NSTA structured datasets on data.gov.uk (Wellbores, Fields, Licences)

The North Sea Transition Authority - the UK body that grants offshore petroleum licences and keeps the shelf's well records - publishes the paperwork of the North Sea as geography: 922 datasets whose layers trace offshore and onshore wellbore origins, top holes and bottom holes; petroleum licence blocks, licences, subareas and relinquished blocks; hydrocarbon field outlines; production reporting points; carbon storage licences; quadrant maps; and exploration round references. Where most regulators publish registers you read, this one publishes surfaces you plot - nearly every layer carries geometry, and the same features appear in four coordinate systems (ED50, WGS84, ETRS89 and British National Grid) so a block drops into a North Sea map built in any of them.

Depth concentrates in the wellbore tables. Each origin record follows PPDM 'What is a well' conventions and carries 101 attributes fed by the Authority's Well Operations Notification System: spud, total-depth and completion dates, the declared intent of the wellbore, its mechanical and operational status, the licence numbers in force at origin and at total depth, the subarea operator behind each, the field or prospect targeted, and surveyed coordinates. Roughly 9,500 offshore top-hole origins make up the headline set, and one of them - an ENI UK appraisal of the Bunter and Hewett Sandstones spudded in May 2025 and abandoned by September - appears verbatim in the sample below.

Get a sample of this dataset - name the layers, quadrants or operators you care about and we return real rows before anything ships.

One captured wellbore origin record, then the shape of the collection around it:

# one wellbore origin record (southern North Sea appraisal)
well_origin_reference       = C48/30- 21
original_wellbore_intent    = Appraisal
spud_date                   = 5/30/2025        # month-first, as recorded
td_reached_date             = 8/6/2025
completion_date             = 9/10/2025
mechanical_status           = Abandoned Phase 3
operational_status          = Decommissioned
prospect                    = Bunter and Hewett Sandstones
field_name                  = No Data Available
subarea_operator_at_td      = ENI UK LIMITED

# shape of the collection
datasets                    = 922
wellbore_attribute_columns  = 101              # PPDM 'What is a well' aligned
offshore_top_hole_origins   = ~9,500
layers                      = wellbore origins | top holes | bottom holes | licence blocks
                              | subareas | relinquished blocks | field outlines
                              | production reporting points | carbon storage licences
                              | quadrant maps | exploration round references

Three things fall out of a single row. First, the campaign timeline is complete: spudded 30 May, total depth reached in early August, completion logged by mid-September 2025 - a full drilling story in four dated columns. Second, field_name reading 'No Data Available' is information, not absence: the well was drilled into a named prospect before any field allocation existed, and the prospect column preserves the geological target regardless. Third, the operator arrives as the legal entity - ENI UK LIMITED - so operator rollups group cleanly instead of fuzzy-matching trading names.

Eleven core fields define the wellbore origin layout in the dictionary below, verified against the payload during cataloging. well_origin_reference is the primary key, with quadrant, block and sequence numbers riding inside the string. The three date columns turn every record into a campaign timeline. The status pair splits cleanly: mechanical status describes the hardware - plugged, abandoned by phase - while operational status describes the lifecycle - drilling, suspended, decommissioned. Licence and subarea-operator columns repeat twice, once at well origin and once at total depth, which is what catches a licence reassignment mid-well.

The full wellbore schema runs to 101 columns under PPDM 'What is a well' conventions, and the non-wellbore layers - blocks, subareas, field outlines, reporting points - carry their own geometries and identifiers. Everything past the verified core sits under additional fields on request and ships with any sample.

Geography - the United Kingdom Continental Shelf across its offshore quadrants, plus the onshore petroleum areas of England and the wider UK landward side. Features resolve as points (well origins, top and bottom holes, production reporting points) and polygons (blocks, subareas, relinquished blocks, field outlines), each published in ED50, WGS84, ETRS89 or British National Grid.

Temporal - a living registry rather than a frozen archive cut. Every record carries its own operational dates, and those run the depth of the UK's offshore drilling era through campaigns as recent as the 2025 Hewett-area appraisal in the sample row. Read the collection as the current published state of each wellbore, block and field, with per-record history living in the date and status columns rather than in versioned snapshots.

Granularity - one record per real-world object: a wellbore origin, a licence, a block, a subarea, a field outline, a reporting point. There is deliberately no rollup layer; aggregation by operator, quadrant or field happens on your side of the join, which is precisely why the keys are kept clean.

API, files, or your warehouse. Daily, weekly, or hourly.

Pick the channel your stack already speaks. Geometry ships flattened - coordinates and grid references as typed columns beside the polygon - so pipelines that never signed up to parse shapes don't have to start.

Every delivery includes the field dictionary above plus the sample rows for validation, and all four coordinate systems travel together so a datum mismatch surfaces during scoping, not in production.

• Licence-map screening - blocks, subareas and relinquished blocks answer who held what, where, and what got handed back; the subarea-operator column turns acreage into named company footprints.
• Decommissioning pipeline sizing - mechanical statuses like 'Abandoned Phase 3' and operational statuses like 'Decommissioned' are the plug-and-abandonment queue itself, well by well, which makes service-demand models arithmetic rather than guesswork.
• Drilling-cadence analytics - spud dates per quarter give an activity signal straight from the regulator's own notification records, no trade-press intermediation.
• Carbon-storage siting - the carbon storage licence layer maps who holds CCS acreage and where it sits relative to field outlines and depleted structures.
• Field-outline joins - polygon boundaries let third-party production, survey or emissions data attach to named fields instead of nearest-point approximations.
• Prospect intelligence - the prospect column ties individual wells to named geological targets (the sample row's Bunter and Hewett Sandstones being exactly such a target).

Deeper persona guidance lives on the investors & quants, data scientists and developers & builders pages for oil & gas equipment & services.

Competitive intelligence and product teams read subarea operators as the ground truth of who actually works a quadrant - licence holders announce intentions, operators drill wells. Investors and quants get a spud-date ledger long enough to build drilling-cycle signals and short enough to load whole. Data scientists inherit a PPDM-aligned schema, which means the column meanings match the industry standard their other sources already speak, and keys that survive joins without hand-holding. GIS and engineering teams get four coordinate systems per feature, so legacy ED50 project grids and modern WGS84 pipelines coexist without conversion surprises. Strategy and market research teams get the decommissioning and CCS pictures in the same atlas as the drilling one - three stages of the basin's life under one geometry.

Naming note - C48/30- 21 bundles quadrant, block and sequence inside one string. Treat it as the join key verbatim; decode it only if your team maintains its own quadrant index.

Datum note - ED50 and WGS84 disagree by more than a hundred metres in parts of the North Sea. Pick one coordinate system per project and hold it; mixing datums quietly relocates platforms.

Date-format note - operational dates arrive month-first (5/30/2025 is 30 May). We normalize on request; unnormalized rows ship flagged.

Catalogue-completeness note - a minority of catalogue entries carry placeholder descriptive text inherited from harvesting. Geometry and attributes are unaffected; we reconcile descriptions during sample scoping rather than shipping boilerplate.

Where to go next:

• NSTA National Data Repository (NDR) - the curated well, survey and document archive behind these registry layers.
• NSTA UKCS Open Data: Wells, Production, Seismic and Infrastructure - the same authority's portal view across wells, production and seismic.
• NSTA UK Data Centre (Open Data) - the predecessor portal serving reports and legacy records.
• OGIM - Oil and Gas Infrastructure Mapping Database v2.5.1 - global infrastructure context beside this UK-shelf registry.
• OPD - Offshore Platform Dataset (Sentinel-1 derived) - satellite-counted platforms to audit against the mapped ones.
• Best oil & gas equipment & services datasets - the ranked shortlist across the vertical.
• Oil & Gas Equipment & Services data hub - the pooled industry view.