TerraSelect | EU Site-Suitability Scorer for AI Gigafactories
Modular geospatial pipeline that overlays 15+ public EU datasets into a single suitability surfaceMay 19, 2026
TerraSelect is a modular geospatial site-suitability scorer built to study where AI Gigafactories could plausibly land across Europe — the same question the EU is asking as it allocates €20 billion in Gigafactory funding.
It overlays public EU datasets — land cover, terrain, power infrastructure, water bodies, transmission lines, internet exchanges, protected areas, regional R&D capacity, and more — onto a grid of candidate cells, scores each layer independently with a customisable rule (distance decay, attribute weighting, normalisation), and combines the layers into a single suitability surface you can explore interactively.
The project ships three surfaces from one codebase: a Streamlit explorer for interactive map-driven analysis, a CLI for batch generation of per-layer score grids, and a Docker image that bundles app + dependencies so data updates don’t require rebuilds. Fifteen-plus loaders cover the EU public-data landscape (Copernicus, Eurostat, OpenStreetMap, Natura 2000, EU-Hydro, OpenAlex, and others), each with its own scoring rule and license-aware attribution.
TerraSelect is open source under Apache-2.0. The reference blog post — “AI Gigafactories: Europe’s €20B Race and the Site Selection Challenge” — walks through the policy context, the methodology, and the practical pitfalls of stitching together EU geospatial data without a unified standard. Pre-built score grids for Germany and EU-wide extents are published as GitHub Releases so you can explore the suitability surface without re-downloading every upstream dataset.