Seagrass Ecosystem Accounting

1 Outcome

This circular provides comprehensive guidance on compiling ecosystem accounts for seagrass meadows, addressing the distinct challenges of accounting for these critically important yet difficult-to-monitor marine ecosystems. Seagrass meadows represent one of the most valuable coastal ecosystem types, providing essential services including carbon sequestration (blue carbon storage), sediment stabilisation, coastal protection, nutrient cycling, and nursery habitat for commercially important fish species.^[1]

Seagrass ecosystem accounts support decision-making across multiple policy domains. Countries implementing this guidance will be able to measure and value carbon sequestration for blue carbon policy and climate commitments, quantify coastal protection services for disaster risk reduction planning, assess nursery habitat contributions to fisheries management, and track water quality through seagrass condition indicators. These accounts provide the evidence base for marine spatial planning decisions that balance competing uses of coastal waters, for evaluating the effectiveness of seagrass restoration investments, and for monitoring progress toward SDG Target 14.2 on sustainable management and protection of marine and coastal ecosystems.^[2]

Upon implementation, countries will be able to: (a) compile seagrass extent accounts; (b) develop condition accounts using seagrass-specific indicators; (c) quantify ecosystem service flows, with emphasis on carbon sequestration, nursery habitat, and coastal protection; and (d) apply monetary valuation and integrate seagrass accounts within national ocean accounting frameworks.^[3]

The carbon stock and sequestration methods described here connect directly to the parallel blue carbon accounting guidance for mangroves and coastal wetlands in TG-6.2 Mangrove and Coastal Wetland Accounting; compilers working across blue carbon ecosystems should apply Fourqurean et al. (2012) and Howard et al. (2014) for consistent measurement protocols. The remote sensing methods described in TG-4.1 Remote Sensing and Geospatial Data provide the foundation for seagrass extent mapping, while the climate indicators developed in TG-2.8 Climate Change Indicators connect seagrass carbon services to national climate accounting systems.

2 Requirements

2.1 Prerequisite Knowledge

This Circular requires familiarity with:

• TG-0.1 General Introduction to Ocean Accounts—for the conceptual framework, terminology, and key components of Ocean Accounts, including the relationship between environmental and economic accounting frameworks that underpin ecosystem-level analysis.

• TG-0.7 Quality Assurance—for the data quality framework, accuracy assessment, and uncertainty documentation applied throughout the compilation procedure (Section 3.5 Step 8) and to deep-meadow undercoverage in extent accounts (Section 3.1.2).

• TG-2.1 Biophysical Indicators—for the indicator-selection framework underpinning the seagrass condition variables in Section 3.2.

• TG-2.4 Ecosystem Goods and Services—for the general methodology of identifying and measuring ecosystem service flows applied in Section 3.3.

• TG-2.8 Climate Change Indicators—for the climate-account linkages applied to seagrass carbon services in Sections 3.3.2 and 3.4.1.

• TG-3.1 Asset Accounts—for the methodology of physical and monetary asset accounts, including the treatment of ecosystem assets as described in Sections 3.3 and 3.4 of that circular. The extent accounting structure (Section 3.4.1) and condition accounting framework (Section 3.4.2) provide the templates that this circular applies to seagrass ecosystems.

• TG-4.1 Remote Sensing and Geospatial Data—for satellite imagery sources (Section 3.1), ecosystem extent mapping methodologies (Section 3.2), and quality assurance procedures applicable to marine ecosystem extent mapping (Section 3.4). Seagrass extent mapping relies on the optical remote sensing platforms and water column correction techniques described in that circular, supplemented by the acoustic survey methods introduced in Section 3.1.3 below.

Readers should note that TG-1.9 Valuation is NOT a direct prerequisite for seagrass accounting, though countries intending to compile monetary accounts should consult that circular for detailed valuation guidance. This is because seagrass ecosystems may be monitored and accounted for in physical terms without requiring monetary valuation as a first step.^[4]

2.2 Institutional Requirements

Countries implementing seagrass ecosystem accounts shall:

a) Establish coordination mechanisms between the National Statistical Office, marine environmental agencies, fisheries authorities, and relevant research institutions, consistent with governance arrangements described in TG-0.1;

b) Designate technical responsibility for seagrass monitoring to agencies with appropriate marine survey and remote sensing capabilities;

c) Develop data sharing arrangements for integrating field survey data with satellite-derived extent estimates, following the data integration principles in TG-4.1 Section 3.4;

d) Maintain quality-assured databases of seagrass extent, condition, and associated ecosystem service flows, applying the quality framework from TG-0.7; and

e) Participate in international seagrass monitoring networks where available, including the Global Seagrass Monitoring Network.^[5]

2.3 Technical Requirements

Countries shall ensure technical capacity for:

a) Processing and analysing satellite imagery and conducting acoustic or field surveys for seagrass extent mapping, following TG-4.1 Ecosystem Extent and TG-2.1 Biophysical Indicators;

b) Conducting underwater visual surveys or acoustic mapping for ground-truthing and detailed extent assessment;

c) Measuring seagrass condition indicators including shoot density, canopy height, and species composition, following the biophysical indicator selection framework in TG-2.1;

d) Estimating carbon stocks in seagrass biomass and underlying sediments, following Fourqurean et al. (2012) and Howard et al. (2014); parallel guidance for mangroves and salt marshes is in TG-6.2; and

e) Modelling ecosystem service flows using biophysical models appropriate for seagrass ecosystems, consistent with the SEEA EA guidelines on biophysical modelling.^[6]

3 Guidance Material

3.1 Extent Accounting

Ecosystem extent accounts record the area of seagrass meadows at the beginning and end of each accounting period, along with changes in extent during the period.^[7] The general methodology for ecosystem extent accounts is described in TG-3.1 Section 3.4; this section provides seagrass-specific guidance. Seagrass extent mapping presents unique challenges compared to terrestrial ecosystems due to their subtidal location, the optical properties of the water column, and the dynamic nature of seagrass distributions.^[8]

3.1.1 Classification Framework

Seagrass meadows are classified within the IUCN Global Ecosystem Typology as ecosystem functional group M1.1 Seagrass meadows within the Marine Shelf biome (M1); for the GET realm/biome/EFG hierarchy and the national crosswalk obligation, see TG-4.1 Remote Sensing and Geospatial Data Section 3.2.4.^[9] The IUCN GET describes seagrass meadows as:

"Seagrass meadows are important sources of organic matter, much of which is retained by seagrass sediments. Seagrasses are the only subtidal marine flowering plants and underpin the high productivity of these systems."^[10]

The IUCN GET further specifies that seagrass ecosystems "have a higher abundance and diversity of flora and fauna, compared to surrounding unvegetated soft sediments and comparable species richness and abundances to most other marine biogenic habitats." This high biodiversity, combined with the three-dimensional structure that provides shelter and binds sediments, distinguishes seagrass meadows as a distinct functional group warranting separate accounting.^[11]

For national ecosystem accounts, countries may develop more detailed classifications based on:

• Dominant species composition (e.g., Posidonia, Zostera, Thalassia, Halophila meadows)
• Depth zone (intertidal versus subtidal meadows)
• Density class (sparse, moderate, dense)
• Associated habitat (meadows on sand, mud, or mixed substrates)

National classifications should crosswalk to GET M1.1 to ensure international comparability (see TG-4.1 Section 3.2.4).^[12] Detailed crosswalk tables between species assemblages and national classification schemes are expected to emerge as more countries compile seagrass accounts; in the interim, countries are encouraged to document their classification decisions to support future harmonisation.

3.1.2 Mapping Challenges

Seagrass mapping faces several methodological challenges that distinguish it from terrestrial ecosystem mapping:

Subtidal visibility constraints. Unlike terrestrial vegetation, seagrass meadows are submerged and their detectability from optical sensors depends on water depth, turbidity, and water column properties. Light attenuation in the water column limits optical remote sensing to depths of approximately 10-15 metres in clear waters, and significantly less in turbid coastal environments.^[13] Deeper seagrass meadows, which may extend to 40-60 metres in exceptionally clear waters, cannot be detected using standard optical methods. TG-4.1 Section 3.1.3 provides guidance on satellite-derived bathymetry techniques that can assist with depth correction but do not resolve this fundamental limitation.

The IUCN GET specifies that "minimum water depth is determined mainly by wave orbital velocity, tidal exposure and wave energy (i.e. waves disturb seagrass and mobilise sediment), while maximum depth is limited by the vertical diminution of light intensity in the water column."^[14] This depth-light relationship creates a detectability gradient that compilers must account for when interpreting extent estimates.

Spectral confusion. The spectral signature of seagrass can be confused with other benthic features including macroalgae, coral, and unvegetated sediments, particularly in mixed habitats. This challenge is compounded by the overlying water column, which absorbs and scatters light differentially across wavelengths.^[15]

Temporal variability. Seagrass meadows exhibit seasonal variation in biomass and spatial extent, particularly in temperate regions. Mapping must account for this variability through appropriate timing of imagery acquisition and, where feasible, multi-temporal analysis.^[16]

Fragmented distributions. Many seagrass meadows occur as patchy or fragmented distributions rather than continuous stands, requiring higher resolution imagery to accurately delineate boundaries.

Because TG-4.1 focuses primarily on satellite and bathymetric data sources, this circular serves as an introductory GOAP reference for acoustic survey methods for seagrass mapping pending a dedicated acoustic methods circular. Section 3.1.3 below describes these methods in the context of the Tier 3 integrated approach.

3.1.3 Mapping Methods

Countries should adopt tiered approaches to seagrass extent mapping based on available resources and the specific characteristics of their seagrass ecosystems, consistent with the tiered approach framework applied throughout the GOAP Technical Guidance:^[17]

Tier 1: Global data products.

At the most basic level, countries may utilise existing global seagrass distribution datasets as starting points. Key resources include:

• UNEP-WCMC Global Distribution of Seagrasses dataset
• Allen Coral Atlas benthic habitat maps (which include seagrass classes, as referenced in TG-4.1 Section 3.2.2)
• Regional seagrass mapping initiatives

Tier 1 approaches provide indicative extent estimates but typically lack the temporal consistency and national verification required for formal ecosystem accounts.^[18]

Tier 2: Satellite imagery classification.

Moderate resolution optical satellite imagery (Sentinel-2, Landsat) can be used to map seagrass extent in suitable conditions. Recommended approaches include:

• Supervised classification using training data from field surveys
• Object-based image analysis to delineate meadow boundaries
• Water column correction algorithms to compensate for depth effects
• Multi-temporal compositing to reduce cloud contamination and capture seasonal variation

Sentinel-2 imagery is particularly suitable for seagrass mapping due to its 10m resolution in visible bands, coastal aerosol band, and 5-day revisit time.^[19] Platform specifications and data access are described in TG-4.1 Section 3.1.1. However, satellite-derived seagrass maps require ground-truthing with in-situ data.

Tier 3: Integrated multi-source mapping.

The most rigorous approach integrates satellite imagery with additional data sources:

• Aerial photography or UAV imagery for high-resolution mapping of accessible areas
• Acoustic surveys (sidescan sonar, multibeam echosounder) for mapping beyond optical depth limits
• Underwater video transects for habitat verification and species identification
• Diver surveys for detailed condition assessment at representative sites

Acoustic methods are particularly valuable for mapping deeper seagrass beds that are beyond optical detection limits. Sidescan sonar at frequencies in the 100-500 kHz range can detect the acoustic contrast between seagrass and bare sediment at depths exceeding 30 metres.^[20] Because TG-4.1 does not currently address acoustic survey techniques, this circular serves as an introductory GOAP reference for acoustic methods in the context of seagrass extent mapping pending a dedicated acoustic methods circular. Countries employing acoustic surveys should document their protocols and calibration procedures to enable reproducibility across accounting periods.

3.1.4 Recording Extent Changes

The ecosystem extent account records opening stock, additions, reductions, and closing stock of seagrass area, following the structure defined in TG-3.1 Section 3.4.^[21] Changes in seagrass extent may result from the categories summarised in Table 3.1.1 below.

For accounting purposes, reductions attributable to anthropogenic pressures should be classified as degradation rather than natural reduction. The principal anthropogenic pressure categories recognised in the published seagrass-loss literature (Waycott et al. 2009; Dunic et al. 2021) are: eutrophication; sediment loading and turbidity from coastal development; physical disturbance from anchoring, mooring chains, propeller scarring, and dredging; destructive fishing gear (trawls, dredges); coastal engineering and reclamation; marine heatwaves and climate-driven dieback; and disease (wasting disease). Edge E1 pressure accounts should record these categories separately where data permit.^[22] The IUCN GET notes that "in eutrophic waters, high nutrient availability can lead to the overgrowth of seagrasses by epiphytes and shading by algal blooms, leading to ecosystem collapse."^[23] This ecological mechanism provides the conceptual basis for distinguishing anthropogenic degradation from natural variability.

Anthropogenic pressure-driven loss is recorded in the extent account under "Other reductions—degradation" per SEEA EA Table 4.2. Compilers may add optional memorandum sub-rows beneath this line to attribute degradation to specific pressure categories (e.g., eutrophication, physical disturbance, coastal engineering), enabling pressure-specific policy analysis. This convention is demonstrated in the worked example (Section 3.6 Step 1) and applied in Step 3 of the compilation procedure (Section 3.5).

3.2 Condition Assessment

Ecosystem condition accounts complement extent accounts by recording the quality or state of seagrass ecosystems relative to a reference condition.^[24] SEEA EA recommends a three-stage approach to condition accounting: recording condition variables, deriving condition indicators, and optionally aggregating into composite condition indices. The general framework is described in TG-3.1 Section 3.4; this section provides seagrass-specific indicators. The principles for selecting biophysical indicators that are policy-relevant, scientifically defensible, and cost-effective to monitor are addressed in TG-2.1 Biophysical Indicators.

Species identity is treated as a stratification variable rather than a condition indicator. Posidonia oceanica, Thalassia testudinum, Zostera marina, Halophila spp., and Enhalus acoroides differ in morphology, longevity, and carbon density by up to an order of magnitude; reference values and service coefficients should be selected per genus rather than applied across an assemblage as if interchangeable.

3.2.1 Condition Variables

Key condition variables for seagrass ecosystems include:

Structural characteristics:

• Shoot density (shoots per square metre)
• Canopy height (centimetres)
• Percent cover within meadow boundaries
• Above-ground and below-ground biomass

Species composition (stratification, not indicator):

• Dominant species identity (used to select species-appropriate reference values; not scored as a condition indicator)
• Species richness (number of seagrass species present)

Sediment characteristics:

• Sediment grain size distribution
• Sediment accumulation rates
• Sulphide concentrations (elevated levels indicate stress)

Sediment organic carbon stock (typically expressed as percent or tonnes per hectare in the top 1 m) is an asset/stock attribute of the seagrass ecosystem, not a condition variable, and is treated in Section 3.3.2 (climate regulation); compilers should follow Fourqurean et al. (2012) and Howard et al. (2014) for stock measurement protocols, with parallel guidance for mangroves and salt marshes in TG-6.2. Where a sediment-carbon-related condition signal is desired, a process metric (e.g., observed sediment accumulation rate against a reference accumulation rate) should be used instead of the stock itself, to avoid double counting between condition and asset accounts.

Water quality and light availability:

• Light attenuation coefficient (Kd)
• Depth of light penetration (Secchi depth)
• Nutrient concentrations in overlying waters
• Chlorophyll-a concentrations (elevated levels indicate eutrophication pressure)

Associated biota:

• Epiphyte loading on seagrass leaves
• Abundance of mesograzers (amphipods, gastropods that control epiphytes)
• Fish community composition and abundance within meadows^[25]

The IUCN GET emphasises that "mesograzers, such as amphipods and gastropods, play an important role in controlling epiphytic algal growth on seagrass," and that "mutualisms with lucinid molluscs may influence seagrass persistence."^[26] These ecological relationships provide the functional basis for including associated biota as condition indicators, not merely as biodiversity metrics.

Table 3.2.1 presents the recommended minimum set of condition variables for international comparability.

Table 3.2.1: Recommended minimum condition variables for seagrass ecosystem accounts

This minimum set balances practicality with the need for consistent condition reporting. Percent cover and shoot density capture structural condition; canopy height reflects growth vigour; epiphyte load serves as a pressure indicator for eutrophication; and species diversity provides a biodiversity dimension. QA check: condition indicators should fall within the range [0, 1]; values outside this range indicate a direction error or reference-level error. Countries should record reference condition values specific to their dominant seagrass species and environmental settings, following the guidance in Section 3.2.2 below.

3.2.2 Reference Conditions

For condition accounting, variable measurements are compared against reference conditions representing ecosystems with minimal anthropogenic disturbance.^[27] For seagrass meadows, establishing reference conditions is challenging because:

• Few pristine seagrass meadows remain globally
• Natural variability in seagrass condition across environmental gradients (depth, latitude, substrate) complicates definition of a single reference state
• Historical baselines are often unavailable due to lack of long-term monitoring

SEEA EA recommends that reference conditions be based on expert assessment of natural or historical conditions, drawing on the best available scientific evidence.^[28] Compilers should follow the SEEA EA reference-condition hierarchy on a document-and-justify basis: natural reference preferred; historical baseline where natural is not determinable; best observed as fallback; policy target where mandated for regulatory reporting. Sources of reference condition information include:

• Protected marine areas with limited human disturbance
• Historical photographs, records, or traditional ecological knowledge
• Palaeoecological evidence from sediment cores
• Scientific literature on seagrass ecology in relatively pristine systems

3.2.3 Condition Indicators

Condition indicators express the state of each ecosystem characteristic relative to the reference condition, typically as a proportion or index value between 0 (complete degradation) and 1 (reference condition).^[29] The illustrative table below uses the Table 3.2.1 minimum set; the simplified ratio shown is for introductory purposes only.

Note: The percent cover, shoot density, canopy height, and species diversity rows use the simplified ratio of current value to reference value, equivalent to the full normalisation formula (defined in TG-2.1 Biophysical Indicators for Ocean Accounts Section 3.4.1) only when VL = 0. The epiphyte load row is an inverse indicator. Compilers should apply the full formula with site-specific VL values, as demonstrated in the worked example (Section 3.6) and the compilation procedure (Step 4). Where additional indicators are illustrated outside the minimum set (e.g., light at depth, sediment process metrics), they should be flagged as supplementary.

3.2.4 Composite Condition Indices

Where condition indicators are aggregated into a single composite index, compilers may apply one of four documented approaches:

GOAP does not endorse a single index. Worked examples in the published literature include SEQI (Indonesia), POMI and BiPo (W Mediterranean), PREI (France WFD), SQI (UK WFD), GBR MMP Seagrass Condition Index, and Chesapeake Bay SAV scorecard.^[30] Compilers selecting an equal-weight default should specify which structural indicator (shoot density or canopy cover) anchors the composite, document any metric excluded or down-weighted, and defer to regulatory indices where the accounting unit overlaps a regulatory reporting unit. The worked example in Section 3.6 uses the equal-weight default for illustrative purposes.

3.3 Ecosystem Services

Seagrass meadows generate a diverse suite of ecosystem services that can be recorded in ecosystem service flow accounts.^[31] Following the SEEA EA classification, services are organised as provisioning, regulating and maintenance, and cultural services. The ecosystem service identification and measurement framework in TG-2.4 Ecosystem Goods and Services provides the general methodology for quantifying these flows.

3.3.1 Provisioning Services

Biomass provisioning. Seagrass meadows support artisanal fisheries through their role as nursery habitat and foraging grounds. The service is measured as the contribution of seagrass ecosystems to fish catch, typically estimated through:

• Catch data from fishing grounds associated with seagrass meadows
• Bio-economic models linking seagrass extent to fisheries productivity
• Meta-analysis of seagrass-fishery relationships^[32]
• Trophic-based seagrass dependency (TB-SGd): food-web network analysis tracing the proportion of each trophic level's biomass that originates from seagrass as a primary producer, enabling attribution of fish catch to seagrass across multiple trophic levels. TB-SGd is computed as the product of diet-contribution proportions along all prey--predator pathways linking seagrass to the target species (Addamo et al. 2024).^[33] This approach is more ecologically rigorous than simple habitat-area ratios and is especially appropriate where trophic data are available for the focal species assemblage.

It is important to attribute only the ecosystem contribution to catch, excluding the contribution of human inputs (labour, capital, technology). The SEEA EA describes methods for estimating ecosystem contributions to biomass provisioning, including residual value and simulation approaches.^[34] Detailed guidance on fisheries valuation may be found in TG-1.9.

Note that the seagrass-attributable share of fish provision may appear numerically small (Addamo et al. (2024) estimate approximately 1% for a Mediterranean food-web model), yet the entire value of dependent catch is at risk if seagrass disappears. Compilers may record this broader flow at risk as a supplementary memorandum entry in the supply table, distinct from the directly attributed flow, to support policy communication on ecosystem dependency.^[35]

Other provisioning. In some regions, seagrasses themselves are harvested for traditional uses (thatching, fertiliser, crafts), though such uses are typically minor in economic terms.

Raw biomass provision (prospective service). Dead seagrass leaves that accumulate on beaches as banquettes represent a prospective provisioning service, with potential uses as biofuel feedstock, soil amendments, and composite materials.^[36] This service is not currently commercially exploited at scale, so a conservative valuation approach is recommended: the avoided cost of removing and disposing of stranded biomass (estimated at approximately EUR 30--155 per tonne depending on jurisdiction; Di Gennaro 2018; Balata and Tola 2016 cited in Addamo et al. 2024). Where primary cost data are unavailable, benefit transfer from Mediterranean studies is permissible with appropriate context adjustments. Compilers should record this service as prospective, flagging that its biophysical basis (total leaf turnover per hectare per year) requires species-specific estimates and that avoided-cost valuation depends on local waste management costs.

3.3.2 Regulating and Maintenance Services

Global climate regulation (carbon sequestration and storage). Seagrass meadows provide two related but distinct blue carbon services: carbon sequestration (Cseq)—the active annual removal of carbon from the atmosphere and water column, captured through photosynthesis in above- and below-ground biomass—and carbon storage (Cstor)—the long-term stock of organic carbon locked in anaerobic sediments and resistant to rapid decomposition (Addamo et al. 2024; Nellemann et al. 2009). For accounting purposes, Cseq is recorded as an annual service flow (tonnes CO2-eq/ha/yr), while Cstor represents an asset stock attribute of the seagrass ecosystem. This distinction matters because Cstor accumulates over centuries to millennia and can be permanently released if meadows are destroyed—a one-way loss not captured by the annual flow alone.^[37]

Seagrass carbon stocks are measured using the three-pool framework (above-ground biomass, below-ground biomass, sediment organic carbon) following Fourqurean et al. (2012) and Howard et al. (2014). The common blue carbon measurement protocol and IPCC Wetlands Supplement Tier framework are described in TG-6.2 §3.3; this section presents the seagrass-specific parameters within that framework. Approximately 90 percent of seagrass carbon stocks are held in soils, with stocks reaching up to 140 Mg C per hectare in the top 1 m of sediment. The carbon sequestration service is measured as the annual carbon flux into long-term storage, typically in tonnes of CO2-equivalent per hectare per year. Key characteristics of seagrass carbon sequestration include:^[38]

• Mean long-term carbon burial rates of approximately 48-138 g C m-2 yr-1 (equivalent to 1.8-5.1 t CO2-eq/ha/yr), with sediment stocks representing centuries to millennia of accumulation; per-unit-area burial rates exceed those of many terrestrial forests. This is distinct from total net primary production, which may exceed 400 g C m-2 yr-1 but is not equivalent to long-term sequestration.
• Long-term storage in anaerobic sediments where decomposition is inhibited
• Risk of carbon release if meadows are degraded or destroyed^[39]

Species-specific defaults differ by an order of magnitude across e.g. Posidonia (high stock and rate) versus Halophila (lower), and compilers should select species- and site-appropriate values from the cited synthesis literature with citation.

For accounting purposes, the climate regulation service should record the annual sequestration flow (addition to stock) and, where seagrass is being lost, the carbon emissions associated with degradation (reduction in service provision or release of stored carbon). The linkage between seagrass condition and carbon service capacity can be quantified through regression models relating shoot density or percent cover to carbon accumulation rates, providing the biophysical model required by SEEA EA.^[40]

Coastal and sediment stabilisation. Seagrass canopies attenuate wave energy and currents, reducing coastal erosion and protecting shorelines.^[41] Monetary valuation of this service follows the replacement cost and avoided-damage approaches described in Section 3.4.2 and TG-1.9 Safe Usage of Monetary Valuation. The IUCN GET notes that seagrass ecosystems "bind sediments and, at fine scales, dissipate waves and currents," providing the ecological foundation for this service.^[42]

The sediment stabilisation service can be measured through:

• Wave attenuation rates across meadows (percentage wave-height reduction per 100 m of meadow width; see footnote ^[41:1])
• Sediment trapping and retention rates
• Comparison of erosion rates between protected and unprotected coastlines

This service provides direct economic benefits by reducing infrastructure damage and avoiding the need for engineered coastal protection structures.

Water purification and nutrient cycling. Seagrass meadows remove nutrients from the water column through uptake and by promoting denitrification in sediments.^[43] This service helps maintain water quality and can reduce eutrophication impacts on other coastal ecosystems including coral reefs. However, the IUCN GET notes that "seagrass growth can be limited by nitrogen and phosphorous availability, but in eutrophic waters, high nutrient availability can lead to the overgrowth of seagrasses by epiphytes and shading by algal blooms, leading to ecosystem collapse."^[44] Nutrient-removal capacity exhibits non-linear threshold and hysteresis behaviour: where epiphyte cover exceeds the threshold in Table 3.2.1 (indicative default: 10%), compilers should apply a condition-adjusted service flow scaling nutrient-removal capacity by canopy health. Compilers with local monitoring evidence supporting a different threshold should document and cite that evidence. Where eutrophication has caused meadow collapse, the water purification service flow is recorded as zero regardless of threshold.

Nursery population and habitat maintenance. Seagrass meadows provide critical nursery habitat for juveniles of many commercially and ecologically important fish and invertebrate species.^[45] The IUCN GET specifies that "the complex three-dimensional structure of the seagrass provides shelter and cover to juvenile fish and invertebrates," and that seagrass systems "support infauna living amongst their roots, epifauna, and epiflora living on their shoots and leaves, as well as nekton in the water column."^[46]

Species dependent on seagrass nursery habitat include:

• Commercially important fish (snapper, sea bream, mullet species)
• Crustaceans (prawns, blue swimmer crabs)
• Molluscs (scallops, queen conch)
• Megafauna (dugongs, green sea turtles)

Nursery service classification as an intermediate ecosystem service and the double-counting avoidance method follow TG-6.2 §3.5. The seagrass GOAP default differs: where stage-structured production-function data are unavailable, the full fisheries contribution is assigned to biomass provisioning (§3.3.1) rather than split. Where a compiler has stage-structured production-function data, the contribution may be apportioned between nursery habitat and biomass provisioning, with the partitioning method documented, cited, and verified to sum without overlap. The IUCN GET notes that "grazing megafauna, such as dugongs, manatees and turtles, can contribute to patchy seagrass distributions, although they tend to 'garden' rather than deplete seagrass," highlighting the reciprocal relationship between habitat provisioning and herbivore pressure.^[47]

3.3.3 Cultural Services

Recreation and tourism. Seagrass meadows support recreational activities including snorkelling, diving, recreational fishing, and wildlife viewing (particularly for charismatic megafauna such as dugongs, manatees, and sea turtles).^[48] These services can be measured through:

• Visitor numbers to seagrass-associated recreation sites
• Expenditure on seagrass-dependent recreational activities
• Travel cost or revealed preference studies
• Seagrass-density and megafauna-density proxies: where direct visitor survey data are unavailable, seagrass shoot density (as a proxy for dive-site attractiveness) and the density of dependent megafauna (turtles, cetaceans) can be combined to rank recreation opportunity across spatial units as high, medium, or low, with net benefit values (e.g. per-visit consumer surplus) applied by class. Addamo et al. (2024) demonstrate this approach for the Mediterranean using logistic and Poisson regression models to predict visit probability and frequency from landscape and demographic covariates, anchored to a per-visit net benefit value from the literature.^[49] As with fish provision, the full nature-based recreation value associated with the presence of seagrass-dependent megafauna may be at risk if seagrass disappears, even where the direct seagrass attribution is a small fraction of total visit value.

Education and research. Seagrass ecosystems serve as sites for marine education programs and scientific research. While challenging to quantify, these services contribute to knowledge generation and environmental awareness.

3.4 Valuation Methods

Monetary valuation enables aggregation of seagrass ecosystem services and assets for incorporation in extended national accounts and policy analysis.^[50] Monetary valuation follows the exchange-value preference hierarchy defined in TG-1.9 Valuation Principles and TG-3.2 Flows from Environment to Economy; the applicable methods for seagrass services are described in the subsections below.

3.4.1 Valuing Carbon Sequestration

The global climate regulation service from seagrass carbon sequestration can be valued using:

Social cost of carbon (SCC). The SCC represents the economic damage caused by an additional tonne of CO2 emissions, or equivalently, the benefit of avoiding that emission. Countries should use the most current SCC estimates from recognised assessment bodies such as the US EPA Interagency Working Group on the Social Cost of Greenhouse Gases or equivalent national bodies; these estimates are subject to revision and the circular does not specify a fixed range. As of 2024-2026, central estimates from authoritative sources typically exceed USD 100/t CO2.^[51]

Carbon market prices. Where carbon credits from blue carbon projects are traded, market prices provide observable exchange values. However, seagrass is currently underrepresented in carbon markets compared to mangroves and salt marshes, and verified seagrass carbon credits remain rare.^[52]

For accounting purposes, the annual carbon sequestration flow is valued by multiplying physical sequestration (tonnes CO2-eq) by the selected price. Consistency with the carbon price used for other blue carbon ecosystems (mangroves, salt marshes) and terrestrial carbon sinks is essential for coherent national accounts, as described in TG-2.8 Climate Change Indicators.

3.4.2 Replacement Cost Approaches

For services where direct valuation is difficult, replacement cost methods estimate the cost of providing equivalent services through engineered alternatives:^[53]

Coastal protection. Coastal protection services are valued using the replacement cost or avoided damage approach; for the full method including seawall unit cost benchmarks and annualisation procedure, see TG-3.2 Flows from Environment to Economy Section 3.5. For seagrass systems, the seagrass-specific parameter to apply within that method is wave attenuation capacity: seagrass canopies attenuate wave energy at approximately 20--40% per 100 m of meadow width (Ondiviela et al. 2014^[41:2]), with the protected coastline length and attenuation rate modelled and documented in the service flow record.

Water purification. The replacement cost for nutrient removal can be estimated from the cost of wastewater treatment infrastructure required to achieve equivalent nutrient load reduction. Tertiary-treatment unit costs vary by orders of magnitude across jurisdictions; compilers must derive unit costs from national wastewater cost data and document the calculation.

Replacement cost approaches should be applied with caution, as they assume the engineered alternative would actually be constructed in the absence of the natural service, and that the replacement provides genuinely equivalent benefits.^[54]

3.4.3 Fisheries Contribution

Per the GOAP default (Section 3.3.2), the full fisheries contribution attributable to seagrass is recorded under biomass provisioning, valued as the seagrass-attributable share of landed catch (ex-vessel price times physical share). Where a compiler has stage-structured production-function data, the contribution may be apportioned between nursery habitat and biomass provisioning using the methods below, with the partitioning method documented and the two service flows verified to sum to total attributed value without overlap.

Production function methods. Econometric models estimate the statistical relationship between seagrass extent and fishery catch, controlling for fishing effort and other inputs. The estimated coefficient represents the marginal contribution of seagrass to production.^[55]

Resource rent approaches. The resource rent (gross value of catch minus all costs of fishing including normal returns to capital) represents the surplus attributable to the natural resource. This rent can be apportioned between fish stocks and supporting habitat (including seagrass nursery areas) based on ecological evidence or model estimates. See TG-3.1 Section 3.3.2 for the treatment of fish stocks in asset accounts.

Benefit transfer. Where primary valuation is not feasible, values from existing studies may be transferred to the site of interest with appropriate adjustments for context.^[56]

3.4.4 Asset Valuation

The monetary value of seagrass ecosystem assets is estimated as the net present value of expected future ecosystem service flows, following the methodology described in TG-3.1 Section 3.2:^[57]

Asset Value = Sum of (Condition-adjusted Annual Service Value / (1 + r)^t) for t = 0 to T

Where r is the discount rate and T is the asset life. For seagrass assets under effective protection and management, an indefinite asset life may be assumed, simplifying the calculation to:

Asset Value = (Annual Service Value x Condition Index) / r

The condition index is applied once, at the physical service flow stage (Step 3 of the compilation procedure), by treating the literature-derived rate as a reference-condition value and multiplying by the composite condition index to obtain an observed-condition flow. The asset valuation step then uses this condition-adjusted annual service total directly without a further condition multiplier; applying the condition index at both Step 3 and Step 4 would double-count the condition adjustment. Where condition is declining, compilers should project forward service flows using trend data rather than applying the perpetuity formula directly, and document the stable-flow assumption explicitly. The choice of discount rate significantly affects asset values, particularly for services with long time horizons such as carbon storage. Countries should apply discount rates consistent with those used for other natural assets in their national accounts.^[58]

3.5 Compilation Procedure

This section provides step-by-step guidance for compiling seagrass ecosystem accounts, integrating the extent, condition, and services guidance from Sections 3.1-3.4 into an operational workflow.

Step 1: Define the ecosystem accounting area and classify seagrass ecosystem types

Delineate the geographic boundary of the national EEZ or sub-national coastal region to be covered by the account, following the spatial boundary guidance in TG-4.1 Remote Sensing and Geospatial Data Section 3.2.6. Within this area, classify seagrass meadows to the IUCN GET functional group M1.1 and stratify by dominant species (e.g. Posidonia, Thalassia, Zostera, Halophila) as a default rather than where data permit, since species identity drives carbon, productivity, and service parameters. Disaggregate further by depth zone as described in Section 3.1.1. Document the crosswalk between any national classification and the IUCN GET to ensure international comparability.

Step 2: Map seagrass extent using tiered approach

Select the appropriate mapping tier (1, 2, or 3) based on available data and technical capacity (Section 3.1.3). For Tier 1, download UNEP-WCMC global seagrass datasets and clip to the national EAA. For Tier 2, acquire Sentinel-2 or Landsat imagery for the accounting period, apply water column correction algorithms, and classify seagrass using supervised classification with field training data. For Tier 3, integrate satellite imagery with acoustic surveys, UAV imagery, and diver transects. Where acoustic or diver survey data are unavailable for deep seagrass (beyond optical detection limits of ~10-15 m), document the estimated proportion of the EAA likely to contain deep meadows based on bathymetric and habitat suitability data, record the mapped extent as a lower bound with a quality-statement note, and where feasible apply a regional-study-based coverage correction. Validate all maps against ground-truth data and document accuracy using confusion matrices as specified in TG-4.1 Section 3.5.1, applying the TG-0.7 quality framework.

Step 3: Detect changes in extent and populate extent account

Compare opening and closing extent maps for the accounting period using change detection techniques. Classify detected changes as managed or natural expansions and reductions following Section 3.1.4, consulting with national marine agencies to distinguish managed interventions (restoration, dredging) from natural processes (storm damage, recovery). Record extent changes in the standard SEEA EA extent account table format (Table 4.2 of SEEA EA—the ecosystem typology table is Table 4.1), ensuring that additions and reductions sum correctly to the net change in extent. Eutrophication-related decline and other anthropogenic pressure-driven loss should be mapped to the appropriate standard category (typically "Other reductions—degradation") with a supplementary memorandum row showing the pressure sub-component for policy analysis.

Step 4: Measure condition variables and derive indicators

Select a representative sample of seagrass meadows stratified by species type, depth, and pressure exposure. At each site, measure the minimum condition variables specified in Table 3.2.1 (percent cover, shoot density, canopy height, epiphyte load, species diversity) using standardised field protocols. Establish reference condition values from protected sites, historical records, or scientific literature following Section 3.2.2. Condition variables are normalised using the standard formula defined in TG-2.1 Biophysical Indicators for Ocean Accounts Section 3.4.1; tag each variable as standard or inverse in a direction column before applying the formula. Optionally aggregate indicators into a composite condition index using one of the four documented approaches in Section 3.2.4; equal weighting is an acceptable documented default where expert weights are unavailable, with the dominant structural indicator (shoot density or cover) anchoring the composite.

Step 5: Quantify ecosystem service flows in physical units

For carbon sequestration, measure sediment carbon accumulation rates using dated sediment cores (Pb-210 or Cs-137) and combine with extent data to calculate the reference-condition annual carbon sequestration (extent x species-appropriate rate); then multiply by the composite condition index derived in Step 4 to obtain the observed-condition flow in tonnes CO2-eq/yr. This is the convention used in Section 3.4.4 and the worked example—the condition adjustment is applied once, here, so that the Step 7 asset valuation uses the condition-adjusted annual total without a further multiplier. For coastal protection, model wave attenuation across meadow width using hydrodynamic models or empirical attenuation coefficients from the literature, expressed as percentage wave-height reduction per 100 m of meadow width (Ondiviela et al. 2014; footnote ^[41:3]); map the length of coastline protected and document the implied attenuation rate in the service flow record. Monetary valuation should then link the attenuation rate to an avoided-damage or replacement-cost calculation (e.g., cost of a seawall providing equivalent wave-height reduction for the same coastal length), with the unit price and derivation documented per Section 3.4.2. For nursery/biomass-provisioning fisheries service, apply the GOAP default of assigning the full seagrass-attributable fisheries contribution to biomass provisioning (kg landed catch x ex-vessel price share) unless stage-structured production-function data permit partitioning per Section 3.3.2. For each service, record both the physical quantity and the natural spatial unit (hectares, kilometres, tonnes), cross-referencing the supplying ecosystem asset in hectares consistent with the extent account, following the SEEA EA ecosystem services flow account structure (Table 7.1 of SEEA EA). For sites where epiphyte load exceeds the condition threshold (Table 3.2.1 indicative default: 10%), apply a condition-adjusted nutrient-removal service flow per Section 3.3.2; where eutrophication has caused meadow collapse, record the water purification service flow as zero.

Step 6: Apply monetary valuation (optional)

For countries compiling monetary accounts, apply the valuation methods in Section 3.4 to convert physical service flows into monetary values. Use current carbon market prices or social cost of carbon for sequestration services (Section 3.4.1), replacement costs or avoided damage approaches for coastal protection (Section 3.4.2), and the biomass-provisioning default (or documented production function/resource rent partitioning) for fisheries contribution (Section 3.4.3). Ensure consistency with carbon prices used in other blue carbon accounts and with discount rates used in national accounts. Record monetary values in the monetary ecosystem services supply-use table (SEEA EA Table 9.3).

Step 7: Compile asset valuation and degradation accounts

Calculate the monetary value of seagrass ecosystem assets as the net present value of expected future service flows using the condition-adjusted formula in Section 3.4.4. For the opening asset value, use the previous period's closing value; for first-time compilations, calculate the opening NPV as the NPV of current observed service flows (not reference-condition flows), scaled by the condition indicators derived in Step 4, and document as a baseline-year value (no degradation charge arises in the first period unless condition or extent changes within that period). For the closing asset value, recalculate NPV based on revised extent and condition.

Record ecosystem degradation as: Degradation = (Opening asset value) - (Closing asset value at opening-period prices and discount rate). Record separately: (a) revaluations due to price or discount rate changes, and (b) degradation attributable to within-period condition and extent decline. Refer to SEEA EA Chapter 10 (Table 10.5) for the standard asset-account structure showing these components. Where significant losses have occurred, disaggregate degradation by driver (eutrophication, storm damage, dredging) to support targeted policy responses.

Step 8: Document methods and quality assurance

Prepare metadata documentation following ISO 19115 standards, describing data sources, processing methods, accuracy assessments, and assumptions for each account component. Apply the quality assurance framework from TG-0.7 Quality Assurance, including coherence checks between extent, condition, and services accounts. For example, verify that reductions in seagrass extent are reflected in reduced carbon sequestration capacity, and that condition declines are consistent with observed pressures such as eutrophication. Publish accounts with accompanying quality statements that transparently report limitations and fitness-for-purpose assessments.

Annex: SEEA EA table cross-reference for compilation steps

3.6 Worked Example

This worked example demonstrates the compilation of seagrass ecosystem accounts for a hypothetical 12,000-hectare seagrass meadow system spanning temperate and tropical waters. The example follows the extent-condition-services-valuation sequence presented in Section 3 and illustrates the key accounting entries and calculations. All monetary values are illustrative composites and should not be used as benchmarks for specific national contexts.

Setting: The ecosystem accounting area (EAA) is a two-embayment coastal zone of approximately 450 km2. Within this geographic frame, seagrass (M1.1 Seagrass meadows) occupies 12,000 hectares as the opening-year asset stock. The EAA is larger than the seagrass extent; the remaining EAA area is unvegetated soft sediment and mixed benthic habitat. The seagrass stock comprises temperate Posidonia beds (7,500 ha, to 15 m depth) and tropical Thalassia-Halophila meadows (4,500 ha, to 12 m depth), distributed across the two embayments.

Step 1: Extent account (year t to t+1)

Step 2: Condition account

Condition indicators are derived from field survey data using species-specific reference levels from protected reference sites. The variable set matches the Table 3.2.1 minimum set:

Note: For epiphyte load, a lower value indicates better condition. Using the inverse normalisation formula (see TG-2.1 §3.4.1): (50 - 18) / (50 - 5) = 0.71.

Composite condition index (equal-weight default per Section 3.2.4): (0.73 + 0.60 + 0.57 + 0.71 + 0.67) / 5 = 0.66

Step 3: Ecosystem services (annual flows)

Notes:

• The carbon sequestration rate of 3.5 t CO2-eq/ha/yr is a mid-range reference-condition value within the 1.8-5.1 t CO2-eq/ha/yr range given in Section 3.3.2 (Fourqurean et al. 2012 synthesis). Compilers should select species- and site-specific values with citation; 1.8 represents a conservative lower bound and 5.1 an upper bound of the cited range.
• The reference-condition flow (12,000 ha x 3.5 = 42,000 t CO2-eq/yr) is adjusted by the composite condition index at Step 3: 42,000 x 0.66 = 27,720 t CO2-eq/yr. This convention treats the literature rate as a reference-condition value and applies the condition adjustment once, at the physical flow stage. The Step 4 asset valuation uses the condition-adjusted annual service total (USD 11,418,000) without a further condition multiplier, avoiding double-counting. See Section 3.4.4. For periods with significant extent change, compilers may use the mean of opening and closing extent.
• The fisheries entry applies the GOAP default rule (Section 3.3.2): the full seagrass-attributable fisheries contribution is assigned to biomass provisioning, not split between nursery habitat and biomass provisioning. Compilers with stage-structured production-function data may apportion per Section 3.4.3, documenting the partitioning method.
• Monetary values for sediment stabilisation and water filtration are illustrative composites. Published seawall replacement costs (USD 1,000-10,000 per linear metre) and tertiary nutrient-removal unit costs vary by orders of magnitude across jurisdictions; compilers must derive their own unit prices from national engineering or wastewater cost data and document the calculation.

Step 4: Asset valuation

Apply the NPV formula from Section 3.4.4 using the condition-adjusted annual service total from Step 3 (USD 11,418,000). Because the condition index has already been applied at the physical flow stage (Step 3), no further condition multiplier is applied here; doing so would double-count the condition adjustment:

Annual service value (condition-adjusted at Step 3): USD 11,418,000

Perpetuity (indefinite horizon, r = 4%): Asset value = 11,418,000 / 0.04 = USD 285,450,000 (illustrative)

25-year finite horizon: Asset value = 11,418,000 x AF, where AF = [1 - (1 + r)^-T] / r = [1 - (1.04)^-25] / 0.04 = 15.62

Asset value (25-year) = 11,418,000 x 15.62 = approximately USD 178,349,000 (illustrative)

Compilers should recalculate the annuity factor (AF) using their national discount rate. Where condition is declining, compilers should project forward service flows using trend data rather than applying the perpetuity formula directly.

This worked example illustrates the full accounting sequence for seagrass ecosystems. Actual compilations will require country-specific data, species-appropriate reference levels, and primary valuation studies for each service type. The carbon price of USD 80/t CO2 is illustrative and should be replaced with the prevailing compliance market price or social cost of carbon applicable in the compiler's jurisdiction. The eutrophication-related decline is classified under "Other reductions—degradation" per SEEA EA Table 4.2, with a memorandum sub-row attributing the pressure, consistent with the guidance in Sections 3.1.4 and 3.5 Step 3.

4 Acknowledgements

This guidance draws on the conceptual framework and methodological recommendations of the System of Environmental-Economic Accounting—Ecosystem Accounting (SEEA EA) and its supporting technical materials. The IUCN Global Ecosystem Typology provides the ecosystem classification framework. Scientific understanding of seagrass ecosystem services draws on extensive research literature, including foundational works by Costanza et al., Duarte et al., Fourqurean et al., and the global seagrass research community.

Authors: [To be confirmed—USER INPUT NEEDED before publication]

Reviewers: [To be confirmed—USER INPUT NEEDED before publication]

5 References

Addamo, A.M., La Notte, A., Ferrini, S., Grilli, G. (2024). 'Marine ecosystem services of seagrass in physical and monetary terms: The Mediterranean Sea case study'. Ecological Economics 226: 108420. DOI: 10.1016/j.ecolecon.2024.108420

Balata, G., Tola, A. (2016). 'Cost-opportunity analysis of the use of Posidonia oceanica as a source of bio-energy in tourism-oriented territories, The case of Alghero'. Journal of Cleaner Production 172: 4085-4098.

De Boer, W.F. (2007). 'Seagrass-sediment interactions, positive feedbacks and critical thresholds for occurrence: A review'. Hydrobiologia 591: 5-24.

Duarte, C.M., Middelburg, J.J., Caraco, N. (2005). 'Major role of marine vegetation on the oceanic carbon cycle'. Biogeosciences 2: 1-8.

Dunic, J.C., Brown, C.J., Connolly, R.M., Turschwell, M.P., Cote, I.M. (2021). 'Long-term declines and recovery of meadow area across the world's seagrass bioregions'. Global Change Biology 27: 4096-4109.

Fourqurean, J.W., Duarte, C.M., Kennedy, H., et al. (2012). 'Seagrass ecosystems as a globally significant carbon stock'. Nature Geoscience 5: 505-509.

Heck, K.L., Hays, G., Orth, R.J. (2003). 'Critical evaluation of the nursery role hypothesis for seagrass meadows'. Marine Ecology Progress Series 253: 123-136.

Howard, J., Hoyt, S., Isensee, K., Pidgeon, E., Telszewski, M. (eds.) (2014). Coastal Blue Carbon: methods for assessing carbon stocks and emissions factors in mangroves, tidal salt marshes, and seagrass meadows. Arlington, VA: CI, IUCN, IOC-UNESCO.

IPCC (2014). 2013 Supplement to the 2006 IPCC Guidelines for National Greenhouse Gas Inventories: Wetlands. Chapter 4. Geneva: IPCC.

Keith, D.A., Ferrer-Paris, J.R., Nicholson, E., Kingsford, R.T. (eds.) (2020). IUCN Global Ecosystem Typology 2.0: Descriptive profiles for biomes and ecosystem functional groups. Gland, Switzerland: IUCN.

Kenny, A.J., Cato, I., Desprez, M., Fader, G., Schuttenhelm, R.T.E., Side, J. (2003). 'An overview of seabed-mapping technologies in the context of marine habitat classification'. ICES Journal of Marine Science 60(2): 411-418.

Larkum, W.D., Orth, R.J., Duarte, C.M. (eds.) (2006). Seagrasses: Biology, Ecology and Conservation. The Netherlands: Springer.

NCAVES and MAIA (2022). Monetary valuation of ecosystem services and ecosystem assets for ecosystem accounting: Interim Version 1st edition. United Nations Department of Economic and Social Affairs, Statistics Division, New York.

Ondiviela, B., Losada, I.J., Lara, J.L., Maza, M., Galvan, C., Bouma, T.J., van Belzen, J. (2014). 'The role of seagrasses in coastal protection in a changing climate'. Coastal Engineering 87: 158-168.

Orth, R.J., Carruthers, T.J., Dennison, W.C., et al. (2006). 'A global crisis for seagrass ecosystems'. BioScience 56(12): 987-996.

Pendleton, L., Donato, D.C., Murray, B.C., et al. (2012). 'Estimating global "blue carbon" emissions from conversion and degradation of vegetated coastal ecosystems'. PLoS ONE 7(9): e43542.

Reynolds, L.K., Waycott, M., McGlathery, K.J., Orth, R.J. (2016). 'Ecosystem services returned through seagrass restoration'. Restoration Ecology 24(5): 583-588.

United Nations (2021). System of Environmental-Economic Accounting—Ecosystem Accounting (SEEA EA). New York: United Nations.

United Nations (2014). System of Environmental-Economic Accounting 2012—Central Framework. New York: United Nations.

United Nations (2022). Guidelines on Biophysical Modelling for Ecosystem Accounting. New York: United Nations Department of Economic and Social Affairs, Statistics Division.

US EPA Interagency Working Group on the Social Cost of Greenhouse Gases (2023). Technical Support Document: Social Cost of Carbon, Methane, and Nitrous Oxide. Washington, DC: US EPA.

Van der Heide, T., Govers, L.L., de Fouw, J., et al. (2012). 'A three-stage symbiosis forms the foundation of seagrass ecosystems'. Science 336(6087): 1432-1434.

Waycott, M., Duarte, C.M., Carruthers, T.J., et al. (2009). 'Accelerating loss of seagrasses across the globe threatens coastal ecosystems'. PNAS 106(30): 12377-12381.

Cross-Reference Summary