Validation at the ingest boundary
Each incoming batch is checked against its declared schema, units and code lists before it is accepted. Failures are quarantined with a reason, and the original is retained unchanged beside the standardised record.
The Global Centre for Risk and Innovation (GCRI)
Data engineering decides whether evidence from different sectors can be combined at all. In this field it covers ingestion, validation, standardisation, provenance and stewardship: source passports, schemas and ontologies, frozen data cuts, and lineage that lets a correction reach every product built on the corrected source.
GCRI · Established · service level R0
Data engineering is the design and operation of the pipelines and stores through which data is ingested, validated, standardised and served. For risk evidence it decides whether data from different sectors can be combined at all, and it uses source passports, schemas, frozen data cuts and lineage so a correction reaches every product built on the data.
Data engineering, interoperability and stewardship cover the pipelines, standards and governance that make data usable and trustworthy. They determine whether information from different sectors can be combined for joint risk assessment. In data and compute and methods and modelling, sound data engineering allows hazard, exposure and financial data to be linked consistently. It supports transparent and repeatable analysis of interconnected risk. The canonical description is on the STEM Commons.
Each incoming batch is checked against its declared schema, units and code lists before it is accepted. Failures are quarantined with a reason, and the original is retained unchanged beside the standardised record.
Indices and loss estimates are computed from versioned, immutable snapshots, so a product can be re-run months later from exactly the data that produced it.
When a source owner issues a correction, the lineage graph lists every derived table, map and index that depends on it, and those products are marked stale until recomputed.
Each term as this page uses it. Where a standard-setting body defines the term, the wording follows that body.
| Term | Definition |
|---|---|
| Frozen data cut | A fixed, versioned snapshot of input data used to produce a product, so it can be reproduced later. |
| STAC | The SpatioTemporal Asset Catalog specification for describing geospatial data such as satellite images. |
| Correction impact analysis | Using lineage to find every product affected by a correction to a source. |
Urban demand centres and the hubs that anchor digital and data infrastructure, shown for Europe as an example geography for this subject.
Data from the source register feeds the methods; calibration and an independent test gate the estimate and its limits; the decision stays with the competent authority.
Fig. 01Nodes carry their type, metrics and state; connectors show what feeds what; the gates are the validation steps; the timeline is one application. Design of the method's use, not a record of a run.
Datasets and records used by data engineering, interoperability and stewardship on this platform, with the role each plays and the label its values carry. Sources are listed in the source register.
| Data | Role | Label |
|---|---|---|
| Source passports | Owner, licence, cadence, authority class and limits | DECLARED |
| Schemas and ontology shapes | Validation at ingest | DECLARED |
| Pipeline run logs | Lineage and failure analysis | OBSERVED |
| Official statistics and registers | Authoritative reference values | OFFICIAL |
| Catalogue metadata in DCAT and STAC | Discovery and reuse | DECLARED |
Pipelines are tested with automated quality checks, schema validation and lineage audits. Reconciliation against source systems confirms completeness.
| Check | How it is done |
|---|---|
| Reconciliation with the source | Row counts, totals and keys are compared with the source system after each load |
| Schema and unit validation | Records are tested against shapes, controlled vocabularies and CF or SI units before acceptance |
| Lineage audit | A sample of published figures is traced back to the original records and the transformation steps are re-run |
| Licence check | Every source and derived product carries a licence that permits its intended use |
Incompatible standards, missing metadata and restrictive licences block integration. Legacy systems and undocumented transformations introduce silent errors.
Data and Compute stewards data engineering, interoperability and stewardship for the Consortium and reviews how it is applied across platforms.
The current service level is R0: information and enquiries. An enquiry receives a reference and names the next step; applications, payments, compute allocations and registrations open only when their team, terms and review capacity exist.