Habitat Mapping: Do It Properly
Digitising may not sound exciting. It’s not a new tool, or a regulation update, or an emerging trend. But if you work with habitat maps, the quality of digitising underpins everything.
That’s why we’ve included digitising standards as a core element in our FRIDAS checklist , as without reliable geometry, even the most accurate condition scores or habitat classifications can fall apart.
We’re seeing more and more assessments where good ecological insight is let down by poor digitising. Boundaries are misaligned, polygons don’t snap, shapes are over-generalised and miss important detail. In some cases, we’re seeing automated outputs that haven’t been checked at all. These might seem like small issues, but they can cause serious problems for biodiversity assessments, compensation calculations, and long-term monitoring.
Digitising is often treated as a task to tick off at the end of a survey. But really, this is where ecological knowledge is converted into spatial data. It’s the bridge between what’s seen on the ground and what’s calculated in the metric. If this step is rushed, automated, or treated as an afterthought, the output becomes distorted. Polygons may not reflect what was actually observed. Habitats may blur into one another or stop short of where they should be. Simple things like snapping to field boundaries, contours, or visible features are often missed.
And when that data is put into the BNG metric calculator, the errors multiply.
BNG units are built on area, distinctiveness, condition, and strategic significance. Area is the starting point, so if it’s wrong, everything downstream is wrong too. A few metres shaved off can mean under-valuing habitat. A slightly overdrawn polygon could over-inflate gain. The risk increases on sites with small, linear, or high-value habitats, or where habitat multipliers are high.
If a site is contentious, or being reviewed by planners or consultees, inaccurate digitising becomes a real liability. It’s also one of the first things picked up in audits or challenges. Poor mapping undermines confidence in the rest of the assessment.
We often see common issues repeating. These include:
Shapes that are too rough or generalised
Gaps or overlaps between polygons
Features not lined up with aerial imagery or OS base mapping
Polygons drawn freehand with no snapping to real features
Inconsistent minimum mapping units
Spatial data that doesn’t match field notes or photos
Topology errors that break area calculations
Each of these can be avoided with clear workflow, simple QA steps, and a trained team. It doesn’t take long. It just needs attention.
In our whitepaper, we set out a practical workflow for BNG digitising, based on real-world experience. In short, good digitising is consistent, traceable, and spatially accurate. Boundaries should reflect what’s there, not what someone guessed. There should be a clear link between what was recorded in the field and what ends up in the GIS. And someone should check it before it goes anywhere near a metric calculator. Getting the digitising right means the data holds up. That gives confidence to clients, regulators, planners, and ecologists. It means the maps are useful for future work, not just the current report. And it means biodiversity assessments are actually grounded in what’s on the ground.
