government · London / England (national platform)
Defra National Land Data Platform — GeoEP
Architected the geospatial data model underpinning Defra's national land data platform — integrating 6 Arm's Length Bodies, 300–500 million feature rows, and a 49-element UK GEMINI 2.3 metadata standard. Reduced data discovery from approximately one month to under 10 minutes.
The Challenge
The Problem We Were Asked to Solve
Defra's Geospatial Engagement Programme (GeoEP) needed to rationalise land data spread across six Arm's Length Bodies (RPA, EA, NE, APHA, FC, MMO) — organisations with conflicting taxonomies, inconsistent data standards, and no common discovery mechanism. Finding a specific land parcel record could take a team member up to a month due to siloed, undocumented datasets.
Key technical challenges: - Cross-ALB taxonomy harmonisation: each body used different classification schemes, attribute names, and coordinate reference systems - Metadata coverage: no consistent metadata existed across the 300–500 million feature rows in scope - Standards compliance: the platform needed to satisfy UK GEMINI 2.3, ISO 19115, INSPIRE, DEFRA DAF, DEFRA Data Modelling Standards v2.3, and UK GDPR/DPIA requirements simultaneously - Data quality: a 40% false-positive rate in SSSI boundary checks and a 14% 'white space' gap in land coverage both needed systematic resolution - Architecture coherence: a Conceptual Data Model (CDM), Logical Data Model (LDM), and Physical Data Model (PDM) all had to be designed in alignment and version-controlled to Defra standards
Project Details
- Client
- Defra (via Informed Solutions)
- Year
- 2025
- Location
- London / England (national platform)
- Sector
- government
Tags
Our Approach
What We Built
Soheil Sotoodeh joined the GeoEP programme as Geospatial Data Architect in November 2025, embedded with Informed Solutions during the Alpha phase and continuing into Beta supervision.
Data Architecture Design - Designed CDM v4.2 (Conceptual Data Model) and LDM v3.1 (Logical Data Model) covering 9 entities with 114 fully-specified attributes - Implemented a linkage-based architecture enabling cross-ALB record resolution without duplicating data across bodies - Applied Third Normal Form (3NF) normalisation throughout the LDM to eliminate data redundancy and protect referential integrity - Designed tiered Physical Data Model (PDM) with Hot / Cold / Archive storage layers for cost-effective at-scale hosting
Metadata Framework - Designed the UK GEMINI 2.3 Defra Land Profile — a 49-element metadata standard covering discovery, lineage, quality, distribution, and governance fields - Mapped 51 functional requirements against the profile, achieving 94% requirements coverage - Integrated QFAIR quality scoring framework across 5 dimensions (Quality, Findability, Accessibility, Interoperability, Reusability)
Taxonomy & Data Discovery Pipeline - Reconciled conflicting land classification taxonomies across all 6 ALBs, including Living England Habitat Map, OS NGD Land Parcel Types, and Soilscapes - Built a production-ready taxonomy discovery pipeline in Python using a modular, reusable architecture - Resolved 40% false-positive rate in SSSI checks through geometry validation and topology QC - Identified and documented 14% white-space land coverage gap for onward data acquisition planning
Standards & Governance - Ensured full compliance with ISO 19115:2003, ISO 19107, ISO 19111, OGC API Records, OGC CSW, INSPIRE Metadata Implementing Rules, and DAMA DMBOK - Implemented PII separation boundaries and DPIA controls in alignment with UK GDPR requirements - Produced structured data lineage documentation covering sources, transformations, and workflows for all integrated datasets
The Results
What Changed
Platform Impact - Reduced data discovery time from approximately 1 month to under 10 minutes — a 99%+ reduction in search time for land data professionals across government - 94% coverage of the 51 mapped programme requirements in the Alpha phase, verified against the CDM/LDM deliverables - Taxonomy conflicts across 6 ALBs reconciled into a single, auditable classification framework - 40% false-positive rate in SSSI checks eliminated through systematic geometry and topology QC - 14% white-space land coverage gap formally documented, enabling targeted data acquisition in the Beta phase
Technical Outputs - CDM v4.2 and LDM v3.1 with 9 entities, 114 attributes, delivered to DEFRA Data Modelling Standards v2.3 - 49-element UK GEMINI 2.3 Defra Land Profile — ready for adoption across ALBs - QFAIR-scored quality framework for all integrated datasets - Production-ready Python taxonomy discovery pipeline (modular, documented, handover-ready) - Tiered PDM architecture specification for Hot/Cold/Archive storage
Explore More