data.europa.eu — EU open data portal

Europe publishes its public data through dozens of national portals and a stack of EU-institution catalogues, and somebody had the good sense to index all of them in one place. That place is data.europa.eu, operated by the Publications Office of the European Union: a DCAT-AP-harmonised shelf of roughly 1.9 million dataset records harvested from EU bodies - Eurostat and the European Environment Agency among them - and from national and regional portals such as opendataportal.at, data.gov.uk, datos.gob.es and data.overheid.nl.

Queried for this industry, the haystack compresses fast: an exact-phrase search for “industrial gases” returned 28 records at verification, while looser unquoted matching fanned out to roughly 54,000. What surfaces is not gas statistics in the neat time-series sense - it is the paper trail around the sector: national inventory reports, air-emissions tables, industrial-registry extracts, classification indices, each carrying multilingual titles, distribution formats and the publishing portal's own identifiers.

One caveat shapes everything else on this page: this is a discovery layer. Every record describes a dataset that still lives on the portal that published it. Get a sample of this dataset and Datadory hands you the resolved industrial-gases slice - records with their fields populated - instead of a list of places to look.

Four of the 28 exact matches at verification, shown flat exactly as the catalogue describes them:

# 4 of 28 exact matches for the phrase “industrial gases”
title   : Austria’s National Inventory Report 2008
catalog : opendataportal.at
country : Austria
formats : PDF

title   : UKCCSRC Call 2 project poster: UK Demonstration of Enhanced Calcium Looping
catalog : data.gov.uk
country : United Kingdom

title   : National classes indices. IPRX-M (Identificador API: 67164)
catalog : datos.gob.es
country : Spain
formats : HTML, CSV, JSON, Excel XLSX

title   : Emissies naar lucht op Nederlands grondgebied; stationaire bronnen, 1990-2021
catalog : Dataportaal van de Nederlandse overheid
country : Netherlands

Read them as four different species of evidence. Row one is Austria's national inventory report - a greenhouse-gas inventory document published as a PDF, the kind of official paperwork that anchors a regulatory filing. Row two is a UK carbon-capture research poster, proof that the phrase reaches adjacent technology programmes, not only heavy industry. Row three is the sleeper: a Spanish national classification index shipping in four formats at once, which makes it the rare record here that drops straight into a parser. Row four is the analytical prize - the Netherlands' stationary-source air-emissions series running 1990 through 2021, three decades of facility-side emissions context for exactly the energy-intensive manufacturing that defines this industry's customer base.

Four countries, four portals, one query. That spread is the whole argument for treating this catalogue as the starting point rather than any single national portal.

Eight documented fields carry every matched record, and they split into three jobs. result.count and country do the slicing - how big is the hit set, and which markets does it span. title, description and catalog do the identification, with the multilingual maps doing quiet work: a record titled only in Dutch still answers an English query, because the language key rides beside the text rather than inside it. And format.label, accrual_periodicity and is_hvd do the triage - which records are machine-readable, which are maintained, and which the EU itself designates as high-value.

Definitions below were checked against live responses during the August 2026 research pass, so the table describes what arrives rather than what a brochure promises. Two further elements ride on each distribution - the re-use terms the publisher attaches and the resolution pointer to the record's home portal - and both fold under additional fields on request so the core schema stays honest about what ships first.

Geography - EU member states, EFTA countries, and EU institutions and agencies, in whatever shape the harvested portals publish. The verified industrial-gases slice alone spans Austria, the United Kingdom, Spain and the Netherlands. The honest framing: coverage is broad but not uniform - a large member state with a strong portal contributes differently from a small one with a sparse catalogue, and the record count tells you which is which.

Temporal - governed entirely by the underlying datasets, and the spread is wide. Many harvested records declare a one-off publication with no maintenance rhythm at all, while others run long historical series; the Dutch air-emissions table verified in this slice spans 1990-2021, and other holdings reach similarly deep. Currency is a per-record property you read off accrual_periodicity rather than something to assume portal-wide.

Granularity - one metadata record per dataset. Beneath that, grain varies from national statistical tables down to facility-level inventories and report documents. Treat the record as the address and the underlying dataset as the building.

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

You pick the channel and the cadence; resolution, cleaning and schema stability are our problem. Each matched record ships with its fields populated - titles, contributing catalogue, country, formats, declared rhythm and designation flags - normalised across however many portals contributed it. Bulk pulls land as files, continuous consumption runs through the API, and warehouse-native loads write straight into your own storage. A sample goes out first, cut to the countries and record types you actually need.

• European regulatory-landscape mapping - national inventory reports and emissions tables surface country by country, giving market-entry teams the paperwork trail around gas-consuming industry before they commit capital.
• Emissions baselining - stationary-source air-emissions series such as the Dutch 1990-2021 table put decades of facility-side environmental context behind any industrial-gases demand model.
• Country screening and territory ranking - the country field splits one pan-European query into per-market reads, collapsing twenty-seven candidate markets into a ranked shortlist.
• High-value dataset monitoring - the is_hvd flag separates records the EU designates as high-value from the long tail, a one-column importance filter most users never think to apply.
• Adjacent-technology tracking - carbon-capture and industrial-decarbonisation programme records surface under the same queries, an early-read on where gas demand may erode or migrate.
• Citation-grade sourcing - every record names its publishing portal and catalogue, so figures survive due diligence with provenance attached.

Market researchers and consultants sweep European public holdings once instead of learning each member state's portal taxonomy; see market researchers use cases. Competitive intelligence and product teams watch registry extracts and inventory publications across markets for signs of where gas-consuming capacity is being documented and permitted; see competitive intel product teams use cases. Data scientists and ML engineers mine emissions and inventory tables as exogenous features, with format labels revealing which records are parseable before any pipeline work; see data scientists use cases. Developers and data-product builders get a uniform DCAT-AP record shape contributed by hundreds of catalogues; see developers builders use cases. Journalists, academics and students cite national publications with the publishing institution named on the record.

Four things, stated plainly.

First, this is a discovery layer. The portal holds descriptions and identifiers; each underlying dataset lives with its publisher. A sample from Datadory means resolved records rather than a pointer list - but the grain beneath is the publishers', not ours.

Second, match counts swing enormously with phrasing. The exact phrase returned 28 records; the unquoted equivalent matched roughly 54,000 through fuzzy matching. Scope the terms first, then read result.count to see what each variation bought you.

Third, recency varies record by record. One-off publications sit beside maintained multi-decade series, sometimes within the same result set. Read accrual_periodicity per record instead of assuming a common rhythm.

Fourth, the score reflects the job. Datadory rates this record 6 out of 10 against a catalog mean of 7.81 - excellent breadth, modest depth on any single topic. When a question needs one authoritative series rather than wide discovery, pair this with Eurostat Comext at 9/10 or the Federal Reserve's G.17 at the same grade. The sample shows the terrain unchanged, caveats included.

Of the 1,744 datasets Datadory catalogs across 159 viable industries, eight cover industrial gases, and this portal is the widest-discovery entry among them. The companions:

• Eurostat International Trade in Goods (Comext) - monthly EU trade values and quantities by CN/HS product code back to January 1988, the strongest single series in the slice at rating 9.
• Industrial Production & Capacity Utilization (G.17) - the Federal Reserve's monthly US industrial production and capacity utilization indexes, machinery detail included.
• eGon - industrial gas demand (CH4 & H2) for Germany - hourly methane and hydrogen demand scenarios for Germany at NUTS-3 resolution, the forward-looking outlier.
• AMT USMTO - monthly US metalworking machinery orders, a demand proxy for gas-using fabrication.
• CECIMO European Machine Tool Sector Data - the European machine-tool association's economic publications, the European counterpart to USMTO.
• IMTMA Machine Tool Industry Data (India) - Indian machine-tool statistics and research reports.
• IndiaMART B2B Marketplace - structured supplier and product listings across Indian industrial platforms.

Browse the industrial gases data hub for the full slice, or see the best industrial-gases datasets ranking.