Biological Condition Measurement

This Circular translates biotic field measurements and remote-sensing products into SEEA EA ecosystem condition indicators (ECT Classes B1—B3 and C1). Upon implementation, compilers will be able to (1) select bio-indicators by ecosystem type, (2) apply standardised in-situ, remote-sensing, and citizen-science protocols, (3) compile condition variable tables conforming to SEEA EA Table 5.2, and (4) apply tiered data approaches in data-poor contexts.

• TG-0.1 — SEEA EA framework.
• TG-0.7 — QA framework.
• TG-3.1 — ecosystem condition account structure (SEEA EA Table 5.2).
• TG-4.4 — citizen-science collection protocols. TG-4.9 governs indicator construction from such data.
• TG-4.8 — canonical for reference-condition framework, indicator selection criteria, variable → indicator → index pathway, and rescaling formulas. TG-4.8 also provides the abiotic envelope for joint attribution analysis.

3.1.1 Defining Biological Condition

The SEEA EA organises ecosystem condition variables into the Ecosystem Condition Typology (ECT), a hierarchical classification of condition characteristics. The ECT groups condition characteristics into three groups and six classes: Group A (abiotic), with classes A1 Physical state and A2 Chemical state; Group B (biotic), with classes B1 Compositional state, B2 Structural state and B3 Functional state; and Group C (landscape), with class C1 Landscape and seascape characteristics. Biological condition variables fall in Group B and, where they describe the biotic seascape, in class C1. Table 3.1.2.1 below summarises the ECT classes within which biological condition variables primarily fall.

Physical and chemical condition variables (ECT Classes A1 and A2) are addressed in TG-4.8. Compilers must classify each biological indicator by its ECT class before entering it in the condition account table.

The four-approach reference-condition framework (historical, reference site, modelled, regulatory target) and the fixed-baseline rule for time-series accounts apply equally to biological and physical condition. See TG-4.8 Physical Condition Measurement §3.1 for the canonical framework. Bio-specific reference sources for individual indicators are listed in §3.2.3.

Boundary with TG-4.8. Boundary-case variables are assigned by ECT class: e.g., turbidity → TG-4.8 (Class A1), and seagrass shoot density (response to turbidity) → TG-4.9 (Class B2).

The SEEA EA indicator-selection criteria (measurability, ecological relevance, sensitivity to change, data availability) and the variable → indicator → index three-level hierarchy are in TG-4.8 §3.2 and apply to biological indicators without modification. Biological selection additionally weights sensitivity to anthropogenic pressures (overfishing, nutrient loading, temperature stress, physical disturbance, invasive species) and interpretability for non-specialist account users. For data-poor contexts, a simpler indicator reliably measured is preferable to a theoretically superior indicator with large sampling uncertainty. Aggregation into composite indices is addressed in TG-3.1 Asset Accounts.

3.2.2 Core Indicators by Ecosystem Type

The following table specifies minimum biological condition indicators by ocean ecosystem type. Indicators are classified by ECT class, assigned a data tier (see §3.4), and linked to their standard measurement protocol.

NIS indicators should reference EU MSFD Descriptor D2 (Non-indigenous species) for harmonisation in European seas.

Sea urchin density (kelp forest / temperate reef) is an ecosystem state-shift indicator — elevated density signals a phase shift to urchin barrens. Compilers should record it as a supplementary condition variable and flag threshold exceedance rather than rescaling it to a 0—1 indicator on the same basis as canopy cover.

3.2.3 Reference Condition Values

For each indicator, compilers must identify a reference condition value (the value expected at optimal biological condition) before rescaling. Reference condition values should be derived using the hierarchy described in §3.1.3. Where an empirical reference is unavailable, the approach used and its limitations should be explicitly documented in the metadata. Global products that may assist in setting reference values include:

• ReefBase historical coral cover data for coral reef reference conditions
• Global Mangrove Watch for mangrove canopy reference values
• CMEMS ocean colour climatologies for phytoplankton biomass baselines
• OBIS species occurrence archives for pre-impact species richness estimates

OBIS historical records carry geographic and taxonomic coverage bias. Compilers should assess record density for the accounting unit and supplement with regional literature or expert elicitation where coverage is sparse.

3.3.1 In-Situ Survey Methods

Transect surveys are the primary method of in-situ biological condition monitoring. Three configurations are standard in ocean environments, summarised in Table 3.3.1.1 below.

Surveys should be replicated (minimum three transects per habitat stratum per site) and sites georeferenced to WGS84 with at least ±10 m accuracy. Survey data should record observer, time, depth, visibility, and sea state as ancillary variables.

BRUVS (Baited Remote Underwater Video Systems): BRUVS provide standardised, non-destructive estimates of reef fish abundance (MaxN metric — the maximum number of individuals of a species visible simultaneously in a frame) and can be deployed in depths and locations inaccessible to diver surveys. BRUVS data are increasingly accepted as Tier 1 data for fish biomass accounts. Compilers should report MaxN per deployment with standard uncertainty metrics. Where BRUVS and UVC transect data are both available, calibration relationships should be developed to enable time-series comparability.

Acoustic methods: Hydroacoustic biomass surveys (scientific echosounders, 38—200 kHz) are the standard for pelagic fish and zooplankton biomass in open-water environments. Passive acoustic monitoring (PAM) using hydrophone arrays provides abundance and distribution indices for cetaceans and acoustic soundscape indicators. Both methods require specialised expertise. NSOs should explore partnership with fisheries research agencies that operate acoustic survey vessels.

eDNA methods: Environmental DNA (eDNA) sampling provides species detection and community characterisation without physical capture of organisms. Two analytical workflows serve different purposes in biological condition accounts:

Biological condition surveys should aim for species-level identification wherever feasible. Where species identification is impractical in the field (e.g., cryptic polychaete taxa, small crustaceans), genus or family-level identification is acceptable, provided this is consistent across time-series measurements and documented in the metadata. Morphospecies categories (e.g., “encrusting coral morphospecies A”) may be used where expertise is limited, but should be flagged as a data quality limitation.

Non-indigenous species (NIS) should be recorded as a dedicated condition indicator. NIS presence/absence (or relative abundance) is a SEEA EA-recognised biotic pressure indicator and corresponds to EU MSFD Descriptor D2. Where a national NIS database exists, cross-reference occurrence records to validate field identifications. NIS data should also be reported to OBIS where feasible.

3.3.3 Remote Sensing Contributions

Remote sensing provides spatial coverage for biological condition monitoring that is not achievable with in-situ surveys alone. Key remote sensing products for biological condition include:

Remote sensing products should be validated against in-situ data before use in accounts. Accuracy assessments should be documented per the QA requirements in TG-0.7 Quality Assurance Principles.

3.3.4 Citizen Science Integration

Citizen science data (collected under protocols governed by TG-4.4 Citizen Science and Community-Based Monitoring) may supply biological condition variables including:

• Reef fish species occurrence and relative abundance (Reef Life Survey, REEF Volunteer Survey)
• Coral condition and bleaching (CoralWatch, Reef Check — trained observer tiers)
• Species occurrence records (iNaturalist, eBird marine species)
• eDNA water samples (community eDNA programmes)

Before integrating citizen science data into condition indicators, compilers should apply the quality control tiers defined in TG-4.4 and assess whether the spatial and temporal coverage is sufficient to represent the accounting unit. Where citizen science data are aggregated with professional survey data, the data source should be recorded as a metadata field for each account row.

3.3.5 Biotic Survey SOP References

The methods above (§3.3.1—3.3.4) describe survey families. Field implementation should follow a published, standardised protocol for the ecosystem × indicator pair being measured. Compilers should cite the SOP applied in account metadata, and document any departure (sample design, gear, replication, taxonomic resolution) against the cited reference. Indicators selected per §3.2 map to the protocols below.

3.3.6 Quality Control Requirements

Apply the QA framework in TG-0.7. Bio-specific requirements:

• Minimum replication: at least three transects or quadrats per habitat stratum per site, with stratified design preferred.
• Detection probability reported for BRUVS (MaxN) and acoustic surveys.
• eDNA contamination controls: field blank results, positive controls, and replication metadata.
• Observer effect: inter-observer reliability assessed for visual survey methods.
• Temporal coverage: annual preferred, biennial accepted where resourcing constrains, and a 3-year rolling average for composite indices to smooth interannual variability.

3.4 Data Requirements and Sources

3.4.1 Tiered Data Source Hierarchy

Most national statistical offices, particularly in small island developing states and lower-income coastal countries, will lack systematic in-situ biological monitoring at full national scale. A tiered data source hierarchy allows compilers to identify the best available data for each indicator and to document the quality limitations of lower tiers:

Where Tier 4 or 5 data are used, the revision log and metadata should record this explicitly and quantify the additional uncertainty introduced.

3.4.2 Variable Data Requirements Table

3.4.3 Key Institutional Data Sources

3.5.2 Integration with Physical Condition (TG-4.8)

Biological condition indicators should be delivered alongside physical condition variables (compiled per TG-4.8) for the same accounting period. Joint reporting enables attribution analysis: for example, a decline in coral cover (biological) co-occurring with elevated Degree Heating Weeks (physical — thermal stress) supports attribution of the biological change to bleaching pressure. Compilers should flag where biological condition indicators are likely responding to changes in physical condition variables observed in the same accounting period.

3.5.3 Handover to Condition Accounts Compilation

The biological condition indicators compiled following this Circular supply the ECT Class B1 and Class B2 rows of the ecosystem condition account, which is compiled in TG-3.1 Asset Accounts Section 3.5.2. Compilers using TG-3.1 should cross-reference the indicator identifiers used in this Circular’s variable table (§3.4.2) with the ECT classification scheme in the account template, to ensure consistent labelling across ecosystem types and accounting periods.