The data problems hiding in BNG maps
- Claire Birnie
- 20 hours ago
- 4 min read
Imagine spending 30 years legally responsible for a habitat boundary that was drawn on the wrong piece of land.
And I don't mean wrong by a field or two.
We're talking wrong because the data underlying the map was never accurate to begin with. It's things like this that make me sit bolt upright at 3 am, sweating like plums wrapped in a polythene bag.
Data problems are undermining Biodiversity Net Gain, which, you lucky pups, is something I got into with Chris Kinnear on his Housing Geospatial Podcast.
What BNG legislation are we talking about here?
In 2024, legislation was introduced requiring every new development in England to achieve at least a 10% net gain in habitat value.
Biodiversity Net Gain is an important piece of legislation.
But the way it's being implemented in practice has a data problem that nobody wants to take accountability for.
Businesses aren’t talking to each other
A typical BNG project involves:
A developer
An architect
A topographic surveyor
An ecologist
And further down the line, someone like me will do the GIS and mapping work. In theory, data should flow cleanly between all of these parties. In reality, it often doesn't.
BNG mapping problems start long before the data reaches my desk. Most days, my first challenge isn't the mapping itself, it's getting usable data to map with:
"A number of companies don't give me anything more than PDFs to work with, which is quite a risky strategy. I just put in caveats to say that I'm working with PDFs, not an exact site outline."
When I do receive CAD files (DXF or DWG), the problems don't necessarily stop there. Architects and engineers work in millimetres. GIS works in metres. If a file hasn't been correctly georeferenced and scaled, I'll load it into QGIS and find a site boundary that appears to be larger than Russia, sitting off the West Coast of Cornwall or somewhere in South Africa.
I hadn't thought about the architect's difference of millimetres versus metres. If you've got a big enough site that would just become massive
One of them was literally larger than the universe. Open Street Map was down in the bottom left and my site was up in the top right. It was huge. I don't know what they'd done!
I can transform the data and move it to the right location, but the moment you start manually repositioning and rescaling geometry, you introduce potential errors.
In BNG, error has consequences that can echo for decades.
And this isn't anyone's fault individually.
Many ecology consultancies are micro-businesses, with just one or two people who aren't responsible for creating the CAD files. They're the intermediary between the developer and me, receiving data from one end of the chain and passing it on to the other.
The redline boundary problem is more serious than it looks
Here's where this gets legally significant. Some BNG sites will have Habitat Management & Monitoring Plans (HMMPs) in place for up to 30 years. And Section 106 agreements can bind landowners and developers to those commitments for the lifetime of those plans.
If you've made a red line boundary on someone else's land accidentally, what are the implications of that? Who's liable for that? Am I liable for that? Because I've made the data.
That's why I'm strict about not mapping on Google Maps or OpenStreetMap
Not because those tools aren't useful. They're excellent for what they're designed for, but they weren’t designed for legally binding spatial data with 30-year implications.
As I said on the podcast and written about in more depth on the Maptastic blog, Better data & mapping for Ecology & Biodiversity Net Gain, the metric is only as good as the geometry underpinning it.
And right now, that geometry isn't being held to a consistent standard.
Ecologists are being asked to become GIS specialists overnight
Then there's a second, quieter problem. One that sits with the people doing ecology work themselves.
Ecologists are nature specialists.
They know habitats, species, survey methodology, legislation and all that lovely stuff.
But BNG means they’re now expected to produce spatially accurate, legally defensible data in GIS.
The bit that really resonates with me is the fact that ecologists aren't data specialists. They're specialists in nature. And they're being asked to do this quite arduous, quite technical process that is also legally liable
GIS is not something you can generally just rock up and do. There are tips and tricks that make it repeatable and consistent to simplify the process, but you have to learn it to know what you’re really doing
BNG was not created with metric data first. The spatial infrastructure around it was an afterthought. Nobody made it easy for non-GIS specialists to produce GIS-quality data.
If you're an ecology practice weighing up whether to build that capability in-house or bring in external support, it's worth reading my article, Should I hire an external GIS training provider?
What would help with BNG mapping problems?
The fix isn't complicated, but it is politically harder to implement.
BNG mapping needs a digital-first mandate from Natural England and the government, with defined, repeatable, documented standards for the spatial data that underpins the metric.
BNG is the right idea, but good intentions don't make good maps.
Precise, accurate, well-documented data needs to be built into the process from the start, but until that happens, here are a couple of things that can help ecologists:
The Fridas white paper by Katy at Ecospatial: this is a good starting point, as it offers digestible geospatial metadata guidance that doesn't require a GIS background to understand it.
Buy OS topography from a partner like Blue Sky: instead of relying on consumer mapping tools, this gives you templates and workflows that build in accuracy from the start, rather than patching problems at the end.
Other providers are available *wink*
Want to hear the full conversation?
If this has got you thinking, the conversation with Chris goes much further. We got into the tools and workflows I use day-to-day, field data collection, and why open-source GIS might be the answer for smaller ecology practices.
Give it a listen because there’s loads more to dig into:
And make sure to connect with Chris on LinkedIn, too.
Catch ya next time,
Claire
