{
 "upstream_attribution": [
  "OpenStreetMap via Geofabrik extracts (ODbL) \u2014 \u00a9 OpenStreetMap contributors, https://www.openstreetmap.org/copyright"
 ],
 "generated": "2026-11-02",
 "plain": "9 of the 37 instruments in the data catalogue, as 13 files (10 tables, 1 of them in numbered parts: water-quality-near-facilities.csv in 4) you can download, open in a spreadsheet and join to your own data. The other 28 have no file here; instruments_without_csv names them. Each row carries the date it was collected and the source it came from, so a row still says where it is from after it has left this site.",
 "compiled_by": "WAPR \u2014 Worldwide AI Proliferation Registry, operated by BATECH LLC",
 "origin": "https://wapr.io",
 "attribution": "Every row carries wapr_compiled_by, wapr_origin, wapr_collected_on, wapr_source and wapr_citation, so a single row pasted elsewhere still says who compiled it and when. Each file has a sibling .LICENCE.txt.",
 "licence": "The underlying facts are public records published by the agencies named per file; WAPR claims no ownership of them and nobody needs our permission to use them. What WAPR contributes is the compilation and the derived columns \u2014 the joins, baselines, anomaly flags, coverage arithmetic and the vintage \u2014 and that is what the attribution is for. Upstream licences are not ours to relax: OpenStreetMap-derived instruments are ODbL, whose share-alike terms follow any derived database, and WRI Aqueduct is CC BY 4.0. Except 1 file(s) whose licence_state says the publisher declares no licence (csv/ireland-datacentre-electricity.csv): that file's LICENCE.txt says what to confirm before republishing.",
 "count": 10,
 "files": [
  {
   "file": "csv/va-deq-regulated-sites.csv",
   "licence_file": "csv/va-deq-regulated-sites.LICENCE.txt",
   "rows": 199,
   "columns": 50,
   "collected_on": "2026-11-01",
   "source": "VA DEQ ArcGIS EDMA layers 294 + 292",
   "note": "Virginia DEQ permitted sites flagged as data centres, with reported emissions by pollutant in short tons. DEQ OVERWRITES this table each year, so the collection date is what makes this a vintage rather than a scrape.",
   "cite_as": "WAPR \u2014 Worldwide AI Proliferation Registry (BATECH LLC). \"va-deq-regulated-sites.csv\", collected 2026-11-01. https://wapr.io/csv/va-deq-regulated-sites.csv",
   "payload": "va-sites.json",
   "collected_on_basis": "payload generated"
  },
  {
   "file": "csv/ireland-datacentre-electricity.csv",
   "licence_file": "csv/ireland-datacentre-electricity.LICENCE.txt",
   "rows": 11,
   "columns": 15,
   "collected_on": "2026-11-02",
   "source": "Central Statistics Office, Ireland; Data Centres Metered Electricity Consumption; https://ws.cso.ie/public/api.restful/PxStat.Data.Cube_API.ReadDataset/MEC02/JSON-stat/2.0/en",
   "note": "Ireland is the only country that officially counts data-centre electricity as its own category. Annual GWh and the share of all metered electricity. The share column is WAPR's own arithmetic over CSO's published figures, not a CSO statistic.",
   "cite_as": "WAPR \u2014 Worldwide AI Proliferation Registry (BATECH LLC). \"ireland-datacentre-electricity.csv\", collected 2026-11-02. https://wapr.io/csv/ireland-datacentre-electricity.csv",
   "licence_state": "not_declared_in_api",
   "licence_note": "The JSON-stat payload names a copyright HOLDER, not a licence. WAPR's source survey assumed CC-BY-4.0 by data.gov.ie convention; that is UNVERIFIED from this endpoint. Confirm on the CSO release page before republishing the values under a licence claim.",
   "payload": "cso-ireland.json",
   "collected_on_basis": "payload generated"
  },
  {
   "file": "csv/neighbour-plant-emissions.csv",
   "licence_file": "csv/neighbour-plant-emissions.LICENCE.txt",
   "rows": 223,
   "columns": 50,
   "collected_on": "2026-11-02",
   "source": "US EPA Clean Air Markets \u2014 apportioned daily emissions from continuous monitoring systems at the units themselves",
   "note": "Measured output of large fossil plants near mapped facilities, from the regulator's own continuous monitors. These plants are NEIGHBOURS, not suppliers. Masses are US short tons, as the regulator reports them. Load is gross, not net of the plant's own consumption. A few plants report emissions to this programme without a metered generation figure; those are marked, because a zero beside a real emissions number means unreported here, never a plant that burned fuel and generated nothing.",
   "cite_as": "WAPR \u2014 Worldwide AI Proliferation Registry (BATECH LLC). \"neighbour-plant-emissions.csv\", collected 2026-11-02. https://wapr.io/csv/neighbour-plant-emissions.csv",
   "payload": "plant-emissions.json",
   "collected_on_basis": "payload generated"
  },
  {
   "file": "csv/water-discharge-permits.csv",
   "licence_file": "csv/water-discharge-permits.LICENCE.txt",
   "rows": 45,
   "columns": 26,
   "collected_on": "2026-11-02",
   "source": "US EPA Enforcement and Compliance History Online \u2014 Clean Water Act permitted facilities reporting sector code 518210 (data processing and hosting)",
   "note": "Clean Water Act permits and general-permit coverages whose holder reports the data-centre sector code, one row per permit, with its permit_type (stormwater_general_coverage marks a stormwater coverage), with compliance state. A stormwater general coverage authorises construction or site runoff and is not an outfall of cooling or process water (discharge.json counts.by_permit_type counts each kind), and several permits can share one location, so rows are not facilities (counts.distinct_facility_locations). A floor, not an inventory.",
   "cite_as": "WAPR \u2014 Worldwide AI Proliferation Registry (BATECH LLC). \"water-discharge-permits.csv\", collected 2026-11-02. https://wapr.io/csv/water-discharge-permits.csv",
   "payload": "discharge.json",
   "collected_on_basis": "payload generated"
  },
  {
   "file": "csv/river-gauges-near-facilities.csv",
   "licence_file": "csv/river-gauges-near-facilities.LICENCE.txt",
   "rows": 998,
   "columns": 19,
   "collected_on": "2026-11-02",
   "source": "US Geological Survey National Water Information System \u2014 instantaneous values from instrumented stream gauges",
   "note": "USGS stream gauges within 10 km of a mapped facility. Exposure, not impact \u2014 distance is not a hydraulic connection.",
   "cite_as": "WAPR \u2014 Worldwide AI Proliferation Registry (BATECH LLC). \"river-gauges-near-facilities.csv\", collected 2026-11-02. https://wapr.io/csv/river-gauges-near-facilities.csv",
   "payload": "rivers.json",
   "collected_on_basis": "payload generated"
  },
  {
   "file": "csv/crane-notices-near-facilities.csv",
   "licence_file": "csv/crane-notices-near-facilities.LICENCE.txt",
   "rows": 680,
   "columns": 23,
   "collected_on": "2026-11-02",
   "source": "FAA Obstruction Evaluation / Airport Airspace Analysis \u2014 Part 77 case archives, all nine regions",
   "note": "Aviation crane notices near mapped facilities, against each site's own 25 km background rate. A watchlist signal, not a finding.",
   "cite_as": "WAPR \u2014 Worldwide AI Proliferation Registry (BATECH LLC). \"crane-notices-near-facilities.csv\", collected 2026-11-02. https://wapr.io/csv/crane-notices-near-facilities.csv",
   "payload": "cranes.json",
   "collected_on_basis": "payload collected"
  },
  {
   "file": "csv/basin-water-stress-by-operator.csv",
   "licence_file": "csv/basin-water-stress-by-operator.LICENCE.txt",
   "rows": 41,
   "columns": 16,
   "collected_on": "2026-11-01",
   "source": "WRI Aqueduct 4.0 baseline water stress (bws), HydroBASINS L6, 1979-2019 baseline \u2014 CC BY 4.0, \u00a9 World Resources Institute. Exposure context, never impact.",
   "note": "WRI Aqueduct baseline water stress of the basins under the mapped sites of the operators listed. Exposure, never usage. Only operators with at least 10 mapped sites are listed (40 of 865 operators); 1,303 attributed sites under the other 825 are not in this file. An operator is WAPR's operator identity: spellings differing only in case or punctuation are one operator, as on the /operator/ pages, and the other spellings are in the aka column. Columns cover the extremely high and high bands and blind sites only; every band for every attributed operator is in operator-index.json.",
   "cite_as": "WAPR \u2014 Worldwide AI Proliferation Registry (BATECH LLC). \"basin-water-stress-by-operator.csv\", collected 2026-11-01. https://wapr.io/csv/basin-water-stress-by-operator.csv",
   "payload": "water-stress.json",
   "collected_on_basis": "payload generated",
   "coverage_state": "computed",
   "operators_listed": 40,
   "operators_not_listed": 825,
   "sites_not_listed": 1303,
   "min_sites_listed": 10,
   "min_sites_basis": "published by the collector (operators_min_sites)",
   "unlisted_at_or_above_floor": 0,
   "grouping": "WAPR's operator identity (build/operators.py op_slug): spellings differing only in case or punctuation are one operator, as on the /operator/ pages",
   "water_rows_not_in_register": 0,
   "register_generated": "2026-11-01",
   "water_frame_vintage": "2026-11-01",
   "frame_matches_register": true
  },
  {
   "file": "csv/substation-voltage-by-facility.csv",
   "licence_file": "csv/substation-voltage-by-facility.LICENCE.txt",
   "rows": 4796,
   "columns": 35,
   "collected_on": "2026-09-15",
   "source": "OpenStreetMap power infrastructure (power=substation, power=plant) via the Overpass API, licensed ODbL 1.0",
   "note": "Every mapped facility with the substations within 5 km of it: how many, the highest-voltage one in reach, the nearest one that records a voltage, and a count by voltage band. A substation near a facility is NOT that facility's substation, and 'none_mapped' describes the map, not the site.",
   "cite_as": "WAPR \u2014 Worldwide AI Proliferation Registry (BATECH LLC). \"substation-voltage-by-facility.csv\", collected 2026-09-15. https://wapr.io/csv/substation-voltage-by-facility.csv",
   "payload": "grid-iron.json",
   "collected_on_basis": "payload generated"
  },
  {
   "file": "csv/high-voltage-substations-near-facilities.csv",
   "licence_file": "csv/high-voltage-substations-near-facilities.LICENCE.txt",
   "rows": 6501,
   "columns": 18,
   "collected_on": "2026-09-15",
   "source": "OpenStreetMap power infrastructure (power=substation, power=plant) via the Overpass API, licensed ODbL 1.0",
   "note": "Substations and power plants at 100 kV and above within 5 km of a mapped facility, each with its nearest facility. The lower-voltage objects are counted in the facility file, not listed.",
   "cite_as": "WAPR \u2014 Worldwide AI Proliferation Registry (BATECH LLC). \"high-voltage-substations-near-facilities.csv\", collected 2026-09-15. https://wapr.io/csv/high-voltage-substations-near-facilities.csv",
   "payload": "grid-iron.json",
   "collected_on_basis": "payload generated"
  },
  {
   "file": "csv/water-quality-near-facilities.part-001.csv",
   "licence_file": "csv/water-quality-near-facilities.LICENCE.txt",
   "rows": 22923,
   "columns": 66,
   "collected_on": "2026-11-01",
   "source": "Water Quality Portal (waterqualitydata.us) \u2014 the joint EPA / USGS / NWQMC store of ambient water-quality samples filed by state agencies and USGS (surface-water samples, field measurements, and field replicates and duplicates of them; blanks, spikes and other quality-control checks, rejected results, other media and non-level statistics are excluded and counted)",
   "note": "Ambient water-quality readings \u2014 one row per station, characteristic and series (a statistic code, or for phosphorus and nitrate a sample fraction, is its own series, so a station can have several rows for one characteristic), the latest value with the day it was taken, its age, and whether it stands apart from that station's own same-season history. Many are years old. Condition of the water, not the effect of any facility on it.",
   "cite_as": "WAPR \u2014 Worldwide AI Proliferation Registry (BATECH LLC). \"water-quality-near-facilities.csv\", collected 2026-11-01. https://wapr.io/csv/water-quality-near-facilities.LICENCE.txt",
   "dataset": "water-quality-near-facilities.csv",
   "parts": [
    {
     "file": "csv/water-quality-near-facilities.part-001.csv",
     "part": 1,
     "rows": 7542,
     "first_row": 1,
     "last_row": 7542,
     "bytes": 7999161
    },
    {
     "file": "csv/water-quality-near-facilities.part-002.csv",
     "part": 2,
     "rows": 7638,
     "first_row": 7543,
     "last_row": 15180,
     "bytes": 7999370
    },
    {
     "file": "csv/water-quality-near-facilities.part-003.csv",
     "part": 3,
     "rows": 7719,
     "first_row": 15181,
     "last_row": 22899,
     "bytes": 7999523
    },
    {
     "file": "csv/water-quality-near-facilities.part-004.csv",
     "part": 4,
     "rows": 24,
     "first_row": 22900,
     "last_row": 22923,
     "bytes": 27125
    }
   ],
   "part_max_bytes": 8000000,
   "reassemble": "the table is in 4 CSV files, each with the same header: read them in the order `parts` lists them and drop the header of every part after the first",
   "payload": "water-quality.json",
   "collected_on_basis": "payload generated"
  }
 ],
 "not_exported": [],
 "withdrawn_not_in_exports": [],
 "instruments_catalogued": 37,
 "instruments_with_csv": 9,
 "instruments_without_csv": [
  "facilities_osm",
  "peeringdb_fac",
  "gppd_plants",
  "lst_gibs_textures",
  "eia930_overnight",
  "degree_days",
  "ercot_gis_queue",
  "hf_creation_feed",
  "identity_endpoints",
  "watchlist",
  "edgar_oem",
  "anomaly_feeds",
  "thermal_ceiling",
  "gridmix",
  "echo_dmr",
  "siting",
  "icis_air",
  "nei_point",
  "eia_gen",
  "wapr_ids",
  "series",
  "transitions",
  "water_causes",
  "lifecycle",
  "prime_register",
  "crw_reef_stations",
  "news_aggregator",
  "usgs_quakes"
 ],
 "files_from_payloads_not_catalogued": [],
 "catalogue_generated": "2026-11-02T03:29:48Z"
}