GCRI · Foundry
Grix evidence
A risk figure is only as good as the record behind it. The Global Risks Index (NXS-GRX), the risk-data and semantic system of Foundry, is designed to register, standardise, link and trace risk data so that every statement can be followed to its source and re-run. This page sets out how evidence degrades through a pipeline, from licence and meaning to place, time and scenario, which NXS-GRX binds in Chronotopes, and what data stewards and reviewers need before a product leaves a project.
Grix evidence on the record.
Grix evidence: how the risk develops.
Data, ontology, Chronotopes and lineage. The dossier below sets out how the risk develops, what is measured, where the data come from, how it is modelled and who decides.
Meaning drifts across sources
Two datasets both say exposure, affected population or asset and mean different things. Harmonising categories makes them combinable and can hide the difference that mattered. An ontology states the mapping so that it can be inspected and contested.
Silent transformation
A unit conversion, a join on a mistyped key, a filled gap or a reprojection happens inside a pipeline step that nobody records. The output looks clean; the error stays invisible until someone tries to reproduce it from the original.
Time and scenario collapse
An asset’s state before an event, after it and under a what-if branch are three different records. Stored as one, history is overwritten. Chronotopes bind each record to place, time and scenario so the states coexist without conflict.
Corrections must propagate
When a source is corrected, every index, map and loss estimate derived from it is stale. Without lineage nobody knows which products to recompute, and the wrong figure keeps circulating in briefings long after the source was fixed.
GIS and modelling workspace.
The screen a team would use to follow this subject: its measures, states and decisions. Names are examples and values are illustrative.
NXS-GRX is designed to hold evidence from every domain with its labels.
NXS-GRX records are designed to carry NS-3 identifiers and NS-3b metadata under the proposed Nexus Standards, and to back every figure in Reports.
- Source agreements.
- Corrections from data owners.
How grix evidence is modelled, measured and tested.
The model, its assumptions, its data and its uncertainty, in the form a scientist or engineer would check. Structures follow published methods. The values drawn are illustrative, not findings.
Model, equations and assumptions
The model structure used for grix evidence, the assumptions it makes and how each is tested.
Ri = f(Hi, Ei, Vi, Ci)hazard, exposure, vulnerability and coping capacityEAL = ∑k pk · Lkover events k with annual probability pkxt+1 = A xt + B ut + εtA carries the couplings, u the shock, ε the error| Assumption | Why it is made | How it is tested |
|---|---|---|
| Hazard and exposure are independent within an event | Keeps the loss integral tractable | Compare with a copula fit on joint records |
| Couplings are linear near the operating point | Lets shocks propagate through A | Perturb beyond the linear range and compare |
| Vulnerability curves transfer between places | Local curves are rarely available | Validate against observed losses where they exist |
- Inputs
- versioned datasets with provenance
- Outputs
- distributions, never single numbers
- Code
- released with the model card and tests
- Reproduction
- same data, code, environment and seed
What is measured, and how well.
Each indicator carries its unit, its cadence, the kind of source behind it and the evidence label a value would carry. No values are shown: a programme supplies them under its own authority.
| Indicator | Unit | Cadence | Source type | Label |
|---|---|---|---|---|
| Sources with a complete source passport | proportion of registered sources | monthly | Source register | DECLARED |
| Records failing shape validation at ingest | count per batch | per ingest | Ontology shape validation | OBSERVED |
| Published products with end-to-end lineage | proportion of products | per release | Lineage graph | OBSERVED |
| Products affected by a source correction | count per correction | per event | Lineage impact query | OBSERVED |
| Time from source correction to recomputation | days | per event | Pipeline and release records | OBSERVED |
| Records carrying an evidence label | proportion of records | per release | Record metadata | OBSERVED |
| Variables conforming to standard names and units | proportion of variables | per ingest | CF and unit checks | OBSERVED |
Where the evidence comes from.
Named public datasets and institutional records, each with its licence class. Sources by agreement are used only under their terms and never republished.
| Source | What it gives | Licence class |
|---|---|---|
| World Bank Open Data | Development indicators with documented definitions and sources | CC BY 4.0 |
| Humanitarian Data Exchange (HDX) | Crisis datasets and common operational datasets from humanitarian partners | Licence per dataset |
| National open data portals and statistical offices | Published statistics, registers and boundaries | Open government licences, per country |
| CF standard name table | Agreed names and units for climate and forecast variables | Free and open |
| Wikidata | Linked identifiers for places, institutions and infrastructure | CC0 |
| W3C PROV and DCAT vocabularies | Standard models for provenance and for dataset catalogues | W3C document licence |
Who decides, and what the evidence serves.
Foundry prepares evidence. The decisions below are taken by the authorities named, under their own rules, and never by the platform.
| Who decides | The question | Evidence this page serves |
|---|---|---|
| Data owners | Whether their data may enter, be standardised, be published or train models | Source passport, intended uses, derived products |
| National statistical offices | Whether a derived figure may sit beside a published statistic, and how it is labelled | Method, evidence label, and the published figure kept separate |
| Review leads | Whether a product is fit for release | Lineage, validation results, labels and open corrections |
| Programme owners | Whether a new index method version is adopted for reporting | Method version history, frozen data cuts and re-run comparisons |
Outputs this topic produces
- Source passports with owner, licence, cadence, authority class and known limits
- Validated records bound to place, time and scenario
- Lineage graphs from source record to published figure
- Correction impact reports listing products to recompute
What data stewards and analysts ask.
- Where did this value come from, under which licence, and what has been done to it?
- Do two datasets mean the same thing by “exposure”, “asset” or “affected population”?
- Which version of the data and which version of the method produced this index?
- When a source is corrected, which products are affected?
Preserve, describe, bind, trace.
- 01
Source passports
Each source is registered with its owner, licence, access policy, update cadence, authority class and known limits before any record is standardised.
- 02
Originals kept
Original records are retained unchanged; standardised fields are derived and linked back, never substituted for them.
- 03
Ontology
The NXS-GRX ontology describes hazards, exposures, vulnerabilities, assets, institutions and their dependencies, with shapes that validate what enters the knowledge graph.
- 04
Chronotopes
Each risk record is bound to a spatial and temporal frame and a scenario branch, so the same asset can carry different states across time and across what-if branches without conflict.
- 05
Lineage
Indices and losses are designed to be computed from frozen data cuts with declared methodology versions, so they can be re-run and matched.
What the engineering can and cannot do.
The approach draws on established practice: provenance models, semantic web standards for ontologies and shape validation, versioned data lakes and reproducible pipelines. Verification is shown as separate dimensions, integrity, conformance, plausibility, corroboration, method, rights and human review, rather than as one badge, because a record can be well-formed and still wrong.
The limits are real. Ontologies encode choices that some disciplines will contest, and harmonising categories can hide meaningful differences between sources. Lineage shows what was done, not whether it was appropriate. Original source authority is retained: Programme estimates stay labelled as estimates and never replace published figures.
How it is studied, and what the methods cannot say.
- 01Data engineering, interoperability and stewardshipValidation at the ingest boundary · Frozen data cuts for reproducible productsApplied here
- 02Systems engineeringCanonical system inventory · Contract-first interfaces · Requirements traced to evidenceApplied here
- 03Geospatial information systemsImagery catalogues through STAC · Exposure overlays with stated aggregationApplied here
- 04Artificial intelligence and machine learningEarth observation models with spatial validation · Model cards and promotion recordsApplied here
Where this page’s references lead.
The bodies and documents this page is read against. Status and editions change; each link goes to the issuing body.
Questions and answers.
What is a Chronotope?
Why keep original records unchanged?
Is verification a single badge?
Does NXS-GRX rank countries?
What happens when a source is corrected?
Are indicator values shown here?
Contribute evidence from this page.
Reports is the Consortium’s quarterly publication. What you know about this subject can become evidence, a case or a correction in the next edition, credited to you and published under CC BY 4.0 where it is your own work.
- EmailGateway GW-GCRISend it from this pageEvidence, a correction, a case, a method, a review, a translation or profile input. It arrives tagged with this page, so the editor who owns it receives it.
- ZenodoOpen depositDeposit it and submit to the communityPublish your own record on Zenodo with your ORCID iD, then submit it to the Reports community there. Curators review it; the DOI stays yours.
- ReportsGCRISee every way to contributeThe contribution types, how each is reviewed, what credit you receive, the licence and what is never accepted.
Never send personal data, sensitive infrastructure detail or anything you have no right to share. A contribution is reviewed before use; sending it confers no authorship or editorial rights, and sponsorship never buys a finding.
Bring a method, a dataset or a question.
Information and enquiries are open today. 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.
Legal documents.
Maps, data and figures on this site are shown for context and general information. They are not advice, forecasts or warnings; the limits of the information are set out in the Disclaimers and Data Notices.