Biodiversity credits need a unit. Habitat hectares, species richness, condition scores, eDNA indices — the metrics on the table, and what each one misses.
A carbon credit rests on a unit everyone already agrees on: a tonne of CO₂-equivalent, the same in Tasmania as in Trondheim. Biodiversity has no such coin. A hectare of kelp forest and a hectare of muddy seabed, a reef quiet with urchin barrens and the same reef roaring with fish, a species-return that's priceless and one that's commonplace — markets need equivalence, and nature stubbornly declines to provide it. Every biodiversity credit scheme in the world, including the ones we set out in our overview of biodiversity credits, has to answer one question first: what is the unit? This piece tours the metrics actually on the table — what each measures, what each misses, and why the answer in good schemes is almost never "one of them".
Before the candidates, the criteria. A biodiversity-credit metric has to do four things that pull against each other:
No single metric hits all four. Every scheme's unit is a trade negotiated between these demands — and the position of the trade is the scheme's design philosophy.
The floor of the market: a unit of area — hectares of habitat created, restored, protected or improved. It is the cheapest, most legible, most comparable unit available, and it is the backbone of regulatory offset schemes on land.
Its virtues are real. Area is remotely sensible, contractible, surveyable at vessel-day prices, and honest about what the buyer is paying for: footprint, not outcome. Its failures are the reason the whole debate exists. A hectare is not an outcome: a hectare of replanted kelp that fails twice then takes, a hectare of urchin barren, and a hectare of old-growth forest are the same unit and not the same nature at all. Area measures effort, not biodiversity, and a market built on area alone eventually prices effort — leaving genuine outcomes to fend for themselves.
The first refinement: multiply area by condition. A scoring rubric — typically a panel of indicators like canopy cover, understorey diversity, grazer loads, structural complexity — translates a reef's ecological state into a 0–1 or 0–100 score. Unit becomes hectare × condition: a hectare-gained, weighted by how good it is.
This is where much of terrestrial biodiversity-offsetting landed, and marine net-gain proposals — the wave summarised in our explainer on marine biodiversity net gain — follow the same grammar. The gains are accuracy without losing comparability: improvement is visible, a partially degraded reef is priced between barren and pristine, and condition rubrics can absorb expert judgement. The costs are judgement's vulnerabilities: rubrics encode the rubric-writer's idea of a good reef; scores blur the line between what was measured and what was assumed; and condition moves year to year in a way area does not — selling "an improved hectare" for a decade means verifying the condition each of the ten years.
Most intuitive of all: count things. Species richness (how many kinds), species abundance (how many individuals), occupancy (which sites hold which species), or focus on named priority species — a sea-star return, a fish biomass. A credit becomes a verified gain in the countable.
Species metrics are meaningful in ways hectares never are, but they are the expensive kind of meaningful. Counts need skilled observers or sensors in the water, they swing with season and survey effort, they are the easiest metric to fake good by surveying where the fish already are, and — the deep problem — species respond to weather. A credit denominated in fish biomass is a credit partly denominated in whether an El Niño year fell inside the crediting period. Species metrics work best as condition indicators within a rubric, or as guardrails — "no credit if priority species X trend is negative" — rather than as the unit itself.
The new entrant: net nature by its genetics. Water and sediment samples, sequenced and matched against reference libraries, yield species lists at a cost-and-coverage no dive team can match; the indices built on them — richness, community composition against a reference, biotic index scores — turn a bottle of seawater into a structured biodiversity readout, as our piece on eDNA surveys of kelp forests explains in the kelp context.
For crediting, eDNA's gifts are scale and comparability: one protocol, many sites, season after season, at a per-point cost field ecology could never reach. Its gaps are that it tells presence, not abundance or age; it inherits every bias of reference libraries; and its readouts are community metrics that drift with weather for reasons as non-climatic as currents shifting a water mass around. Most working designs treat eDNA indices as the consistency layer — a cheap, machine-checkable stream of community state — not the unit.
The honest answer most mature schemes arrive at: stack them. A unit constructed to need several things true at once — area, plus a condition score, plus an eDNA or species-state check, plus a permanence horizon — is harder to game because each component compensates for another's blindness. The price is complexity buyers have to read, and methodologies have to explain — which is where the market's transparency burden shifts: from the search for one perfect metric to the discipline of explaining which bundle one chose, and why.
Marine metrics, far more than terrestrial ones, are measured on someone's tenure. On Indigenous sea country the question "what is the unit of nature?" has been answered, with a longer track record than any market, by the people who hold Law for the reef — and a credit metric imposed over those answers without consent is not a measurement choice but a governance failure. The principle runs parallel to the rules of tenure and consent in blue carbon itself; our piece on benefit sharing and FPIC sets out the ground this must stand on.
How does a bundling look in practice? An illustrative construction of the kind Tidewright reports through Canopy Ledger-style monitoring for a kelp-restoration biodiversity unit — illustrative, not a quote from any standard — might denominate one unit as:
Note what this bundle is: an area unit wearing a condition, molecular and guardrail harness, denominated in maintained condition-years rather than hectares. Most serious marine schemes are converging on some version of this shape.
No. Unlike carbon's tonne of CO₂e, no single metric has converged internationally; schemes define their own units (area-based, condition-weighted, species-based or composite) under their own standards. Comparability between schemes is one of the field's open problems.
Richness is intuitive but treacherous as a unit: it swings with survey effort, weather and season; it's easy to inflate by surveying biodiverse spots; and it can rise for wrong reasons — an urchin barren holds plenty of species. It works much better as a guardrail or indicator than as a coin.
For extent and some condition elements (canopy cover, habitat structure), increasingly well. For what lives in the water column and in the sediment, no — and no credible scheme claims it. Satellites measure the geometry; divers, traps and eDNA measure the life.
Duration and verification: a credit denominated in maintained condition over years, checked by a stack of independent measurements, with issuance suspended when guardrails trip. Single-survey credits are where paper nature breeds.
Corporate nature-disclosure frameworks (TNFD and its kin) act as a demand engine: they force hundreds of companies to state which metrics track their impact on nature. That pressure standardises the measurement layer the credit market is built on — the connective rationale explored in our TNFD-and-the-ocean explainer.
Biodiversity credits promise funding for nature restoration that carbon markets can't reach. Here's how they work, what can go wrong, and where marine ecosystems fit.
Priya Raman ·
Marine biodiversity net gain is moving from land into coastal planning. Here's how BNG works, what it means for developers and councils, and where kelp fits.
Priya Raman ·
The TNFD framework is pushing companies to measure and disclose nature-related risks. Here's what it means for coasts, oceans and marine business.
Priya Raman ·