Priority questions for the next decade of blue carbon science

The top ten priority questions, identified and ranked by 28 global experts, highlight a field striving to balance technical rigour with practical implementation to ensure credible and scalable blue carbon solutions (Fig. 1). From an initial pool of 116 submissions, the highest-ranked question (Q1) stands as a cornerstone, articulating the challenge of managing BCEs at scale while sustaining coastal livelihoods. Half of the questions (Q3, Q5, Q6, Q7, Q8) focus on strengthening the precision, comparability and scalability of carbon data, underscoring the need for robust evidence to underpin policy development and market mechanisms. The remaining questions (Q2, Q4, Q9, Q10, together with Q1) address the enabling conditions for effective blue carbon governance and finance, including restoration methods, natural capital accounting, crediting standards and science communication. Notably, most questions were rated as highly policy-relevant (Fig. 1; see Table 1 for definitions of each dimension and tier) and considered achievable within a 3–5-year timeframe at moderate cost (US$500,000–2 million) and research complexity, often requiring specialized techniques.

Q1. How can we manage BCEs while supporting the livelihoods of coastal communities?

BCEs and their coastal stewards are deeply interconnected. Management strategies that incorporate local knowledge enhance effectiveness, ensure sustainability and create mutually beneficial outcomes. Early research in marine conservation and biodiversity highlighted the limitations of top-down approaches and underscored the importance of community engagement for conservation success²⁸. In response, the integration of local (or traditional) ecological knowledge, rooted in generations of direct interaction with BCEs, has been increasingly acknowledged as a means to improve research and management outcomes^28,29. Fiji’s locally managed marine areas are frequently cited as an example of participatory management supporting both conservation objectives and sustainable livelihoods³⁰. Although these arrangements have strengthened community engagement and local governance, recent analyses suggest that they do not necessarily result in clear ecological or socio-economic gains³¹. This highlights the complexity of linking conservation interventions with measurable outcomes and the importance of new resources, such as Including Local Ecological Knowledge (LEK) in Mangrove Conservation & Restoration³², which offers practical guidance and case studies for the ethical integration of local and traditional ecological knowledge into research and project design.

To ensure both the long-term persistence of BCEs and the livelihoods they support, future conservation and management must integrate local and traditional ecological knowledge with scientific research, also known as academic ecological knowledge. This fusion enables the refinement of best practices within a modern context^29,33 (Fig. 2). Sustainable BCE conservation depends on knowledge sharing, capacity building and inclusive approaches that prioritize local needs and participation. Achieving this requires moving away from one-size-fits-all management models (for example, blanket no-take marine protected area) and ensuring that research and funding deliver tangible benefits to local communities rather than external stakeholders.

Conservation efforts should recognize and manage the full range of ecosystem services BCEs provide³⁴, particularly those that directly underpin coastal livelihoods and deliver tangible, recurring benefits. The sustainable use of BCEs will vary across locations, as the relative importance of ecosystem services can vary according to factors such as coastal geomorphology and cultural practices. Transparency in blue carbon projects is essential to ensure that efforts support ecosystem health while promoting more equitable livelihoods³⁵. Poorly designed projects can inadvertently worsen inequities if they fail to account for local socio-economic dynamics. While traditional ecological knowledge-based conservation inherently adopts a holistic approach to managing entire watersheds²⁸, this perspective is not consistently reflected in research-driven management plans or blue carbon project development. Landscape-scale approaches, such as ‘ridge to reef’ or ‘coastal corridors’, offer valuable opportunities to connect ecosystems and land-use categories beyond man-made boundaries.

Investments in blue carbon projects should align with community priorities, available coastal resources and varying government actions to ensure holistic and equitable outcomes³⁴. Priority areas for blue carbon projects should not be solely determined by resource (ecosystem) availability. Even resource-rich regions may face lower conservation success if investments and development plans are inequitable, harmful or fail to integrate local priorities. Sustainability in BCE management and equitable participation in the blue economy rely on understanding and respecting the stewardship practices of local communities and Indigenous peoples, ensuring that conservation efforts are culturally appropriate, mutually beneficial and contribute to long-term environmental and socio-economic resilience^29,36.

Q2. How can we develop affordable, high-quality methods for implementing restoration?

Effective recovery of BCEs requires first addressing the drivers of ecosystem decline, which should always precede any active restoration efforts^37,38. Once the ecological conditions for restoration are in place, targeted active restoration may accelerate BCE recovery. Testing and refining low-cost, effective restoration approaches is essential to increase the likelihood of restoring ecosystem structure and function, including carbon storage.

Despite growing interest, the costs, successes and failures of restoration projects are often underreported^39,40,41. Restoration costs strongly influence project feasibility and scalability and vary widely across countries, with lower costs in the global south (that is, regions with typically lower incomes and research capacity) reflecting reduced labour expenses³⁹. Costs also depend on both methods and scale. While larger projects often achieve lower costs per hectare³⁹, restoration success strongly depends on the recovery of site-specific ecological condition and function, rather than the scale of investment. Among BCEs, mangrove restoration is the least expensive (median: US$9,000 ha⁻¹)^39,42, while seagrasses, tidal marshes and emerging BCEs (for example, macroalgae) can be substantially more costly to restore³⁹.

Mangrove restoration has traditionally relied on planting, but this is now discouraged as a primary strategy and is instead used to support natural regeneration where hydrology, elevation and soil quality are suitable⁴³. Ecological mangrove restoration is now preferred, as it restores tidal flows and biophysical conditions to promote natural regeneration⁴⁴. This approach supports faster biodiversity recovery⁴⁵ and costs are comparable to planting, with planting averaging US$1,191 ha⁻¹ (ref. ³⁹) in the global south, and ecological restoration averaging US$1,388 ha⁻¹ in Indonesia⁴⁵. Final costs vary with seedling price, land preparation, permitting and monitoring effort.

Seagrass active restoration (for example, transplantation or seeding) is typically undertaken when natural recolonization is limited by propagule supply. These methods can accelerate recovery, but success is still highly variable and often constrained by scale and cost^37,46,47. Outcomes depend on method, labour and local environmental conditions³⁹, with larger-scale interventions generally performing better³⁷. Methods effective in one region may fail elsewhere due to differences in species life histories, structural traits and functional roles^48,49. Most methodologies have been developed in wealthier nations, where reproductive strategies (for example, flowering and seed production) are well understood^50,51. Combined with high costs, low success rates (11% in the global south versus higher elsewhere) and inadequate long-term support, knowledge gaps restrict the scaling of seagrass restoration worldwide³⁹.

Tidal marsh restoration is typically achieved through managed realignment, hydrological reconnection or restoring tidal elevation⁵². Costs vary widely with method and context: global syntheses estimate US$9,000–90,000 ha⁻¹, with a mean of ~US$38,000 ha⁻¹ (ref. ³⁹). More recent reviews suggest median costs of ~US$24,000 ha⁻¹, though projects involving major earthworks or land purchase can exceed US$200,000 ha⁻¹ (ref. ⁵³). Outcomes vary substantially, with some sites regaining vegetation and carbon burial rates within 5–10 years⁵⁴, while others take decades depending on sediment supply, tidal range and grazing pressure⁵⁵.

Delivering cost-effective, high-quality BCE restoration depends on local economic and logistical context, the degree of degradation and the suitability of restoration approaches. Costs can be reduced by involving local communities and volunteers, and by developing adaptive techniques using locally available materials and species^56,57. Pre-feasibility assessments that identify drivers of degradation can guide targeted site-specific restoration actions, while building local capacity for monitoring, reporting and verification (MRV) is also essential for long-term success.

Q3. Can we forecast the future GHG balance of BCEs in response to global change?

Forecasting the GHG balance of BCEs under changing conditions is essential for carbon financing and other activities that rely on predicting the permanence of sequestered carbon and continued atmospheric CO₂ removal. This is particularly challenging under anthropogenic pressures such as climate change, which strongly affect GHG budgets. While reference sites can provide implicit forecasts for restoration goals, numerical models are needed to represent key mechanisms and processes across diverse hydrogeomorphic settings (distinct from models that assess the GHG effects of human impacts in Q4). Accurate predictions also require high-quality data on organic carbon accumulation (or net CO₂ exchange), nitrogen cycling⁴⁸ and GHG fluxes⁴⁹ across global biogeographic zones and diverse coastal environmental settings (Q8). Understanding how carbon dynamics respond to biophysical, chemical and environmental drivers (for example, nutrients/sediments, salinity, temperature, sea-level-driven accommodation space), including their relative importance, interactions and timescales, is also crucial as carbon accumulation rates and GHG fluxes are influenced by these factors.

Several reviews have focused on predicting changes in the spatial distribution of BCEs, organic carbon accumulation and stocks under anthropogenic and climate change scenarios^{11,58,59,60,61}. For instance, sea-level rise is projected to drive coastal squeeze in tidal marshes and other coastal wetlands, reducing habitat space and limiting carbon storage and accumulation potential^62,63. This growing mechanistic understanding has improved forecasting of organic carbon accumulation and habitat distribution, providing proxies for evaluating BCE climate mitigation potential⁶⁴ and informing accredited blue carbon methodologies, some of which incorporate risk assessments for stock losses due to factors such as project management, land tenure or extreme weather events⁶⁵.

Advances in mechanistic understanding have enhanced process-based models, which, due to their reliance on well-understood relationships between GHG fluxes, organic carbon accumulation and drivers, are valuable tools for forecasting GHG balance under changing environmental conditions or modifications. Several soil cohort models designed for tidal marshes and mangroves predict organic matter and carbon accumulation, as well as sediment accretion, with some explicitly simulating responses to sea level rise^{66,67,68,69,70}. The next step is expanding these models to predict the GHG balance, incorporating CH₄ and N₂O fluxes, as well as lateral exchanges of dissolved GHGs and carbon across habitats (Q6, Q8)^71,72.

Unlike in terrestrial wetlands, comprehensive process-based models for forecasting emissions under changing environmental conditions remain underdeveloped in BCEs. Some progress exists, such as the PEPRMT-tidal model for marsh ecosystems⁷³, which predicts CO₂ and CH₄ emissions and carbon accumulation by coupling with the cohort marsh equilibrium model⁶⁹, and the denitrification–decomposition model, which simulates carbon and nitrogen dynamics in mangroves⁷⁴ and tidal marshes⁷⁵, providing estimates of biomass, soil carbon and GHG fluxes. However, models that incorporate lateral carbon exchange in all BCEs (Q6) or forecast GHG budgets beyond soil and biomass stock changes for submerged aquatic vegetation (for example, seagrasses and macroalgae) still need to be developed. Application beyond classical BCEs is also critical to pace science advancements in additional coastal wetlands⁷⁶. Lateral carbon transport is important for the GHG balance, particularly in habitats with submerged aquatic vegetation, but the impact of climate-change-induced changes in ocean hydraulic structure (for example, currents and stratification) and air–water gas re-equilibrium on GHG balance remains unknown⁷⁷.

Q4. How can we improve estimates of human pressures and management on carbon cycling of BCEs?

Human activities in BCEs generally increase net GHG emissions that contribute to radiative forcing⁷⁸ (Fig. 3). These emissions create both challenges and opportunities; while degradation accelerates emissions, protecting and restoring these ecosystems can support climate mitigation and deliver multiple co-benefits. Realizing these benefits requires robust data collection and analysis to quantify baseline conditions and project GHG reductions. Limited access to data, analytical tools and measurement technologies at the necessary spatial and temporal scales remains a major barrier to advancing blue carbon science and applications.

Despite important advancements over the past decade, critical knowledge gaps persist in assessing the effectiveness of BCEs for GHG mitigation. Soil and biomass carbon pools are among the best-constrained parameters, having been synthesized across multiple spatial scales and incorporated into coherent databases^79,80,81,82. However, coverage remains uneven, particularly for seagrasses and BCEs in the global south^79,83. Existing maps for upscaling point data to broader seascape units are relatively reliable for mangroves, but remain less developed or unavailable for other BCEs, limiting verification of their effectiveness in carbon accumulation and refinement of global estimates. Current datasets are biased towards intact ecosystems, with limited information on how land use and land-use change (LULUC) and forestry influence carbon dynamics, particularly following disturbance or restoration.

A major challenge in blue carbon inventories is the scarcity of data on carbon and GHG fluxes (Q8), particularly CH₄ and N₂O, which have high global warming potential and introduce the largest uncertainty into estimates of LULUC and forestry effects on radiative forcing⁸⁴. Syntheses of chamber-based and eddy flux data have improved organic carbon budgets at continental and global scales^85,86, yet remain insufficient for quantifying emissions at small-project scales where field measurements are not feasible.

While continued research on the mechanisms driving soil carbon accumulation is essential⁸⁷ (Q3, Q8, Q9), priority should also be given to developing flux-relevant proxies for these processes, including local sea-level rise⁸⁸, geomorphic setting⁸¹, vegetation structure and productivity⁸⁹, and suspended sediment⁹⁰. These processes are partially captured in robust numerical models that predict tidal marsh elevation changes in response to sea-level rise⁹⁰, warming⁹¹ and elevated CO₂ (ref. ⁹²). However, existing models mainly apply to tidal marshes and mangroves, without being directly transferable to other BCEs with woody vegetation (that is, tidal freshwater forested wetlands) or coastal plants (that is, seagrasses), limiting their applicability for forecasting LULUC and restoration impacts on radiative forcing across BCEs.

The basic processes governing CH₄ and N₂O emissions are well known, but our ability to model their spatial and temporal variability remains limited. Salinity is a strong predictor of CH₄ emissions at broad spatial scales⁸⁵, but local variations often depend on additional factors, such as distance to tidal creeks, plant traits and microbial community composition^93,94,95. Data on N₂O flux drivers are even more limited, though emissions tend to be low in the absence of external nitrate loading⁹⁶. This could change as high-intensity agriculture continues to expand into BCE-adjacent areas, increasing nutrient inputs and potentially altering emission dynamics. Least understood are the processes that govern hydrologic fluxes of organic carbon and GHGs, the consequences of LULUC on these fluxes and the fate of exported compounds in adjacent marine ecosystems.

Quantifying the effects of LULUC on BCEs is scientifically and technically challenging, requiring sustained investment in research. Future efforts should focus on ecosystem features that can be remotely sensed and integrated with data from sensor networks, empirical measurements and scale-appropriate models, to produce high-resolution maps for applications ranging from site-level restoration projects to national inventories.

Progress will be greatly accelerated by improved mapping capability and open-access platforms for trusted data sharing that support MRV (for example, ref. ⁷⁹), particularly where remote-sensing data from managed networks is integrated with flux-relevant proxies and numerical models⁹⁷. Data sharing should allow information to be easy to find, accessible, interoperable across systems and reusable⁹⁸, particularly for underrepresented and emerging BCEs. Addressing these challenges will be essential for fully integrating BCEs into global climate strategies.

Q5. How can we advance natural capital accounting in BCEs to include a more comprehensive range of co-benefits and trade-offs?

Understanding the full range of co-benefits and trade-offs in natural capital accounting for BCEs requires a robust framework that integrates ecosystem dynamics, service valuation and long-term monitoring. The System of Environmental-Economic Accounting (SEEA) is the international standard for quantifying spatial and temporal relationships and dynamics between ecosystem extent, condition, services provided and economic value⁹⁹. SEEA informs economic and environmental policies¹⁰⁰, business accounting¹⁰¹ and multiple global conventions. It typically focuses on individual environmental components such as carbon, water or biodiversity, which can be aggregated from local to national and global natural capital accounts¹⁰⁰. For instance, in ref. ¹⁰² the authors use country-specific social costs of carbon to estimate that BCEs contribute US$190.67 billion per year in global wealth.

To fully capture the co-benefits and trade-offs of BCEs, SEEA frameworks require coordinated assessments, stakeholder engagement, clear institutional mandates and sustained resources for data collection and MRV. A key challenge is transitioning from valuing individual ecosystem services to aggregating them at the ecosystem level, given the complex assessment requirements for doing so¹⁰³. Clear guidance on how to achieve this is needed to harmonize data across countries, alongside strengthened technical capacity in natural capital accounting, ecosystem valuation and sustainable management, particularly in the global south.

Australia recently developed a guide for applying the SEEA framework to BCEs, detailing the methodologies for assessing restoration benefits¹⁰⁴. The guide outlines approaches to measuring and valuing various ecosystem services, including carbon accumulation, water purification, coastal protection and cultural services, with example SEEA-aligned tables for tracking ecosystem changes due to restoration. Its application is demonstrated in the Hunter River estuary in New South Wales and East Trinity Inlet in Queensland. In the former, restoration efforts improved tidal marsh and supratidal forest ecosystems, leading to increased biomass, benefits to fisheries and recreation, and carbon abatement through avoided emissions and enhanced accumulation¹⁰⁵. In the latter, restoration reduced acid sulfate soil impacts, improved water quality, expanded mangrove and tidal marsh areas, and strengthened ecosystem connectivity. Cultural services for the Mandingalbay Yidinji people were also incorporated¹⁰⁶.

Despite this substantial progress, underrepresented and emerging BCEs remain excluded from global frameworks, markets and natural capital accounting¹⁰⁷. Expanding research to verify their effectiveness in delivering a wide range of ecosystem services will be critical for refining natural capital accounts and ensuring that BCE co-benefits and trade-offs are accurately represented. Practical management techniques and frameworks are also needed to facilitate their inclusion in conservation and climate strategies. As financial interest in blue carbon accounting grows (estimated at US$10 billion or more)¹⁰⁸, aligning ecosystem service benefits with funding mechanisms may help support informed, equitable and actionable decisions regarding sustainable development, climate adaptation and BCE conservation.

Q6. Which innovative techniques, analytical tools and new data or proxies may improve the accuracy of blue carbon flux estimates?

Quantifying blue carbon requires an integrated approach that combines remote sensing, in-situ measurements of above- and below-ground biomass and soil organic carbon, and machine learning techniques, ideally encompassing both stores and flows of dissolved and particulate organic and inorganic carbon, as well as the associated gas fluxes. These fluxes occur vertically and laterally, driven by natural biogeochemical processes within coastal ecosystems and their adjacent environments. Although vertical and lateral carbon fluxes can be substantial in some BCEs, their high spatial and temporal variability makes them difficult to quantify (see Q4 for a discussion of these constraints).

The traditional approach to estimating the organic carbon density of BCEs (that is, soil organic carbon, below- and above-ground biomass per unit area) relies on point-based field sampling, sediment coring and laboratory analysis of biomass organic carbon and soil organic carbon, often combined with sediment dating to assess long-term carbon accumulation. While highly accurate, this approach is time-intensive, costly and limited in spatial coverage. To overcome these limitations, integrating remote sensing with in-situ measurements and new machine learning techniques offers a promising, scalable and cost-effective alternative for mapping carbon stocks and fluxes across BCEs^109,110. However, remote sensing alone cannot estimate soil organic carbon accumulation rates, which determine the long-term accumulation of atmospheric carbon. The high spatial and temporal variability of these accumulation rates, even within a single BCE, further constrains large-scale extrapolation^20,83.

A growing suite of in-situ sensors and flux networks enables continuous, multi-scale observation of CO₂, CH₄ and N₂O in BCEs, such as eddy-covariance systems coupled to infrared or laser spectrometers¹¹¹. Eddy-covariance methods from atmospheric science are increasingly adapted to tidal marshes and mangroves, where combining tower fluxes with burial and lateral exchanges can close the net ecosystem carbon balance^112,113. Practical guidance now emphasizes gas-specific method selection, chamber design and deployment frequency, along with quality assurance and control procedures to reduce bias¹¹⁴. For stock and emission-factor work, standardized field protocols remain essential for comparability across mangroves, tidal marshes and seagrasses¹¹⁵. Critically, non-CO₂ gases can alter net climate benefit (for example, seagrass CH₄ can reduce, whereas N₂O dynamics can enhance apparent sinks), so integrated GHG measurement is required¹¹⁶. Recent guidance calls for standardized protocols, transparent uncertainty analysis and long-term distributed observatories to support credible MRV and policy uptake¹¹⁴.

Recent advances in remote sensing, including multispectral, hyperspectral and synthetic aperture radar imagery, generate rich spectral, spatial and multi-temporal data on BCEs¹¹⁷. As these sensors capture complementary structural attributes of BCEs, machine learning approaches that integrate multimodal Earth observations through data fusion models are increasingly important for scaling of carbon stocks and fluxes¹¹⁰. Cloud computing platforms, such as the Google Earth Engine, further support scalable processing of large remote-sensing datasets^110,118, and species-level classification, ensemble-based decision trees and deep learning approaches have proven effective for improving retrieval accuracy and tracking carbon changes over time¹¹⁰. Continued advances in remote sensing and artificial intelligence, alongside collaboration between researchers, policymakers and stakeholders, will be critical for refining global BCE carbon assessments and supporting investment in blue carbon projects¹¹⁰. Ensuring equitable access to these technologies through capacity building and training, particularly in regions with extensive BCEs, is essential for reducing data disparities between the global south and the global north.

Q7. Can we simplify blue carbon crediting, while maintaining appropriate integrity standards?

Although historically rates of BCE loss have declined, total area loss still outpaces restoration and creation efforts^17,119,120. Carbon financing could support conservation, restoration and creation of BCEs, but project uptake remains low due to numerous technical, financial and social barriers^108,121. A major challenge is the complexity of quantifying organic carbon stocks and fluxes, which requires specialized expertise and evidential support. While several methodologies exist to simplify carbon accounting, most remain too complex or costly for widespread community implementation. This raises the question of whether blue carbon crediting methodologies can be further simplified without compromising scientific rigour.

Such simplification may be feasible without compromising project integrity and standards. Existing frameworks, including Verra and Plan Vivo, could streamline monitoring protocols by leveraging wider data availability or adopting tiered verification procedures similar to the Intergovernmental Panel on Climate Change’s (IPCC) tier system. The key challenge is linking the primary drivers of organic carbon accumulation and the magnitude of GHG fluxes, ensuring that underlying assumptions are well supported by empirical data. Developing reliable default values at specific spatial scales requires high-quality datasets that capture diverse geomorphic, hydrological, hydrodynamic and ecological conditions, including species composition and stand structure. For some BCEs, such as mangroves, existing datasets on organic carbon stocks and GHG fluxes provide representative default values for national, regional or species-specific baseline assessments^122,123,124 and support high-quality models to estimate stocks and fluxes under different management scenarios^8,12. However, such models may not fully capture site-specific mechanisms driving blue carbon dynamics. The treatment of allochthonous organic carbon also remains a critical challenge for assessing additionality, underscoring the need for robust observational and experimental approaches to support blue carbon crediting frameworks¹²⁵.

Recent progress includes the synthesis of high-quality datasets and the development of comprehensive databases, such as the Coastal Carbon Library and Atlas⁷⁹ and the EURO-CARBON database¹²⁶. These resources provide baseline reference data and highlight underrepresented BCEs and regions with data deficiencies¹²⁷. However, many regions still lack the capacity to generate the high-quality datasets needed to improve the accuracy and inclusivity of carbon accounting^82,128. Building capacity through global and regional training centres, research hubs and context-appropriate methodologies could help bridge these gaps by enhancing technical expertise, facilitating the development of spatially unbiased models, and establishing robust baseline carbon stock and flux values. These efforts would ultimately support the broader adoption of blue carbon crediting and strengthen the representation of BCEs in global carbon markets.

Q8. Which regions and flux types need priority measurement to improve blue carbon budgets?

Most BCE research has focused on quantifying carbon stocks in soils and biomass^79,80, with comparatively fewer studies addressing fluxes^72,86, despite their importance for understanding net carbon balance. This gap is particularly acute in restored BCEs, where limited comparisons of carbon fluxes with reference sites constrain assessments of restoration additionality (but see ref. ¹²⁹).

Long-term monitoring systems that integrate local- and national-scale data are essential, yet improving carbon flux estimates is constrained by the logistical and financial burden of long-term monitoring, particularly for highly variable ecosystem-scale GHG fluxes. Techniques such as eddy covariance flux towers provide valuable continuous measurements but are expensive and rarely deployed across BCEs. In addition, data on lateral carbon exchange remain sparse, particularly for particulate organic carbon, dissolved oganic and inorganic carbon, and total alkalinity export. These lateral fluxes, especially total alkalinity export, may account for 25–40% of the carbon budgets in mangroves and tidal marshes^72,130. Although carbon fluxes are needed for conservative estimates of carbon uptake and long-term removal, they are more site-specific than carbon stocks and, therefore, poorly represent the global diversity of coastal geomorphic and climate settings where BCEs occur (Fig. 4). Accounting for timescales is also critical, as they affect estimates of organic carbon preservation^21,22,131. Addressing many of these data gaps would benefit from new protocols to estimate lateral carbon fluxes¹³² and sustained investment in global-scale monitoring networks.

The availability and distribution of GHG flux measurements vary substantially across regions and BCEs (Fig. 4). Mangrove and seagrass flux estimates largely rely on episodic measurements (for example, chambers, headspace equilibration, seawater–air exchange), whereas tidal marshes are more frequently monitored with eddy covariance flux systems that provide higher temporal resolution of net ecosystem exchange. However, monitoring efforts are heavily concentrated in the global north, particularly in subtropical and temperate regions, resulting in notable gaps for tropical tidal marshes and seagrasses (Fig. 4), as well as Nordic/Baltic, subarctic and arctic BCEs^133,134,135. Likewise, while soil organic carbon accumulation rates (based on ²¹⁰Pb and/or ¹³⁷Cs) are relatively well-documented in mangroves and seagrasses, data for tropical tidal marshes¹³⁶ and emergent BCEs are largely missing.

Efforts to characterize the ecological and geomorphic drivers of BCE carbon dynamics across contrasting geographies have advanced, yet critical gaps persist⁶⁰. While key environmental drivers of carbon stocks in seagrasses and carbon accumulation in mangroves have been identified across coastal geomorphic settings (for example, river-dominated to carbonate coastlines)^{14,81,137,138,139}, a unified typology spanning multiple BCEs would improve comparability and predictive modelling. Species composition strongly influences seagrass carbon stocks¹⁴, and global patterns in tidal marsh soil organic carbon stocks are emerging¹⁴⁰. In contrast, the role of coastal geomorphic settings on GHG emissions⁸⁶ and lateral carbon exchange remains poorly understood^72,141, with limited observations preventing clear global patterns. Addressing these data gaps requires an internationally coordinated effort to establish long-term observatory networks across diverse climate zones and coastal geomorphic settings, enabling conservative estimates of net ecosystem carbon balance. Such monitoring would also improve assessments of BCE services at scale and anticipate future data needs to ensure robust, accurate, site-specific and globally representative blue carbon assessments that support conservation, restoration and climate mitigation strategies. Improved monitoring also enhances the reliability of regional assessments, which directly inform the upscaling of blue carbon estimates across scales (Q9). An additional way to improve estimates of carbon uptake and long-term removal is to integrate BCEs into national GHG inventories, following the IPCC’s wetlands supplement⁶, as demonstrated in Australia, Costa Rica and the USA.

Q9. How can we enhance the accuracy of upscaling blue carbon estimates across scales?

The development of robust methods for collecting and synthesizing observational data has improved global and national assessments of carbon stocks, GHG fluxes and the distribution of BCEs^{85,122,142,143}. However, with many regions and BCEs remaining data-limited (Q8), effective upscaling techniques are needed to translate available observations into reliable, scalable estimates while accounting for spatial heterogeneity across scales (Q9). One approach is statistical upscaling, whereby blue carbon features are predicted from geological, hydrological and biogeochemical parameters. Ideally, these models account for spatial heterogeneity from local to regional scales, but identifying predictors that remain consistent across multiple scales remains challenging. Scale-independent processes could offer promising insights by making reliable, generalizable predictions from measured data^110,117. Observational-based upscaling benefits from using direct measurements of carbon stocks or process rates, which can then be extrapolated. When blue carbon features exhibit nonlinear relationships with multiple predictors, machine learning may offer higher-precision upscaling⁶⁴. In this case, models require comprehensive datasets with complete blue carbon feature values and multiple predictor variables, which are often unavailable. Standardized archiving of datasets across regions is, therefore, essential⁷⁹.

Another approach involves first estimating the spatial extent of BCEs and then upscaling carbon stocks or fluxes accordingly. Traditionally, large-scale estimation of BCE extent has relied on remote sensing, but challenges remain, particularly for submerged ecosystems such as macroalgal forests and seafloor habitats¹¹⁰. Expanding high-precision observational techniques and acoustic methods can improve the accuracy of remote-sensing-based BCE estimates¹¹⁰ (Q6).

Beyond remote sensing, numerical modelling provides an alternative approach to estimating BCE extent. Such models fall into statistical or mechanistic categories¹⁴⁴. Statistical models use geo-referenced species observations and environmental parameters to define a multivariate space of suitable environmental conditions, which are then used to parameterize species distributions^{143,145,146,147,148}. While useful, these models simplify complex ecological processes and can be difficult to interpret, particularly for detecting change over time (that is, inference). In contrast, mechanistic models incorporate species traits (for example, morphology, physiology, demography) to establish direct links between environmental conditions and species distributions¹⁴⁴. By integrating ecological understanding, mechanistic models offer more robust, long-term and large-scale predictions¹⁴⁹ and are now beginning to emerge for key BCE species^150,151.

Upscaling carbon stocks and fluxes can be achieved by estimating BCE extent (via remote sensing or modelling) and multiplying it by a known measured carbon stock value (for example, carbon content) or process rates (for example, carbon accumulation rates)¹⁵². Alternatively, spatial extent can be coupled with physical models to estimate carbon fluxes linked to BCEs across ocean domains¹⁵³. Distribution maps are available for classical BCEs and (to an extent) macroalgae forests (Table 2), but only local-to-regional data are available for underrepresented or emerging BCEs. Improving spatial coverage across all BCEs (classical and emerging) and updating maps to account for LULUC (Q4) are, therefore, essential¹⁶.

Q10. How can we ensure blue carbon data and communication methods effectively inform climate policy?

The term ‘blue carbon’ was initially introduced as a policy and marketing strategy to promote conservation, restoration and management of coastal vegetated ecosystems based on their carbon storage and climate change mitigation potential¹. Over time, it evolved into a commonly used noun, reflecting its acceptance and integration into scientific and policy discourse. However, communicating carbon stock and flux findings and integrating them into policy remain key challenges.

Policymakers require robust and scientifically credible metrics to inform decision-making¹⁵⁴, yet the complexity and variability of BCEs, along with data gaps and methodology standardization, have made science translation challenging. Blue carbon science would benefit from coordinated international research and standardized monitoring to address data gaps, particularly in the global south, underrepresented regions and emerging BCEs. International guidelines should be updated and expanded to reflect the latest high-quality data and ensure equitable access to data and methodologies. While mangroves, tidal marshes and seagrasses have been included in the IPCC’s 2014 guidelines²⁴, emerging BCEs remain excluded because of insufficient documentation of additionality and permanence of carbon storage (for example, tidal flats, macroalgae forests^4,155). Uncertainty in carbon accumulation rates and reference values further constrain the development of robust blue carbon accounting frameworks. Hence, the IPCC’s guidelines should be regularly updated using new high-quality data generated through standardized protocols across classical and emerging BCEs. Revising the IPCC’s tier values to reflect the vast increase in data over the past decade would further improve inventory reliability, ensuring a more accurate representation of blue carbon in global climate strategies^9,10 and carbon markets¹²¹.

Clear, consistent and concise communication strategies are critical to accurately convey the benefits and limitations of blue carbon to policymakers and the public. Transparent, data-driven messaging builds public trust and reduces misperceptions about the role of BCEs in carbon accumulation and emissions²¹. Equally important is highlighting co-benefits beyond carbon storage, including coastal protection, biodiversity enhancement and nutrient cycling^156,157, which can support the inclusion of BCEs in climate adaptation strategies. Interdisciplinary collaboration is, therefore, needed to integrate co-benefits and socio-economic considerations into blue carbon strategies¹⁵⁸.

Closer collaboration with policymakers would enable blue carbon science to more directly inform policy decisions, international agreements and national climate action plans. Global initiatives, such as the United Nations Decade on Ecosystem Restoration (2021–2030) and the Kunming–Montreal Global Biodiversity Framework, offer opportunities to align blue carbon with international policy agendas¹⁵⁹. However, stronger coordination across conventions like the United Nations Framework Convention on Climate Change, the Convention on Biological Diversity and the Ramsar Convention on Wetlands of International Importance, is needed to enhance policy coherence and implementation¹⁰. Integrating blue carbon accounting into national climate strategies and expanding its role in carbon markets will require concerted efforts to standardize methodologies and improve data accessibility.