Irreplaceable habitats and why net gain cannot compensate

Irreplaceable habitat net gain is a category error, and the statutory scheme says so. Where development affects an irreplaceable habitat, the biodiversity metric does not generate units that can offset the loss, the 10 per cent requirement cannot absorb it, and the biodiversity gain plan has to set out bespoke compensation agreed with the local planning authority instead. The policy position runs the same way: loss or deterioration of irreplaceable habitat should be refused unless a very narrow policy exception applies.

For a planning consultant, the practical consequence is that these sites are an avoidance problem, not an accounting problem. The moment ancient woodland or a veteran tree is confirmed on or adjacent to a site, the question changes from how many units to how the layout keeps away from it.

What counts as irreplaceable

The Biodiversity Gain Requirements (Irreplaceable Habitat) Regulations 2024 list the habitats treated as irreplaceable for the purposes of the biodiversity gain condition. The list includes ancient woodland, blanket bog, limestone pavement, coastal sand dunes, spartina saltmarsh swards, Mediterranean saltmarsh scrub and lowland fen.

National policy is wider in its framing. The National Planning Policy Framework (August 2026 edition) treats irreplaceable habitats as including ancient woodland and ancient or veteran trees, and the Framework's test applies whether or not the biodiversity gain condition is engaged.

Two definitional points come up on almost every site. Ancient woodland means land continuously wooded since at least 1600, which includes ancient semi-natural woodland and plantations on ancient woodland sites. The second is that a veteran tree is identified by its features, not its species or age alone: decay columns, hollowing, water pockets, large dead limbs and bark loss. A 180-year-old oak with extensive deadwood habitat can qualify where a larger, younger tree does not.

Why the metric does not apply

The statutory biodiversity metric works by scoring habitat distinctiveness, condition and strategic significance, then applying multipliers for difficulty of creation, time to target condition and spatial risk. The arithmetic assumes that habitat can in principle be recreated.

Ancient woodland cannot. The soil profile, the mycorrhizal networks, the seed bank and the ground flora assemblage are the product of centuries of continuity, and nothing in a 30-year management plan reproduces them. So the regulations take irreplaceable habitat outside the metric rather than assigning it an impossibly high multiplier.

That has a practical effect on the gain plan. The 10 per cent requirement still applies to the rest of the development site, calculated as usual with the multipliers described in the note on spatial risk multipliers in the biodiversity metric. Alongside it, the plan must identify the irreplaceable habitat affected, describe the adverse effect, and set out the bespoke compensation. A Biodiversity Net Gain Assessment that reports a positive percentage while staying silent on an adjacent ancient woodland is incomplete, and the local planning authority cannot approve the plan on that basis.

Exemptions are a separate question, and the grounds set out in the note on biodiversity net gain exemptions do not exempt a scheme from the policy test on irreplaceable habitat.

Bespoke compensation and what it is worth

Bespoke compensation means a package negotiated with the authority rather than a number of units bought on a register. Typical components are new woodland planting adjacent to the retained habitat, long-term management of the existing habitat including deer and grey squirrel control, removal of conifers from a plantation on an ancient woodland site, soil translocation, buffer creation, and controls on public access and dogs.

Be honest about what each element achieves. Translocating ancient woodland soil has a poor record and should be presented as salvage rather than replacement. New planting adjacent to existing ancient woodland has real value, because it buffers the edge and increases the core area, which is a different claim from replacing what is lost.

Secure it properly. Bespoke compensation that depends on management for decades belongs in a section 106 agreement with a monitoring regime and a default mechanism, not in a condition requiring a scheme to be submitted later. The authority has to be satisfied the compensation will happen before it grants permission.

Buffers and standoff distances

The published standing advice from Natural England and the Forestry Commission recommends a buffer of at least 15 metres between development and ancient woodland, with a larger buffer where the development is substantial or the ground conditions make edge effects worse.

For ancient and veteran trees the recommended buffer is at least 15 times the stem diameter, or five metres beyond the extent of the canopy, whichever is greater. That calculation surprises people: a tree with a one metre diameter stem generates a 15 metre radius standoff before the canopy is considered.

What has to stay out of the buffer is the part applicants get wrong. The buffer is not a landscape strip. It should exclude built form, hard surfacing, drainage runs, soakaways and attenuation features, service trenches, construction compounds, material storage, lighting columns and anything that changes the hydrology or soil chemistry. A Tree Survey to BS 5837:2012 should show root protection areas and the ancient woodland buffer as separate overlays, because they are different things and the larger one governs.

The policy test for loss or deterioration

Deterioration is the word that does the work, and it catches indirect effects. Hydrological change from a new drainage regime, nitrogen deposition from a road, recreational pressure from housing without alternative green space, light spill into a woodland edge, root severance, and the loss of a dark corridor used by bats all count.

An ecological impact assessment therefore has to address indirect effects with the same rigour as direct loss, and the significance framework set out in the note on significance of effects in an ecological impact assessment is the right structure for it.

The wholly exceptional reasons test is a high bar. The Framework gives examples of the kind of case that might qualify, and they are nationally important infrastructure and mineral working where no alternative source exists. Housing on an allocated site is not in that territory, and an application that argues otherwise usually generates a refusal and a costs risk at appeal.

Evidence that establishes status

Status disputes are common, and they are winnable in both directions.

1. Check the Ancient Woodland Inventory through the Defra MAGIC service as the starting point. Treat it as a record rather than a boundary: the original inventory excluded small woods, and the boundaries were digitised at a coarse scale.
2. Review historic Ordnance Survey editions and estate maps for continuity of woodland cover back as far as the record allows, together with the 1600 test.
3. Survey the ground flora for ancient woodland indicator species, with the caveat that indicator assemblages vary regionally.
4. Examine soils, banks, boundary features and coppice stools for physical evidence of long continuity.
5. For trees, record veteran features individually with photographs and a stem diameter at 1.5 metres, rather than relying on a canopy category.

Where the inventory is wrong, say so with the evidence and ask the authority to accept a corrected boundary. Where the inventory is silent and the evidence shows continuity, the honest position is that the woodland is ancient regardless of the inventory, and an assessment that relies on the absence of a mapped designation will not survive scrutiny by the county ecologist.

Designing the question out

The simplest resolution is layout. Pull the development footprint back beyond the buffer at concept stage, put the open space and the sustainable drainage downstream rather than in the woodland edge, route the spine road away, and use the buffer as managed woodland edge rather than as amenity grass with a path through it.

Where a site is so constrained that development inevitably touches the habitat, the honest advice to a client is that the scheme is unlikely to be consented and that the land value assumption needs revisiting. That conversation early is worth more than an ecology budget spent proving the point.

Ecological surveys, the statutory metric and the biodiversity gain plan are work for a competent person: the metric must be completed by someone competent to do so and the gain plan approved by the local planning authority, and protected species work requires a suitably licensed ecologist. An AI-assisted draft is a starting point for that professional rather than a substitute, and for specialist technical reports, qualified professional input may be required.

This is general guidance, not legal advice; check 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.