INNOVATIONS IN THE FACE OF CLIMATE CHANGE

468 Coastal infrastructure: a typology for the

next century of adaptation to sea-level rise
Kristina Hill

Categorizing the choices in coastal infrastructure that are available to policy makers will allow for comparisons of
their potential impacts on ecosystems and of their value in preparation for long-term sea-level rise. Although sim-
ilar approaches have been described elsewhere in different policy contexts, this article focuses on evaluating phys-
ical infrastructure types – including hybrid structures that combine landforms with concrete and steel elements –
based on historical differences in engineering practices. Such structures can be optimized for different phases of
coastal adaptation and can provide multiple benefits (eg supporting ecosystems as well as minimizing flooding in
coastal cities). Key factors in a geomorphological, ecological, and land-use context must be taken into account
when selecting various infrastructure strategies, to ensure that they function as intended. The San Francisco Bay
region provides an example of how this typology can be applied to help policy makers choose more successful
strategies as coastal areas plan for sea-level rise.
Front Ecol Environ 2015; 13(9): 468–476, doi:10.1890/150088

H umans have altered coastal areas by introducing arti-

ficial structures dating back to at least 2580 BCE, on
the shores of the Red Sea in modern Egypt (Tallet and
them to live in a tidally flooded environment (Henry
1855). The initial driver for mound construction in sandy
tidal areas was access to small-scale fishing and trade,
Marouard 2014). The first coastal structures that appear in while the main driver of later dike construction was urban
the archaeological record were rock breakwaters, built to growth, which led to more land being dedicated to inten-
protect harbor entrances from wave energy. Artificial har- sive food production (Charlier et al. 2005).
bors shaped by rock walls were also an early invention, In past centuries, the term “infrastructure” referred pri-
often contained within early cities by gates built both for marily to masonry and metal constructions, but more
protection from attacking navies and to control the pas- recently it has come to signify any structures (eg power-
sage of goods and travelers (Blackman 1982). In 79 CE, lines, floodwalls, wetlands) that support or alter the spa-
Pliny the Elder reported encountering people on the tial and temporal distribution of resources and risks for
northern coasts of present-day Germany and the human benefit. Several new terms have emerged to serve
Netherlands who built artificial mounds that enabled this new definition of infrastructure; for instance, “land-
scape infrastructure” alludes to the capacity of topogra-
phy, soils, and entire ecosystems to support human needs
In a nutshell: (Beach 2003; Hill 2011), whereas “green infrastructure”
• Typologies are useful when many examples of alternative typically relates to the use of plants and soils to provide
infrastructure design strategies exist, and a high-level catego- ecosystem services (eg Arcadis 2014). Coastal wetlands,
rization allows planners to perceive the pattern of alternatives sand dunes, beaches, and freshwater ponds are treated as
under consideration
• Many coastal regions have begun to plan for adaptation to supporting structures for flood management, co-existing
sea-level rise, and are in need of a clear overview of options in hybrid systems with levees, breakwaters, seawalls,
that can be discussed with policy makers, advocacy organiza- floodwalls, tide gates, storm-surge barriers, pumps, and
tions, and the public pipes. The US Army Corps of Engineers (USACE) has
• A mix of adaptation strategies often provides the broadest developed guidelines for what they call “natural and
suite of benefits, including newer approaches that involve liv-
ing system components such as wetlands, sandy beaches, nature-based features” (NNBFs) that are now considered
sandbars, or living breakwaters by federal agencies as functional components of coastal
• Some regions have limited experience with these new infrastructure designs (Bridges et al. 2015).
approaches and may benefit from using decision-support tools
that identify such ecosystem-based strategies
• Investment may need to be in phases to accommodate higher n Drivers of new investment
rates of sea-level rise over time (ie investing in a new seawall
or floodwall structure that may require replacement or reloca- In the second half of the 20th century, new residential
tion in the future) and recreational land uses have been the primary drivers
of investment in engineering projects at local and
regional scales in coastal areas in the US (Beach 2003;
Department of Landscape Architecture and Environmental Planning, Hill 2011). Marina development, breakwater or jetty
University of California, Berkeley, Berkeley, CA (kzhill@berkeley.edu) construction to protect marinas and harbors, and the

www.frontiersinecology.org © The Ecological Society of America

K Hill Coastal adaptation to sea-level rise

addition of highways, bridges, and pipelines have all been vided safe harbors for ships. In large bays and estuaries, 469
major public projects, along with the development of artificial structures can alter salinity dynamics as well as
deep-water ports with dredged shipping lanes (Pilkey and sediment concentrations and transport (Kadiri et al.
Dixon 1996). Tide gates and upriver dams used for flood 2012). Changes in sediment or wave energy dynamics
protection, irrigation, and hydroelectric power genera- can lead to erosion of wetlands and other ecosystems
tion have also had major impacts on sediment dynamics along estuary shores, together with changes in water qual-
in coastal areas, often accelerating erosion by depriving ity if the flux of ocean water into the estuary is reduced
coastal landscapes of sand or silt (Giannico and Souder (Eelkema et al. 2013). Exotic species, which can have
2005). Factors that are likely to increase investments in widespread negative impacts on the biodiversity of urban
coastal infrastructure over the next century include estuary systems, often enter estuaries via ships’ ballast
increased vulnerability of developed areas to flooding water (Ruiz et al. 2000). Activities at commercial or mili-
(Aerts et al. 2011), higher rates of salinization of water tary ports, and the industrial sites typically associated
supplies and a corresponding and growing need for fresh with those ports, introduce chemical pollutants into estu-
water (Sekovski et al. 2012), ecosystem losses from ero- aries that can be dissolved or suspended in the water col-
sion and development (Gedan et al. 2009; Jennerjahn umn or deposited in sediments. Major energy-generation,
and Mitchell 2013), and intensified international trade desalinization, and sewage treatment facilities in bays and
connections via ever-larger ships (Bruun 2005). estuaries can also affect biota through pollution or by
altering physical characteristics such as water tempera-
n Impacts of coastal infrastructure on natural ture (Schifter et al. 2011). Marine borers and other organ-
systems isms may also damage or cause failure in coastal infra-
structure systems by blocking flows or removing material
Artificial coastal structures, along with their construction (Borges 2014). Sea-level rise is expected to have major
and maintenance, have had a variety of effects on the impacts on infrastructure in the shore zone of estuaries
geomorphology and ecology of coastal systems (Bulleri (Flood and Cahoon 2011; Biging et al. 2012).
and Chapman 2010; Nordstrom 2014). For instance,
these structures often interfere with the spatial dynamics n A typology of infrastructure strategies
of sediment transport, salinity, flooding, and animal
movement or reproduction. At the same time, the physi-
Structures
cal and biological systems of the marine environment
have extensive impacts on built structures (see Burcharth One method of gaining insight into the advantages and
et al. 2014). disadvantages of different coastal infrastructure types, and
On open sandy coasts, structures such as groins, chan- how they might be applied in a given environmental and
nels, and breakwaters typically alter wave energy regimes land-use context, is to organize a typology based on the
and sediment supply (Nordstrom 2014), which affect fun- history of coastal engineering practices. Typologies are
damental processes of longshore sediment transport that defined here as “conceptually derived interrelated sets of
influence levels of turbidity and rates of accretion and ideal types” (Doty and Glick 1994), which can be used to
erosion. This in turn generates changes in barrier island develop hypotheses about the causes of deviations from
and beach dynamics, dune growth and migration, and graded-membership ideal types. Graded-membership
inlet locations, even where these landscapes are pro- types are defined by a best example, but are grouped by
tected from human development (Louters et al. 1991). degrees of similarity. For instance, “walls” are typically
The geomorphological impacts of coastal infrastructure represented as solid barriers, but they can also be built to
(those that alter patterns of sediment erosion and deposi- allow water to filter through them, and would still be
tion) often lead to changes in biotic communities above, considered walls. Leaky walls are sufficiently similar to
within, and below the intertidal zone (Mattheus et al. impermeable walls to be considered as the same type of
2010). The material, shape, anchoring method, and sur- structure. A typology can be a valuable heuristic tool in
face roughness of coastal infrastructure can also influence decision theory, used to reveal omissions of important
diversity and population sizes within biotic communities options within sets of alternatives that occupy a solution
(Perkol-Finkel et al. 2012). Tide gates, seawalls, bridge space (Chernoff and Moses 1959; Mees et al. 2014). The
ramps, and roadways may all cease to function as designed historical record of coastal engineering practices provides
when relative sea levels rise or when storm-driven flood- a starting point for the development of a simple typology,
ing becomes more frequent or more severe (Johnston et consisting of four ideal types of coastal strategies that may
al. 2014). Bluffs may become highly eroded by changes in become more useful over time as new innovations and
the relative height of waves, which can destabilize coastal hybrid strategies are introduced.
infrastructure located outside the actual tidal zone (eg The top-tier categories in this typology have their
roads; Barton et al. 2014). origins in the history of structures associated with
Historically, coastal cities were usually built in naturally rocky shore environments – such as walls and breakwa-
sheltered bays or on the banks of tidal rivers that pro- ters – rather than sandy/marshy environments, where

© The Ecological Society of America www.frontiersinecology.org

 Coastal adaptation to sea-level rise K Hill

470 Figure 1. Seawall in San Francisco, California. Seawalls are

fixed walls made of reinforced concrete or steel, often combined
with wooden pilings treated with creosote. They are designed to
retain unconsolidated fill on the landward side, which contains the
footings of buildings and other infrastructure. Relieving platforms
constructed of wood, or concrete and steel, hold the walls in
place. These walls were essential for bringing ships directly into
urban quays to efficiently load and unload cargo or passengers.

ing of approaches that may be underutilized relative to

their potential.

Walls: static or dynamic

The structures that developed from design practices in

rocky shore environments with fortified harbors tend to
be wall-based and rigid; be made of stone, concrete, and
metal; and function as either movable gates or static
walls (Figure 1). In contemporary practice, these struc-
tures typically consist of inflexible walls made of con-
crete and steel, and are fixed structures, such as a seawall
or dam. Tidal energy dams, or “barrages”, may be built as
a hybrid of fixed concrete-and-steel walls and earthen
dikes, but like other dams they alter the fluxes between
downstream brackish water and upstream fresh water as
silt, sand, and gravel materials were used to shape new they generate power (Kadiri et al. 2012). At the
landforms such as mounds and dikes. The second tier Annapolis Royal tidal barrage in eastern Canada, for
of the typology is defined by whether the structure is instance, researchers have documented higher rates of
static, meaning it is not designed to move, or dynamic, shore erosion in the freshwater zone upstream of the bar-
meaning it is designed to move in specific ways. These rage, most likely the result of wind-driven waves that
two simple tiers create four classes of coastal infra- have become more powerful after the construction of the
structure strategies, which can be used to generate barrage (Morris 2013).
hybrids that combine elements of the four classes. The Walls can also be designed to function dynamically,
purpose of this typology is to gain a better understand- moving into position only when needed – tide gates, which
are usually closed during the highest tides to pre-
vent inundation, are simple examples. The
Thames Barrier outside London, UK, is a much
more complex version of a dynamic wall, formed
by long steel sections between towers that “rest”
flat in the riverbed and are rotated into place to
block storm surges (Figure 2). The Eastern
Arpingstone/Wikimedia Commons CC BY-SA 3.0

Scheldt Barrier in the Dutch Deltaworks system

uses vertical gates in a similar way, whereas the
Maeslant Barrier in the Dutch city of Rotterdam
consists of two triangular frames with a curved
wall at the end of each. These move laterally on
tracks in the riverbed using hydraulic systems,
dropping into the channel once they are in posi-
tion to close off storm surges
While providing immediate functional bene-
Figure 2. Thames Barrier, with one gate raised for maintenance. The fits, these technically complex structures may
Thames Barrier consists of a series of moveable walls, built in response to a offer a narrower range of functions because they
major North Sea flood in 1953. The Barrier protects valuable land in central do not typically provide any additional natural
London. Its walls rotate into place between towers placed at approximately habitat or recreational opportunities. They may
200-ft intervals across the Thames River near Woolwich, in response to storm also be less adaptable to future needs given that
surge predictions. Plans are being discussed to replace or enlarge the barrier as they often require complete replacement if it
sea levels rise (see Reeder and Ranger 2011), which will be extremely costly. becomes necessary to enlarge their height or

www.frontiersinecology.org © The Ecological Society of America

K Hill Coastal adaptation to sea-level rise

extent. Although well-suited to 471

small construction spaces, such
structures also introduce new risks
associated with mechanical or elec-
trical failures; the weaknesses
resulting from relying on walls and
pumps were evident in New
Orleans, Louisiana, during and after
Hurricane Katrina (Reid 2013).

Landforms: static or dynamic

In the past, coastal infrastructures in

sandy and marshy areas were more
likely to be landforms composed of
materials that could be excavated
and mounded but which required

M Hook
frequent labor for maintenance, and
could be deformed and redistributed Figure 3. A superdike. This fixed landform is innovative because, unlike most earthen
either incrementally by human labor dikes, it is designed to allow buildings and trees to be built on the top level and on its back
or suddenly by natural processes terraces. This gives residents a water view, increasing property values as well as raising
associated with an extreme weather awareness of the dynamic environment that surrounds them. This superdike was built in
event (Charlier et al. 2005). The Osaka, Japan; it gains its unique structural qualities from being much wider (approximately
contemporary analogs of these 1200 ft) than a normal dike, which might be 400 ft wide to support 30 ft of height.
ancient structures are designed
either as rigid landforms that require frequent monitoring Engine” or “Sand Motor”; Figure 4): a massive project ini-
and maintenance, as in the case of levees or raised mounds, tiated in 2011 – in which an artificial sand delta was
or as dynamic forms that are intended to be altered by pre- dredged and positioned on the ocean coast between
dictable forces over time (VanKoningsveld et al. 2008). In Rotterdam and The Hague, where wind and wave energy
the Netherlands, for instance, static landforms include are expected to move the sand north and south – in order
many different types of dikes, such as the Afsluitdijk, built to widen the protective dune and beach shore zone of that
in 1932; this 32-km dike closed off a saltwater inlet of the region (Aarninkof et al. 2010). This dynamic sand land-
Zuiderzee that was consequently transformed into the form is intended as a replacement for the annual nourish-
freshwater lake known today as the Ijsselmeer (Van de Ven ment of beaches and dunes (ie the addition of more sand
1993; VanKoningsveld et al. 2008). Likewise, in the 1990s, following erosion) using heavy equipment, and is already
Japanese engineers built an ultra-wide platform, the producing benefits for birds and plants associated with less-
superdike, to allow construction of an urban district on top disturbed sandy habitats along the Dutch coast, as well as
of the dike (Figure 3). providing recreational surfing opportunities for humans
The Dutch have also embraced dynamic landforms, most (van Slobbe et al. 2013).
notably the Zandmotor (referred to in English as the “Sand While dynamic landforms, such as beaches or marshes,

Figure 4. The Sand Engine. This 128-ha dynamic landform was built using 21 million cubic meters of dredged sand on the Dutch
coast near The Hague in 2010. If it performs as designed, it will add 200 ha of beach along 10–20 km of coastline over 20 years as a
result of wind and wave action, while adding habitat and recreational value (Stive et al. 2013). The dredged sand delta is intended to
erode, feeding beaches and dunes up and down the ocean coast. If it is successful, more sand may be placed in this location to continue
the strategy of mega-nourishment for the Dutch coast. This panoramic view from 2012 was taken from the dunes on the landward
side of the Sand Engine, looking toward the sea.

© The Ecological Society of America www.frontiersinecology.org

 Coastal adaptation to sea-level rise K Hill

472 these contextual issues is the fragility or

robustness of the adjacent landscape, given
that alternative strategies offer different
levels of protection over time and space.
If the area inland of the coastal infrastruc-
ture is vulnerable, meaning that failure of
the infrastructure could easily result in loss
of human lives, extreme property damage,
or the destruction of unique and sensitive
ecosystems, then the infrastructure strategy
must be robust to prevent such conse-
quences (eg Sterr 2008). Alternatively, the
landscape adjacent to the infrastructure
could be resilient to disturbances, in the
sense that it is able to recover from a given
range of frequency and intensity of events,
such as flooding (Barroca et al. 2015).
A third alternative is that the adjacent
land uses or ecosystems might be designed to
be adaptive in themselves, so that they will
Figure 5. Illustration of the typology as four ideal types that establish gradients of not be greatly affected by an event that
similarity to the ideal, defining four quadrants of a solution space. The vertical axis overtops or results in the temporary failure
is defined by the percentage of the physical infrastructure proposal that uses walls of a coastal protective structure (Restemeyer
versus landforms; the horizontal axis is defined by the percentage of design et al. 2015). This more robust strategy would
components that are dynamic versus static. This version of the diagram can be used provide substantial benefits in other areas,
to generate a wide range of alternative structural and non-structural proposals. such as ecosystem support and recreation,
because the infrastructure investment does
may not provide a consistent level of function over time not need to be utilized solely to produce a very high relia-
and space (flood protection levels, for instance, may vary bility of protection from flood events. This type of infra-
in specific locations during different stages of erosion and structure has been characterized as “safe-to-fail” (as
accretion; Stevens et al. 2014), they may successfully opposed to “fail-safe”) although the “failure” in this case
deliver multiple benefits, including habitat, recreation, refers to a temporary loss of some functionality, rather than
and other ecosystem services. These dynamic landforms a catastrophic or permanent loss of all functional capacity
may also be easier to build in phases as sea levels continue (Lister 2007). One example is the use of wide
to rise. Because they are made of loose material, it may be wetland/beach/dune complexes in shore zones to protect
possible to enlarge them without replacing the original urban districts or critical habitat areas from inundation in
structure, as future conditions require a higher or wider most – though not all – weather events, while simultane-
structure (SPUR 2012; Clevenger et al. 2014). ously providing more land area for coastal ecosystems and
These four simple types – static and dynamic walls, and more opportunities for recreation (van Slobbe et al. 2013).
static and dynamic landforms – represent ideals; actual Figure 6 illustrates these strategic choices in the pairing of
built or proposed structures can “mapped” onto gradients infrastructure with urban districts that represent different
that are defined by similarity to these four ideal types. levels of vulnerability due to the design characteristics of
Figure 5 represents the typology as a solution space for an their roads and buildings.
optimization problem with four quadrants. Gradients that
represent similarity to the ideal types in the “corners” of n Applying the typology
this diagram are defined using both the percentage of
walls versus landforms contained in a specific infrastruc-
Application to decision making: San Francisco Bay
ture project or proposal, and the percentage of fixed ver-
sus dynamic components. A useful typology of coastal infrastructure strategies is one
that can serve as a heuristic tool in planning and policy
Identifying coastal infrastructure strategies making, meaning that it allows for a more thorough
appropriate for specific contexts exploration of a set of solutions and the generation of a
complete range of alternatives within a defined set of
The intent of the typology is to represent the range of variables. Application of the typology can provide
choices that could be applied, but the selection of the spe- insights into whether some strategies may be overlooked
cific types of infrastructure that should be applied requires or pre-judged without adequate consideration, since
a review of the specific context. The most important of many coastal planning efforts will give less attention to

www.frontiersinecology.org © The Ecological Society of America

K Hill Coastal adaptation to sea-level rise

certain quadrants of the typology Column A Column B 473

than to others. For example, civil Types of
coastal infrastructure
Adjacent uplands

engineers who are more familiar

with levees and floodwalls may I. Walls I. Urban districts II. Ecosystem examples

ignore wetland and beach strategies A. Vulnerable to any flooding

as viable alternatives. A. Fixed
Floodwalls and seawalls
The process of planning coastal A. Riparian and
groundwater wetlands
infrastructure in the US is described B. Dynamic
B. Resilient to temporary flooding

in Part 5, Chapter 1 of the Movable or temporary

gates B. Intertidal wetlands
USACE’s Coastal Engineering
Manual (Housley and Thompson II. Landforms

2008). This publication notes that a C. Adaptive for permanent flooding C. Sandy beaches,
dunes, and sandbars
wide range of different alternatives A. Fixed
Mounds, dikes, and
canals
should be considered during a 1

“reconnaissance” planning stage, B. Dynamic D. Rocky coasts

Beaches, dunes, and
and that selected alternatives wetlands 2120
2070
2
should be examined in greater
detail through a feasibility study. E. Contaminated soils

However, the description of struc-

tural and non-structural infrastruc- Figure 6. The four basic infrastructure types can be paired with adjacent land uses and
ture designs is unbalanced, provid- landscapes, including urban districts and ecosystems (eg wetlands, rocky shores, sandy
ing greater detail in defining beaches, contaminated soils). The typology is a heuristic, meaning that it enables a user to
structural types than non-structural generate many alternative pairings as a way of studying options, rather than producing a
types. The USACE manual uses a single pairing between Column A and Column B.
flow chart to describe the coastal
infrastructure planning process, which includes 11 stan- will contain elements that overlap along the shore,
dard alternatives (from “do nothing”, to non-structural requiring two points to be used to represent a single pro-
options such as marshes, to structural options including ject within the diagram; for instance, an earthen levee
breakwaters and seawalls; see Figure V-1-1 in Housley and with an additional floodwall on top would require two
Thompson 2008). This list of alternatives does not distin- points to represent both the static landform (the levee)
guish between dynamic strategies that move mechanically and the static wall (the floodwall). The diagram’s simplic-
(eg a movable storm-surge barrier) or function as a result ity also means that it should be limited to use in what the
of change over time (eg “sacrificial” beaches, which are USACE refers to as the “reconnaissance” stage of review-
designed to erode as they supply sediment to areas farther ing a broad set of alternative proposals. It is also limited
down the coast), and static strategies that are maintained to biophysical adaptation strategies, and does not address
exactly as they were initially built and are fixed in space specific opportunities for re-aligning coastal development
(eg a permanent floodwall). Moreover, although the man- by a combination of removing structures in some loca-
ual includes information about non-structural options, it tions and adding new structures in others.
does not identify specific examples as it does for the struc- Recent planning efforts in the San Francisco Bay
tural options. region of California illustrate how the typology presented
Unlike the standard USACE manual, the typology here might be applied, and the insights it can offer. The
described here provides equal consideration to the struc- San Francisco Bay Commission for Development and
tural, non-structural, dynamic, and static categories of Conservation (BCDC) was created in 1965 to regulate
coastal infrastructure designs; this allows a broader assess- the artificial shoreline position of the bay, after decades of
ment of alternatives without pre-judgment of whether urban fill deposition. BCDC has assumed a coordinating
one specific subset of alternatives is more feasible than role in planning for adaptation to sea-level rise in the
another. The typology is simple in that it uses only four region. In 2009, BCDC organized an international com-
ideal types to describe an inclusive solution space and can petition to generate design proposals for sea-level rise
be used to generate alternatives by exploring options from adaptation projects in the San Francisco Bay area, with
the four quadrants of the typology diagram (Figure 5), but five bay-centered entries selected as winners by an inde-
it can also be used to categorize actual proposals and built pendent jury. A review of those winning entries reveals
projects – planners can describe examples in a specific that all but one included “wall” proposals (including the
region by plotting actual projects on the diagram. By construction of a dynamic barrier under the Golden Gate
measuring the percentage of the proposed project’s length Bridge; a series of permanent or temporary barriers sepa-
along the shoreline that would be dedicated to each of rating smaller bays or critical urban sites, such as airports,
the ideal types, a specific built or proposed project can be from the main bay; a light installation that would repre-
assigned a location within the diagram. Some projects sent the location of needed barriers along the shore).

© The Ecological Society of America www.frontiersinecology.org

 Coastal adaptation to sea-level rise K Hill

474 (a) described in this paper, where the crite-

ria for inclusion in categories are repre-
sented as gradients rather than discrete
thresholds. This diagram reveals that the
category of “dynamic landforms” as
coastal infrastructure strategies is empty
(see Figure 7a). Static landforms
appeared in the proposals, as did mov-
able and static walls. The absence of
proposals in the fourth quadrant of the
typology diagram, which includes
dynamic landforms such as marshes and
beaches, suggests that there is potential
in exploring other options for coastal
infrastructure.
The dynamic landform options
(b) include sandbars, beaches, dunes, and
wetlands, all of which could provide
ecosystem services, including flood pro-
tection. Interestingly, none of the 2009
competition-winning alternatives con-
sidered the option of pairing coastal
infrastructure of any kind with floodable
urban districts, or took into account the
likelihood that many coastal wetlands
will be lost to higher sea levels. The crit-
ical relationships between coastal infra-
structure choices and adjacent land
areas went unexplored, aside from pro-
posals to raise the elevation of some
coastal areas.
In contrast, in preparation for a spring
Figure 7. The typology can also be used to “map” specific proposals within the four 2015 workshop sponsored by BCDC,
quadrants, based on the percentage of the shore that is occupied by each of the four staff from the San Francisco Estuary
ideal types (static and dynamic walls and landforms). This can reveal the areas of the Institute (SFEI) developed a list of alter-
diagram’s solution space that might be unexplored: for example, in a case where non- natives for coastal infrastructure that
structural alternatives, such as a wetland or an oyster reef, are not included. were intended for application along dif-
Different regional approaches or different eras in time can be compared side-by-side ferent segments of the bay shoreline
using the diagram as a map of specific alternatives. The dashed line represents the (SFEI 2015). That list of alternatives,
concept of a solution space, which contains voids that depict unexplored alternatives. when mapped onto the same typology
(a) BCDC’s Rising Tides competition winners (BCDC 2009). The six winning diagram, produces a different pattern
proposals are mapped onto the diagram with approximate positions, since specific within the solution space represented by
lengths of shore were not specified for each design. As the diagram shows, dynamic the four quadrants (Figure 7b). This
landforms were not well-explored. (b) BCDC’s Bay Policies workshop, exploring reveals a shift in strategy toward
alternatives for adaptation (SFEI 2015). Specific dimensions were not proposed at dynamic landforms, including beaches,
the workshop, but ranges were discussed and are used to represent the proposals dunes, and marshes, that are expected to
within the diagram, extending their shapes. Mixes of dynamic walls and landforms grow with managed nourishment over
were not as frequently discussed as “pure” strategies. time. Dynamic walls, such as tide gates,
were considered but were not developed
Three of the winning entries also included proposals for as proposals by the participants at the workshop. Static
expanding marsh edges at locations where marshes or salt walls, such as seawalls and reinforced concrete “lid” struc-
ponds already exist, to be supported by reconfigured tures that would allow urban development to expand over
regional water systems of various kinds. None of the existing highways and connect to the future shoreline
entries recognized that intertidal wetlands are at risk of from a higher elevation, were also included.
submergence and collapse. The use of this typology reveals that the emphasis in
The winning proposals can be mapped onto a diagram May 2015 was quite different than that in 2009, perhaps
defined by the four types included in the typology because of SFEI’s knowledge about the effects that mov-

www.frontiersinecology.org © The Ecological Society of America

K Hill Coastal adaptation to sea-level rise

able barriers would have on tidal wetlands, as well as range of options, with a greater interest in using dynamic 475
about their potential costs and feasibility. The reasons for landforms as engineered components of infrastructure.
the change in priorities are not stated explicitly in the The financial cost of all of these adaptation proposals is
BCDC/SFEI workshop materials, but guidance to the high (Jonkman et al. 2013), and there will be intense
workshop participants stated that “strategies should max- competition for future funding. Policy makers, planners,
imize nature-based adaptation solutions where appropri- and ecologists need ways to explain these choices to the
ate” (Case Study 1.1 in SFEI 2015). Recent adaptation public that will help them understand the range of alter-
proposals for Ocean Beach in San Francisco (SPUR native strategies, and increase public support for both
2012) and the Metropolitan Transportation Commission long- and short-term investments that may help to sus-
(Clevenger et al. 2014) that pre-dated the 2015 workshop tain coastal ecosystems and urban districts under new cli-
also made dynamic landforms a priority as strategies for mate regimes.
coastal infrastructure, suggesting that the region may be
developing a preference for multi-benefit strategies that n Acknowledgements
provide both protection and habitat.
Discussions at BCDC’s 2015 workshop also considered I thank the consulting engineers who educated me about
the relative vulnerability of adjacent land areas to engineering practices, and the public agency leaders who
increased flooding, including both wetlands and urban dis- included me in strategy discussions. Without the opportu-
tricts. Workshop participants were encouraged to promote nity to learn from both private consultants and public
equitable solutions that increase resilience in communi- agency staff grappling with storm surges and sea-level rise,
ties” and “restore and enhance diversity of Bay ecosystems I would have been unable to develop the typology
and wildlife” (Case Study 1.1 in SFEI 2015). The relative described here.
vulnerability of shore zone ecosystems and urban land areas
was reflected in participants’ choices about the need for n References
different levels of robustness in the coastal infrastructure Aarninkof SGJ, van Dalfsen JA, Mulder JPM, and Rijks D. 2010.
that would be paired with those land areas. The option of Sustainable development of nourished shorelines: innovations
creating floodable urban development areas, such as those in project design and realisation. In: Proceedings of the PIANC
that exist in Hamburg, Germany, or in London (see MMX Congress 2010. Liverpool, UK: Waterman.
Greater London Authority, London [England] and Mayor Aerts JCJH, Wouter Botzen WJ, Bowman M, et al. (Eds). 2011.
Climate adaptation and flood risk in coastal cities. Amsterdam,
of London 2008; Clevenger et al. 2014), was also discussed.
the Netherlands: Earthscan Climate.
This change represents a considerable shift in strategy from Arcadis. 2014. Coastal green infrastructure research plan for New
the 2009 competition results. York City. A report prepared for the Hudson River Estuary
Program, New York State Department of Environmental
n Conclusion Conservation. Amsterdam, the Netherlands: Arcadis. www.
dec.ny.gov/docs/remediation_hudson_pdf/cginyc.pdf. Viewed
Over the next century and beyond, policy makers, plan- 12 Oct 2015.
ners, and ecologists will likely be forced to deal with Barroca B, Becue V, Serre D, and Balsells M. 2015. Making urban
major changes in coastal ecosystems and coastal commu- flood resilience more operational: current practice. P I Civil Eng
– Wat M 168: 57–65.
nities. A simple typology of coastal infrastructure strate- Barton ME, Brown S, and Nicholls RJ. 2014. Shoreline response of
gies based on long-term historical models in coastal engi- eroding soft cliffs due to hard defences. P I Civil Eng – Mar En
neering can be used to support both the analysis and 167: 3–14.
generation of a complete range of alternative proposals BCDC (San Francisco Bay Commission on Development and
for adaptation to sea-level rise. Specifically, the omission Conservation). 2009. Rising Tides Competition. www.rising
of some strategies and emphasis on others is evident when tidescompetition.com/risingtides/Home.html. Viewed 12 Oct
sets of alternatives are represented in a typological solu- 2015.
tion space. Important questions remain about the dynam- Beach D. 2003. Coastal sprawl: the effects of urban design on
aquatic ecosystems in the United States. Philadelphia, PA: The
ics of specific coastal systems, which need to be under- Pew Charitable Trusts.
stood before shore zones can be classified in ways that Biging G, Radke J, and Lee JH. 2012. Impacts of predicted sea level
would allow a rational match to be made between infra- rise and extreme storm events on the transportation infrastruc-
structure strategies and natural processes. Similarly, clas- ture in the San Francisco Bay region. CEC-500-2012-040.
sifications of the vulnerability of land areas that will be Sacramento, CA: California Energy Commission.
subject to more frequent flooding in the future must Blackman DJ. 1982. Ancient harbors in the Mediterranean. Int J
evolve with greater sophistication to facilitate the match- Naut Archaeol 11: 79–104, 185–211.
Borges LMS. 2014. Biodegradation of wood exposed in the marine
ing process between infrastructure choices and the social, environment: evaluation of the hazard posed by marine wood-
biological, and physical conditions of urban districts and borers in fifteen European sites. Int Biodeter Biodegr 96: 97–104.
ecosystems. Bridges TS, Wagner PA, Burks-Copes KA, et al. 2015. Use of nat-
Proposals for coastal infrastructure as adaptations to ural and nature-based features for coastal resilience. Wash-
sea-level rise are shifting toward the inclusion of a wider ington, DC: US Army Corps of Engineers. www.nad.usace.

© The Ecological Society of America www.frontiersinecology.org

 Coastal adaptation to sea-level rise K Hill

army.mil/Portals/40/docs/NACCS/NNBF%20FINAL.pdf. Louters T, Mulder JPM, Postma R, and Hallie FP. 1991. Changes in
476 Viewed 12 Oct 2015. coastal morphological processes due to the closure of tidal
Bruun P. 2005. Recent developments in shipping ports and termi- inlets in the SW Netherlands. J Coast Res 7: 635–52.
nals. J Coastal Res S46: 1–21. Mattheus CR, Rodriguez AB, McKee BA, and Currin CA. 2010.
Bulleri F and Chapman MG. 2010. The introduction of coastal Impact of land-use change and hard structures on the evolution
infrastructure as a driver of change in marine environments. J of fringing marsh shorelines. Estuar Coast Shelf Sci 88: 365–76.
Appl Ecol 47: 26–35. Mees HLP, Dijk J, van Soest D, et al. 2014. A method for the delib-
Burcharth HF, Lykke Andersen T, and Lara JL. 2014. Upgrade of erate and deliberative selection of policy instrument mixes for
coastal defence structures against increased loadings caused by climate change adaptation. Ecol Soc 19: 58.
climate change: a first methodological approach. Coast Eng 87: Morris RKA. 2013. Geomorphological analogues for large estuar-
112–21. ine engineering projects: a case study of barrages, causeways
Charlier RH, Chaineuxt MCP, and Morcos S. 2005. Panorama of and tidal energy projects. Ocean Coast Manage 79: 52–61.
the history of coastal protection. J Coastal Res 21: 79–111. Nordstrom KF. 2014. Living with shore protection structures: a
Chernoff H and Moses LE. 1959. Elementary decision theory. New review. Estuar Coast Shelf Sci 150: 11–23.
York, NY: John Wiley and Sons. Perkol-Finkel S, Ferrario F, Nicotera V, and Airoldi L. 2012.
Clevenger C, Feng T, LaClair J, et al. 2014. Climate change and Conservation challenges in urban seascapes: promoting the
extreme weather adaptation options for transportation assets in growth of threatened species on coastal infrastructures. J Appl
the Bay Area: pilot project. San Francisco, CA: Metropolitan Ecol 49: 1457–66.
Transportation Commission. Pilkey OH and Dixon KL. 1996. The Corps and the shore.
Doty DH and Glick WH. 1994. Typologies as a unique form of the- Washington, DC: Island Press.
ory building: toward improved understanding and modeling. Reeder T and Ranger N. 2011. How do you adapt in an uncertain
Acad Manage Rev 19: 230. world? Lessons from the Thames Estuary 2100 project.
Eelkema M, Wang ZB, Hibma A, and Stive MJF. 2013. Washington, DC: World Resources Institute.
Morphological effects of the Eastern Scheldt storm surge bar- Reid RL. 2013. Defending New Orleans. Civil Eng 83: 43–83.
rier on the ebb-tidal delta. Coast Eng J 55: 1350010. Restemeyer B, Woltjer J, and van den Brink M. 2015. A strategy-
Flood JF and Cahoon LB. 2011. Risks to coastal wastewater collec- based framework for assessing the flood resilience of cities – a
tion systems from sea-level rise and climate change. J Coastal Hamburg case study. Planning Theor Pract 16: 45–62.
Res 274: 652–60. Ruiz GM, Fofonoff PW, Carlton JT, et al. 2000. Invasion of coastal
Gedan KB, Silliman BR, and Bertness MD. 2009. Centuries of marine communities in North America: apparent patterns,
human-driven change in salt marsh ecosystems. Annu Rev Mar processes, and biases. Annu Rev Ecol Syst 31: 481–531.
Sci 1: 117–41. Schifter I, González-Macías C, Salazar-Coria L, and González-
Giannico G and Souder J. 2005. Tide gates in the Pacific Northwest: Lozano C. 2011. Pollution in estuarine and bay sediments at a
operation, types and environmental effects. Corvallis, OR: refinery complex located on the Mexican Pacific Ocean. Int J
Oregon Sea Grant. Research Paper ORESU-T-05-001. Environ Stud 68: 83–106.
Greater London Authority, London (England) and Mayor of Sekovski I, Newton A, and Dennison WC. 2012. Megacities in the
London. 2008. East London green grid framework: London coastal zone: using a driver–pressure–state–impact–response
plan (consolidated with alterations since 2004) supplementary framework to address complex environmental problems. Estuar
planning guidance. London, UK: Greater London Authority. Coast Shelf Sci 96: 48–59.
Henry T. 1855. Natural history of Pliny the Elder. London, UK: SFEI (San Francisco Estuary Institute). 2015. Case Study 1.1,
HG Bohn. BCDC Policies for a Rising Bay Project. San Francisco, CA:
Hill K. 2011. Climate-resilient urban waterfronts. In: Aerts J, SFEI.
Botzen W, Bowman M, et al. (Eds). Climate adaptation and SPUR. 2012. Ocean beach master plan. San Francisco, CA: SPUR.
flood risk in coastal cities. London, UK: Routledge. Sterr H. 2008. Assessment of vulnerability and adaptation to sea-
Housley J and Thompson E. 2008. Planning and design process. In: level rise for the coastal zone of Germany. J Coastal Res 242:
Coastal engineering manual. Washington, DC: US Army 380–93.
Corps of Engineers. Stevens R, Hill K, Burgess N, and Grady A. 2014. New beach
Jennerjahn TC and Mitchell SB. 2013. Pressures, stresses, shocks designs as urban adaptation to sea level rise. Landscape Res Rec
and trends in estuarine ecosystems – an introduction and syn- 1: 176–87.
thesis. Estuar Coast Shelf Sci 130: 1–8. Stive MJF, de Schipper MA, Luijendijk AP, et al. 2013. A new
Johnston A, Slovinsky P, and Yates KL. 2014. Assessing the vulner- alternative to saving our beaches from sea-level rise: the Sand
ability of coastal infrastructure to sea level rise using multi-cri- Engine. J Coastal Res 290: 1001–08.
teria analysis in Scarborough, Maine (USA). Ocean Coast Tallet P and Marouard G. 2014. The Harbor of Khufu on the Red
Manage 95: 176–88. Sea Coast at Wadi al-Jarf, Egypt. Near East Archaeol 77: 4–14.
Jonkman SN, Hillen MM, Nicholls RJ, et al. 2013. Costs of adapt- Van de Ven GP. 1993. Man-made lowlands: a history of water man-
ing coastal defences to sea-level rise – new estimates and their agement and land reclamation in the Netherlands. Utrecht,
implications. J Coast Res 290: 1212–26. the Netherlands: Matrijs.
Kadiri M, Ahmadian R, Bockelmann-Evans B, et al. 2012. A VanKoningsveld M, Mulder JPM, Stive MJF, et al. 2008. Living
review of the potential water quality impacts of tidal renewable with sea-level rise and climate change: a case study of the
energy systems. Renew Sust Energ Rev 16: 329–41. Netherlands. J Coastal Res 242: 367–79.
Lister N-M. 2007. Sustainable large parks: ecological design or van Slobbe E, de Vriend HJ, Aarninkhof S, et al. 2013. Building
designer ecology? In: Czerniak J and Hargreaves G (Eds). Large with nature: in search of resilient storm surge protection strate-
parks. Princeton, NJ: Princeton Architectural Press. gies. Nat Hazards 66: 1461–80.

www.frontiersinecology.org © The Ecological Society of America