GIS, Mapping, and Spatial Data for BNG

Why mapping matters so much for BNG

The DEFRA Statutory Biodiversity Metric 4.0 is a multiplication formula. Area (or length for hedgerows and watercourses) is the base measure — everything else multiplies against it. This means that a small error in how you delineate or measure a habitat parcel compounds through every subsequent factor. A 10% underestimate of a habitat area that scores highly on distinctiveness and condition can translate to a significant undercount of biodiversity units, potentially causing a development to fail its 10% uplift requirement before the survey ecologist has even entered the data.

Beyond the maths, spatial data quality has direct legal consequences. A Biodiversity Gain Plan submitted to an LPA must accurately represent the pre-development baseline and the post-development commitment. Errors in the baseline map — missing habitat parcels, incorrect boundary placement, wrong habitat types — can result in a gain plan being rejected, the ecologist being required to re-survey, and the development programme being delayed.

The key spatial data sources

A competent BNG habitat survey draws on a combination of national datasets and site-specific field survey. The following are the most commonly used spatial data sources in England.

OS MasterMap Topography Layer

Ordnance Survey MasterMap is the most detailed large-scale national topographic dataset available for England, covering every physical feature at a minimum scale of 1:1,250 in urban areas and 1:2,500 in rural areas. For BNG purposes it provides:

• Accurate site boundary delineation
• Land parcel boundaries (useful as a starting framework, though not habitat-specific)
• Built feature mapping (buildings, roads, hard standing) for distinguishing non-habitat from habitat areas

OS MasterMap is a licensed product — it is available free of charge through the Public Sector Geospatial Agreement (PSGA) for eligible public sector and education organisations, and is available commercially for others. Ecological consultancies typically access it through their own OS licences.

Phase 1 and Phase 2 habitat survey data

Phase 1 habitat surveys are the standard methodology for rapid habitat characterisation in England, based on the JNCC Handbook for Phase 1 Habitat Survey. The survey produces a polygon-based habitat map with each polygon assigned a Phase 1 broad habitat type. For BNG, Phase 1 survey data provides the starting framework for habitat parcel delineation.

Phase 2 surveys go further, characterising vegetation communities in greater botanical detail (typically using National Vegetation Classification methodology). For some habitat types — particularly higher distinctiveness habitats like traditional orchards, lowland meadows, or ancient woodland — Phase 2 data is needed to accurately determine the habitat type and condition score.

DEFRA MAGIC (Multi-Agency Geographic Information for the Countryside)

MAGIC is DEFRA's free web-based GIS tool that aggregates spatial data on protected sites and priority habitats across England. For BNG practitioners it is indispensable for:

• Designated sites: SSSIs, SACs, SPAs, Ramsar sites, NNRs, and their notification boundaries
• Priority habitats: The Priority Habitat Inventory (PHI) maps habitats that are classified as requiring conservation action — these often align with higher distinctiveness scores in the metric
• Ancient woodland: The Ancient Woodland Inventory, a key input for the highest-distinctiveness habitats
• Local Nature Recovery Strategies (LNRS): Once LNRS maps are published from late 2025, MAGIC will show which habitats attract Strategic Significance multipliers under the metric
• National Character Areas (NCAs): The 159 NCA boundaries that determine geographic trading rules

MAGIC data can be exported as shapefiles for use in desktop GIS applications, and an API is available for programmatic access.

CEH Land Cover Map

The UKCEH Land Cover Map provides a national-scale classification of land cover types across Great Britain, updated periodically (the current version covers 2023). It is derived primarily from satellite imagery classification and has a minimum mapping unit of around 0.5 ha in most versions.

The Land Cover Map is useful as a national-scale starting point for understanding the broad land cover context of a site, but it is not suitable as the primary input for BNG metric calculations. The resolution and classification system do not align well enough with the DEFRA habitat typology used in the metric, and its minimum mapping unit is too coarse for accurate parcel delineation on most development sites.

LiDAR (Light Detection and Ranging)

The Environment Agency publishes a national LiDAR data collection for England, freely available for download. LiDAR provides highly accurate elevation models (typically 0.5–2m point spacing) that are valuable for:

• Distinguishing canopy cover from ground-level vegetation — particularly useful for woodland boundary delineation
• Identifying hydrological features — drainage ditches, ponds, and wet ground that may indicate wetland or watercourse habitats
• Detecting hedgerow structure — height and continuity indicators that inform condition assessment
• Topographic context — slope and aspect data relevant to habitat characterisation in upland sites

LiDAR supplements field survey — it does not replace it. But it can significantly improve the accuracy of habitat parcel delineation before the ecologist visits the site, reducing survey time and improving consistency.

Aerial and satellite imagery

High-resolution aerial imagery (typically 25cm resolution or better from aerial photography surveys) is widely used as a backdrop for habitat mapping. The most commonly used sources include:

• Getmapping / Bluesky aerial photography — used as the standard OS background in many GIS environments
• Google Earth / Google Maps — widely used due to accessibility, but variable resolution (see the caution below)
• Sentinel-2 satellite imagery — free, multispectral, and useful for large-scale vegetation analysis, but limited to 10m resolution
• Planet Labs / Maxar — commercial high-resolution satellite imagery (0.3–0.5m) for detailed site analysis

GIS tools used in BNG practice

Habitat data for the BNG metric is typically captured, managed, and analysed in a Geographic Information System (GIS). The most commonly used platforms in ecological consultancy practice in England are:

For most smaller ecological consultancies and independent ecologists, QGIS is the practical standard — it is free, capable, and has excellent documentation and community support. If you are new to QGIS or want to build your skills specifically for ecological survey work, Ecology Training offer QGIS courses tailored to ecologists. ArcGIS Pro remains common in larger consultancies and public sector organisations that have existing Esri licences.

Common mapping errors and how to avoid them

These are the errors most likely to cause a metric calculation to be challenged, either by the LPA or by Natural England during Biodiversity Gain Plan assessment:

1. Relying on remote sensing without field verification

Satellite and aerial imagery can be used to inform habitat parcel boundaries before field survey, but cannot determine habitat type or condition without ground-truthing. The DEFRA metric guidance requires habitat surveys to be conducted by a suitably qualified ecologist, and while it does not prescribe specific survey standards, the expectation of field verification is clear. An entirely desk-based metric is unlikely to withstand scrutiny from an experienced LPA ecologist.

2. Incorrect minimum mapping unit

The DEFRA metric guidance recommends a minimum mapping unit of 0.01 ha for area habitats and 10m for linear habitats (hedgerows and watercourses). Mapping at a coarser resolution — for example, aggregating adjacent habitat types into a single large parcel — can significantly underestimate the complexity of a site and miss higher-distinctiveness habitat patches that should be mapped separately.

3. Lumping adjacent parcels of different condition

A common shortcut is to map a large area of grassland as a single parcel and assign it a single condition score. Where the grassland has spatially distinct zones of different condition — for example, a well-managed central area adjacent to a degraded margin — these should be mapped as separate parcels with separate condition scores. Combining them into one parcel and averaging the condition artificially dampens the score of the better-condition area.

4. Misidentifying habitat types at the boundary

The metric's habitat typology maps to the DEFRA Habitat Classification. Some habitat-type boundaries are clear (woodland vs. grassland); others require botanical or ecological expertise to determine accurately (for example, the distinction between 'modified grassland' and 'lowland meadow', which have very different distinctiveness scores). Where the habitat type affects the distinctiveness score significantly, errors in classification have a proportionally large effect on the unit value.

5. Failing to map all habitat types, including linear features

Sites often contain hedgerows, ditches, or watercourses that are easy to overlook on desk-based review but must be included in the metric if they are present. Hedgerow units and watercourse units are separate from area units — a development that destroys hedgerow must compensate with hedgerow units and cannot substitute area habitat units. Missing linear features from the baseline means the developer's post-development gain plan may not accurately reflect what has been lost.

Survey standards and qualifications

The metric documentation does not mandate specific survey qualifications, but best practice guidance from professional bodies sets a clear expectation:

• Chartered Institute of Ecology and Environmental Management (CIEEM) guidance recommends that metric assessments be undertaken by ecologists with relevant experience in habitat survey and the use of the metric tool. CIEEM membership (MCIEEM) or equivalent is a common benchmark.
• Phase 1 habitat survey is the standard baseline methodology. Some consultancies use their own proprietary equivalents, but the JNCC Phase 1 handbook is the accepted reference.
• Rapid assessment vs. full walkover: For small or low-complexity sites, a rapid assessment using phase 1 methods may be sufficient. For larger, more complex, or higher-distinctiveness sites, a full walkover with representative vegetation recording is more appropriate.

LPAs are increasingly specifying in their local validation requirements that BNG supporting information must be prepared by a suitably qualified ecologist. Some LPAs are also requiring peer review of metric calculations for major applications. Where this is a requirement, it will be stated in the LPA's pre-application guidance.

How spatial data flows into the metric

Understanding the data workflow helps identify where errors can enter:

1. 1Site boundary agreed. The planning application boundary (or a defined study area) is established in GIS. All habitats inside this boundary form the baseline.
2. 2Background data layers loaded. OS MasterMap, aerial imagery, LiDAR, MAGIC layers (PHI, designated sites, NCA) loaded into the GIS project.
3. 3Desk-based pre-screen. Potentially sensitive or high-distinctiveness features identified. Phase 1 habitat types provisionally assigned where clearly identifiable from imagery and national datasets.
4. 4Field survey conducted. Qualified ecologist visits the site, ground-truths and corrects habitat boundaries, records habitat types, and assesses condition using the DEFRA metric condition assessment modules.
5. 5Habitat parcel polygons finalised in GIS. Field data is incorporated into the GIS habitat map. Each polygon is attributed with habitat type and condition. Areas are calculated automatically in GIS.
6. 6Data entered into the statutory metric spreadsheet. Area, habitat type, distinctiveness, condition, strategic significance, and other factors are entered into the DEFRA metric Excel tool. The tool calculates biodiversity unit values.
7. 7Output reviewed and the gain plan prepared. The ecologist reviews the outputs for sense-checking, prepares the Biodiversity Gain Information Report, and supports the developer in preparing the Biodiversity Gain Plan for LPA submission.

GIS and the off-site gain site market

Spatial data plays an equally important role in the off-site BNG market. Developers searching for off-site units need to find sites within the same National Character Area or LPA as their development — a spatial constraint that cannot be resolved without mapping.

The Biodiversity Gain Sites Register provides grid references for all registered gain sites, enabling basic geographic analysis. However, the public register does not provide detailed spatial data — habitat parcel boundaries, habitat types, or condition maps are not published. This is one of the reasons why platforms enriching the register data (linking metric spreadsheet data to spatial location) are proving useful in the market.

When to involve a GIS specialist

Most metric calculations for small to medium-sized sites can be carried out by an experienced ecologist using standard GIS tools. However, there are circumstances where involving a specialist GIS analyst or remote sensing specialist adds significant value:

• Large or complex sites where the habitat map involves hundreds of polygons and quality control of the spatial data is essential
• Sites with significant woodland, hedgerow, or watercourse networks where LiDAR analysis can substantially improve the accuracy of linear feature delineation
• Multiple connected sites (e.g., a developer with off-site gain land in addition to the development site) where spatial consistency and cross-site data management are important
• Planning appeals or legal disputes where the technical robustness of the spatial data and survey methodology may be subject to expert scrutiny
• Gain site registration where the landowner is creating a habitat bank across a large area and needs accurate, publication-quality spatial data for the Natural England registration process