Deep Sea and ABNJ Accounting

This Circular is classified as Emerging: methodology is still developing, and readers should anticipate that guidance will evolve as scientific understanding improves, governance frameworks mature, and practical implementation experience accumulates. The BBNJ Agreement creates new governance structures and information requirements that will shape ABNJ accounting in coming years. This Circular should be revised following COP1 (see Section 3.6 for review schedule).

Asset accounts for seabed minerals (described in TG-3.1 Asset Accounts) inform national natural capital budgets. Ecosystem extent accounts for deep-sea protected areas feed MEA indicator reporting against SDG 14.5.1 and GBF Target 3 (see TG-2.10 MEA Indicators). Condition accounts for chemosynthetic ecosystems connect to valuation frameworks in TG-1.9 Valuation. The conceptual framework is established in TG-0.1 General Introduction to Ocean Accounts and methodological foundations of asset accounting in TG-3.1 Asset Accounts.

This Circular requires familiarity with:

• TG-0.1 General Introduction to Ocean Accounts — provides the conceptual framework and key components of Ocean Accounts, including the relationship between accounting boundaries and jurisdictional frameworks essential for understanding the boundary challenges posed by ABNJ.

• TG-3.1 Asset Accounts — for the methodological foundations of physical and monetary asset accounts, ecosystem extent and condition accounts, and the concepts of depletion and degradation that this Circular applies to deep sea and ABNJ contexts.

Readers may also benefit from TG-0.2 Overview of Relevant Statistical Standards for grounding in the international statistical standards context relevant to the UNCLOS and BBNJ legal frameworks discussed here.

UNCLOS establishes the legal framework governing ocean jurisdiction and defines distinct maritime zones with different rights and obligations. For countries that have not ratified UNCLOS, many key provisions are considered to reflect customary international law and remain relevant to accounting boundary determination.

As of 17 January 2026, the BBNJ Agreement has entered into force (UN Treaty Collection). That date is the single point of reference for the Agreement’s operational status throughout this Circular.

Maritime zones under UNCLOS

Territorial Sea extends up to 12 nautical miles from baselines. Coastal States exercise full sovereignty over the territorial sea, including the seabed and subsoil.

Exclusive Economic Zone (EEZ) extends up to 200 nautical miles from baselines. Coastal States have sovereign rights for exploration, exploitation, conservation and management of natural resources, both living and non-living. The SEEA CF notes that “a country’s EEZ may extend up to 200 nautical miles from the country’s normal baselines”.

Continental Shelf may extend beyond 200 nautical miles to the outer edge of the continental margin, subject to limits defined in UNCLOS Article 76. Coastal States exercise sovereign rights for exploring and exploiting natural resources, including sedentary species. The Commission on the Limits of the Continental Shelf (CLCS) reviews submissions from coastal States.

The High Seas comprise all parts of the sea not included in the EEZ, territorial sea, or internal waters of a State, and are open to all States subject to obligations for conservation of living resources.

The Area designates the seabed, ocean floor and subsoil beyond the limits of national jurisdiction, declared “the common heritage of mankind” and administered by the ISA.

Implications for accounting boundaries

The SEEA CF states that “the SEEA Central Framework includes all natural resources, cultivated biological resources and land within a country of reference (including resources within a country’s exclusive economic zone)”. Resources in ABNJ are explicitly excluded from individual country asset boundaries.

Compilers should distinguish three states of extended continental shelf (ECS) claims when delineating accounting boundaries. The distinction is a methodological choice made on prudential grounds, and it does not determine legal rights. UNCLOS Article 77(3) recognises that coastal-State rights exist ab initio and do not depend on a CLCS recommendation, and the accounting treatment recommended here does not contest that legal position. It reflects the principle that assets should be included in national balance sheets only when their extent is sufficiently certain to support reliable measurement and inter-country comparability. Pending final CLCS confirmation, compilers should apply a conservative framing and disclose the uncertainty per SEEA EA para 5.20.

• Approved CLCS recommendations: The ECS area is included within the national accounting boundary in the same way as the EEZ seabed.
• Submitted but pending claims: The ECS area should be excluded from the core national accounting boundary and disclosed only as a memorandum item with a clear note on jurisdictional uncertainty.
• No submission: The area is treated as ABNJ for accounting purposes.

Compilers should seek legal advice on the status of any ECS claim before including the area in compiled accounts.

The BBNJ Agreement establishes new frameworks for conservation and sustainable use of marine biodiversity in ABNJ. Key provisions relevant to ocean accounting are summarised in Table 3.1.1 below.

The development of monitoring frameworks for BBNJ-designated MPAs could provide baseline data for ecosystem extent and condition accounts in ABNJ. This Circular should be updated following COP1.

3.2 Deep Sea Ecosystems

Deep sea ecosystems are classified within the IUCN GET across two biomes: M3 Deep Sea Floors (benthic ecosystems) and M2 Pelagic Ocean Waters (water column ecosystems extending to depth). 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.

M3 Deep sea floors biome

M3.1 Continental and island slopes extend from the shelf break (~250 m) to abyssal basins (4,000 m), characterised by strong depth gradients in pressure, temperature, light and food availability, . Continental slopes within EEZs fall within national accounting boundaries, whilst slopes on the extended continental shelf may also be attributable to coastal States.

M3.2 Submarine canyons function as “dynamic flux routes for resources between continental shelves and ocean basins”. They enhance nutrient transport and support dense communities of cold-water corals and sponges.

M3.3 Abyssal plains extend between 3,000 and 6,000 m and are “covered by thick layers… of fine sediment”. They host high biodiversity despite low productivity. The vast majority of abyssal plains lie in ABNJ.

M3.4 Seamounts, ridges and plateaus: approximately 171,000 seamounts documented worldwide, covering 2.6% of the sea floor, with ridges covering approximately 9.2%. Many seamounts are targets for fishing and potential mining activities.

M3.5 Deepwater biogenic beds are formed by sessile suspension-feeders such as cold-water corals, sponges, and bivalves that create structurally complex habitats. These slow-growing systems are particularly vulnerable to physical disturbance. See TG-6.1 Coral Reef Accounts for cold-water coral characteristics. Energy source and growth dynamics differ between the two systems.

M3.6 Hadal trenches and troughs extend from 6,000 to 11,000 metres depth, comprising 27 disjoint deep-ocean trenches and 20 additional features. Endemism in these systems is high.

M3.7 Chemosynthetic-based ecosystems (CBEs) include hydrothermal vents, cold seeps, and organic falls. These systems derive primary productivity from chemoautotrophy using reduced compounds such as hydrogen sulphide and methane, characterised by “high faunal biomass” but “low diversity and high endemism”. CBEs are of particular interest for bioprospecting. The BBNJ Agreement’s marine genetic resources provisions may alter how chemosynthetic-based ecosystems are valued.

M2 Pelagic ocean waters (deep component)

M2.3 Bathypelagic ocean waters (1,000-4,000 m depth) are characterised by permanent darkness, low temperatures (1-3C), and dependence on organic matter flux from surface waters.

Compilers should treat this section as a target framework for structuring accounts as data become available. It is not operational guidance. The Emerging badge reflects the implementation gap between the account structures described and the data currently available to populate them. The measurement protocols for the variables themselves — how each is observed, at what frequency and to what tolerance — are owned by TG-4.8 Physical Condition Measurement and TG-4.9 Biological Condition Measurement; this Section selects and interprets them for deep sea ecosystems, where almost every protocol is constrained by platform access rather than by method.

“Less than 1% of the seafloor has been investigated biologically”. Key challenges include:

1. Inaccessibility: Most deep sea ecosystems cannot be observed directly except through expensive submersibles and ROVs.
2. Scale: The deep sea floor covers approximately 300 million km², so full mapping is impractical with current technology.
3. Ecosystem delineation: Many deep sea ecosystems grade into one another along environmental gradients, which complicates boundary definition.
4. Temporal variability: Some features, particularly hydrothermal vents, are geologically ephemeral, with active venting occurring on timescales of years to decades.

For guidance on remote sensing and spatial data sources, see TG-4.1 Remote Sensing Data. The Seabed 2030 initiative aims to complete global bathymetric mapping by 2030.

Deep sea data priority assessment

Table 1: Indicative data priorities for deep sea accounting

For deep sea ecosystem types within national jurisdiction (EEZ and extended continental shelf), the extent account structure follows SEEA EA methodology as described in TG-3.1 Asset Accounts. For ABNJ, similar structures could be applied at regional or global scales.

Table 2: Structure of deep sea ecosystem extent account

For hydrothermal vents and other ephemeral features, “natural expansion” and “natural reduction” may be substantial as geological processes create and extinguish active vent sites. Anthropogenic reductions may occur through seabed mining, bottom trawling (on seamounts), or other human activities.

• Gradual decline in venting activity over years to decades (the typical case) should be recorded as natural unmanaged reduction: geologically driven (natural), not the result of management action, and gradual rather than sudden.
• Sudden cessation of a large vent field following a major tectonic or volcanic event within a single accounting period should be classified as catastrophic loss consistent with para. 5.54.

Compilers should document the basis for classification (rate of change, scale of affected field) in account metadata.

Relevant condition variables by ECT class include:

Physical state: hydrostatic pressure, temperature, substrate type and stability, sedimentation rates.

Chemical state: oxygen concentration, pH and carbonate saturation (relevant for cold-water corals), nutrient concentrations, pollutant concentrations.

Compositional state: species richness and endemism, community composition, presence/absence of indicator taxa, microbial diversity (particularly for CBEs).

Structural state: megafaunal biomass, structural complexity (coral framework, vent chimneys), bioturbation indicators.

Functional state: chemosynthetic productivity (for CBEs), organic matter flux and remineralisation rates, carbon sequestration in sediments.

For guidance on selecting condition variables and constructing indicators, see TG-2.1 Biophysical Indicators.

For ECS areas, apply the three-state CLCS treatment described in Section 3.1. Obtain legal advice on the status of any ECS claim before including the area in compiled accounts.

Step 2: Acquire bathymetric and geomorphology data

Obtain the best available bathymetric data. Priority sources:

• National hydrographic surveys (where available)
• GEBCO global compilation
• Multibeam sonar surveys from research expeditions
• Satellite-derived bathymetry for depths <1000m

Use geomorphometric analysis to delineate ecosystem types following IUCN GET categories: identify continental slopes (depth gradient analysis), detect seamounts and ridges (topographic prominence algorithms), delineate abyssal plains (slope <0.1 degrees, depth 3000-6000m), and map submarine canyons. For guidance on geospatial data processing, see TG-4.1 Remote Sensing and Geospatial Data.

Step 3: Compile ecosystem extent baseline

For each ecosystem type, measure opening extent in km². Assign a data quality rating following TG-0.7 Quality Assurance protocols:

• High: Multibeam survey with ground-truthing
• Medium: Satellite-derived bathymetry or predictive modelling validated by samples
• Low: GEBCO compilation or unvalidated predictive models

Populate the opening stock row of the extent account (Table 2 structure).

Where V is the observed value, VH is the reference (high) value, and VL is the degraded (low) value. For inverse variables (e.g., pollutant concentrations, disturbance percentage): (VL - V) / (VL - VH), where the denominator must be evaluated using the actual reference value of VH for the variable in question (which is 0 only for variables whose reference state is zero). See Section 3.7 Table 4 for a worked example with VH explicitly stated.

Aggregate indicators into composite condition indices using equal weighting or expert-weighted aggregation. Document the aggregation method.

Step 7: Populate extent and condition accounts

Enter compiled data into the account structures in Section 3.3. Ensure the closing stock = opening stock + additions - reductions. Document all data sources, measurement dates, quality ratings, and estimation methods in metadata.

Step 8: Link accounts to policy frameworks

Cross-reference compiled accounts to decision use cases:

• For deep-sea mining assessment: highlight extent and condition for areas subject to ISA exploration contracts
• For BBNJ MPA designation: identify ecosystem types with high endemism or low condition scores as priority conservation areas
• For MEA reporting (via TG-2.10 MEA Indicators): extract protected area coverage and condition trends for SDG 14 and GBF indicators

3.5 Ecosystem Services from the Deep Sea

Genetic resources: Deep sea organisms, particularly from extreme environments such as hydrothermal vents, harbour unique genetic resources with potential applications in biotechnology and pharmaceuticals. The BBNJ Agreement establishes frameworks for access and benefit-sharing related to marine genetic resources from ABNJ, which may enable future monetary valuation. The CEPA classification includes “Scientific, cultural and educational services” (CEPA division 15) which may encompass genetic resources.

Fisheries resources: Some commercial fish stocks occur on or near deep sea features, particularly seamounts. High seas fish stocks subject to RFMOs may be attributed based on quota allocations. See TG-6.7 Fisheries Stock Assessment and TG-3.1 Asset Accounts for detailed guidance.

Mineral resources: The Area contains large mineral deposits, including polymetallic nodules, cobalt-rich ferromanganese crusts, and seafloor massive sulphides. UNCLOS defines “polymetallic nodules” as “one of the resources of the Area consisting of any deposit or accretion of nodules, on or just below the surface of the deep seabed, which contain manganese, nickel, cobalt and copper”. The ISA has issued exploration contracts covering approximately 1.5 million km² of seabed, though no commercial exploitation has yet occurred. See TG-3.10 Offshore Energy Accounts for methodological parallels.

Carbon sequestration: Abyssal sediments are a major long-term carbon sink via the biological carbon pump. Deep sea carbon sequestration contributes to global climate regulation but is not currently quantified in ecosystem service accounts. For carbon-related accounting approaches, see TG-6.2 Mangrove and Coastal Wetland Accounting, which sets the blue carbon pricing rule for the publication.

Climate regulation: Deep water formation and thermohaline circulation influence global climate patterns at scales not readily attributed to specific ecosystem assets.

Nutrient cycling: Deep sea ecosystems contribute substantially to global biogeochemical cycles. Remineralisation of organic matter returns nutrients that eventually upwell to support surface productivity.

Challenges in service valuation

1. Spatial attribution: Many deep sea services operate at global scales that cannot be attributed to discrete ecosystem assets. See TG-2.2 Macro Dependencies for macro-scale service flow approaches.
2. Temporal scales: Carbon sequestration benefits may extend over millennia, challenging NPV discounting approaches. See TG-1.9 Valuation for discount rate selection.
3. Non-use values: Much deep sea value may be non-use value (existence, bequest, option values) not captured in market transactions.
4. Irreversibility: Deep sea ecosystems are slow to recover from disturbance, and losses may therefore be irreversible.

Monetary valuation of deep sea ecosystem services should be treated as experimental, with transparent documentation and appropriate caveats.

3.6 Accounting Challenges and Methodological Gaps

1. The commons problem: ABNJ resources are not owned by any country. Whilst activities in ABNJ can be attributed to countries based on flag state, sponsoring state, or quota allocation, the underlying assets remain collective.
2. Migratory species: Highly migratory species transit between EEZs and high seas areas. Current conventions attribute stocks to countries “during the period when those stocks inhabit its EEZ”, creating discontinuous records.
3. Transboundary flows: Ecosystem services from ABNJ (e.g., carbon sequestration, nutrient cycling) benefit all countries but cannot be attributed to specific beneficiaries.

Although no institution holds a global ABNJ accounting mandate, a regional seas convention whose convention area extends past the outer limit of member EEZs already has the three things a compiler needs: a standing accounting area that spans the jurisdictional boundary, a data platform, and a periodic reporting cycle. Where an ABNJ area falls inside such a convention area, compilers should treat the convention rather than any single coastal State as the natural compiling entity, consistent with the “regional sea” jurisdictional basis listed under Experimental ABNJ accounts below.

Data scarcity is the principal challenge for deep sea accounting. Compilers should not attempt to produce deep sea accounts without first addressing data availability. Key gaps:

1. Baseline gaps: For most deep sea ecosystems, no baseline inventory of extent or condition exists.
2. Monitoring gaps: Ongoing monitoring is minimal, so changes over time are difficult to track.
3. Taxonomic gaps: Many deep sea species remain undescribed, and biodiversity assessment is correspondingly limited.
4. Economic data gaps: Limited information on economic activities occurring in ABNJ or their values.

Experimental ABNJ accounts: presentation guidance

This Circular should be reviewed on a 2-year cycle (next review: early 2028), with key triggers for interim updates including the adoption of monitoring frameworks at BBNJ COP, approval of a first commercial seabed mining contract by ISA, and publication of major deep sea mapping advances.

3.7 Worked Example: Deep-Sea Mining Impact Assessment

This worked example demonstrates compilation of deep sea ecosystem accounts for a hypothetical 200,000 km² deep-sea area (depths exceeding 1,000 m) within a coastal State’s extended continental shelf and adjacent ABNJ. Given the Emerging status of this Circular and severe data limitations, compilers should treat the methods and values below as illustrative only.

Setting and decision context

Accounting area: A deep-sea accounting area of 200,000 km² spanning the continental slope and adjacent abyssal plain (bathymetry 1,000-4,500 m), including 12 mapped seamounts (M3.4, total area 3,200 km²), 3 confirmed active hydrothermal vent fields (M3.7, total area 8 km²), and extensive abyssal plain sediments (M3.3). Of the total, 140,000 km² lies within the State’s extended continental shelf claim (approved by CLCS), and 60,000 km² lies in ABNJ. One seamount hosts an ISA exploration contract for polymetallic sulphides.

Decision context: The sponsoring State is evaluating whether to support progression of the ISA exploration contract to the exploitation phase, requiring assessment of: (1) the economic value of mineral resources, (2) the environmental cost of ecosystem degradation from mining, and (3) the opportunity cost of foregone ecosystem services.

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

Extent data compiled from multibeam bathymetry (seamounts, abyssal plains) and ROV surveys (hydrothermal vents). Extent changes reflect discovery of a new vent field (natural expansion), cessation of one known vent (natural reduction confirmed by temperature monitoring), and seabed disturbance from mineral exploration test mining (anthropogenic reduction).

Table 3: Extent account for deep-sea accounting area

Interpretation: The 3 km² abyssal plain reduction represents sediment disturbance from environmental baseline surveys and small-scale test mining under the ISA exploration contract. The 0.5 km² seamount reduction reflects localised disturbance from geological sampling on the seamount hosting the mineral deposit. The +1 km² net change in vent extent reflects discovery of a larger new vent field (2 km²) offset by cessation of a smaller known vent (1 km²). The vent cessation is recorded as natural unmanaged reduction (gradual geological process) consistent with the classification criteria in Section 3.3.

Step 2: Condition account

Condition indicators derived from ROV surveys (structural variables), water column sampling (chemical variables), and ISA exploration contract environmental monitoring data. Reference conditions drawn from undisturbed control sites within the same depth range and biogeographic province. Variables are classified against the SEEA EA Ecosystem Condition Typology (ECT) variable classes as set out in SEEA EA Chapter 5.

Table 4: Condition account for deep-sea accounting area

Calculation notes: For benthic disturbance (inverse variable), VH = 0% and the indicator is: (VL - V) / (VL - VH) = (15 - 0.8) / (15 - 0) = 14.2 / 15 = 0.95. The denominator reflects the actual reference value of VH and is not a default substitution. For variables whose reference value is non-zero, VH must be inserted explicitly. For species discovery rate: (V - VL) / (VH - VL) = (12 - 3) / (18 - 3) = 0.60.

Composite condition index: (0.95 + 0.60 + 0.74 + 0.64) / 4 = 0.73

Interpretation: The composite index of 0.73 indicates relatively good condition (73% of reference condition), consistent with limited human disturbance to date. The species discovery rate indicator (0.60) suggests biodiversity may be less intact than other condition dimensions, potentially due to historical fishing impacts on seamounts. Compilers should note that even small condition reductions may be effectively irreversible on human timescales due to slow deep-sea recovery rates.

Step 3: Ecosystem services (annual flows)

Ecosystem service flows estimated from scientific literature, ISA environmental baseline reports, and expert judgement. Values carry substantial uncertainty and should be interpreted as illustrative orders of magnitude.

Table 5: Ecosystem service flows from deep-sea accounting area

Valuation notes:

Step 4: Asset valuation

Asset values calculated as the net present value of expected future ecosystem service flows using a 4% real discount rate. For deep-sea ecosystems dominated by carbon sequestration, a single 25-year horizon understates the carbon benefit substantially. This example presents two horizons — 25 years (conventional) and 100 years (long-horizon sensitivity). Consult TG-1.9 Valuation on discount-rate and horizon choice.

The carbon sequestration component is held at the midpoint of the market reference range (USD 100/t CO2e × 600,000 t CO2/yr = USD 60,000,000/yr) to isolate horizon sensitivity.

Conservative service flow (carbon sequestration midpoint + scientific value) = USD 63,200,000/yr Including genetic resource potential (midpoint) = USD 63,650,000/yr

Sensitivity: Extending the horizon from 25 to 100 years increases asset value by approximately 57% under either scenario. This reflects the dominant share of long-horizon carbon sequestration in total asset value. The difference between scenarios that include or exclude genetic resource potential (less than 1% at either horizon) is small relative to horizon sensitivity. The choice of valuation horizon, together with the underlying carbon price, is therefore the dominant driver for deep-sea asset values. If scientific value is reclassified as a memorandum item, the A-scenario annual flow drops to USD 60,000,000/yr and the B-scenario to USD 60,450,000/yr, with corresponding reductions in asset value.

Step 5: Decision support analysis

• Extent loss: 85 km² of abyssal plain directly disturbed (0.043% of total area), plus 120 km² indirect disturbance from sediment plume
• Condition decline: Composite condition index projected to decline from 0.73 to 0.52 in mined area, and to 0.68 in indirectly affected area
• Service flow reduction: Annual service losses of USD 61,500 (carbon sequestration in the 205 km² of disturbed sediments: 615 t CO2/yr at 3.0 t CO2/km²/yr and the USD 100 midpoint, an upper bound that assumes burial stops entirely) + USD 240,000 (lost scientific value from altered community composition) = approximately USD 301,500/yr

Ecosystem degradation cost (SEEA EA-compliant method): SEEA EA paras. 6.44-6.48 define ecosystem degradation cost as the change in ecosystem asset value attributable to a decline in condition below a sustainable reference level. The correct measure is the NPV difference in service flows between the mining and no-mining scenarios, not the sum of a proportional asset share and a separately capitalised service loss (which would double-count, because the asset value is itself the NPV of those service flows. See SEEA EA para. 7.14).

Applying this method using the 25-year conservative horizon (Scenario A1):

Degradation cost (NPV of foregone flows, 4%, 25 years) = USD 301,500 × 15.62 ≈ USD 4,700,000

Economic benefit from mining: Gross in-situ ore value of USD 3-4 billion minus extraction costs (estimated 70% of gross value for deep-sea mining) yields net value of USD 900M-1.2B. The ISA’s benefit-sharing regime for exploitation remains unadopted as of the publication date of this Circular. Three illustrative State-share scenarios:

Table 7: Illustrative State benefit from mining under hypothetical ISA benefit-sharing scenarios. No ISA benefit-sharing regime has been finalised, so figures are presented to support sensitivity analysis only.

1. Irreversibility: Deep-sea recovery timescales exceed 100 years, so the 25-year and 100-year degradation cost horizons are lower bounds. Under strict irreversibility, the perpetuity-equivalent degradation cost (USD 301,500/yr ÷ 4%) would be approximately USD 7,500,000 — consistent with the 100-year estimate. See SEEA EA paras. 6.44-6.48 and 7.20-7.22.
2. Option value: Delaying mining preserves future options for genetic resource discovery or alternative uses.
3. Existence value: Non-use values are not captured in service flow valuation.
4. Risk: Sediment plume impacts may exceed projections.

The accounting framework transparently presents trade-offs but does not prescribe the decision.

Step 6: Upward connections to policy frameworks

TG-1.1 National Ocean Budgets: The degradation cost (USD 4.7M at 25-year horizon and USD 7.4M at 100-year horizon) informs natural capital budgeting. If mining proceeds, this cost should be deducted from gross mineral revenues when calculating environmentally adjusted NDP.

TG-2.10 MEA Indicators: Extent accounts feed SDG 14.5.1 (MPA coverage) if the deep-sea area includes protected zones. The 0.8% benthic disturbance metric supports SDG 14.2.1. Condition index trends inform BBNJ monitoring requirements.

TG-1.3 Marine Spatial Management: The spatial distribution of high-condition areas (seamounts, vent fields) versus disturbed areas guides MPA designation and mining exclusion zones. The connectivity index (0.65), classified as a structure ECT variable, informs network design for deep-sea MPAs.