Grossberg, M., Gladkova, I., Aizenman, H., Syvitski, J. P. M., & Foufoula-Georgiou, E. (2015). Profiling risk and sustainability in coastal deltas of the world. Science, 349(6248), 638–643.
Teuchies, J., Vandenbruwaene, W., Carpentier, R., Bervoets, L., Temmerman, S., Wang, C., et al. (2013). Estuaries as Filters: The Role of Tidal Marshes in Trace Metal Removal. PLoS ONE, 8(8), 1–11. https://doi.org/10.1371/journal.pone.0070381
Thampanya, U., Vermaat, J. E., Sinsakul, S., & Panapitukkul, N. (2006). Coastal erosion and mangrove progradation of Southern Thailand. Estuarine, Coastal and Shelf Science, 68(1), 75–85. https://doi.org/10.1016/j.ecss.2006.01.011
The World Bank. (2017). Implementing nature-based flood protection Principles and implementation guidance, 7–31. Retrieved from http://documents.worldbank.org/curated/en/739421509427698706/pdf/120735-REVISED-PUBLIC-Brochure-Implementing-nature-based-flood-protection-web.pdf
Tian, B., Wu, W., Yang, Z., & Zhou, Y. (2016). Drivers, trends, and potential impacts of long-term coastal reclamation in China from 1985 to 2010. Estuarine, Coastal and Shelf Science, 170, 83–90. https://doi.org/10.1016/j.ecss.2016.01.006
UNEP-WCMC. (2006). In the front line: shoreline protection and other ecosystem services from mangroves and coral reefs. United Nations Environmental Programme- World Conservation Monitoring Centre International Coral Reef Action Network, (24), 40. https://doi.org/10.1089/bfm.2011.0071
References
206
United Nations. (2012). World Population Prospects: the 2010 Revision. Waste Management Research, 27(8), 800–812. https://doi.org/10.1553/populationyearbook2010s77
United Nations Department of Economic and Social Affairs Population Division. (2016). The World’s Cities in 2016. Data Booklet (ST/ESA/SER.A/392). Retrieved from www.unpopulation.org.
UT BATTELLE LLC. (n.d.). LandScan frequently asked questions. Retrieved November 16, 2016, from http://web.ornl.gov/sci/landscan/landscan_faq.shtml
Vafeidis, A. T., Boot, G., Cox, J., Mcfadden, L., & Nicholls, R. J. (2005). The diva database documentation, 1–33.
Valiela, I. (2006). Global Coastal Change. Environmental Conservation (Vol. 33). Oxford, UK: Blackwell Publishing. https://doi.org/DOI: 10.1017/S0376892907253652
Valiela, I., Bowen, J. L., & York, J. K. (2001). Mangrove Forests: One of the World’s Threatened Major Tropical Environments. BioScience, 51(10), 807. https://doi.org/10.1641/0006-3568(2001)051[0807:MFOOTW]2.0.CO;2
Valiela, I., Kinney, E., Culbertson, J., Peacock, E., & Smith, S. (2009). Global Loss of Coastal Habitats Rates, Causes and Consequences. In C. M. Duarte (Ed.), Global Loss of Coastal Habitats Rates, Causes and Consequences (pp. 108–142).
Vermaat, J. E., & Thampanya, U. (2006). Mangroves mitigate tsunami damage: A further response. Estuarine, Coastal and Shelf Science, 69(1–2), 1–3. https://doi.org/10.1016/j.ecss.2006.04.019
Vitousek, S., Barnard, P. L., Fletcher, C. H., Frazer, N., Erikson, L., & Storlazzi, C. D. (2017). Doubling of coastal flooding frequency within decades due to sea-level rise. Scientific Reports, 7(1), 1–9. https://doi.org/10.1038/s41598-017-01362-7
Voortman, H. G., Van Gelder, P. H. a J. M., & Vrijling, J. K. (2003). Risk-Based Design of Large-Scale Flood Defence Systems. Coastal Engineering 2002 Solving Coastal Conundrums Proceedings of the 28th International Conference. https://doi.org/10.1142/9789812791306_0199
Vos, P. C. (2015). Origin of the Dutch coastal landscape. Universiteit Utrecht.
Vuik, V., Jonkman, S. N., Borsje, B. W., & Suzuki, T. (2016). Nature-based flood protection: The efficiency of vegetated foreshores for reducing wave loads on coastal dikes. Coastal Engineering, 116, 42–56. https://doi.org/10.1016/j.coastaleng.2016.06.001
Vuik, V., van Vuren, S., Borsje, B. W., van Wesenbeeck, B. K., & Jonkman, S. N. (2018). Assessing safety of nature-based flood defenses: Dealing with extremes and uncertainties. Coastal Engineering, 139(April), 47–64. https://doi.org/10.1016/j.coastaleng.2018.05.002
Wamsley, T. V., Cialone, M. a., Smith, J. M., Ebersole, B. a., & Grzegorzewski, A. S. (2009). Influence of landscape restoration and degradation on storm surge and waves in southern Louisiana. Natural Hazards, 51(1), 207–224. https://doi.org/10.1007/s11069-009-9378-z
Wamsley, T. V., Cialone, M. A., Smith, J. M., Atkinson, J. H., & Rosati, J. D. (2010). The potential of wetlands in reducing storm surge. Ocean Engineering, 37(1), 59–68. https://doi.org/10.1016/j.oceaneng.2009.07.018
Wang, W., Liu, H., Li, Y., & Su, J. (2014). Development and management of land reclamation in China. Ocean & Coastal Management, 102, 415–425. https://doi.org/10.1016/j.ocecoaman.2014.03.009
References
207
Webster, P. J., Holland, G. J., Curry, J. A., Chang, H.-R., Sciences, A., Webster, P. J., et al. (2005). Changes in Tropical Cyclone Number , Duration , and Intensity in a Warming Environment. Sciences, Atmospheric, 437(1996), 2003–2006. https://doi.org/10.1126/science.1116448
van Wesenbeeck, B. K., Mulder, J. P. M., Marchand, M., Reed, D. J., De Vries, M. B., De Vriend, H. J., & Herman, P. M. J. (2014). Damming deltas: A practice of the past? Towards nature-based flood defenses. Estuarine, Coastal and Shelf Science, 140, 1–6. https://doi.org/10.1016/j.ecss.2013.12.031
van Wesenbeeck, B. K., de Boer, W., Narayan, S., van der Star, W. R. L., & de Vries, M. B. (2017). Coastal and riverine ecosystems as adaptive flood defenses under a changing climate. Mitigation and Adaptation Strategies for Global Change, 22(7), 1087–1094. https://doi.org/10.1007/s11027-016-9714-z
White, A. T., Vogt, H. P., & Arin, T. (2000). Philippine coral reefs under threat: The economic losses caused by reef destruction. Marine Pollution Bulletin, 40(7), 598–605. https://doi.org/10.1016/S0025-326X(00)00022-9
Wilkinson, C., Lindén, O., Cesar, H., Hodgson, G., Rubens, J., & Strong, A. E. (1999). Ecological and Socioeconimoc Impacts of 1998 Coral Mortality in the Indian Ocean: An ENSO Impact and a Warning of Future Change? Ambio, 28(2), 188–196. https://doi.org/10.5363/tits.6.3_37
Wolanski, E., & Elliott, M. (2015). Estuarine Ecohydrology: An Introduction. Elsevier Science, Amsterdam, 322pp.
Wolff, W. J. (1993). Netherlands-Wetlands. Hydrobiologia, 265(1–3), 1–14. https://doi.org/10.1007/BF00007260
Woodruff, J. D., Irish, J. L., & Camargo, S. J. (2013). Coastal flooding by tropical cyclones and sea-level rise. Nature, 504(7478), 44–52. https://doi.org/10.1038/nature12855
World Population History. (n.d.). World Population. Retrieved September 4, 2017, from http://worldpopulationhistory.org/map/306/mercator/1/0/25/
Yepsen, M., Baldwin, A. H., Whigham, D. F., McFarland, E., LaForgia, M., & Lang, M. (2014). Agricultural wetland restorations on the USA Atlantic Coastal Plain achieve diverse native wetland plant communities but differ from natural wetlands. Agriculture, Ecosystems and Environment, 197, 11–20. https://doi.org/10.1016/j.agee.2014.07.007
Ysebaert, T., Yang, S. L., Zhang, L., He, Q., Bouma, T. J., & Herman, P. M. J. (2011). Wave attenuation by two contrasting ecosystem engineering salt marsh macrophytes in the intertidal pioneer zone. Wetlands, 31(6), 1043–1054. https://doi.org/10.1007/s13157-011-0240-1
Ysebaert, T., van der Hoek, D. J., Wortelboer, R., Wijsman, J. W. M., Tangelder, M., & Nolte, A. (2016). Management options for restoring estuarine dynamics and implications for ecosystems: A quantitative approach for the Southwest Delta in the Netherlands. Ocean and Coastal Management, 121, 33–48. https://doi.org/10.1016/j.ocecoaman.2015.11.005
Ysebaert, T., Jansen, H. M., & Poelman, M. (2017). Coastal protection and seafood security in Bangladesh : Food for Thought Nature-based solutions at the land-ocean interface : Building with Nature Introduction The coastal area of Bangladesh Ecosystem-based coastal protection. In Ocean/Blue Economy for Bangladesh Workshop (pp. 22–23). Dhaka.
References
208
Zhang, K., Liu, H., Li, Y., Xu, H., Shen, J., Rhome, J., & Smith, T. J. (2012). The role of mangroves in attenuating storm surges. Estuarine, Coastal and Shelf Science, 102–103, 11–23. https://doi.org/10.1016/j.ecss.2012.02.021
Zhao, H., Cui, B., Zhang, H., Fan, X., Zhang, Z., & Lei, X. (2010). A landscape approach for wetland change detection (1979-2009) in the Pearl River Estuary. Procedia Environmental Sciences, 2(5), 1265–1278. https://doi.org/10.1016/j.proenv.2010.10.137
Zhao, Q., Bai, J., Huang, L., Gu, B., Lu, Q., & Gao, Z. (2016). A review of methodologies and success indicators for coastal wetland restoration. Ecological Indicators, 60, 442–452. https://doi.org/10.1016/j.ecolind.2015.07.003
Zhu, M. S., Sun, T., & Shao, D. D. (2016). Impact of Land Reclamation on the Evolution of Shoreline Change and Nearshore Vegetation Distribution in Yangtze River Estuary. Wetlands, 36, 11–17. https://doi.org/10.1007/s13157-014-0610-6