Irreplaceable habitats and why the metric cannot trade them

Irreplaceable habitat sits outside the biodiversity net gain trading system entirely. The statutory metric records it, but it does not generate units you can lose and replace, because the regime's starting assumption is that these habitats cannot be recreated on any timescale that matters. Where a development would affect one, the biodiversity gain plan has to set out bespoke compensation agreed with the local planning authority instead of a metric calculation showing a 10 per cent uplift.

For a consultant scoping a site, this changes the sequence. You establish whether irreplaceable habitat is present before you run the metric, not after.

What the regulations define as irreplaceable habitat

The definition for biodiversity gain purposes is set by regulations made under Schedule 7A to the Town and Country Planning Act 1990, and it is a closed list rather than a judgement call. The habitats specified include ancient woodland, ancient and veteran trees, blanket bog, limestone pavement, coastal sand dunes, lowland fen and spartina saltmarsh swards.

Two features of that list are worth noting. It is narrower than the everyday professional sense of irreplaceable, so species-rich grassland that took a century to develop is not on it. And it is fixed by statutory instrument, which means it can be amended, so check the current regulations rather than relying on a summary written in 2024.

The list also cuts across other designations. An ancient woodland may or may not be a site of special scientific interest, and a veteran tree may or may not be covered by a Tree Preservation Order. The irreplaceable habitat status is independent of both and applies wherever the habitat meets the definition.

Where the classification is contested, the classification is what the case turns on, so resolve it early with survey evidence rather than at committee with assertions.

Why the metric excludes these habitats

The statutory metric works by assigning biodiversity units to habitat parcels according to type, condition, distinctiveness and strategic significance, then comparing pre-development and post-development totals. That arithmetic only works if habitat of a given type can in principle be created elsewhere.

Ancient woodland cannot. Its value lies in centuries of undisturbed soil development, seed banks, deadwood continuity and associated species assemblages. Planting an equivalent area of trees produces woodland, not ancient woodland, and the metric would flatter the outcome if it were allowed to say otherwise.

So the metric treats these parcels separately. They are recorded on the baseline, they are excluded from the trading calculation, and no number of units bought elsewhere compensates for their loss. A Biodiversity Net Gain Assessment that shows a positive percentage while a corner of ancient woodland is being cleared has not answered the question.

That is a design constraint rather than an accounting problem. The layout has to avoid the habitat.

The bespoke compensation requirement

Where loss or deterioration cannot be avoided, the gain plan must set out the steps taken to avoid and minimise the effect and the compensation arrangements agreed with the planning authority for the irreplaceable habitat. There is no formula, no standard ratio and no market from which to buy units.

In practice authorities and their ecology advisers look for compensation that is substantial in scale, secured for a very long period, located close to the affected habitat, and designed to deliver something functionally related to what is lost. Woodland creation adjacent to retained ancient woodland, with buffer planting and long-term management, is a recognisable shape of proposal. A cheque is not.

Two points consultants raise with clients too late. Agreement with the authority is required, so this is a negotiation with a body that has every incentive to say no, and it takes months rather than weeks. And soil translocation, which is sometimes offered as mitigation for ancient woodland, does not turn an irreplaceable habitat into a replaceable one in policy terms.

Ancient woodland, veteran trees and buffers

National policy treats development resulting in the loss or deterioration of irreplaceable habitats as something that should be refused unless there are wholly exceptional reasons and a suitable compensation strategy exists. That test is deliberately severe. Verify the current policy references in the National Planning Policy Framework published on 17 August 2026, because the Framework moved to policy codes in that edition.

The published standing advice on ancient woodland and ancient and veteran trees supplies the working numbers that officers use.

• A buffer of at least 15 metres between development and ancient woodland, larger where the scheme is substantial
• For an ancient or veteran tree, a root protection area based on a multiple of the stem diameter, with a buffer extending beyond the canopy edge
• Consideration of indirect effects, including changes to hydrology, light spill, recreational pressure, dust and air quality

Deterioration matters as much as loss. A scheme that retains every tree but drains the fen beside it, or that puts 40 houses backing onto ancient woodland with no buffer and predictable garden encroachment, causes deterioration and is caught by the same policy test.

Buffers are also where the arithmetic bites commercially. Fifteen metres around an irregular woodland edge removes a surprising amount of developable area, and finding that out after the land is under offer is an expensive way to learn it.

Evidence that a habitat is not irreplaceable

The Ancient Woodland Inventory is the usual starting point, and it is indicative rather than definitive. Parcels are added and removed as evidence improves, and the inventory historically excluded small woodlands below a threshold size, so an unlisted wood is not automatically outside the definition.

Building a case that a parcel falls outside the definition means documentary and field evidence together: historic map regression across successive Ordnance Survey editions and earlier estate maps, the ground flora assemblage, soil profile, coppice stool and boundary bank evidence, and the structure of the woodland edge. Plantation on an ancient woodland site remains ancient woodland, so evidence of conifer planting proves nothing on its own.

For trees, a veteran tree assessment considers girth relative to species, the presence of decay features, cavities, deadwood and epiphytes, rather than age alone. Record the assessment properly, because an officer who disagrees will ask for the method.

This work belongs in the Preliminary Ecological Appraisal stage, before the layout is drawn. Where protected species are also present, a Protected Species Survey runs in parallel on its own seasonal timetable, and the constraints usually overlap on the same parcels.

Policy tests running alongside the gain condition

The biodiversity gain condition and the development plan operate independently. Satisfying the gain condition does not satisfy policy on irreplaceable habitat, and a scheme can hold an approved gain plan and still be refused.

Three other regimes commonly engage on the same sites. The habitats regulations where the parcel forms part of a European site. The Wildlife and Countryside Act where the site is notified as a site of special scientific interest. And local plan policies on ancient woodland buffers, which in some authorities exceed the national standing advice.

Set all of this out as one narrative rather than as separate appendices. Officers reading an ecology chapter that answers the metric but not the policy test will ask for a further submission, and the delay lands in the determination period.

How this changes the site layout

Work in this order. Map the irreplaceable habitat and its buffer first. Take that area out of the developable envelope entirely, including for drainage features, roads, gardens and construction compounds. Then design the scheme in what remains. Then run the metric on the remaining parcels.

Designing first and testing afterwards produces layouts that have to be rebuilt, and rebuilt layouts lose viability and time. Ecology surveys, including any Bat Survey work on veteran trees, should be commissioned to suit that sequence rather than to suit the planning programme.

The statutory metric is expected to be completed by a competent person, and the biodiversity gain plan is approved by the local planning authority. A licensed or competent professional is legally required for this work, and an AI-assisted draft is a starting point for that professional rather than a substitute.

Browse the ecology report types on PlanScope to scope a site with irreplaceable habitat constraints.

This is general guidance rather than legal advice. Check the current regulations and the position with your local planning authority or a qualified professional before relying on it. PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.