Biodiversity metric errors councils find most often

Most biodiversity metric calculations that come back from a council with questions fail on the same handful of points. Not on ecology, and rarely on arithmetic, because the statutory tool does the sums. They fail because the habitat map does not match the site, the condition scores are not evidenced, or the post-development sheet claims habitat that the layout builds over.

Councils have had since the mandatory regime began in February 2024 for major development, and April 2024 for small sites, to build up their auditing routine. Many now run the submitted tool themselves, change one input and see what happens. That changes what a submission has to withstand.

Why calculations are audited

The biodiversity gain condition is deemed to be attached to most planning permissions in England, and the biodiversity gain plan has to be approved by the local planning authority before development begins. The metric output is the evidence for the 10 per cent figure, so it sits at the centre of a condition the council must discharge and can be held to account for.

There is a second reason. Off-site units and statutory credits cost money, and the difference between a calculation showing a 10.4 per cent gain and one showing 8.9 per cent is a purchase the developer would rather not make. Officers know where the pressure sits.

Habitat misclassification and its effect

The single most common finding. A parcel recorded as "other neutral grassland" that a site visit shows to be modified grassland, or amenity grassland recorded where a species-rich sward exists, changes the baseline units in both directions and undermines confidence in everything else.

The classification has to follow the UK Habitat Classification as the metric requires, at the correct level, with the habitat name used exactly as the tool lists it. Common faults include: recording a habitat at too coarse a level to capture what is there, using a habitat type that does not correspond to the survey evidence, splitting one parcel into several to smooth an inconvenient condition score, and merging parcels of clearly different quality into a single average.

Where the baseline includes hedgerows, watercourses or trees, remember that the metric runs three separate calculations, and each has to show a gain in its own right. A scheme that delivers a surplus in area units and a shortfall in hedgerow units has not met the requirement.

Condition scores without supporting evidence

Condition assessment is the input with the most influence and the least paperwork behind it. Each habitat type has its own condition sheet with defined criteria, and the assessment is a pass or fail against each criterion rather than a judgement call.

What a reviewing officer looks for: a completed condition sheet for every parcel, dated, with the criteria marked individually, the surveyor named, and the survey date within the appropriate season for that habitat type. A grassland condition assessment carried out in January is not an assessment.

The characteristic error runs in one direction. Baseline condition assessed as poor when the evidence supports moderate, and proposed condition assessed as good on habitat that has not been created yet. Both inflate the gain. Where the created habitat's target condition is good, the management plan has to show how it will get there, because the metric applies difficulty and time-to-target multipliers that punish optimism.

An ecologist appointed as the competent person for the assessment should hold the condition sheets on file. Ask for them before submission rather than after the query arrives.

Retention claimed for habitat that will be lost

This is the error with the biggest numerical consequence, and it usually comes from drawing the post-development habitat layer over the wrong base.

Habitat inside the construction footprint is lost, not retained. So is habitat under the site compound, the haul road, the crane oversail area and the level change. Trees retained on the layout still lose the ground beneath their crown to compaction unless the root protection area is fenced and kept clear, and a tree survey prepared to BS 5837:2012 provides the protection area that the habitat plan should respect.

Two checks catch most of it. Overlay the post-development habitat plan on the engineering layout, including drainage, attenuation, services and access, and confirm that no retained parcel sits under an engineered feature. Then overlay it on the construction phase plan, because habitat retained in the final scheme but destroyed during construction has to be entered as lost and recreated, with the temporal multiplier that follows.

Trading and temporal rules ignored

The metric's trading rules stop a developer replacing distinctive habitat with easy habitat. Losses of high and very high distinctiveness habitat must be compensated with the same habitat type, or a type of equal or higher distinctiveness, rather than offset by a larger area of something ordinary. Irreplaceable habitats sit outside the metric entirely and need bespoke compensation agreed with the authority.

Temporal risk is the other rule applied loosely. Habitat that takes 20 years to reach target condition is discounted accordingly, and the multiplier is severe for slow-forming habitats. A calculation showing a gain delivered by woodland reaching good condition quickly should be checked against the tool's own time-to-target values rather than an assumption.

Spatial risk applies to off-site units, with a reduction where compensation sits outside the local planning authority area or the national character area. Off-site units also need a legal mechanism, either a conservation covenant or a section 106 obligation, and registration on the biodiversity gain sites register before they can be relied on.

Areas that do not reconcile with the layout

Arithmetic reconciliation is the check an officer can run without any ecological knowledge, which is why it gets run first.

• The baseline area total should equal the red line area, unless the assessment explains why it does not.
• The post-development area total should equal the same figure.
• Hardstanding, buildings and roads should appear as urban habitat types, at the areas shown on the layout.
• Areas in the metric should match the areas in the landscape strategy and the planting plan.
• Attenuation basins should be classified according to what they will be, not as generic grassland.
• Green roofs and other urban habitat types should match the roof plan areas rather than the building footprint.

Where the numbers do not reconcile, expect every other input to be examined. Where they do, the assessment starts from a position of credibility.

Presenting the calculation so it can be checked

The submission that gets approved first time contains the tool file itself rather than a summary alone. Include the completed metric in its native format, the habitat plans for baseline and post-development at a readable scale with the parcel references used in the tool, the condition assessment sheets, the survey methodology and dates, and a short note explaining any judgement the reviewer might question.

Write that note deliberately. If a parcel was assessed as moderate rather than good on a marginal criterion, say which criterion and why. If the baseline survey was outside the optimal season and a precautionary classification was applied, say so. Reviewers respond well to an assessment that flags its own weak points, and badly to one that hides them.

The biodiversity net gain assessment should also set out the management and monitoring proposals for the 30-year period, because the gain plan will not be approved on the calculation alone. Where the baseline rests on a preliminary ecological appraisal carried out for a different purpose or in an earlier year, say when it was done and whether it was updated.

The statutory metric must be completed by a competent person under the Environment Act 2021 regime, and an AI-assisted draft is a starting point for that ecologist rather than a substitute for their work.

PlanScope drafts the narrative and structure around a biodiversity net gain submission, with the metric outputs and site data flagged for your ecologist to supply. Enter your postcode to see which reports your site needs. Reports should be reviewed before submission, and we do not guarantee planning approval.