← Cases
United KingdomPublic sector2025 – ongoing

Natural England

AI4Peat mapping of peat drainage and restoration features

Natural England used computer vision to produce a national map of upland peat drainage features

WorkingCompany's own claimOfficial ↗

Problem

England lacked a national record of upland peat drainage features and previous restoration work. Existing mapping relied on manual work with uneven geographic coverage.

Approach

Natural England trained segmentation models on aerial imagery and LiDAR to map drainage and erosion features. A separate object detector located restoration dams. The models produced downloadable layers for the England Peat Map.

Results

No published figures

Takeaway

The baseline map required expert regional knowledge when informing policy. Future monitoring depended on rerunning the models with updated imagery.

Categories

Business function
Production and quality
Innovation type
Process innovation
AI technology
Computer vision
Evidence
Company's own claim

Sources

  1. Natural England : AI4Peat
    GOV.UK30 Oct 2025Company publication

Source summaries

[1]GOV.UK30 Oct 2025

Natural England documented the models used to create its peat surface-feature map. It confirmed a completed national mapping run and explained that the outputs supported decisions rather than making them.

Full details are in the original sources.

Added 1 Nov 2026 · Updated 1 Nov 2026
Open case page ↗