REVIEW ARTICLE

PUBLISHED ONLINE: 29 OCTOBER 2013 | DOI: 10.1038/NCLIMATE1970

The role of coastal plant communities for climate

change mitigation and adaptation
Carlos M. Duarte1,2,3,*, Iñigo J. Losada4, Iris E. Hendriks2, Inés Mazarrasa2 and Núria Marbà2
Marine vegetated habitats (seagrasses, salt-marshes, macroalgae and mangroves) occupy 0.2% of the ocean surface, but
contribute 50% of carbon burial in marine sediments. Their canopies dissipate wave energy and high burial rates raise the
seafloor, buffering the impacts of rising sea level and wave action that are associated with climate change. The loss of a third
of the global cover of these ecosystems involves a loss of CO2 sinks and the emission of 1 Pg CO2 annually. The conservation,
restoration and use of vegetated coastal habitats in eco-engineering solutions for coastal protection provide a promising
strategy, delivering significant capacity for climate change mitigation and adaption.

T
he loss of natural CO2 sinks and reservoirs results in about (Intertidal marshes) in the Ramsar Classification System for Wetland
12–20% of anthropogenic greenhouse gas emissions1. Strategies Type combines freshwater and salt marshes (Recommendation 4.7
to mitigate climate change that are based on actions to prevent as amended by Resolution VI.5 of the Ramsar Conference of the
this loss have focused on the conservation of terrestrial sinks, pri- Contracting Parties), which are reported jointly providing an imped-
marily tropical forests2. Reports that vegetated coastal habitats rank iment to assessments of the area occupied by salt marshes alone.
among the most intense carbon sinks in the biosphere3 lead to so- About 25% to 50% of the area covered by vegetated coastal
called blue carbon strategies4 to explore their potential for mitigating habitats has been lost in the past 50 years (Table 1). Losses of sea-
climate change, stimulating an increase in papers on the topic from grass, which are accelerating globally 8, have been mostly caused
30 studies published in 2005 to 110 papers in 2012. In parallel, the by increased nutrient inputs and coastal transformation9, and salt-
role of vegetated marine ecosystems in fighting climate change can marshes and mangroves have been lost due to changes in land use,
be developed beyond conservation of CO2 sink capacity4–6 by con- coastal transformation and reclamation10,11. Losses of seagrass12,13 and
sidering their contribution to both mitigation of CO2 emissions and kelps14 associated with heat waves indicates that climate change may
adaption to sea-level rise, increasing wave energy and storm surges. lead to loss of seagrass in some areas, such as the Mediterranean15.
Here we present the scientific basis and opportunities for a com- On the other hand, mangroves have been reported to extend their
prehensive strategy to use vegetated coastal habitats to mitigate range polewards with climate change16.
and adapt to climate change. We focus specifically on two aspects: Submerged canopies reduce flow and turbulence17,18, increase the
the capacity of these habitats to act as CO2 sinks, referring to recent bottom shear stress and dampen wave energy 19,20 and flow velocity 21,
reviews for further detail, and their ability to protect the coast against thereby promoting sedimentation and reducing sediment resuspen-
erosion from sea-level rise and increasing wave action (as well as pro- sion22. Partially submerged vegetation, such as salt marshes and man-
viding sources of biodiesel). These are roles that hold considerable grove forests in tidal areas, also affect flow speed23, and reduce wave
potential for climate change adaptation and mitigation strategies but action24 and sediment deposition23–26. More specifically, mangroves
that have not yet received sufficient attention. may trap about 80% of suspended sediment 25. Through their high
productivity and capacity to enhance sediment accretion, seagrass,
Vegetated coastal habitats in the biosphere salt marshes and mangroves build large carbon deposits while rais-
Characterized by the presence of macrophytes, both submerged (sea- ing the sea floor (Tables 2 and 3), acting as important carbon sinks
grass and macroalgae) and partially emerged (mangroves and salt- and mitigating the impacts of sea-level rise on coastline.
marshes), these habitats occupy a narrow fringe — from the upper The ability of vegetated coastal habitats to engineer their envi-
intertidal zone to about 40 m depth — along the shores of all conti- ronment underpins their remarkable capacity for supplying ecosys-
nents. Globally, they extend over approximately 2.3–7.0 million km2, tems services26. A pioneer survey of ecosystem services across the
with macroalgae being the largest contributors and mangroves biosphere ranked vegetated coastal habitats among the most valu-
accounting for the smallest area (Table 1). The global area and trends able ecosystems on Earth27, primarily for their capacities to regulate
in area change of mangrove forests are reasonably well estimated, nutrient fluxes, provide habitat 26 and climatic regulation, and for
whereas those for seagrass meadows, which cannot be retrieved from their function as CO2 sinks28 and coastal protection26.
remote sensing products, have much greater uncertainty. However,
global estimates for salt marshes suffer from severe, 20-fold uncer- Carbon sequestration
tainties (Table  1). This is surprising as salt marshes can be easily Despite the small fraction of the ocean surface occupied by salt
assessed by remote sensing, and these uncertainties probably derive marsh, mangrove and seagrass ecosystems, they account for 46.9%
from the amalgamation of salt marshes in the broader and ambigu- of the total carbon burial in ocean sediments3,4. Most macroalgal
ous category of wetlands in the Ramsar Convention (Convention stands develop on hard, rocky substrates, and — despite their high
on Wetlands of International Importance)7. Specifically, category H productivity (Table 1) and capacity to trap suspended particles  — do

UWA Oceans Institute, University of Western Australia, 35 Stirling Highway, 6009 Crawley, Western Australia, Australia, 2Department of Global Change
1

Research, Instituto Mediterráneo de Estudios Avanzados (CSIC-UIB), Miquel Marqués 21, 07190 Esporles, Spain, 3Faculty of Marine Sciences, King
Abdulaziz University, PO Box 80207, Jeddah 21589, Saudi Arabia,4Instituto de Hidráulica Ambiental, Universidad de Cantabria, Isabel Torres 15, 39011
Santander, Spain. *e-mail: [email protected]

NATURE CLIMATE CHANGE | VOL 3 | NOVEMBER 2013 | www.nature.com/natureclimatechange 961

© 2013 Macmillan Publishers Limited. All rights reserved

REVIEW ARTICLE NATURE CLIMATE CHANGE DOI: 10.1038/NCLIMATE1970

Table 1 | Extension, production and losses of vegetated coastal ecosystems.

Global extension Local net primary Global net primary Global loss rate Percentage of area lost since
Ecosystem (km2) production (g C m−2 yr−1) production (Pg C yr−1) (% yr−1) the Second World War

Salt marshes 22,000–400,0005 44029 0.01–0.18 1–261 –

200,00098
Mangroves 137,760–152,3615 40029 0.06 1–361 30–5010,99
152,308100 0.7–1.798 >25100
Seagrasses 177,000–600,0005 27829 0.05–0.17 0.98 308
Macroalgae 2,000,000100–6,800,00029 9429 0.19–0.64 – –
Superscript numbers indicate the reference sources of data. For some parameters and habitats more than one estimate (or range of estimates) is available.

Table 2 | Carbon burial and soil stocks in vegetated coastal ecosystems.

Local C burial rate Local C stock in soil Global C burial rate Global C stock in soil
Ecosystem (g C m−2 yr−1) (Mg C ha−1) (Tg C yr−1) (Pg C)
Salt marshes 218±245 162 (259)65 4.8–87.35 0.4–6.5
Mangroves 163 35
255 (683.4)
64 38
22.5–24.9 35
9.4–10.4
Seagrasses 138±385 139.7 (372)39 48.0–1125 4.2–8.439
Mean and, when available, standard error of the mean (±s.e.m.) of organic carbon (C) burial and stock within the top 1 m of soil. Maximum local C stock is provided in brackets. Global C stocks are estimated from
local C stocks and ecosystem extension (Table 1) unless indicated. Superscript numbers indicate the reference sources of data.

Table 3 | Sediment accretion and elevation rates in vegetated coastal ecosystems.

Ecosystem Accretion rates (mm yr−1) Elevation rates (mm yr−1)
Range Average Median s.d. n Range Average Median s.d. n
Salt marshes 0.39–61.1 6.73 5.5 0.7 98 –6.92–25 3.76 2.01 1.33 22
Mangroves 0.34–20.8 5.47 4.5 0.38 123 –9.5–11.3 1.87 1.4 0.53 58
Seagrasses 0.61–6 2.02 1.48 0.44 12 –5.2–10.2 –0.08 –5.17 5.14 3
Range, average, median, standard deviation (s.d.) and number of observations (n) for a compilation of representative values of sediment accretion and elevation rates compiled from the published literature101.

not develop significant carbon deposits. Community primary century 35. Moreover, sediment accretion responds to climate change
production generally exceeds respiration in vegetated coastal habi- through feedbacks that involve increased plant growth and produc-
tats3,28 leading to their capacity for producing excess organic carbon tion, which are conducive to faster accretion rates with increasing CO2
and acting as CO2 sinks (Fig. 1). Carbon sequestration in vegetated (ref. 36) and sea-level rise37. Indeed, recent models indicate that cli-
coastal habitats is further enhanced by their unique ability to trap mate change will increase salt marsh carbon burial and accretion rates
particles from the water flow and store them in the soil29 (Fig.  1). in the first half of the twenty-first century 38.
As a result, burial rates of organic carbon in salt marsh, mangrove The long-term preservation and continuous accretion of carbon in
and seagrass ecosystems are exceptionally high (Table  2), exceed- the soil of coastal habitats with sea-level rise leads to the development
ing those in the soils of terrestrial forests by 30–50 fold5. Globally, of organic carbon deposits several metres thick30,38. The magnitude of
coastal vegetated habitats bury a similar amount of organic carbon carbon deposits under the top metre of soil in a salt marsh or seagrass
to terrestrial forests annually, despite the extent of coastal marine meadow is similar, on average, to that in the upper 1-m soil in ter-
vegetation being less than 3% of that of forests. restrial forests (Table 2), whereas the top metre of soil in mangrove
The carbon buried in coastal vegetated ecosystems can be pre- forests stores more than three times the organic carbon contained in
served over millennia, as demonstrated by radiocarbon dating of the upper soil under forests on land39. Globally, salt marsh, mangrove
seagrass30, salt marsh31 and mangrove soils32. The efficient preserva- and seagrass ecosystems store about 10 Pg C each in their top 1-m soil
tion of the carbon under these habitats is due to: slow decomposi- layer 39 (Table 2). This is one order of magnitude lower than the soil
tion rates33; low nitrogen and phosphorous concentrations in plant carbon stock under terrestrial forests, but still large enough to play a
tissues; low oxygen levels in the sediments; and the allocation of a role in the global carbon cycle.
large fraction, often exceeding 50%, of plant biomass production to
roots and rhizomes that are buried into the soil34. In addition, the Protection against coastal flooding and erosion
entangled network of roots (and rhizomes) and the dense canopy of The risks of accelerated sea-level rise with climate change are fur-
coastal vegetation protect soil carbon deposits from erosion (Fig. 1). ther enhanced by associated increases in the frequency of extreme sea
Indeed, some vegetated coastal habitats can support organic-rich level, waves and the strength of storm surges40, resulting in a higher
soils5 that deserve conservation measures. intensity and frequency of flooding and erosion of vulnerable coastal
Seagrass, salt marshes and mangroves accumulate enough carbon areas. Observations and numerical reanalysis have shown that signifi-
and mineral particles to support characteristic sediment accretion rates cant wave-height variations are clearly linked to climate modes41,42 and
exceeding 10 cm per century, with the highest accretion rates found in that wave heights have increased in the North Pacific, North Atlantic
salt marshes (Table 3, Fig. 1). Recent (that is, twentieth century) accre- and Southern Ocean during the past century 43–45. Sea level has been
tion rates in mangrove forests have been reported to average 28 cm per rising globally at an average rate of 1.6±0.2 mm yr–1 since 190146, and

962 NATURE CLIMATE CHANGE | VOL 3 | NOVEMBER 2013 | www.nature.com/natureclimatechange

© 2013 Macmillan Publishers Limited. All rights reserved

NATURE CLIMATE CHANGE DOI: 10.1038/NCLIMATE1970 REVIEW ARTICLE
moderate emission scenarios project a future global mean sea-level
rise of 0.21–0.48  m by 2100 (Intergovernmental Panel on Climate Energy dissipation/wave attentuation
Change scenario SRES A1B for 2090–2099)47, with some recent pro- Particle trapping
jections exceeding a global mean sea-level rise of 1 m (ref. 48). Even
if wave projections carry substantial uncertainty, the direct connec- CO2 uptake
tion between wind and storminess indicates that climate change is
Carbon deposition
likely to have a significant impact on wave heights and other wave
parameters49. As a result, coastal flooding and erosion will be, and
are already becoming, a major threat to coastal areas, demanding the
introduction of sustainable measures to cope with this problem.
Coastal protection measures against wave action are purely based Carbon burial
on the principle that the incident wave energy flux — defined as the
Opportunity for
wave energy multiplied by the velocity of a wave group — has to be expansion/conservation
balanced by the total energy reflected, transmitted and dissipated
at the protection element 50. Formulated in a simple way, the wave Reef
energy flux that is not reflected offshore or dissipated is transmitted formation Sediment
onshore, thereby inducing flooding, erosion or damages to infra- stabilization
structure, goods and services, or loss of lives. Vegetated coastal hab-
CO2 uptake
itats, through their capacity to provide coastal protection26, could
Energy dissipation/wave attentuation
assist in mitigating the impacts of sea-level rise and the associated
Particle trapping
increase in wave action.
The function of vegetated coastal habitats for coastal protec-
tion involves the attenuation of wave transmission onshore (Fig. 1),
Carbon deposition
which can be achieved by: (1) inducing wave breaking as the main
damping mechanism19; (2) dissipating energy through flow separa-
tion51; (3) dissipating energy through friction on rough surfaces;
(4) dissipating energy through porous friction52,53; (5) producing a
barrier effect that reflects energy in the offshore direction — and a Opportunity for
Carbon burial
combination of the above mechanisms. expansion/conservation
The capacity of vegetated coastal habitats for protecting the coast
against the different dynamics considered (waves, storm surges, tsu-
namis and currents) is highly dependent on both the large- and small- Sediment stabilization Sediment
scale characteristics of these ecosystems (Supplementary Table 1). accretion
The relevant elements of natural ecosystems are their location and
geometry with respect to the incoming dynamics, which may affect CO2 uptake
shoaling, refraction, diffraction, blocking or breaking. Wave atten-
uation will be dependent on the freeboard of the vegetated field,
namely the relationship between the submergence and total water
depth (h); the relative wave height (H/h, where H is wave height), the
width (b) of the field with respect to the incoming wavelength (L) (the
longer the wave, the wider the domain needs be to achieve damping),
the vegetation density, the nature of the substrate and the quality and
abundance of the aboveground biomass. The geometry of each indi-
vidual plant (roots, stems and canopies), its buoyancy, stiffness and Energy dissipation/wave attentuation
degrees of freedom also affect wave attenuation50,54. Carbon burial
Wave breaking is closely controlled by the relationship between
H and h, the ratio of which increases as waves propagate towards
the shore by the combined effect of wave shoaling and shallower Sediment accretion
depths. Breaking takes place when a certain threshold of H/h has
been reached. Therefore, changes in the bathymetry may contrib-
Opportunity for Sediment stabilization
ute to dissipation. Hence, the capacity of vegetated coastal habitats expansion/conservation
to raise the sea floor at speeds that can match or exceed current
sea-level rise (Table 3), thereby counterbalancing the effect of sea-
Figure 1 | Key processes of vegetated coastal habitats for climate change
level rise on h, allows them to remain effective in breaking waves
mitigation and adaptation. Processes that affect the capacity for climate
with moderate to high scenarios of sea-level rise, in areas where
change mitigation (CO2 sinks) and adaptation (shore line protection from
subsidence and other processes that lower the elevation of the
rising sea level) are shown for seagrass meadows (upper panel), salt
shore are not important.
marshes (middle panel) and mangrove forests (lower panel). Blue arrows
Flow separation occurs at the edges of large structures, or
indicate transport of atmospheric or dissolved material, red arrows show
on the lee side of small structures. The turbulent wake gener-
transport of particulates and purple arrows indicate vegetative growth.
ated is a sink of energy. Flow separation is controlled by the
Images reproduced with permission: Top, Posidonia meadow, water colour
Reynolds number (ratio  between inertial and viscous forces),
by Miquel Alcaraz; middle, Spartina in Rattekaai salt marsh, photo by
the Keulegan–Carpenter number (ratio between drag and iner-
Iris Hendriks; bottom, Mangrove Forest, photo by Rohan Arthur.
tia  forces) and the relative roughness of the body surface. The
parameters considered to define the contribution of each of the or mangroves are typical sources of this kind of dissipation55.
forces depend on the process considered. Fields of slender, verti- Vegetated coastal habitats can thus dissipate wave energy through
cal rigid or flexible elements such as seagrass, kelp, salt-marshes flow separation.

NATURE CLIMATE CHANGE | VOL 3 | NOVEMBER 2013 | www.nature.com/natureclimatechange 963

© 2013 Macmillan Publishers Limited. All rights reserved

REVIEW ARTICLE NATURE CLIMATE CHANGE DOI: 10.1038/NCLIMATE1970
Seagrass and other vegetated coastal habitats also provide Blue Carbon initiatives could broaden the participation in the
protection by the dissipation of wave energy thanks to friction result- current REDD+ scheme, as small island states — who are eager to
ing from their presence increasing the bottom roughness, reducing mitigate the climate change that threatens their livelihoods through
near-bed flow velocity and elevating the bottom boundary layer56. sea-level rise — have reduced opportunities at present to partici-
Vegetated coastal habitats with their particular stem density and flex- pate in REDD+ due to limited land space. However, island states
ibility also provide a porous medium with a large energy dissipation often have extensive shallow platforms (for example, the Bahamas,
capacity 57. Turbulent and laminar flow inside porous structures is the Maldives and Cuba) that support vegetated coastal habitats,
an important sink of wave energy that is controlled by the Reynolds where conservation initiatives funded by a Blue Carbon extension
number, the size of the porous elements and the porosity 58. of REDD+ could be successfully deployed and contribute to cli-
Hence, vegetated coastal habitats act on all of the mechanisms mate change mitigation. As vegetated coastal habitats are intense
that dissipate wave energy (Supplementary Table 1). However, their carbon sinks (Table 2), even conservation and revegetation projects
contribution to bathymetric changes through sediment accumu- that involve relatively limited areas can be significant.
lation (Supplementary Table  1 and ref.  59) and shoreline accretion Unlike lost forests, which were largely transformed to cropland
is key to shoreline protection, aside from direct damping of the and pastures, most lost coastal habitats have not been transformed
incoming waves. to other uses and generally lack property rights. They are there-
fore available for recolonization by coastal vegetation. Indeed,
Climate change mitigation vegetated coastal habitats can be restored at a large scale. The best
The conservation and protection of ecosystems that act as carbon demonstration is the complete reforestation of the Mekong Delta
sinks are among the cheapest, safest and easiest solutions to reduce mangrove forests (which were completely destroyed by the US
greenhouse gas emissions and promote adaptation to climate Air Force in the war 40  years ago) through the initiative of the
change6,60. High loss rates of vegetated coastal habitats (Table 2), ten Vietnamese government, arguably the largest ecosystem restora-
times faster than those of tropical forests61, represent a major loss tion ever undertaken67. Likewise, large-scale mangrove afforesta-
of natural CO2 sink capacity and coastal protection and therefore tion programmes have been successful in Thailand and — more
contribute to the component of increased greenhouse gas emis- recently — India; and large-scale salt marsh restoration schemes
sions that is termed land-use change1. Specifically, by combining are in place in China68 and the USA69. Seagrass transplants can,
specific loss rates (Table 1) with estimates of global CO2 sink capac- because of their clonal nature, return several million shoots per site
ity (Table 2), the loss of vegetated coastal habitats is found to repre- after a few years70, even at a typical success rate of 30%. No carbon
sent a loss of CO2 sink capacity of approximately 7–20 Tg CO2 yr–1. burial estimates are available for seagrass revegetation projects, but
The fate of soil carbon stored in tidal wetlands that are submerged one study found that sediment carbon accumulation by a revege-
when vegetation is lost is an open question that requires research. tated seagrass meadow doubled that in bare sediments 9 years after
However, there is evidence that soil carbon stores can be destabi- planting 71. However, studies conducted over the past decade have
lized when mangroves and salt marsh cover is removed, leaving shown that carbon burial rates in restored mangrove forests72 and
soils exposed to the atmosphere — the unvegetated habitats act as salt marshes73 are similar to those in undisturbed habitats, despite
sources of CO2 and CH4 to the atmosphere62–64. Pendleton et al.65 sediment carbon pools in revegetated mangroves, salt marshes
estimated that 0.15–1.02 Pg CO2 are being released annually from and seagrass meadows being lower than in undisturbed habitats
loss or conversion of vegetated coastal habitats, assuming that all 9 (ref. 71), 20 (ref. 72) and 28 (ref. 73) years after planting. Hence,
of the organic carbon in biomass and the top metre of soils is lost. restoration of coastal vegetation has been proposed to be a sound
This estimate, which carries considerable uncertainty, is equiva- strategy to mitigate climate change74.
lent to 3–19% of that from deforestation globally, and results in However, extending the REDD+ programme to encompass
economic damages of US$6–42 billion annually from loss of CO2 coastal vegetation is not without challenges. Current REDD+
sequestration alone65, not accounting for the damages associated accountings only consider the carbon stored as aboveground bio-
with the loss of coastal protection capacity. The uncertainty in the mass, where much of carbon capture by forest is bound. In contrast,
loss of carbon sink capacity and in emissions released by habitat coastal habitats store most of the carbon that is sequestered in sedi-
losses is dominated by uncertainties in the global extent and loss ments (for example, 68% for mangroves and 95% for seagrasses39).
rates of these habitats, with the rough estimates available (Table 1), Accounting for soil carbon within REDD+ also requires knowledge
which are particularly coarse in the case of salt marshes, propagat- of the origin of the carbon stored in sediments. Although this is
ing unverified across citation networks. Improved estimates of the possible through stable isotope analysis of sediment carbon in sea-
global extent and loss rates of vegetated coastal habitats are urgently grass meadows28, apportioning carbon between allochthonous and
needed to assess the potential of conservation and restoration of autochthonous sources in salt marsh and mangrove sediments is not
these habitats as an element of climate change mitigation strategies. as straightforward.
The use of vegetated coastal ecosystems to protect and restore Expanding REDD+ to vegetated coastal habitats requires the
lost CO2 sink capacity and prevent the loss of deposits to mitigate development and acceptance of protocols for measuring, report-
climate change — Blue Carbon initiatives — was proposed in 20094. ing, verifying and monitoring the carbon stored in sediments.
One aim is to encourage various carbon trading programmes to Revegetation programmess, which are an option for coastal wet-
credit activities in tidal wetland and seagrass ecosystems. For exam- lands, are not currently considered in REDD+, which focuses on
ple, there is interest in extending the REDD+ (Reducing Emissions conservation. Revegetation projects require that suitable conditions
from Deforestation and Forest Degradation), including conserva- be re-established and, in the case of seagrass, can be met with mixed
tion, sustainable management of forests and enhancement of for- success. The need for monetary compensation may also preclude
est carbon stocks) programme — a UN programme drawing from revegetation programmes where habitats have been converted to
voluntary payments and market CO2 taxes to pay land-owners to other uses (for example, aquaculture farms, coastal infrastructure
conserve forests66 — to the coastal ocean. Avoidance of losses in or reclaimed areas). Finally, although payments under REDD+ con-
vegetated coastal habitats that are threatened by local human activ- servation approaches can be immediate, revegetation methods may
ity (for example, aquaculture, waste-water discharges to coastal require decades before carbon credits can be collected. An extended
water, coastal tourism developments and reclamation) could help REDD+ programme may not be applicable to coastal habitats that
to maintain CO2 sinks and therefore be considered within the are particularly vulnerable to sea-level rise, as losses of the associated
REDD+ mechanism. carbon sinks cannot be avoided through conservation. Specifically,

964 NATURE CLIMATE CHANGE | VOL 3 | NOVEMBER 2013 | www.nature.com/natureclimatechange

© 2013 Macmillan Publishers Limited. All rights reserved

NATURE CLIMATE CHANGE DOI: 10.1038/NCLIMATE1970 REVIEW ARTICLE
although carbon accumulation in salt-marshes increases with structures, vegetated ecosystems can naturally adapt to changes in
sea-level rise, it does so until a critical rate is reached, beyond which sea levels and wave storminess if they are not severely affected by
the marsh vegetation is drowned, halting carbon accumulation75. human action, as demonstrated by their capacity to adjust accretion
The fate of carbon deposits lost to submergence is not known, but rates to sea-level rise37. Hence, the adaptive capacity of vegetated
the question is important for understanding the full impact of tidal coastal habitats helps maintain their capacity for coastal protection
wetland feedbacks on climate. at a negligible cost, while conserving the ecosystems and maintain-
ing their services84,85. The potential of a coastal ecosystem to pro-
Blue biofuels. Macroalgal beds dominate the global area and pro- tect the coastline does not increase linearly with its size, but varies
duction of vegetated coastal habitats (Table 1), but most macroalgal nonlinearly 19,86. The effectiveness of sediment accretion by vegetated
communities grow on hard substrates and do not contribute to car- coastal habitats in adapting to sea-level rise is dependent, however,
bon sequestration except for the biomass that may be exported to on both the rates of accretion, which vary 10-fold within habitat
the deep sea76. Yet macroalgae can play a role in mitigating climate types (Table 3), and local processes such as compaction, subsidence
change if either wild or aquaculture crops are used to derive biofuels. and local rates of sea-level rise. Indeed, reported elevation rates for
The development of biofuels from mass aquaculture of macroalgae is vegetated coastal habitats tend to be lower than accretion rates, and
a new, vibrant area of research77,78, which has the potential to contrib- the two available estimates that report elevation rates for seagrass
ute to climate change mitigation. The production of ‘blue biofuels’ meadows indicate net subsidence (Table 3 and ref. 59).
has many advantages over that based on land crops (green biofuels), Adaptation strategies that include the conservation, restoration or
generating multiple environmental and societal impacts79,80. Blue introduction of vegetated coastal ecosystems provide a cost-effective
biofuels do not compete for arable land and water with food crops, option for addressing the increased risk from flooding and ero-
whereas green biofuels have co-opted both essential resources from sion under climate change in vulnerable areas. Moreover, produc-
agriculture, becoming a threat to food security 81. Moreover, blue ing vegetated coastal protections, unlike cement-based structures,
biofuels are not currently staples of global significance, and so the generates limited CO2 emissions and in fact removes atmospheric
production does not negatively affect food prices in the global mar- CO2. However, CO2 emissions may be generated where machinery
ket. In contrast, the development of green biofuels has diverted crops is involved in preparing the terrain and revegetating the coast. In
otherwise used as staples (corn or palm oil) away from the food mar- addition, salt marshes and mangrove forests, particularly newly
ket into fuel production, becoming an element in the recent global created ones, can release CH4 and N2O that may partially offset
rise of food prices81. Moreover, neither pesticides nor fertilizers are the carbon sequestered87. Accordingly, best practices to minimize
used in most seaweed farms, which remove nutrients often found in greenhouse gas emissions from eco-engineering projects need be
excess in coastal waters, generating environmental benefits in coastal developed. The adaptation of vegetated coastal habitats to increased
areas affected by eutrophication. In areas where nutrient availabil- CO2 and higher sea level involves an increase in sediment accretion,
ity is insufficient for macroalgae farming, such as in the Yellow Sea also enhancing their capacity to act as carbon sinks. Lastly, vege-
(China), large kelp production is supported by multi-species polycul- tated coastal habitats have a high capacity to produce carbonates
tures, where kelps benefit from nutrients released by scallops, oysters and other materials that contribute to sediment accretion, beach
and/or mussels. Fertilizer production and application are a signifi- nourishment and to dune formation on land54,88, further preventing
cant component of greenhouse gas emissions, as the Haber–Bosch coastal erosion.
reaction — used to produce ammonia for fertilizers  — is an energy- Coastal vegetation may not always offer sufficient protection, as
intensive process and excess fertilizer application is a major source the capacity of natural ecosystems for shoreline protection varies
of N2O (ref.  82), a major greenhouse gas. Finally, phosphorus res- due to seasonal and interannual variations in the development of
ervoirs are being depleted, so replacing land with marine biofuels the vegetation. In addition, the requirements for successful devel-
without fertilizer application will allow more phosphorus to be used opment of coastal vegetation — such as elevation, currents and
to produce food. Hence there is a potential for wild and cultured wave exposure85,89 — are not met everywhere, restricting the areas
macroalgae to help mitigate climate change, while generating signifi- where protection based on coastal vegetation can be successful.
cant additional benefits. Finally, the particular circumstances of an event (that is, tidal con-
ditions and the track of a storm) can also affect the effectiveness of
Climate change adaptation coastal vegetations in shoreline protection.
Vegetated coastal ecosystems are important in protecting the coast
against flooding and erosion due to waves and storm surges Costs. Conserving and restoring vegetated coastal habitats is also
under mean and extreme conditions, including hurricanes relatively inexpensive and is affordable to all countries, includ-
(Supplementary Table 1 and ref. 59). Seagrasses have a particularly ing developing ones. According to the International Federation of
high capacity to dissipate wave energy, whereas salt marshes and the Red Cross and Red Crescent (ref.  90), replanting mangroves
mangroves have a high capacity to protect from surges. Moreover, in Vietnam has helped to reduce the cost of dyke maintenance by
these ecosystems often occur in juxtaposition with seagrass in US$7.3 million a year for an investment of US$1.1 million over the
subtidal areas and salt marshes or mangroves (depending on lati- period 1998–2002. Indeed, existing coastal wetlands in the USA have
tude) in the intertidal zone, thereby increasing their combined been estimated to provide a value of US$23.2 billion a year at present
effectiveness in protecting from waves and surges. in storm protection services91 at only the (minor) cost of conserving
these habitats.
Benefits of ecosystem-based coastline protection. Artificial Artificial coastal structures provide limited or no extra benefits
coastal protection structures are constructed with an expected beyond the function for which they were built, and may generate
service life spanning several decades. Whereas extreme events are negative impacts — possibly including a role in promoting jellyfish
considered in their design, the statistics of extreme events is now blooms92. In contrast, vegetated coastal habitats add benefits such as
affected by climate change and may alter during their operational nutrient cycling, food provision and biodiversity regulation to their
life. As climate change was not generally considered in the design capacity for coastal protection26, thereby further increasing the value
of most coastal defences that are now in place, an intense upgrade and positive externalities of defence strategies that involve these habi-
of coastal defence structures during the coming decades will be tats. For instance, mangrove expansion has been shown to improve
needed worldwide, requiring a huge investment to provide adap- the yield of fisheries93. In turn, conserving vegetated habitats for their
tation to an uncertain level of climate change83. Unlike artificial capacity to sequester CO2 or their role in enhancing biodiversity will

NATURE CLIMATE CHANGE | VOL 3 | NOVEMBER 2013 | www.nature.com/natureclimatechange 965

© 2013 Macmillan Publishers Limited. All rights reserved

REVIEW ARTICLE NATURE CLIMATE CHANGE DOI: 10.1038/NCLIMATE1970

Table 4 | Eco-engineering solutions for coastal areas in the Netherlands.

Environment Problem Eco-dynamic design Area
Tidal Erosion of the Surplus Sand nourishment a
Delfland a
intertidal area Shellfish reefsb Eastern Scheldt
Circle-shaped nourishmentc Eastern Scheldt

Coastal erosion Sand dunes b

Wetlands

Flooding (waves) Optimizing texture of dykes by North Sea c

ecological growth

Non-tidal Flooding (storms) Semi-natural floodplains IJsselmeer d

Willow floodplainsd Noordwaard

a–d refer to the photographic examples of eco-dynamic design shown on the right. Eco-dynamic designs are based on the dynamics of the natural environment: developing hydraulic engineering infrastructure and
at the same time creating opportunities for nature and the environment. Sand nourishment is a technique where extra sand is deposited in an (intertidal) area to counter losses from erosion. In the Delfland Sand
Engine experiment, a larger than normal (surplus) nourishment of 21.5 million m3 of sand was introduced, rising up to 7 m above mean sea level. The sand is gradually redistributed by natural processes over the
shoreface beach and dunes102. http://www.ecoshape.nl/en_GB/delfland-sand-engine.html Images courtesy of: 1, Joop van Houdt/Rijkswaterstaat; 2–4 Ecoshape.

increase coastal protection as a side benefit. Therefore, the restoration A comprehensive coastal eco-engineering programme could
and preservation of these ecosystems can be considered a cost- strike a rational balance between mitigation and adaptation instru-
effective strategy due to the combined services provided for climate ments based on protecting and restoring or introducing different
change mitigation and adaptation. vegetated coastal ecosystems to maximize the potential synergies
between them. Indeed, the separation between adaptation and miti-
Conclusion gation strategies may lead to lost opportunities and to underestimate
For decades vegetated coastal ecosystems have remained the poor the value of conservation, as is clearly the case for a REDD+ exten-
relations of biological conservation61. However, recent findings on sion focused on vegetated coastal ecosystems, which would need
their remarkable capacity for CO2 sequestration and storage, and to account for their role in coastal protection, with a value likely
their capacity for sediment accretion and coastal protection, have to exceed that of CO2 sequestration by at least an order of magni-
converged to identify these habitats as essential elements of a strat- tude. Moreover, cost–benefit analyses should not focus on benefits
egy that combines both climate change adaptation and mitigation. that are associated with climate change mitigation and adaptation
Unfortunately, the role of vegetated coastal habitats as a valid alone, but should encompass the broad suite of services that veg-
alternative in the portfolio of measures for climate change mitiga- etated coastal habitats provide26. Eco-engineering approaches that
tion and adaptation has not been sufficiently considered by coastal involve wetland creation can provide habitat for threatened species,
managers, who still opt for hard adaptation measures and on-land as exemplified by the refugia that mangrove forest offers to critically
mitigation options. Strategies that involve vegetated coastal habi- endangered felids and primates in Africa and Asia97, thereby deliv-
tats are now being included in an eco-engineering approach to ering conservation benefits beyond those associated with coastal
climate change. Eco-engineering emerged in the early 1960s94, but protection or carbon stocks. The eco-engineering approach could
it has only recently gained broad recognition as a new paradigm. become societally and economically efficient and may offer greater
The original starting point to use natural energy sources as the pre- opportunities for countries — especially developing ones — to
dominant input to manipulate and control environmental systems achieve sustainable targets even under limited financial resources
was broadened by Mitsch and Jørgensen95 to ‘the design of sus- and capacity.
tainable ecosystems that integrate human society with its natural Coastal eco-engineering through vegetated coastal ecosystems
environmental to promote both’. Examples of experiments and suc- represents a new paradigm, whose significance can be best under-
cessful implementations of eco-engineering in coastal protection stood by drawing a parallel with material science. Coastal engineer-
projects89,96 can be found for example in the Netherlands, a country ing has introduced a new material whose production, unlike that of
challenged by sea-level rise (Table 4). The preceding discussion cement, does not lead to CO2 emissions, but rather CO2 removal. It
points at a huge opportunity to develop projects and training cur- can achieve comparable efficiency for coastal protection to cement-
ricula in coastal eco-engineering through options that are based on based solutions; can repair itself; can grow; and can adapt to shift-
vegetated coastal habitats to mitigate and adapt to the impacts of ing conditions. This newly discovered material is none other than
climate change. marine plants.

966 NATURE CLIMATE CHANGE | VOL 3 | NOVEMBER 2013 | www.nature.com/natureclimatechange

© 2013 Macmillan Publishers Limited. All rights reserved

NATURE CLIMATE CHANGE DOI: 10.1038/NCLIMATE1970 REVIEW ARTICLE
Received 20 October 2012; accepted 4 July 2013; 32. Filho, P. W. M. et al. Holocene coastal evolution and facies model of the
published online 29 October 2013. Braganca macrotidal flat on the Amazon mangrove coast, northern Brazil.
J. Coast. Res.39, 306–31 (2006).
References 33. Enríquez, S., Duarte, C. M. & Sand-Jensen, K. Patterns in decomposition rates
1. Le Quéré, C. et al. Trends in the sources and sinks of carbon dioxide. among photosynthetic organisms: The importance of detritus C: N:P content.
Nature Geosci. 2, 831–836 (2009). Oecologia 94, 457–471 (1993).
2. Agrawal, A., Nepstad, D. & Chhatre, A. Reducing emissions from 34. Duarte, C. M. et al. Root production and belowground seagrass biomass.
deforestation and forest degradation. Annu. Rev. Environ. Resour. Mar. Ecol. Prog. Ser. 171, 97–108 (1998).
36, 373–396 (2011). 35. Breithaupt, J. L., Smoak, J. M., Smith, T. J. III, Sanders, C. J. & Hoare, A.
3. Duarte, C. M., Middelburg, J. & Caraco, N. Major role of marine vegetation on Organic carbon burial rates in mangrove sediments: Strengthening the global
the oceanic carbon cycle. Biogeosciences 2, 1–8 (2005). budget. Glob. Biogeochem. Cycles 26, GB3011 (2012).
4. Nellemann, C. et al. Blue Carbon: A Rapid Response Assessment 36. Langley, J. A., McKee, K. L., Cahoon, D. R., Cherry, J. A. & Megonigal, J. P.
(United Nations Environment Programme, 2009). Elevated CO2 stimulates marsh elevation gain, counterbalancing sea-level rise.
5. McLeod, E. et al. A blueprint for blue carbon: Towards an improved Proc. Natl Acad. Sci. USA 106, 6182–6186 (2009).
understanding of the role of vegetated coastal habitats in sequestering CO2. 37. Kirwan, M. L. & Mudd, S. M. Response of salt-marsh carbon accumulation to
Front. Ecol. Environ. 9, 552–560 (2011). climate change. Nature 489, 550–553 (2012).
6. Jones, H. P., Hole, D. G. & Zavaleta, E. S. Harnessing nature to help people 38. Donato, D. C. et al. Mangroves among the most carbon-rich forests in the
adapt to climate change. Nature Clim. Change 2, 504–509 (2012). tropics. Nature Geosci. 4, 293–297 (2011).
7. http://www.ramsar.org 39. Fourqurean, J. W. et al. Seagrass ecosystems as a significant global carbon
8. Waycott, M. et al. Accelerating loss of seagrasses across the globe threatens stock. Nature Biogeosci. 5, 505–509 (2012).
coastal ecosystems. Proc. Natl Acad. Sci. USA 106, 12377–12381 (2009). 40. Menéndez, M. & Woodworth, P. L. Changes in extreme high water levels based
9. Orth, R. J. et al. A global crisis for seagrass ecosystems. BioScience on a quasi-global tide-gauge data set. J. Geophys. Res. 115, C10011 (2010).
56, 987–996 (2006). 41. Izaguirre, C., Mendez, F. J., Menendez, M., Luceño, A & Losada, I. J. Extreme
10. Valiela, I., Bowen, J. L. & York, J. K. Mangrove forests: One of the world’s wave climate variability southern Europe using satellite data. J. Geophys. Res.
threatened major tropical environments. BioScience 51, 807–815 (2001). 115, C04009 (2010).
11. Adam, P. Saltmarshes in a time of change. Environ. Conserv. 29, 39–61 (2002). 42. Menéndez, M., Mendez, F. J., Losada, I. J. & Graham, N. E. Variability
12. Reusch, T. B. H., Ehlers, A., Hämmerli, A. & Worm, B. Ecosystem recovery of extreme wave heights in the northeast Pacific Ocean based on buoy
after climatic extremes enhanced by genotypic diversity. Proc. Natl Acad. Sci. measurements. Geophys. Res. Lett. 35, L22607 (2008).
USA 102, 2826–2831 (2005). 43. Hemer, M. A., Church, J. A. & Hunter, J. R. Variability and trends in the
13. Marbà, N. & Duarte, C. M. Mediterranean warming triggers seagrass directional wave climate of the Southern Hemisphere. Int. J. Climatol.
(Posidonia oceanica) shoot mortality. Glob. Change Biol. 30, 475–491 (2010).
16, 2366–2375 (2010). 44. Izaguirre, C., Mendez, F. J., Menendez, M. & Losada, I. J. Global extreme wave
14. Wernberg, T., Thomsen, M. S., Tuya, F. & Kendrick, G. A. height variability based on datellite data. Geophys. Res. Lett. 38, L10607 (2011).
J. Exp. Mar. Biol. Ecol. 400, 264–271 (2011). 45. Young, R. R., Zieger, S. & Babanin. A. V. Global trends in wind speed and
15. Jordà, G., Marbà, N. & Duarte, C. M. Mediterranean seagrass vulnerable to wave height. Science 332, 451–455 (2011).
regional climate warming. Nature Clim. Change 2, 821–824 (2012). 46. Church, J. A. & White, J. W. Sea level rise from the late 19th to early
16. Sorte, C. J. B., Williams, S. L. & Carlton, J. T. Marine range shifts and species 21st century. Surv. Geophys. 32, 585–602 (2011).
introductions: Comparative spread rates and community impacts. 47. Meehl, G. A. et al. in Climate Change 2007: The Physical Science Basis
Glob. Ecol. Biogeog. 19, 303–316 (2010). (eds Solomon, S. et al.) Ch. 10 (IPCC, Cambridge Univ. Press, 2007).
17. Ackerman, J. D. & Okubo, A. Reduced mixing in a marine macrophyte canopy. 48. Vermeer, M. & Rahmstorf, S. Global sea level linked to global temperature.
Funct. Ecol. 7, 305–309 (1993). Proc. Natl Acad. Sci. USA 106, 21527–21532 (2009).
18. Koch, E. W. et al. Non-linearity in ecosystem services: Temporal and spatial 49. Seneviratne, S. I. et al. in Special Report on Managing the Risks of Extreme
variability in coastal protection. Front. Ecol. Environ. 7, 29–37 (2009). Events and Disasters to Advance Climate Change Adaptation (SREX) (eds
19. Mendez, F. J. & Losada, I. J. Transformation of random and non-random Field, C. B. et al.) Ch. 3 (IPCC, Cambridge Univ. Press, 2012).
breaking waves over vegetation fields. Coast. Eng. 51, 103–118 (2004). 50. Méndez, F. J., Losada, I. J. & Losada, M. A. Hydrodynamics induced by wind
20. Bouma, T. J. et al. Trade-offs related to ecosystem engineering: A case study on waves in a vegetation field. J. Geophys. Res. Oceans 104, 18383–18396 (1999).
stiffness of emerging macrophytes. Ecology 86, 2187–2199 (2005). 51. Mazda, Y. et al. Drag force due to vegetation in mangrove swamps.
21. Koch, E. W., Ackerman, J. D., Verduin, J. & van Keulen, M. in Seagrasses: Mangroves Salt Marsh 1, 193–199 (1997).
Biology, Ecology and Conservation (eds Larkum, A. W. D., Orth, R. J. & 52. Ghisalberti, M. & Nepf, H. Shallow flows over a permeable medium: The
Duarte, C. M.) 193–225 (Springer, 2006). hydrodynamics of submerged aquatic canopies. Transport Porous Med.
22. Gacia, E. & Duarte, C. M. Sediment retention by a Mediterranean Posidonia 78, 385–402 (2009).
oceanica meadow: The balance between deposition and resuspension. 53. Chen, Z., Ortiz, A., Zong, L. & Nepf, H. The wake structure behind a porous
Estuar. Coast. Shelf Sci. 52, 505–514 (2001). obstruction and its implications for deposition near a finite patch of emergent
23. Peralta, G., van Duren, L. A., Morris, E. P. & Bouma, T. J. Consequences of vegetation. Wat. Resour. Res. 8, W09517 (2012).
shoot density and stiffness for ecosystem engineering benthic macrophytes in 54. Hemminga, M. A. & Duarte, C. M. Seagrass Ecology
flow dominated areas: A hydrodynamic flume study. Mar. Ecol. Prog. Ser. (Cambridge Univ. Press, 2000).
368, 103–115 (2008). 55. Tanino, Y. & Nepf, H. Laboratory investigation oF mean drag in a random
24. Yang, S. L., Shi, B. W., Bouma, T. J., Ysebaert, T. & Luo, X. X. Wave attenuation array of rigid, emergent cylinders. J. Hydraul. Eng. 134, 4–41 (2008).
at a salt-marsh margin: A case study of an exposed coast on the Yangtze 56. Bryan, K. R., Tay, H. W., Pilditch, C. A., Lundquist, C. J. & Hunt, H. L.
estuary. Estuar. Coasts 35, 169–182 (2012). The effects of seagrass (Zostera muelleri) on boundary-layer hydrodynamics in
25. Furukawa, K., Wolanski, E. & Mueller, H. Currents and sediment transport in Whangapoua Estuary, New Zealand. J. Coast. Res. 50, 668–672 (2007).
mangrove forests. Estuar. Coast. Shelf Sci. 44, 301–310 (1997). 57. Vo-Luong, P. & Massel, S. Energy dissipation in non-uniform mangrove
26. Barbier, E. B., Hacker, S. D., Kennedy, C., Koch, E. W., Stier, A. C. forests of arbitrary depth. J. Mar. Syst. 74, 603–622 (2008).
& Silliman, B. R. The value of estuarine and coastal ecosystem services. 58. Losada, I. J., Losada, M. A. & Martin, F. Experimental study of wave-induced
Ecol. Monogr. 81, 169–193 (2011). flow in porous media. Coast. Eng. 26, 77‑98 (1995).
27. Costanza, R. et al. The value of the world´s ecosystem services and natural 59. Mazarrasa, I., Marbà, N., Hendriks, I. E., Losada, I. J. & Duarte, C. M.
capital. Nature 387, 253–260 (1997). Estimates of Average Sediment Accretion Rates in Vegetated Coastal Habitats
28. Duarte, C. M. & Cebrián, J. The fate of marine autotrophic production. Around the World (Digital CSIC, 2013); http://hdl.handle.net/10261/77396
Limnol. Oceanogr. 41, 1758–1766 (1996). 60. Turner, W. R., Oppenheimer, M. & Wilcove, D. S. A force to fight global
29. Kennedy, H. et al. Seagrass sediments as a global carbon sink: Isotopic warming. Nature 462, 278–279 (2009).
constraints. Glob. Biogeochem. Cycles 24, GB4026 (2010). 61. Duarte, C. M., Dennison, W. C., Orth, R. J. W. & Carruthers, T. J. B.
30. Mateo, M. A., Romero, J., Pérez, M., Littler, M. M. & Littler, D. S. Dynamics The charisma of coastal ecosystems: Addressing the imbalance. Estuar. Coasts
of millenary organic deposits resulting from the growth of the Mediterranean 31, 233–238 (2008).
seagrass Posidonia oceanica. Estuar. Coast. Shelf Sci. 44, 103–110 (1997). 62. Cahoon, D. R. et al. Mass tree mortality leads to mangrove peat collapse at Bay
31. Ward, L. G., Zaprowski, B. J., Trainer, K. D. & Davis, P. T. Stratigraphy, pollen Islands, Honduras after Hurricane Mitch. J. Ecol. 91, 1093–1105 (2003).
history and geochronology of tidal marshes in a Gulf of Maine estuarine 63. Lovelock, C. E., Ruess, R. W. & Feller, I. C. CO2 efflux from cleared mangrove
system: Climatic and relative sea level impacts. Mar. Ecol. 256, 1–17 (2008). peat. PLoS ONE 6, e21279 (2011).

NATURE CLIMATE CHANGE | VOL 3 | NOVEMBER 2013 | www.nature.com/natureclimatechange 967

© 2013 Macmillan Publishers Limited. All rights reserved

REVIEW ARTICLE NATURE CLIMATE CHANGE DOI: 10.1038/NCLIMATE1970
64. Siikamäki, J., Sanchirico, J. N. & Jardine, S. L. Global economic 86. Barbier, E. B. et al. Coastal ecosystem–based management with nonlinear
potential for reducing carbon dioxide emissions from mangrove loss. ecological functions and values. Science 319, 321–323 (2008).
Proc. Natl Acad. Sci. USA 109, 14369–14374 (2012). 87. Adams, C. A., Andrews, J. E. & Jickells, T. Nitrous oxide and methane fluxes
65. Pendleton, L. et al. Estimating global “blue carbon” emissions from conversion versus carbon, nitrogen and phosphorous burial in new intertidal and
and degradation of vegetated coastal ecosystems. PLoS ONE 7, e43542 (2012). saltmarsh sediments. Sci. Total Environ. 434, 240–251 (2012).
66. Report of the Conference of the Parties on its Fifteenth Session, Held in 88. Temmerman, S., Govers, G., Wartel, S. & Meire, P. Modelling estuarine
Copenhagen from 7 to 19 December 2009 (UNFCCC, 2009). variations in tidal marsh sedimentation: Response to changing sea level and
67. Arnaud-Haond, S. et al. Genetic recolonization of mangrove: Genetic suspended sediment concentrations. Mar. Geol. 212, 1–19 (2004).
diversity still increasing in the Mekong Delta 30 years after Agent Orange. 89. Borsje, B. W. et al. How ecological engineering can serve in coastal protection.
Mar. Ecol. Prog. Ser. 390, 129–135 (2009). Ecol. Eng. 37, 113–122 (2011).
68. An, S. et al. China’s natural wetlands: Past problems, current status, and future 90. Mangrove planting saves lifes and money in Viet Nam. International
challenges. Ambio 36, 335–342 (2007). Federation of the Red Cross and Red Crescent (19 June 2002);
69. http://www.globalrestorationnetwork.org/ecosystems/coastal/ http://go.nature.com/B6ZVwU
estuaries‑marshes‑mangroves/case-studies 91. Costanza, R. et al. The value of coastal wetlands for hurricane protection.
70. Sintes, T., Marbà, N., Duarte, C. M. & Kendrick. G. A. Non-linear processes in Ambio 37, 241–248 (2008).
seagrass colonisation explained by simple clonal growth rules. Oikos 92. Duarte, C. M. et al. Is global ocean sprawl a trojan horse for jellyfish blooms?
108, 165 –175 (2005). Front. Ecol. Environ. 11, 91–97 (2013).
71. McGlathery, K. J. et al. Recovery trajectories during state change from bare 93. Aburto-Oropeza, O. et al. Mangroves in the Gulf of California increase fishery
sediment to eelgrass dominance. Mar. Ecol. Prog. Ser. 448, 209–221 (2012). yields. Proc. Natl Acad. Sci. USA 105, 10456–10459 (2008).
72. Osland, M. J. et al. Ecosystem development after mangrove wetland creation: 94. Odum, H. T. Experiments with engineering of marine ecosystems.
Plant-soil change across a 20-year chronosequence. Ecosystems Publ. Inst. Mar. Sci. Univ. Texas 9, 374–403 (1963).
15, 848–866 (2012). 95. Mitsch, W. J. & Jørgensen, S. E. Ecological engineering: A field whose time has
73. Craft, C. et al. The pace of ecosystem development of constructed come. Ecol. Eng. 20, 363–377 (2003).
Spartina alterniflora marshes. Ecol. Appl. 13, 1417–1432 (2003). 96. Fiselier, J. et al. Perspectief Natuurlijke Keringen (Ecoshape, 2011).
74. Irving, A. D., Connell, S. D. & Russell, B. D. Restoring coastal plants to 97. Nowak, K. Mangrove and peat swamp forests: Refuge habitats for primates
improve global carbon storage: Reaping what we sow. and felids. Folia Primatol. 83, 361–376 (2013).
PLoS ONE 6, e18311 (2011). 98. Cai, W. J. Estuarine and coastal ocean carbon paradox: CO2 sinks or sites of
75. Mudd, S. M., Howell, S. M. & Morris, J. T. Impact of dynamic feedbacks terrestrial carbon incineration? Annu. Rev. Mar. Sci. 3, 123–45 (2011).
between sedimentation, sea-level rise, and biomass production on near- 99. Alongi, D. M. Present state and future of the world’s mangrove forests.
surface marsh stratigraphy and carbon accumulation. Estuar. Coast. Shelf Sci. Environ. Conserv. 29, 331–349 (2002).
82, 377–389 (2009). 100. Spalding, M., Kainuma, M. & Collins, L. World Atlas of mangroves
76. Dierssen, H. M., Zimmerman, R. C., Drake, L. A. & Burdige, D. J. Potential (Earthscan, 2010).
export of unattached benthic macroalgae to the deep sea trough wind-driven 101. Gattuso, J. P., Frankignoulle, M. & Wollast, R. Carbon and carbonate metabolism
Langmuir circulation. Geophys. Res. Lett. 36, L04602 (2009). in coastal aquatic ecosystems. Annu. Rev. Ecol. Syst. 29, 405–434 (1998).
77. Hossain, A. B. M. S., Salleh, A., Boyce, A. N., Chowdhury, P. & Naqiuddin, M. 102. http://www.ecoshape.nl/en_GB/delfland-sand-engine.html
Biodiesel fuel production from algae as renewable energy.
Am. J. Biochem. Biotechnol. 4, 250–254 (2008).
Acknowledgements
78. Hussain, K., Nawaz, K., Majeed, A. & Feng, L. Economically effective potential This study was funded by the Spanish Ministry of Economy and Competitiveness
of algae for biofuel production. World Appl. Sci. 9, 1313–1323 (2010). (projects MEDEICG, CTM2009-07013 and ESTRESX, CTM2012-32603), the EU FP7
79. Sawyer, D. Climate change, biofuels and eco-social impacts in the Brazilian projects Opera (contract number 308393) and MedSEA (contract number -265103)
Amazon and Cerrado. Phil. Trans. R. Soc. B 363, 1747–1752 (2008). and the CSIRO Coastal Carbon Cluster. N.M. was supported by a Gledden Fellowship
80. Phalan, B. The social and environmental impacts of biofuel in Asia: An from the Institute of Advanced Studies of the University of Western Australia, I.H. was
overview. Appl. Energy 86, S21–S29 (2009). supported by a Posdoctoral contract of the JAE programme of CSIC and I.M. by a PhD
81. Eide, A. The Right to Food and the Impact of Liquid Biofuels (Agrofuels) scholarship of the Government of the Balearic Islands (Spain).
(FAO, 2008).
82. Harrison, R. & Webb, J. A review of the effect of N fertilizer type on gaseous
emissions. Adv. Agron. 73, 65–108 (2001). Additional information
83. Hallegatte, S. Strategies to adapt to an uncertain climate change. Supplementary information is available in the online version of the paper. Reprints
and permissions information is available online at www.nature.com/reprints.
Glob. Environ. Change 19, 240–247 (2009).
Correspondence should be addressed to C.M.D.
84. Feagin, R. Vegetation’s role in coastal protection. Science 320, 176–177 (2008).
85. Gedan, K. B., Kirwan, M. L., Wolanski, E., Barbier, E. B. & Silliman, B. R. The
present and future role of coastal wetland vegetation in protecting shorelines: Competing financial interests
Answering recent challenges to the paradigm. Climatic Change 106, 7–29 (2010). The authors declare no competing financial interests.

968 NATURE CLIMATE CHANGE | VOL 3 | NOVEMBER 2013 | www.nature.com/natureclimatechange

© 2013 Macmillan Publishers Limited. All rights reserved