PERSPECTIVE Corrected: Author correction https://doi.org/10.1038/s41467-019-11693-w OPEN The future of Blue Carbon science Peter I. Macreadie et al.# The term Blue Carbon (BC) was first coined a decade ago to describe the disproportionately large contribution of coastal vegetated ecosystems to global carbon sequestration. The role of BC in climate change mitigation and adaptation has now reached international prominence. To help prioritise future research, we assembled leading experts in the field to agree upon the 1234567890():,; top-ten pending questions in BC science. Understanding how climate change affects carbon accumulation in mature BC ecosystems and during their restoration was a high priority. Controversial questions included the role of carbonate and macroalgae in BC cycling, and the degree to which greenhouse gases are released following disturbance of BC ecosystems. Scientists seek improved precision of the extent of BC ecosystems; techniques to determine BC provenance; understanding of the factors that influence sequestration in BC ecosystems, with the corresponding value of BC; and the management actions that are effective in enhancing this value. Overall this overview provides a comprehensive road map for the coming decades on future research in BC science. B lue Carbon (BC) refers to organic carbon that is captured and stored by the oceans and coastal ecosystems, particularly by vegetated coastal ecosystems: seagrass meadows, tidal marshes, and mangrove forests. Global interest in BC is rooted in its potential to mitigate climate change while achieving co-benefits, such as coastal protection and fisheries enhance- ment1–3. BC has attracted the attention of a diverse group of actors beyond the scientific community, including conservation and private sector organizations, governments, and inter- governmental bodies committed to marine conservation and climate change mitigation and adaptation. The momentum provided by these conservation and policy actors has energized the scientific community by challenging them to address knowledge gaps and uncertainties required to inform policy and management actions. The BC concept was introduced as a metaphor aimed at highlighting that coastal ecosystems, in addition to terrestrial forests (coined as green carbon), contribute significantly to organic carbon (C) sequestration1. This initial metaphor evolved to encompass strategies to mitigate and adapt to climate change through the conservation and restoration of vegetated coastal ecosys- tems1,2. As BC science consolidates as a paradigm, some aspects are still controversial; for instance, contrasting perspectives on the role of carbonate production as a component of BC4 and whether seaweed contributes to BC5,6. We propose an open discussion to refocus the current research agenda, reconcile new ideas with criticisms, and integrate those findings into a stronger scientific framework (Box 1). This effort will address the urgent need for refined understanding of the role of vegetated coastal ecosystems in climate change mitigation and adaptation. There is, therefore, a need to establish a comprehensive research program on BC science that addresses current gaps while continuing to respond to immediate policy and managerial needs. Furthermore, this research program can inform policy directions based on new knowledge, thus Correspondence and requests for materials should be addressed to P.I.M. (email: p.macreadie@deakin.edu.au). #A full list of authors and their affiliations appears at the end of the paper. NATURE COMMUNICATIONS | (2019) 10:3998 | https://doi.org/10.1038/s41467-019-11693-w | www.nature.com/naturecommunications 1 PERSPECTIVE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-11693-w Box 1. | Evidence underpinning the science The role of seagrasses and marine macroalgae as major C sinks in the ocean was first proposed by Smith who suggested that seagrasses and marine macroalgae were overlooked C sinks7; however, at the time, there was minimal uptake of the concept within climate change mitigation efforts. In 2003 the first global budget of C storage in soils of salt marshes and mangroves brought light to the importance of these coastal ocean sink. By 2005, it was shown that seagrass, mangrove, and tidal marsh sediments represent 50% of all C sequestered in marine sediments8. This mounting evidence for such a major role in C sequestration provided the impetus for the Blue Carbon report1, where the term “Blue Carbon” was first coined, and that led to the development of international and national BC initiatives (e.g., http://thebluecarboninitiative.org). This led to research efforts to propose emissions factors from loss and restoration of BC ecosystems for C accounting9, provide empirical evidence of emissions following disturbance and C drawdown from restoration10,11,12, map the C density of mangrove soils globally13, and explore the potential of BC ecosystems to support climate-change adaptation2. playing a role in setting the management agenda and not simply and the space available to accumulate sediment21. Thus, sea level responding to it. Here we identify, based on a broad effort by the rise ranks among the most important factors that will influence leading research academics in BC science, key questions and future BC stocks and sequestration. Sea level rise can result in BC challenges that need to be addressed to consolidate progress in gains, with increasing landward areal extent of ecosystems where BC science and inform current debate. We do so through three possible22, and enhanced vertical accretion of sediments and C main steps. First, we briefly summarize the elements of BC sci- stocks18,23; and losses, with losses of ecosystem extent24, failure of ence that represent the pillar of this research program. Second, we restoration25, remineralization of stored organic matter26 that identify key scientific questions by first surveying the scientific result in greenhouse gas emissions to the atmosphere (Table 1). community. Then we clustered these questions into common Intense storms17, marine heat waves, 27, elevated CO228, and themes, which develop research goals and agendas. Last, we altered availability of freshwater29 have also all been implicated as provide guidance as to how these questions can be best articulated important factors affecting the distribution, productivity, com- into a new research agenda as a path for progress. munity composition and C sequestration of BC ecosystems over a range of locations (Table 1). Geographic variation in exposure to climate change is high. Rates of sea level rise and land sub- Scientists’ perspectives on the 10 key fundamental questions sidence30, which enhances relative rates of sea level rise, vary in BC science geographically18. Additionally, rates of temperature change and We identified and selected scientists from among the leading and changes in the frequency of intense storms and rainfall vary senior authors of the 50 most-cited papers on BC science (ISI regionally15–17. Geomorphic models have provided first pass Web of Science access date 22 June 2017), together with the assessments of the global vulnerability of BC ecosystems to sea participants in a workshop on BC organized at King Abdullah level rise20,31, and for restoration success32, but local scale University of Science and Technology, Saudi Arabia, in March descriptors of changes in exposure of BC ecosystems to climate 2017. We did not attempt to identify any scientists’ area of spe- change and impacts on C stocks are often incomplete or missing. cialisation to avoid bias. Among these authors, we surveyed those For instance, storm associated waves are important for deter- affiliated with academic or research institutions. A group of mining the persistence and recruitment of BC ecosystem33, yet 50 scientists were asked to contribute from their perspective the local assessments are not widely available. top pending questions (up to 10) in BC science. Specifically, the Responses of adjacent ecosystems to climate change may invitees were asked to “Email your ten most important questions influence the exposure and sensitivity of BC ecosystems and their (or fewer) relevant to improving our understanding of blue car- C stocks to climate change. For example, degradation of coral bon science and its application to climate change mitigation”. We reefs could increase wave heights within lagoons which may lead did not ask scientists to prioritise their questions, or target any to losses of seagrass or mangroves within lagoons with rising sea particular geographical area, but we did ask them to focus on levels as waves increase34, or decreases of carbonate sediments mangrove, tidal marsh, macroalgal, and seagrass ecosystems. The due to ocean acidification, may reduce the ability of some BC answers received (35 total respondents, see Supplementary ecosystems to keep up with sea level rise35. Additionally, the Note 1) and were then clustered into ten themes (by grouping sensitivity of BC ecosystems to climate change is also likely questions that were similar) that were subsequently articulated influenced by human activities in the coastal zone. For example, into individual, overarching research questions: deterioration in water quality may increase the impacts of sea Q1. How does climate change impact carbon accumulation level rise on seagrass36 and decreased sedimentation from dam- in mature Blue Carbon ecosystems and during their ming of rivers, hydrological modifications and presence of sea- restoration? walls may negatively affect BC stocks in mangroves and tidal The impacts of climate change on BC ecosystems and their C marshes20,31. stocks are dependent on the exposure to climate change factors. Q2. How does disturbance affect the burial fate of Blue This is influenced by both the frequency and intensity of stres- Carbon? sors, and the sensitivity and resilience of the ecosystem14. The effect of disturbance on BC production and storage has Question 1 reflects uncertainties associated with the rate and become a topic of intense interest because of an increasing desire magnitude of climate change15–17 as well as uncertainties about to protect or enhance this climate-related ecosystem service. There the impacts of climate change on current and restored BC eco- are three key issues, all beginning to be addressed by BC systems, their rates of C sequestration and the stability of C researchers, but requiring further study: (1) the depth in the soil stocks, which are likely to vary with past sea level history18, over profile to which the disturbance propagates, (2) the proportion of geographic locations, among BC ecosystems, and within disturbed C that is lost as CO2, and (3) the extent to which issues 1 ecosystems. and 2 are context dependent. The first global estimates of potential BC ecosystems mainly occupy the intertidal and shallow water losses of BC resulting from anthropogenic disturbance combined environments, where their distribution, productivity and rates of changes in the global distribution of BC ecosystems with simple vertical accretion of soils are strongly influenced by sea level19,20 estimates of conversion (remineralisation) of stored BC per unit 2 NATURE COMMUNICATIONS | (2019) 10:3998 | https://doi.org/10.1038/s41467-019-11693-w | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-11693-w PERSPECTIVE Table 1 Examples of gains and losses for BC stocks with a range of climate change factors Ecosystem Sea level rise Extreme storms Higher temperatures Extra CO2 Altered precipitation Mangrove Landward expansion Canopy damage, reduced Minimal impacts An increase in Canopy dieback due to increases area and C recruitment and soil anticipated, although atmospheric CO2 drought stocks subsidence resulting in increased decomposition benefits plant Losses of C stocks due to Losses of low intertidal losses of C stocks of soil C possible productivity of some remineralization and reduced forests and coastal Soil elevation gains due Poleward spread of species which could productivity squeeze could reduce C to sediment deposition mangrove forests at alter C stocks Increased rainfall may stocks increasing C stocks and, expense of tidal marshes result in increased Increasing reducing effects of sea increases C stocks productivity and C accommodation space level rise Change in dominant sequestration increases C species could influence C sequestration sequestration Tidal Marsh Landward expansion Loss of marsh area and Increased temperatures An increase in Reduced above and increased area and C C stocks may increase atmospheric CO2 belowground production due stocks Enhanced sedimentation decomposition of soil benefits plant to drought reducing C Losses of low intertidal and soil elevation organic matter, but offset productivity of some sequestration marsh and coastal squeeze increasing C stocks and, by increased productivity species which could Possible losses of C stocks could reduce C stocks reducing effects of sea of tidal marsh vegetation alter C stocks due to remineralization Increasing level rise Poleward expansion of Impact could be greater in accommodation space mangroves will replace areas that already have increases C tidal marsh and increase scarce or variable rainfall sequestration C storage Poleward expansion of bioturbators, may decrease soil C stocks Seagrass Loss of deep water Some extreme storms cause Thermal die-offs leading to An increase in Most seagrasses are tolerant seagrass the erosion of seagrasses losses of C stocks dissolved inorganic of acute low salinity events Landward migration in and loss of seagrass C stocks Species turnover C benefits plant associated with high rainfall, areas where seawater but some seagrass species Colonization of new productivity but some are negatively floods the land (into are resistant to these major poleward regions increasing C stocks affected and potential mangrove or tidal events Increased productivity Ocean acidification interactions with disease may marsh ecosystem) Flood events associated leads to loss of lead to losses of C stocks with extreme rainfall may seagrass biodiversity, Reduced rainfall increases result in mortality, but decreasing C stocks light availability which could also increase increases productivity and sediment accretion and C C sequestration sequestration Seaweed Loss of deep water Reduces seaweed cover, but Major retraction in kelp Increased biomass Little effect overall seaweeds could lead to sequestration forest C stores at non-polar and productivity of Regional effects on Seaweeds are expected of C stocks as detritus sinks range edges; kelp where water seaweed flora in areas with to colonise hard Expected expansion at temperatures remain high land run off/rivers substrata that become polar range edges. cool enough flooded, increasing C stocks Bold text indicate potential positive effects on BC stocks, italic text indicate negative effects with roman text indicating where effects could be positive or negative area37. The estimated annual CO2 emission from the disturbance ecosystems, exposing deeply buried sediments to oxygen triggered of BC ecosystems was estimated at 0.45 Petagrams CO2 globally37. microbial breakdown of ancient BC43. At this stage, there is some The generalised assumptions necessary for such global assess- evidence that disturbances can diminish BC stocks, for example: ments—e.g., remineralization within only the top 1 m of soil, and oil spills44, seasonal wrack deposition42, aquaculture45, eutro- 100% loss of BC—provide little guidance at a local management phication46, altered tidal flows46, and harvesting of fisheries scale and gloss over the variability of effects from different dis- resources38,47. Such knowledge is key for the construction of turbance types38. This deficiency has led to a more nuanced Emissions Factors for modelling. But examples in the literature theoretical framework accounting for the intensity of disturbance, are often specific for a particular disturbance or ecosystem setting, especially whether the disturbance affects only the habitat-forming and do not yet offer the generalised understanding necessary to plant (e.g., clearing, eutrophication, light reduction, toxicity) or build a comprehensive framework guiding management projects. whether it also disturbs the soil (e.g., erosion, digging, reclama- Finally, although there is widespread agreement that a changing tion)39,40. The duration of disturbance is another important pre- climate directly affects BC production and storage, we recom- dictor of disturbance effects on BC remineralisation because, over mend a clearer focus on the interacting effects of climate and time, more soil BC is exposed to an oxic environment41. direct anthropogenic disturbances. We have a nascent understanding of the processes by which Q3. What is the global importance of macroalgae, including natural and human disturbances alter C decomposition. Die-off calcifying algae, as Blue Carbon sinks/donors? of below-ground roots and rhizomes in tidal marshes, for Macroalgae are highly productive (Table 2) and have the lar- example, changes the chemical composition of BC and associated gest global area of any vegetated coastal ecosystem48. Yet only in a microbial assemblages, subsequently increasing decomposition relatively few cases have macroalgae been included in BC and decreasing stored C (by up to 90% (ref. 42)). In seagrass assessments. Unlike angiosperms, which grow on depositional NATURE COMMUNICATIONS | (2019) 10:3998 | https://doi.org/10.1038/s41467-019-11693-w | www.nature.com/naturecommunications 3 PERSPECTIVE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-11693-w Table 2 Estimates of global net primary productivity, CO2 release from calcification and C sequestration (Tg C yr−1) for three benthic marine systems System Global CO2 (as C) Global CO2 (as C) release Global net organic C Global C References fixation in NPP from calcification, assimilation = NPP sequestration assuming 0.6 CO2-C per minus C as CO2 CaCO3-C produced produced in calcification Benthic 960–2000 – – 60–1400 Charpy-Roubard & Sournia71; macroalgae Krause-Jensen & Duarte6; (calcified and Duarte49; Raven50 uncalcified) Calcified coralline 720 120 600 – Van den Heijden & red algae Kamenos53, who do not mention CO2 release from CaCO3 formation Coral reefs 0 84–840 84–840 0a Ware et al.150; Smith & Mackenzie151 aAssuming CaCO3 ultimately sinks below the lysocline, where CaCO3 dissolves, and upwelling ultimately (102–103 years) brings the resulting HCO3− back to the sea surface soils2, macroalgae generally grow on hard or sandy substrata that knowledge of their current and recent-past global distributions. have no or only limited C burial potential6. However, a recent The best constrained areal estimates exist for mangroves, which meta-analysis has estimated that macroalgae growing in soft occur in tropical and subtropical regions, generally where winter sediments have a global C burial rate of 6.2 Tg C yr−1 (ref. 6), seawater isotherms exceed 20 °C55. Overall, the global spatial which is comparable to the lower range of estimates for tidal extent of mangroves, and patterns and drivers of their temporal marshes. Furthermore, several studies show that macroalgae act change, are relatively well understood, especially when compared as C donors3,6,49–51, where detached macroalgae are transported with other BC ecosystems. Still, Giri et al.56 estimated a global by currents, and deposited in C sinks beyond macroalgae habitats. area of mangroves of ca. 140,000 km2 in the year 2000 and Recent first-order estimates have suggested that up to 14 Tg C Hamilton and Casey57 83,495 km2 in 2000 and 81,849 km2 in yr−1 of macroalgae-derived particulate organic C is buried in 2012. Both studies used Landsat data but different methodologies. shelf sediments and an additional 153 Tg C yr−1 is sequestered in Mangroves occur in 118 countries worldwide, but ~75% of total the deep ocean6. These calculations suggest that macroalgae may coverage is located within just 15 countries, with ~23% found in be supporting higher global C burial rates than seagrass, tidal Indonesia alone56. Total mangrove extent during the second half marshes, and mangroves combined. This research highlights that of the 20th century declined at rates 1–3% yr−1 mainly due to if we are to incorporate macroalgal systems into BC assessments aquaculture, land use change and land reclamation58. There are we need a better understanding of the fate of C originating from uncertainties in the area of mangrove that are scrub forms and these systems. Furthermore, if we are to scale up from local which are therefore often not considered as forests despite their measurements of C-sequestration to the global level, more refined importance in arid and oligotrophic settings and often their large estimates of the global surface area of macroalgal-dominated soil C stocks59,60. Since the beginning of the 21st century, man- systems are needed. grove loss rates are 0.16–0.39% yr−1 (ref. 57), probably reflecting Most estimates of C-sequestration by marine vegetated eco- changes in aquaculture and conservation efforts. systems refer solely to organic C even though calcifying organ- Tidal marshes are primarily found in estuaries along coasts of isms are also important components of such ecosystems52. For Arctic, temperate and subtropical coastal lagoons, embayments, calcifying algae, whether they serve as C-sinks or sources is and low-energy open coasts, although they also occur in some debated4, especially where calcifying organisms form and become tropical regions61. Woodwell et al.62 estimated global tidal marsh buried within seagrass meadows4,5. Carbonate production results extent of 380,000 km2 using the fraction of global coastline in the release of 0.6 mol of CO2 per mol of CaCO3 precipitated53, occupied by estuaries and the assumption that ~20% of estuaries suggesting that calcifying algae are sources of CO2 that counteract supported tidal marshes48. However, tidal marsh area has been C-sequestration in these ecosystems. However, co-deposition of mapped in only 43 countries (yielding a total habitat extent of ca. organic and inorganic C may also have interacting effects on C- 55,000 km2), which represents just 14% of the potential global sequestration4. Carbonate may help protect and consolidate area63. Tidal marsh extent is well documented for Canada, Eur- organic C sediment deposits, and CO2 release from mineraliza- ope, USA, South Africa and Australia63–65 but remains unknown tion of organic matter may stimulate carbonate dissolution and to a large extent in regions, including Northern Russia and South hence, CO2 removal48,53,54. Burial of inorganic carbon in seagrass America. An historical assessment of 12 estuaries and coastal seas and mangrove ecosystems is also to a large extent supported by worldwide indicated that >60% of wetland coverage has been inputs from adjacent ecosystems rather than by local calcification. lost66 mostly due to changes in land use, coastal transformation Furthermore, mass balances highlight that such Blue Carbon and land reclamation61. The minimum global rate of loss of tidal ecosystems are sites of net CaCO3 dissolution54. More studies are marsh area is estimated at 1–2% yr−1 (ref. 67). needed to assess the net effect of organic and inorganic C Despite the widespread occurrence of seagrass across both deposition on C sequestration in calcifying systems. temperate and tropical regions, the global extent of seagrass area Q4. What is the global extent and temporal distribution of is poorly estimated48. The total global area was recently updated BC ecosystems? to 350,000 km2 (ref. 68), although estimates range from 300,000 Our attempts to upscale BC estimates and model changes (ref.) to 600,000 km2 (ref. 69), with a potential habitable area for across large spatial and temporal scales is hindered by poor seagrass of 4.32 million km2 (ref. 70). Available distribution data 4 NATURE COMMUNICATIONS | (2019) 10:3998 | https://doi.org/10.1038/s41467-019-11693-w | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-11693-w PERSPECTIVE are geographically and historically biased, reflecting the imbal- lower surface water CO2 concentration and be a direct sink for ance in research effort among regions71, and most data have been atmospheric CO276,77. Lowering of surface water CO2 con- collected since the 1980s72. The total global seagrass area has centration is facilitated if allochthonous Corg ((3) in Fig. 1) and decreased by ~29% since first reported in 1879—with ~7-fold DIC inputs ((4) in Fig. 1) are low. Reactions of the inorganic C faster rates of decline since 1990 (ref. 72)—due to a combination (Cinorg) cycle can also change the CO2 concentration in surface of natural causes, coastal anthropogenic pressure and climate water and therefore influence net exchange of CO2 with the change73. atmosphere4,5,78. Formation of calcium carbonate minerals (cal- Producing accurate estimates of the global extent of BC eco- cification) results in an increase of CO2 in the water column ((5) systems is therefore a prerequisite to assess their contribution in in Fig. 1) while dissolution of carbonate minerals decreases CO2 the global carbon cycle. In addition, given the fast rate of decline ((6) in Fig. 1). These processes may critically affect air–water CO2 reported for many BC ecosystems, regular revision of these esti- gas exchange. Although recent studies related to the role of BC in mates is needed to track any changes in their global extent and climate change mitigation are beginning to address the abun- importance. Extensive mapping, with particular focus on under- dance and burial rate of Cinorg in soils4,5,54,78–80, studies investi- studied areas that may support critical BC ecosystems, that gating the full suite of key processes for air–water CO2 fluxes, combines acoustic (i.e., side scan sonar and multi-beam eco- such as carbonate chemistry and Corg dynamics in shallow coastal sounder) and optical (i.e., aerial photography and satellite images) waters and sediments, are still scarce (but see76,77,81,82). In par- remote sensing techniques with ground truthing (by scuba diving ticular, relevance of carbonate chemistry to the overall spatio- or video images) should be undertaken to map and monitor their temporal dynamics of Corg and Cinorg pools and fluxes (e.g., extent and relative change over time74. origin, fate, abundance, rate, interactions) and air–water CO2 Q5. How do organic and inorganic carbon cycles affect net fluxes is largely uncertain for BC ecosystems4. CO2 flux? Therefore, in addition to Corg related processes occurring in Even though BC ecosystems are significant Corg reservoirs, sediments and vegetation, future BC science should also quantify depending on Corg and Cinorg dynamics they could also be net other key processes, such as air-water CO2 fluxes and Corg and emitters of CO2 to the atmosphere through air-water CO2 gas Cinorg dynamics in water, to fully understand the role of BC exchange75. For instance, in submerged BC ecosystems (i.e., ecosystems in climate change mitigation83. seagrasses), Corg storage is not directly linked with the removal of Q6. How can organic matter sources be estimated in BC atmospheric CO2 because the water column separates the atmo- sediments? sphere from benthic systems. BC science gaps exist in complex Coastal ecosystems, mangroves, seagrasses and tidal marshes, inorganic and organic biogeochemical processes occurring within occupy the land-sea interface and are subject to convergent inputs the water column and determining CO2 sequestration of organic matter from terrestrial and oceanic sources as well as functioning. transfers to and from nearby ecosystems84. However, the most Photosynthesis lowers the CO2 concentration in surface water basic requirement of quantifying organic matter inputs, and dif- as dissolved inorganic C (DIC) is incorporated into Corg ((1) in ferentiating between allochthonous and autochthonous sources of Fig. 1), and respiration and remineralization increases the CO2 Corg, remains a challenge. This limitation has particular relevance concentration ((2) in Fig. 1). Net autotrophic ecosystems would because of interest in financing the restoration of coastal Terrestrial systems CO2 (gas) (4) DIC input Source Sink DIC pool (4) CO32− HCO3 − CO2 (aq) Cinorg and Corg input (1) (2) (3) Photosynthesis (5) Respiration Calcification Dissolution remineralization (3) (6) Cinorg Corg (Skeltons and shells, carbonate minerals) (Flesh body, detritus) ? Burial Burial Fig. 1 Conceptual diagram showing the biogeochemistry of carbon associated with air-water CO2 exchanges. Blue lines indicate the processes that enhance the uptake of atmospheric CO2, and red lines indicate those that enhance the emission of CO2 into the atmosphere. The CO2 concentration in surface water is primarily responsible for determining the direction of the flux. The concentration of surface water CO2 is determined by carbonate equilibrium in dissolved inorganic carbon (DIC) and affected by net ecosystem production (the balance of photosynthesis, respiration, and remineralization), which directly regulate DIC (1 and 2), allochthonous particulate and dissolved organic carbon (Corg), particulate inorganic carbon (Cinorg), and DIC inputs from terrestrial systems and coastal oceans (3 and 4), net ecosystem Cinorg production (the balance of calcification and dissolution), directly regulating both DIC and total alkalinity (TA) (5, 6), and temperature (solubility of CO2). Calcification produces CO2 with a ratio (released CO2/precipitated Cinorg) of approximately 0.6 in normal seawater54 NATURE COMMUNICATIONS | (2019) 10:3998 | https://doi.org/10.1038/s41467-019-11693-w | www.nature.com/naturecommunications 5 PERSPECTIVE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-11693-w ecosystems through the sale of BC offset-credits85. Policy fra- respect to biological factors, it remains unclear how primary meworks such as the Verified Carbon Standard Methodology producer diversity and traits (e.g., biochemical composition, VM0033 (ref. 86) stipulate that offset-credits are not allocated productivity, size and biomass allocation) influence BC98,99. under the framework for allochthonous Corg because of the risk of However, it is likely that the suite of macrophytes present in BC duplicating C sequestration gains that may have been accounted ecosystems is critical to the mass of C available to be captured and for in adjacent ecosystems. New methods are emerging that have preserved (as suggested for tidal marshes100). Equally, it is greater potential to quantify the contribution of different primary uncertain how fauna influence the production, accumulation or producers to sedimentary organic carbon in marine ecosystems87. preservation of Corg via top-down processes such as herbiv- Natural abundance of stable isotopes, most commonly 13C, ory38,101–103. Similarly, predators can regulate biomass, persis- 15N and 34S, have been used to trace and quantify allochthonous tence and recovery of seagrasses, marshes and mangroves by and autochthonous Corg sources and their relative contributions triggering trophic cascades38. In addition, the functional diversity to carbon burial. The costs are low, the methodology for sample and activity of the microbial decomposer community, and how preparation and analysis is relatively easy and the validity of the they vary with depth and over time, is only just beginning to be technique has been widely, and generally successfully tested88. examined104 and will need to be linked to BC burial rates. Most However, the diversity of organic matter inputs can result in likely this microbial community will be more important in complex mixtures of Corg that are not well resolved based on the defining the fate of Corg entering BC soils than its production and isotopic separation of the sources. Isotopic values of different sedimentation. species may be similar, or may vary within the same species with The general effects of hydrodynamics on carbon sequestration microhabitats, seasons, growth cycle or tissue type89,90. in BC ecosystems are understood, yet there is much we still do The use of bulk stable isotopes must be improved by addi- not understand which could explain the variability in sequestra- tionally analysing individual compounds with a specific taxo- tion we see across BC ecosystems. We know that hydrodynamics, nomic origin. Biomarkers such as lignin, lipids, alkanes and mediated by biological properties of BC ecosystems (e.g., canopy amino acids, have proven useful for separating multiple-source size and structure), affect particle trapping105–107 and, pre- inputs in coastal sediments88,91. Leading-edge studies, using sumably, Corg sedimentation rates. For example, increasing den- compound-specific stable isotopes, employ both natural and sity of mangrove stands positively affects affect wave attenuation, radiocarbon analyses, providing the added dimension of age to enhancing the accumulation of fine grained material108, which taxonomic specificity92,93. Oxygen and hydrogen stable isotopes promotes Corg accumulation (silts and clays retain more Corg than could also be used to improve resolving power, but up to now sands109,110. However, significant variation in soil Corg has been they have been used mainly in foodweb studies and their utility in observed within seagrass meadow111, pointing to complex determining sedimentary sources in coastal systems still needs to canopy-hydrodynamic interactions which we do not understand be validated87. Studies using both bulk and compound-specific but which could affect our ability to develop robust estimates of isotopes must consider how decomposition may alter species- meadow-scale BC burial. For example, a study of restored sea- specific signatures89,90,94 Other, alternative fingerprinting tech- grass meadow found strong positive correlations between Corg niques are emerging. The deliberate stable isotope labelling of stocks and edge proximity leading to gradients in carbon stocks at organic matter and tracing its fate is a powerful approach that scales of >1 km112. Elsewhere, flexible canopies have been shown overcomes some of the limitations of natural abundance studies to interact with wave dynamics, increasing turbulence near the (e.g., source overlap), but has only looked at short-term Corg sediment surface113. This could explain the loss of fine sediments, burial to-date95. The use of environmental DNA (eDNA) has and presumably Corg, in low shoot density meadows compared to been used to describe community composition in marine systems, high density meadows114, with implications for carbon seques- but the potential to quantify the taxonomic proportions of plant tration over time following restoration of BC ecosystems and the sources in sediments has rarely been tested87,96. development of canopy density. Because these types of hydro- Overall, projects using 13C and 15N stable isotopes will likely dynamic interaction can affect the spatial and temporal patterns continue to dominate the investigation of organic matter sources, in carbon accumulation they need to be better understood in especially in simple two end member systems. While there is a order to design stock and accumulation assessments and to growing suite of organic matter tracers, the ability to distinguish predict the temporal development of stocks following manage- between specific blue carbon sources such as marsh vegetation ment actions. and seagrass still remains a challenge. Sample size requirement, The basic biogeochemical controls on Corg accumulation analytical time and cost implications, will be crucial in the within soils are understood (e.g., biochemical nature of the Corg selection of the most appropriate tracers for the characterisation inputs which vary among primary producers115–117 and the and quantification of the molecular complexity in blue carbon chemistry of their decomposition products)110, but it remains sediments. In general, applications of most compound specific unclear what controls the stability of stored Corg in BC soils and tracers have focused on environments other than those sup- whether these factors vary across ecosystems or under different porting blue carbon ecosystems88,93,97, and more work is needed environmental conditions (incl. disturbance). With the exception to apply the same research tools to these systems. We recom- of one recent paper43, we know little about the Corg -mineral mend, wherever possible, that complementary methods such as associations in BC ecosystems, how these affect the recalcitrance compound-specific isotopes and eDNA that take advantage of of soil Corg or whether specific forms are protected more by this methodological advances in distinguishing species contributions, mechanism than others, though this is clearly the case in other be used in conjunction with bulk isotopes. ecosystems118–120. Undoubtedly the anaerobic character of BC Q7. What factors influence BC burial rates? soils places a significant control on in situ rates of Corg decom- BC ecosystems have an order of magnitude greater C burial position and remineralisation. However, the time organic mate- rates than terrestrial ecosystems3. This high BC burial rate is a rials are exposed to oxygen before entering the anaerobic zone of product of multiple processes that affect: the mass of C produced BC soils will impact the quantity and nature of Corg as will the and its availability for burial; its sedimentation; and its sub- redox potential reached within the soil. The amount of time sequent preservation. A host of interacting biological, biogeo- organic matter is exposed to oxygen explains the observation that chemical and physical factors, as well as natural and Corg concentrations in tidal marshes globally are higher on anthropogenic disturbance (see Q2), affect these processes. With coastlines where relative sea level rise has been rapid compared to 6 NATURE COMMUNICATIONS | (2019) 10:3998 | https://doi.org/10.1038/s41467-019-11693-w | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-11693-w PERSPECTIVE those where sea level has been relatively stable18. Moreover, 0.012 Tg N2O-N yr−1 and a 50% loss of mangroves would result exposure of BC to oxygen has been recently shown trigger in a global reduction in emissions of 0.017 Tg N2O-N yr−1 microbial attack, even ancient (5000-year-old) and chemically (ref. 130). recalcitrant BC43. Enhancing our understanding of oxygen Q9. How can we reduce uncertainties in the valuations of exposure times and critical redox potentials will help explain Blue Carbon? variations in Corg accumulation rates and preservation within Studies into BC increasingly include a valuation aspect, different BC ecosystems. focussed on coastal sites135 but more recently also including From the above, there is increasing evidence that we do not offshore sites136, showing a range of values for different ecosys- understand the complex interactions among influencing envir- tems as depicted in Fig. 2. Differences in values are driven by onmental factors well enough to predict likely Corg stocks in soils, differences in BC sequestration and storage capacity and/or including temperature, hydrodynamic, geomorphic and hydro- potential avoided emissions through conservation and restoration logic factors that can affect biogeochemical processes or mediate of ecosystems. There is also variation in BC values due to biological processes, and this leads to apparent contradictions. uncertainties in the calculation of C sequestration and perma- For example, the influence of nutrient availability on Corg stocks nence of C storage, as is required for valuation. The wide range of is unclear with one study reporting an increase in soil Corg stocks C valuation methods, including social costs of C111, marginal along a gradient of increasing phosphate availability121, another abatement costs112, and C market prices, also enhances the reporting no effect122, and yet others121,123 finding that increas- uncertainty and variation in valuation estimates. ing nutrient availability led to lower soil Corg. Some empirical Valuation of BC enables its inclusion in policy and manage- studies have examined interactive effects or evoked them to ment narratives113, facilitating the comparison of future socio- explain difference in Corg stock101,124,125. However, these studies economic scenarios, including mitigation and adaptation inter- are rare and limited by the complexity or the interactions being ventions137, and raises conservation interests as an approach to examined. We conclude that gaining insights into these inter- mitigate climate change and offset CO2 emissions2. For example, active effects is more likely to be advanced through modelling BC budgets can be incorporated into national greenhouse gas approaches. inventories138. Alternatively, demonstrable gains in C sequestra- Q8. What is the net flux of greenhouse gases between Blue tion and/or avoided emissions through conservation and Carbon ecosystems and the atmosphere? restoration activities can be credited within voluntary C markets BC ecosystems are generally substantial sources or sinks of or through the Clean Development Mechanism of the United greenhouse gases (GHGs) (CO2, CH4, N2O), though we cannot Nations Framework Convention on Climate Change (UNFCCC) construct accurate global BC budgets due to uncertainties in net 86. Voluntary market methodologies for BC ecosystems have been fluxes. The C budget is best constrained for mangroves, with released within the American Carbon Registry139 and within the mangroves globally taking up 700 Tg C yr−1 through Gross Verified Carbon Standard86, while some countries are developing Primary Production, and respiring 525 Tg C yr−1 (75%) back to BC-focussed climate change mitigation schemes that provide the atmosphere as CO2126. However, large uncertainty exists in economic incentives. However, on the international scale, BC budgets due to poorly constrained mineralization pathways ecosystems have previously not been consistently incorporated linked to CO2 efflux119. into frameworks for climate change mitigation that offer We lack robust global C budgets for other BC ecosystems due economic reward for the conservation of C sinks, such as the to insufficient empirical evidence127. For example, while we have REDD + program140, possibly as there was insufficient informa- estimated global soil Corg stocks128 and accumulation rates for tion for its inclusion. Avoiding degradation of mangroves, seagrasses, this is insufficient to create a budget129 because we tidal marshes and seagrasses could globally offer up to 1.02 Pg lack representative data on community metabolism and GHG CO2-e yr−1 in avoided emissions37. Developing countries with fluxes, particularly for CH4 and N2O emissions. Thus, we need to BC resources have the opportunity to use BC for the NDC, for better quantify sink/source balances, e.g., the net balance between example Indonesia, where BC contribution to reduce emissions primary production vs. emissions from ecosystem degradation could be as much as 0.2 Pg CO2-e yr−1 or 30% of national land- and pelagic, benthic, forest floor and canopy respiration126. We based emission while mangrove deforestation only contributes to also need to understand how source/sink dynamics change 6% of national deforestation141. budgets over time and how environmental parameters affect To reduce uncertainty in BC values and encourage use of GHG fluxes129,130, allowing us to estimate thresholds that flip BC values in future policy and management, we recommend ecosystems from GHG sinks to sources. improved interdisciplinary research, combining ecological and Budgets generally focus on CO2 fluxes, though we must better economic disciplines to develop standardised approaches to understand fluxes of other GHGs such as CH4 and N2O, and their improve confidence in the valuation of BC. Ideally this should be contribution to the global BC budget131. Global estimates show undertaken alongside studies which recognise the additional that CH4 emissions can offset C burial in mangroves by 20% values of conserving BC ecosystems, for example the benefits because CH4 has a higher global warming potential than CO2 on generated from fisheries enhancement, nutrient cycling, support a per molecule basis132. CH4 emissions may also offset C burial in to coastal communities and their livelihoods2 and coastal pro- seagrasses, though these estimates have not been made. In con- tection, which is considered a cost-effective method compared to trast, some mangroves are N2O sinks133 which would enhance the hard engineering solutions142. value of the C burial as a means to mitigate climate change. Q10. What management actions best maintain and promote Overall, CH4 and N2O biogeochemistry is understudied in BC Blue Carbon sequestration? ecosystems. Research over the past decade has improved estimates of C Finally, we must understand how GHG fluxes change as BC dynamics at a range of spatial scales. This has enabled modelling ecosystems replace each other, such as when mangroves expand of potential emissions from the conversion of seagrass, mangrove onto marshes at their latitudinal limits134, or are planted on and tidal marsh to other uses41, and estimates of rates of and seagrass meadows in Southeast Asia. We also need to understand hotspots for CO2 emissions resulting from ecosystem loss. The how emissions may change with loss of BC ecosystems. For development of policy, implementation of management actions example, it has been coarsely estimated that a 50% loss of sea- and the demonstration of BC benefits (including payments), grass would result in a global reduction in N2O emissions of however, are still in their infancy. NATURE COMMUNICATIONS | (2019) 10:3998 | https://doi.org/10.1038/s41467-019-11693-w | www.nature.com/naturecommunications 7 PERSPECTIVE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-11693-w 91,000 Ecosystem service value per ha (US dollars) 15,000 10,000 5000 0 Tidal marsh Mangrove Seagrass Coastal Deep sea plankton Fig. 2 Estimates of the economic value of blue carbon ecosystems per hectare. Data from ref. 1 and references therein. Symbols and images are courtesy of the Integration and Application Network, University of Maryland Center for Environmental Science (ian.umces.edu/symbols/) There are three broad management approaches to enhance C ecosystems in many countries (e.g., USA and EU). These gen- mitigation by BC ecosystems: preservation, restoration and erally imply the creation of BC ecosystems to replace those lost creation. Preserving ecosystem extent and quality—for exam- through development. Such approaches should be treated with ple, through legislative protection and/or supporting alternative caution, however, since there is confusion about terminol- livelihoods—has the two-fold benefit of avoiding the reminer- ogy141, lack of enforcement and limited capacity to recreate the alisation of historically sequestered C, while also protecting qualities of pristine sites. future sequestration capacity. Preservation may include direct Tools for the accounting and crediting of C payments now or indirect approaches to maintain or enhance biogeochemical exist for coastal wetland conservation, restoration and creation processes, such as sedimentation and water supply46. Restora- under the voluntary C market86,147. Several small-scale projects tion pertains to a range of activities seeking to improve (e.g., Mikoko Pamoja in Kenya) are now using these frame- biophysical and geochemical processes—and therefore seques- works to generate C credits with others projects in develop- tration capacity—in BC ecosystems. Examples include passive ment148. Few jurisdictions have adopted their own mechanisms and/or active reforestation of logged and degraded mangrove for the accounting and/or trading of BC, though some have forests143; earthwork interventions to return aquaculture ponds undertaken preliminary research to identify BC policy to mangrove ecosystems141; and the restoration of hydrology to opportunities149. drained coastal floodplains144. Managed realignment is a par- Technical, financial and policy barriers remain before local ticular option for creating or restoring tidal marshes as part of a initiatives can be scaled-up to make large impacts—such as strategy to achieve sustainable coastal flood defence together through national REDD + initiatives. Significant barriers include: with the provision of other services, including C benefits145; biases in the geographic coverage of data; approaches for robust, other similar options include: regulated tidal exchange131 and site-specific assessment and prediction of some C pools (e.g., beneficial use of dredged material146. Although restoration may below-ground C and atmospheric emissions); high transaction re-establish C sequestration processes, it is important to note costs; and ensuring that equity and justice are achieved. In that it may not prevent large amounts of fossil C being lost addition, most demonstrated efforts are recent actions with little following future disturbance or intervention. ‘No net loss’ quantification of C mitigation benefits (or societal outcomes) policies have been now developed and applied to wetland beyond the scale of a few years. 8 NATURE COMMUNICATIONS | (2019) 10:3998 | https://doi.org/10.1038/s41467-019-11693-w | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-11693-w PERSPECTIVE Despite such barriers, we now have the fundamental knowl- of BC science to engage scientists from different disciplines edge to justify the inclusion of BC protection, restoration and working within a wide range of ecosystems and nations. creation in C mitigation mechanisms. While there remain knowledge gaps—both in science, policy and governance—these Received: 6 November 2018 Accepted: 31 July 2019 will partly be addressed through the effective demonstration, monitoring and reporting of existing and new BC projects. Toward a research agenda on the role of vegetated coastal References ecosystems on climate change mitigation and adaptation 1. Nellemann C., et al. (eds) Blue Carbon. A Rapid Response Assessment. United The questions above are not short of challenges and therefore, Nations Environment Programme (GRID-Arendal, 2009). This report was the provide ample scope for decisive experiments to be designed and first to use the term ‘blue carbon’ 2. Duarte, C. M., Losada, I. J., Hendriks, I. E., Mazarrasa, I. & Marba, N. The role conducted, current hypotheses to be rejected or consolidated and of coastal plant communities for climate change mitigation and adaptation. new ideas and concepts to unfold. Emerging questions that are Nat. Clim. Change 3, 961–968 (2013). Reviewed data on blue carbon burial, not yet supported by robust observations and experiments, stocks, accretion rates and potential losses. include, for example: the estimation of allochthonous C (organic 3. McLeod, E. et al. A blueprint for blue carbon: toward an improved and inorganic) contributions to BC, which remains challenging understanding of the role of vegetated coastal habitats in sequestering CO2. Front. Ecol. Environ. 9, 552–560 (2011). Identified key areas of uncertainly due to availability of markers able to quantitatively discriminate and specific actions needed to understand the role of vegetated coastal among the different carbon sources; and the net balance of GHG ecosystems as carbon sinks. emissions, which remains challenging as it requires concurrent 4. Macreadie Peter, I., Serrano, O., Maher Damien, T., Duarte Carlos, M. & measurements across relevant time and spatial scales of all major Beardall, J. Addressing calcium carbonate cycling in blue carbon accounting. GHGs (CO2, CH4, NO2), for which not a single estimate is Limnol. Oceanogr. Lett. 2, 195–201 (2017). Argued that calcium carbonate cycling has been ignored in blue carbon offset schemes, but warrants serious available to-date. The core questions that capture much of current attention. research efforts in BC science include the role of climate change 5. Howard, J. et al. Clarifying the role of coastal and marine systems in climate on C accumulation, efforts to improve the precision of global mitigation. Front. Ecol. Environ. 15, 42–50 (2017). estimates of the extent of BC ecosystems, factors that influence 6. Krause-Jensen D., Duarte C. M. Substantial role of macroalgae in marine sequestration in BC ecosystems, with the corresponding value of carbon sequestration. Nat. Geosci. 9, 737 (2016). Demonstrated that macroalgal export can make an important contribution to deep sea carbon BC, and the management actions that are effective in enhancing sequestration, where it can be sequestered from the atmosphere. this value. The preceding text provides a summary of current 7. Smith, S. V. Marine macrophytes as a global carbon sink. Science 211, 838–840 research efforts and future opportunities in addressing these key (1981). Seminal paper on the role of marine macrophytes in sequestering questions. CO2; suggested that the role of seagrasses were overlooked. Three questions are long-standing, controversial, and need 8. Duarte, C. M., Middelburg, J. J. & Caraco, N. Major role of marine vegetation on the oceanic carbon cycle. Biogeosciences 2, 1–8 (2005). This paper showed resolution in order to properly constrain the BC paradigm. The that vegetated marine habitats are responsible for a previously unaccounted, first is the effect of disturbance on GHG emissions from BC 50% of C burial in ocean sediments. ecosystems, where the initial assumption, that the top meter of 9. Kennedy H., et al. in Supplement to the 2006 IPCC Guidelines for National the soil C stock is likely to be emitted as GHG following dis- Greenhouse Gas Inventories: Wetlands (eds Hiraishi, T., et al.) (IPCC, 2014). turbance37,128, continues to be carried across papers without 10. Arias-Ortiz, A. et al. A marine heatwave drives massive losses from the world’s largest seagrass carbon stocks. Nat. Clim. Change 8, 338 (2018). being challenged or verified. The second is whether macroalgae-C 11. Marbà, N. et al. Impact of seagrass loss and subsequent revegetation on carbon can be considered BC. The term BC refers to C sequestered in the sequestration and stocks. J. Ecol. 103, 296–302 (2015). oceans1, and the focus on seagrass, mangroves and tidal marshes 12. Macreadie, P. I. et al. Losses and recovery of organic carbon from a seagrass is justified by the intensity of local C sequestration these eco- ecosystem following disturbance. Proc. Biol. Sci. 282, 1–6 (2015). systems support. If macroalgae provide intense C sequestration, 13. Atwood, T. B. et al. Global patterns in mangrove soil carbon stocks and losses. Nat. Clim. Change 7, 523 (2017). whether in the ecosystem or beyond, they need to be dealt with in 14. IPCC. Climate Change. 2007. Impacts, Adaptation and Vulnerability. this context. And the third controversy is whether carbonate Contribution of Working Group II to the Fourth Assessment Report of the accumulation in BC ecosystems render them potential sinks of Intergovernmental Panel on Climate Change (Intergovernmental Panel on CO2 following disturbance. It is clear that there are far too many Climate Change, Geneva, Switzerland, 2007). key uncertainties4 to resolve this at the conceptual level, since 15. Cabanes, C., Cazenave, A. & Le Provost, C. Sea level rise during past 40 years determined from satellite and in situ observations. Science 294, 840–842 empirical evidence to provide a critical test is as yet lacking. We (2001). propose that a research program including key observational and 16. Hansen, J. et al. Global temperature change. Proc. Natl Acad. Sci. USA 103, experimental tests designed to resolve the mass balance of car- 14288–14293 (2006). bonate (e.g., balance between allochthonous and autochthonous 17. Knutson, T. R. et al. Tropical cyclones andclimate change. Nat. Geosci. 3, production and dissolution)—and then the coupling between BC 157–163 (2010). 18. Rogers, K. et al. Wetland carbon storage controlled by millennial-scale ecosystems and the atmosphere—is needed. In the case of all variation in relative sea-level rise. Nature 567, 91–95 (2019). three controversies, we believe that the positive approach to 19. Kirwan, M. L. & Megonigal, J. P. Tidal wetland stability in the face of human address these questions, is to pause the current discussion, which impacts and sea-level rise. Nature 504, 53–60 (2013). are largely rooted in the lack of solid, direct empirical evidence, 20. Lovelock, C. E. et al. The vulnerability of Indo-Pacific mangrove forests to sea- and recognize that further science is required before any con- level rise. Nature 526, 559–563 (2015). 21. Woodroffe, C. D., et al. Mangrove sedimentation and response to relative sea- clusion can be reached. level rise. Annu. Rev. Mar. Sci. 8, 243–266 (2016). In summary, the overview of questions provided above por- 22. Schuerch, M. et al. Future response of global coastal wetlands to sea-level rise. trays BC science as a vibrant field that is still far away from Nature 561, 231–234 (2018). reaching maturity. Apparent controversies are a consequence of 23. Kelleway, J. J. et al. Seventy years of continuous encroachment substantially this lack of maturity and need to be resolved through high quality, increases ‘blue carbon’ capacity as mangroves replace intertidal salt marshes. Glob. Change Biol. 22, 1097–1109 (2016). scalable and reproducible observations and experiments. We 24. Albert, S., et al. Winners and losers as mangrove, coral and seagrass believe the questions above inspire a multifarious research agenda ecosystems respond to sea-level rise in Solomon Islands. Environ. Res. Lett. 12, that will require continued broadening the community of practice 094009 (2017). NATURE COMMUNICATIONS | (2019) 10:3998 | https://doi.org/10.1038/s41467-019-11693-w | www.nature.com/naturecommunications 9 PERSPECTIVE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-11693-w 25. Lee, S. Y., Hamilton, S., Barbier, E. B., Primavera, J. & Lewis, R. R. Better 56. Giri, C. et al. Status and distribution of mangrove forests of the world restoration policies are needed to conserve mangrove ecosystems. Nat. Ecol. using earth observation satellite data. Glob. Ecol. Biogeogr. 20, 154–159 Evol. 3, 870–872 (2019). (2011). 26. Ellison, J. C. Mangrove retreat with rising sea-level, Bermuda. Estuar. Coast 57. Hamilton, S. E. & Casey, D. Creation of a high spatio-temporal resolution Shelf Sci. 37, 75–87 (1993). global database of continuous mangrove forest cover for the 21st century 27. Wernberg, T. et al. An extreme climatic event alters marine ecosystem (CGMFC-21). Glob. Ecol. Biogeogr. 25, 729–738 (2016). structure in a global biodiversity hotspot. Nat. Clim. Change 3, 78–82 (2013). 58. Valiela, I., Bowen, J. L. & York, J. K. Mangrove forests: one of the world’s 28. Reef, R. et al. The effects of elevated CO2 and eutrophication on surface threatened major tropical environments. Bioscience 51, 807–815 (2001). elevation gain in a European salt marsh. Glob. Change Biol. 23, 881–890 59. Adame, M. F. et al. Carbon stocks of tropical coastal wetlands within the (2017). Karstic landscape of the Mexican Caribbean. PLoS One 8, e56569 (2013). 29. Asbridge, E., Lucas, R., Ticehurst, C. & Bunting, P. Mangrove response to 60. Lugo, A. E. Old-growth mangrove forests in the United States. Conserv Biol. environmental change in Australia’s Gulf of Carpentaria. Ecol. Evol. 6, 11, 11–20 (1997). 3523–3539 (2016). 61. Adam, P. Saltmarshes in a time of change. Environ. Conserv 29, 39–61 (2002). 30. Syvitski, J. P. M. et al. Sinking deltas due to human activities. Nat. Geosci. 2, 62. Woodwell, G.M., Rich, P.H., Mall, C.S.A. Carbon in the Biosphere. In: Woodwell, 681–686 (2009). G.M., Pecan, E.V. (eds.) Proceedings of the 24th Brookhaven Symposium in 31. Spencer, T. et al. Global coastal wetland change under sea-level rise and Biology pp. 221–240 (USAEC, Springfield, Virginian, 1973). related stresses: the DIVA Wetland Change Model. Glob. Planet. Change 139, 63. McOwen, C. J. et al. A global map of saltmarshes. Biodiversity Data J. 55, 15–30 (2016). e11764 (2017). 32. Balke, T. & Friess, D. A. Geomorphic knowledge for mangrove restoration: a 64. Chmura, G. L., Anisfeld, S. C., Cahoon, D. R. & Lynch, J. C. Global carbon pan-tropical categorization. Earth Surf. Process. Landf. 41, 231–239 (2016). sequestration in tidal, saline wetland soils. Glob. Biogeochem. Cycles 17, 1111 33. Leonardi, N., Ganju, N. K. & Fagherazzi, S. A linear relationship between wave (2003). power and erosion determines salt-marsh resilience to violent storms and 65. Macreadie, P. I., et al. Carbon sequestration by Australian tidal marshes. Sci. hurricanes. Proc. Natl Acad. Sci. USA 113, 64–68 (2016). Rep. www.nature.com/articles/srep44071 (2017). 34. Saunders, M. I. et al. Interdependency of tropical marine ecosystems in 66. Lotze, H. K. et al. Depletion, degradation, and recovery potential of estuaries response to climate change. Nat. Clim. Change 4, 724–729 (2014). and coastal seas. Science 312, 1806–1809 (2006). 35. Saderne, V. et al. Accumulation of carbonates contributes to coastal vegetated 67. Duarte, C. M., Dennison, W. C., Orth, R. J. W. & Carruthers, T. J. B. The ecosystems keeping pace with sea level rise in an Arid Region (Arabian charisma of coastal ecosystems: addressing the imbalance. Estuaries Coasts 31, Peninsula). J. Geophys. Res. 123, 1498–1510 (2018). 233–238 (2008). 36. Saunders, M. I. et al. Coastal retreat and improved water quality mitigate 68. UNEP-WCMC. Global distribution of seagrasses (version 4.0). Fourth update losses of seagrass from sea level rise. Glob. Change Biol. 19, 2569–2583 (2013). to the data layer used in Green and Short (2003) (UNEP-WCMC, Cambridge, 37. Pendleton, L. et al. Estimating global “blue carbon” emissions from conversion 2016). and degradation of vegetated coastal ecosystems. PLoS ONE 7, e43542–e43542 69. Charpy-Roubaud, C. & Sournia, A. The comparative estimation of (2012). Estimated that up to 1 billion tonnes of CO2 is emitted each year to phytoplanktonic, microphytobenthic and macrophytobenthic primary the atmosphere following destruction of blue carbon ecosystems. production in the oceans. Mar. Microb. Food Webs 4, 31–57 (1990). 38. Atwood, T. B. et al. Predators help protect carbon stocks in blue carbon 70. Gattuso, J. P. et al. Light availability in the coastal ocean: impact on the ecosystems. Nat. Clim. Change 5, 1038–1045 (2015). distribution of benthic photosynthetic organisms and their contribution to 39. Bouma, T. J. et al. Identifying knowledge gaps hampering application of primary production. Biogeosciences 3, 489–513 (2006). intertidal habitats in coastal protection: opportunities & steps to take. Coast. 71. Short, F. T. et al. Extinction risk assessment of the world’s seagrass species. Eng. 87, 147–157 (2014). Biol. Conserv 144, 1961–1971 (2011). 40. Lovelock, C. E. et al. Assessing the risk of carbon dioxide emissions from blue 72. Waycott, M. et al. Accelerating loss of seagrasses across the globe threatens carbon ecosystems. Front. Ecol. Environ. 15, 257–265 (2017). coastal ecosystems. Proc. Natl Acad. Sci. USA 106, 12377–12381 (2009). 41. Lovelock, C. E., Fourqurean, J. W. & Morris, J. T. Modeled CO2 emissions 73. Orth, R. J. et al. A global crisis for seagrass ecosystems. Bioscience 56, 987–996 from coastal wetland transitions to other land uses: tidal marshes, mangrove (2006). forests, and seagrass beds. Front. Mar. Sci. 4, 1–11 (2017). 74. Pham, D. T. et al. A review of remote sensing approaches for monitoring Blue 42. Macreadie, P. I., Hughes, A. R. & Kimbro, D. L. Loss of ‘blue carbon’ from Carbon ecosystems: mangroves, seagrasses and salt marshes during coastal salt marshes following habitat disturbance. PLoS One 8, 1–8 (2013). 2010–2018. Sensors 19, E1933 (2019). 43. Macreadie, P. I., et al. Vulnerability of seagrass blue carbon to microbial attack 75. Regnier, P. et al. Anthropogenic perturbation of the carbon fluxes from land to following exposure to warming and oxygen. 686, 264–275 (2019). ocean. Nat. Geosci. 6, 597–607 (2013). 44. Silliman, B. R. et al. Degradation and resilience in Louisiana salt marshes after 76. Maher, D. T. & Eyre, B. D. Carbon budgets for three autotrophic Australian the BP-deepwater horizon oil spill. Proc. Natl Acad. Sci. USA 109, estuaries: implications for global estimates of the coastal air-water CO2 flux. 11234–11239 (2012). Glob. Biogeochem. Cycles 26, GB1032 (2012). 45. Sidik, F. & Lovelock, C. E. CO2 efflux from shrimp ponds in Indonesia. PLoS 77. Tokoro, T. et al. Net uptake of atmospheric CO2 by coastal submerged aquatic One 8, e66329 (2013). vegetation. Glob. Change Biol. 20, 1873–1884 (2014). 46. Macreadie, P. I. et al. Can we manage coastal ecosystems to sequester more 78. Howard Jason, L., Creed Joel, C., Aguiar Mariana, V. P. & Fourqurean James, blue carbon? Front. Ecol. Environ. 15, 206–213 (2017). Proposed three key W. CO2 released by carbonate sediment production in some coastal areas may management actions for maximising blue carbon sequestration within offset the benefits of seagrass “Blue Carbon” storage. Limnol. Oceanogr. 63, existing coastal vegetated ecosystems. 160–172 (2017). 47. Coverdale, T. C. et al. Indirect human impacts reverse centuries of carbon 79. Mazarrasa, I. et al. Seagrass meadows as a globally significant carbonate sequestration and salt marsh accretion. PLoS One 9, e9396 (2014). reservoir. Biogeosciences 12, 4993–5003 (2015). 48. Duarte, C. M. Reviews and syntheses: hidden forests, the role of vegetated 80. Fodrie, F. J., et al. Oyster reefs as carbon sources and sinks. Proc. Biol. Sci. 284, coastal habitats in the ocean carbon budget. Biogeosciences 14, 301–310 (2017). 20170891 (2017). 49. Raven, J. A. The possible roles of algae in restricting the increase in 81. Watanabe, K. & Kuwae, T. How organic carbon derived from multiple sources atmospheric CO2 and global temperature. Eur. J. Phycol. 52, 506–522 (2017). contributes to carbon sequestration processes in a shallow coastal system? 50. Trevathan-Tackett, S. M. et al. Comparison of marine macrophytes for their Glob. Change Biol. 21, 2612–2623 (2015). contributions to blue carbon sequestration. Ecology 96, 3043–3057 (2015). 82. Bauer, J. E. et al. The changing carbon cycle of the coastal ocean. Nature 504, 51. Hill, R. et al. Can macroalgae contribute to blue carbon? An Australian 61–70 (2013). perspective. Limnol. Oceanogr. 60, 1689–1706 (2015). 83. Kuwae, T. et al. Blue carbon in human-dominated estuarine and shallow 52. van der Heijden, L. H. & Kamenos, N. A. Reviews and syntheses: calculating coastal systems. Ambio 45, 290–301 (2016). the global contribution of coralline algae to total carbon burial. Biogeosciences 84. Hyndes, G. A. et al. Mechanisms and ecological role of carbon transfer within 12, 6429–6441 (2015). coastal seascapes. Biol. Rev. 89, 232–254 (2013). 53. Smith, S. V. Parsing the Oceanic Calcium Carbonate Cycle: a Net Atmospheric 85. Murray, B., Pendleton, L., Jenkins, W. & Sifleet, S. Green Payments for Blue Carbon Dioxide Source, or a Sink? (Association for the Sciences of Limnology Carbon: Economic Incentives for Protecting Threatened Coastal Habitats and Oceanography (ASLO) L&O e-Books, 2013). (Nicholas Institute for Environmental Policy Solutions, Duke University, 54. Saderne, V. et al. Role of carbonate burial in Blue Carbon budgets. Nat. Durham, 2011). Commun. 10, 1106 (2019). 86. Emmer, I., et al. Methodology for Tidal Wetland and Seagrass Restoration. in 55. Along, D. M. The Energetics of Mangrove Forests (Springer Science and Verified Carbon Standard.VM0033 (2015) The first voluntary market Business Media BV, 2009). methodology for blue carbon ecosystems 10 NATURE COMMUNICATIONS | (2019) 10:3998 | https://doi.org/10.1038/s41467-019-11693-w | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-11693-w PERSPECTIVE 87. Geraldi, N. R. et al. Fingerprinting Blue Carbon: rationale and tools to 115. Trevathan-Tackett, S. M. et al. A global assessment of the chemical determine the source of organic carbon in marine depositional environments. recalcitrance of seagrass tissues: Implications for long-term carbon Front. Mar. Sci. 6, 263 (2019). sequestration. Front. Plant Sci. 8, 925 (2017). 88. Bianchi, T. S., et al. Redox effects on organic matter storage in coastal 116. Torbatinejad, N. M., Annison, G., Rutherfurd-Markwick, K. & Sabine, J. R. sediments during the holocene: a biomarker/proxy perspective. Annu. Rev. Structural constituents of the seagrass Posidonia australis. J. Agric. Food Chem. Earth Planet. Sci. 44, 295–319 (2016). 55, 4021–4026 (2007). 89. Kramer, M. G., Lajtha, K. & Aufdenkampe, A. K. Depth trends of soil organic 117. Kogel-Knabner, I. The macromolecular organic composition of plant and matter C:N and 15N natural abundance controlled by association with microbial residues as inputs to soil organic matter. Soil Biol. Biochem 34, minerals. Biogeochemistry 136, 237–248 (2017). 139–162 (2002). 90. Canuel, E. A. & Hardison, A. K. Sources, ages, and alteration of organic matter 118. Yeasmin, S., Singh, B., Johnston, C. T. & Sparks, D. L. Organic carbon in estuaries. Annu. Rev. Mar. Sci. 8, 409–434 (2016). characteristics in density fractions of soils with contrasting mineralogies. 91. Upadhayay, H. R. et al. Methodological perspectives on the application of Geochim. Cosmochim. Acta 218, 215–236 (2017). compound-specific stable isotope fingerprinting for sediment source 119. Lehmann, J. & Kleber, M. The contentious nature of soil organic matter. apportionment. J. Soils Sediment. 17, 1537–1553 (2017). Nature 528, 60–68 (2015). 92. Wakeham, S. G. & McNichol, A. P. Transfer of organic carbon through 120. Baldock, J. A. & Skjemstad, J. O. Role of the soil matrix and minerals in marine water columns to sediments – insights from stable and radiocarbon protecting natural organic materials against biological attack. Org. Geochem. isotopes of lipid biomarkers. Biogeosciences 11, 6895–6914 (2014). 31, 697–710 (2000). 93. Canuel, E. A. & Hardison, A. K. Sources, ages, and alteration of organic matter 121. Armitage, A. R. & Fourqurean, J. W. Carbon storage in seagrass soils: long- in estuaries. Annu. Rev. Mar. Sci. 8, 409–434 (2016). term nutrient history exceeds the effects of near-term nutrient enrichment. 94. Oreska Matthew, P. J., Wilkinson Grace, M., McGlathery Karen, J., Bost, M. & Biogeosciences 13, 313–321 (2016). McKee Brent, A. Non‐seagrass carbon contributions to seagrass sediment blue 122. Howard, J. L., Perez, A., Lopes, C. C. & Fourqurean, J. W. Fertilization changes carbon. Limnol. Oceanogr. 63, S3–S18 (2018). seagrass community structure but not blue carbon storage: results from a 30- 95. Oakes, J. M. & Eyre, B. D. Transformation and fate of microphytobenthos year field experiment. Estuaries Coasts 39, 1422 (2016). carbon in subtropical, intertidal sediments: potential for long-term 123. Martinez-Crego, B., Olive, I. & Santos, R. CO2 and nutrient-driven changes carbon retention revealed by 13C-labeling. Biogeosciences 11, 1927–1940 across multiple levels of organization in Zostera noltii ecosystems. (2014). Biogeosciences 11, 7237–7249 (2014). 96. Reef, R. et al. Using eDNA to determine the source of organic carbon in 124. Janousek, C. N. et al. Inundation, vegetation, and sediment effects on litter seagrass meadows. Limnol. Oceanogr. 62, 1254–1265 (2017). decomposition in Pacific Coast tidal marshes. Ecosystems 20, 1296–1310 97. Close, H. G. Compound-specific isotope geochemistry in the ocean. Annu. (2017). Rev. Mar. Sci. 11, 27–56 (2019). 125. Weiss, C. et al. Soil organic carbon stocks in estuarine and marine mangrove 98. Handa, I. T. et al. Consequences of biodiversity loss for litter decomposition ecosystems are driven by nutrient colimitation of P and N. Ecol. Evol. 6, across biomes. Nature 509, 218 (2014). 5043–5056 (2016). 99. Chapin, F. S. Effects of plant traits on ecosystem and regional processes: a 126. Alongi, D. M. Carbon cycling and storage in mangrove forests. Annu. Rev. conceptual framework for predicting the consequences of global change. Ann. Marine Sci. 6, 195–219 (2014). Bot. 91, 455–463 (2003). 127. Duarte, C. M., et al. Seagrass community metabolism: assessing the carbon 100. Kelleway, J. J., Saintilan, N., Macreadie, P. I., Baldock, J. A. & Ralph, P. J. sink capacity of seagrass meadows. Global Biogeochem. Cycles 24, GB4032 Sediment and carbon deposition vary among vegetation assemblages in a (2010). coastal salt marsh. Biogeosciences 14, 3763–3779 (2017). 128. Fourqurean, J. W. et al. Seagrass ecosystems as a globally significant carbon 101. Thomas, C. R. & Blum, L. K. Importance of the fiddler crab Uca pugnax to salt stock. Nat. Geosci. 5, 505–509 (2012). marsh soil organic matter accumulation. Mar. Ecol. Prog. Ser. 414, 167–177 129. Macreadie, P. I., Baird, M. E., Trevathan-Tackett, S. M., Larkum, A. W. D. & (2010). Ralph, P. J. Quantifying and modelling the carbon sequestration capacity of 102. Johnson, R. A., Gulick, A. G., Bolten, A. B. & Bjorndal, K. A. Blue carbon seagrass meadows—a critical assessment. Mar. Pollut. Bull. 83, 430–439 stores in tropical seagrass meadows maintained under green turtle grazing. Sci. (2014). Rep. 7, 13545 (2017). 130. Murray, R. H., Erler, D. V. & Eyre, B. D. Nitrous oxide fluxes in estuarine 103. He, Q. & Silliman, B. R. Consumer control as a common driver of coastal environments: response to global change. Glob. Change Biol. 21, 3219–3245 vegetation worldwide. Ecol. Monogr. 86, 278–294 (2016). (2015). 104. Liu, S. L. et al. Sediment microbes mediate the impact of nutrient loading on 131. Kroeger, K. D., Crooks, S., Moseman-Valtierra, S. & Tang, J. W. blue carbon sequestration by mixed seagrass meadows. Sci. Total Environ. 599, Restoring tides to reduce methane emissions in impounded wetlands: a 1479–1484 (2017). new and potent Blue Carbon climate change intervention. Sci. Rep. 7, 11914 105. Gacia, E. & Duarte, C. M. Sediment retention by a mediterranean Posidonia (2017). oceanica meadow: the balance between deposition and resuspension. Estuar. 132. Rosentreter, J. A., Maher, D. T., Erler, D. V., Murray, R. H. & Eyre, B. D. CH4 Coast Shelf Sci. 52, 505–514 (2001). emissions partially offset ‘Blue Carbon’ burial in mangroves. Sci. Adv. 4, 106. Hansen, J. C. R. & Reidenbach, M. A. Wave and tidally driven flows in eelgrass eaao4985 (2018). beds and their effect on sediment suspension. Mar. Ecol. Prog. Ser. 448, 133. Erler, D. V. et al. Applying cavity ring-down spectroscopy for the 271–287 (2012). measurement of dissolved nitrous oxide concentrations and bulk nitrogen 107. Wilkie, L., O’Hare, M. T., Davidson, I., Dudley, B. & Paterson, D. M. Particle isotopic composition in aquatic systems: correcting for interferences and field trapping and retention by Zostera noltii: a flume and field study. Aquat. Bot. application. Limnol. Oceanogr. Methods 13, 391–401 (2015). 102, 15–22 (2012). 134. Bianchi, T. S. et al. Historical reconstruction of mangrove expansion in the 108. Horstman, E. M., Dohmen-Janssen, C. M., Narra, P. M. F., van den Berg, N. J. Gulf of Mexico: Linking climate change with carbon sequestration in coastal F. & Siemerink, M. Hulscher SJMH. Wave attenuation in mangroves: a wetlands. Estuar. Coast Shelf Sci. 119, 7–16 (2013). quantitative approach to field observations. Coast. Eng. 94, 47–62 (2014). 135. Barbier, E. B. et al. The value of estuarine and coastal ecosystem services. Ecol. 109. Keil, R. G. & Hedges, J. I. Sorption of organic matter to mineral surfaces and Monogr. 81, 169–193 (2011). the preservation of organic matter in coastal marine sediments. Chem. Geol. 136. Barange, M. et al. The cost of reducing the North Atlantic Ocean biological 107, 385–388 (1993). carbon pump. Front. Mar. Sci. 3, 290 (2017). 110. Burdige, D. J. Preservation of organic matter in marine sediments: controls, 137. van den Bergh, J. & Botzen, W. J. W. Monetary valuation of the social cost of mechanisms, and an imbalance in sediment organic carbon budgets? Chem. CO2 emissions: a critical survey. Ecol. Econ. 114, 33–46 (2015). Rev. 107, 467–485 (2007). 138. Hiraishi, T. et al. 2013 Supplement to the 2006 IPCC Guidelines for National 111. Ricart, A. M. et al. Variability of sedimentary organic carbon in patchy Greenhouse Gas Inventories: Wetlands. (IPCC, Switzerland, 2014). seagrass landscapes. Mar. Pollut. Bull. 100, 476–482 (2015). 139. Deverel, S., et al. Restoration of California Deltaic and Coastal Wetlands — 112. Oreska, M. P. J., McGlathery, K. J. & Porter, J. H. Seagrass blue carbon spatial American Carbon Registry. 1–186 (2017). patterns at the meadow-scale. PLoS One 12, e0176630 (2017). 140. Scholz, I. & Schmidt, L. Reducing Emissions from Deforestation and 113. Abdolahpour, M., Ghisalberti, M., McMahon, K. & Lavery, P. S. The impact of Forest Degradation in Developing Countries: Meeting the Main Challenges flexibility on flow, turbulence, and vertical mixing in coastal canopies. Limnol. Ahead. Briefing Paper 6. (Deutsches Institut für Entwicklungspolitik, German Oceanogr. 63, 2777–2792 (2018). Development Institute, Bonn 2008). 114. van Katwijk, M. M., Bos, A. R., Hermus, D. C. R. & Suykerbuyk, W. Sediment 141. Murdiyarso, D. et al. The potential of Indonesian mangrove forests for global modification by seagrass beds: muddification and sandification induced by climate change mitigation. Nat. Clim. Change 5, 1089–1092 (2015). plant cover and environmental conditions. Estuar. Coast. Shelf Sci. 89, 142. Narayan, S. et al. The effectiveness, costs and coastal protection benefits of 175–181 (2010). natural and nature-based defences. PLoS One 11, e0154735 (2016). NATURE COMMUNICATIONS | (2019) 10:3998 | https://doi.org/10.1038/s41467-019-11693-w | www.nature.com/naturecommunications 11 PERSPECTIVE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-11693-w 143. Tamooh, F. et al. Below-ground root yield and distribution in natural and was supported by Ramon y Cajal Fellowship RYC2014-14970, co-funded by the Con- replanted mangrove forests at Gazi bay, Kenya. Ecol. Manag. 256, 1290–1297 selleria d’Innovació, Recerca i Turisme of the Balearic Government and the Spanish (2008). Ministry of Economy, Industry and Competitiveness. The University of Dundee is a 144. Howe, A. J., Rodrigues, J. F. & Saco, P. M. Surface evolution and carbon registered Scottish charity, no. 015096. J.P.M. was supported by the Smithsonian Insti- sequestration in disturbed and undisturbed wetland soils of the Hunter tution and the National Science Foundation Long-Term Research in Environmental estuary, southeast Australia. Estuar. Coast. Shelf Sci. 84, 75–83 (2009). Biology Program (DEB-0950080, DEB-1457100, DEB-1557009). 145. Luisetti, T. et al. Coastal and marine ecosystem services valuation for policy and management: Managed realignment case studies in England. Ocean Coast. Author contributions Manag. 54, 212–224 (2011). P.I.M., A.A., J.A.R. and C.M.D. designed the study. P.I.M., A.A., J.A.R., N.B., R.M.C., D. 146. Bolam, S. G. & Whomersley, P. Development of macrofaunal communities on A.F., J.J.K., H.K., T.K., P.S.L., C.E.L., D.A.S., E.T.A., T.B.A., J.B., T.S.B., G.L.C., B.D.E., J. dredged material used for mudflat enhancement: a comparison of three W.F., J.M.H.-S., M.H., I.E.H., D.K.-J., D.L., T.L., N.M., P.M., K.J.M., P.J.M., D.M., B.D.R., beneficial use schemes after one year. Mar. Pollut. Bull. 50, 40–47 (2005). R.S., O.S., B.R.S., K.W. and C.M.D. contributed to the writing and editing of the 147. Plan Vivo. The Plan Vivo standard for community payments for ecosystem manuscript. services programmes. Available at http://www.planvivo.org/wp-content/ uploads/Plan-Vivo-Stan dard-2013.pdf.(2013). 148. Wylie, L., Sutton-Grier, A. E. & Moore, A. Keys to successful blue carbon Additional information projects: lessons learned from global case studies. Mar. Policy 65, 76–84 Supplementary Information accompanies this paper at https://doi.org/10.1038/s41467- (2016). 019-11693-w. 149. Kelleway, J. et al. Technical Review of Opportunities for Including Blue Carbon in the Australian Government’s Emissions Reduction Fund. (Department of the Competing interests: The authors declare no competing interests. Environment and Energy, Canberra, 2017). 150. Ware J. R., Smith S.V., Reaka-Kudla M.L. Coral reefs: sources or sinks of Reprints and permission information is available online at http://npg.nature.com/ atmospheric CO2? Coral Reefs 11, 127–130 (1992) reprintsandpermissions/ 151. Smith S.V., Mackenzie F.T. The Role of CaCO3 Reactions in the Contemporary Oceanic CO2 Cycle. Aquatic Geochemistry 22, 153–175 (2016) Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Acknowledgements P.I.M. and C.E.L. were supported by an Australian Research Council Linkage Project (LP160100242). C.M.D. was supported by baseline funding from King Abdullah Uni- Open Access This article is licensed under a Creative Commons versity of Science and Technology. T.K. and K.W. were supported by JSPS KAKENHI Attribution 4.0 International License, which permits use, sharing, (18H04156) and the Environment Research and Technology Development Fund (S-14) adaptation, distribution and reproduction in any medium or format, as long as you give of the Ministry of the Environment, Japan. B.D.E. was supported by Australian Research appropriate credit to the original author(s) and the source, provide a link to the Creative Council grants DP160100248 and LP150100519. D.A.S. was supported by the UK Nat- Commons license, and indicate if changes were made. The images or other third party ural Environment Research Council (NE/K008439/1), and D.K.J. was supported by the CARMA project (8021-00222B), funded by the Independent Research Fund Denmark. material in this article are included in the article’s Creative Commons license, unless Funding was provided to P.M. by the Generalitat de Catalunya (MERS, 2017SGR 1588) indicated otherwise in a credit line to the material. If material is not included in the and an Australian Research Council LIEF Project (LE170100219). This work is con- article’s Creative Commons license and your intended use is not permitted by statutory tributing to the ICTA ‘Unit of Excellence’ (MinECo, MDM2015-0552). O.S. was sup- regulation or exceeds the permitted use, you will need to obtain permission directly from ported by an ARC DECRA (DE170101524). N.M. was supported by the Spanish Ministry the copyright holder. To view a copy of this license, visit http://creativecommons.org/ of Economy, Industry and Competitiveness (MedShift project). N.B. was supported by licenses/by/4.0/. the UK Research Councils under Natural Environment Research Council award NE/ N013573/1. J.W.F. was supported by the US National Science Foundation through the © The Author(s) 2019 Florida Coastal Everglades Long-Term Ecological Research program under Grant No. DEB-1237517. R.S. had the support of FCT, project FCT UID/MAR/00350/2018. I.E.H. Peter I. Macreadie 1, Andrea Anton 2, John A. Raven3,4,5, Nicola Beaumont6, Rod M. Connolly 7, Daniel A. Friess8, Jeffrey J. Kelleway 9, Hilary Kennedy10, Tomohiro Kuwae11, Paul S. Lavery 12, Catherine E. Lovelock 13, Dan A. Smale 14, Eugenia T. Apostolaki 15, Trisha B. Atwood16, Jeff Baldock 17, Thomas S. Bianchi18, Gail L. Chmura 19, Bradley D. Eyre 20, James W. Fourqurean 5,21, Jason M. Hall-Spencer 22,23, Mark Huxham24, Iris E. Hendriks 25, Dorte Krause-Jensen 26,27, Dan Laffoley 28, Tiziana Luisetti 29, Núria Marbà 25, Pere Masque 12,30,31, Karen J. McGlathery32, J. Patrick Megonigal 33, Daniel Murdiyarso34,35, Bayden D. Russell 36, Rui Santos 37, Oscar Serrano 12, Brian R. Silliman38, Kenta Watanabe11 & Carlos M. Duarte 2 1 Deakin University, School of Life and Environmental Sciences, Center for Integrative Ecology, Geelong, VIC 3125, Australia. 2King Abdullah University of Science and Technology, Red Sea Research Center and Computational Bioscience Research Center, Thuwal, Saudi Arabia. 3Division of Plant Sciences, University of Dundee at the James Hutton Institute, Invergowrie, Dundee DD2 5DQ, UK. 4Climate Change Cluster, University of Technology Sydney, Ultimo, NSW 2007, Australia. 5School of Biological Science, University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia. 6Plymouth Marine Laboratory, Prospect Place, Plymouth PL1 3DH, UK. 7Australian Rivers Institute—Coast & Estuaries, School of Environment and Science, Griffith University, Gold Coast, QLD 4222, Australia. 8Department of Geography, National University of Singapore, 1 Arts Link, Singapore 117570, Singapore. 9School of Earth, Atmospheric and Life Sciences, University of Wollongong, Wollongong, NSW 2522, Australia. 10School of Ocean Sciences, Bangor University, Menai bridge, Bangor LL59 5AB, UK. 11Coastal and Estuarine Environment Research Group, Port and Airport Research Institute, 3-1-1 Nagase, Yokosuka 239-0826, Japan. 12School of Science, Centre for Marine Ecosystems Research, Edith Cowan University, 270 Joondalup Drive, Joondalup, WA 6027, Australia. 13School of Biological Sciences, The University of Queensland, St Lucia, QLD 4072, Australia. 14Marine Biological Association of the United Kingdom, Citadel Hill, Plymouth PL1 2PB, UK. 15Institute of Oceanography, 12 NATURE COMMUNICATIONS | (2019) 10:3998 | https://doi.org/10.1038/s41467-019-11693-w | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-11693-w PERSPECTIVE Hellenic Centre for Marine Research, PO Box 2214, 71003 Heraklion, Crete, Greece. 16Department of Watershed Sciences and Ecology Center, Utah State University, Logan, UT 84322-5210, USA. 17CSIRO Agriculture and Food, Private Mail Bag, Glen Osmond, SA 5064, Australia. 18Department of Geological Sciences, University of Florida, Gainesville, FL 32611-2120, USA. 19Department of Geography, McGill University, 805 Sherbrooke St W, Montreal, QC H3A 0B9, Canada. 20Centre for Coastal Biogeochemistry, School of Environment, Science and Engineering, Southern Cross University, Lismore, NSW 2480, Australia. 21Department of Biological Sciences and Center for Coastal Oceans Research, Florida International University, 11200 SW8th St, Miami, FL 33199, USA. 22School of Biological and Marine Sciences, University of Plymouth, Plymouth, UK. 23Shimoda Marine Research Center, University of Tsukuba, Tsukuba, Japan. 24School of Applied Sciences, Edinburgh Napier University, Edinburgh EH11 4BN, UK. 25Global Change Research Group, IMEDEA (CSIC-UIB), Institut Mediterrani d’Estudis Avançats, Miquel Marquès 21, Esporles 07190, Spain. 26 Department of Bioscience, Aarhus University, Vejlsøvej 25, Silkeborg 8600, Denmark. 27Arctic Research Centre, Department of Bioscience, Aarhus University, Ny Munkegade 114, bldg. 1540, Århus C 8000, Denmark. 28World Commission on Protected Areas, IUCN, Gland, Switzerland. 29 Centre for Environment, Fisheries, and Aquaculture Science, Lowestoft, UK. 30The Oceans Institute and Department of Physics, The University of Western Australia, 35 Stirling Highway, Crawley, WA, Australia. 31Departament de Física & Institut de Ciència i Tecnologia Ambientals, Universitat Autònoma de Barcelona, Bellaterra 08193, Spain. 32Department of Environmental Sciences, University of Virginia, Charlotttesville, VA 22903, USA. 33 Smithsonian Environmental Research Center, 647 Contees Wharf Road, Edgewater, MD 21037, USA. 34Center for International Forestry Research (CIFOR), Jl. CIFOR, Situgede, Bogor 16115, Indonesia. 35Department of Geophysics and Meteorology, Bogor Agricultural University, Kampus Darmaga, Bogor 16680, Indonesia. 36Swire Institute of Marine Science, School of Biological Sciences, University of Hong Kong, Hong Kong SAR, China. 37Center of Marine Sciences, CCMAR, University of Algarve, Faro 8005-139, Portugal. 38Nicholas School of the Environment, Duke University, 135 Duke Marine Lab Road, Beaufort, NC 28516, USA NATURE COMMUNICATIONS | (2019) 10:3998 | https://doi.org/10.1038/s41467-019-11693-w | www.nature.com/naturecommunications 13