Natural Gas Pipeline Proximity Screener — US EIA/NTAD Lines

Natural gas pipeline map proximity check: give this actor a list of sites (lat/lon) and it returns the nearest natural-gas transmission pipeline segment for each one — operator, interstate/intrastate type, status, distance in miles, compass bearing, and a right-of-way (ROW) encroachment band (<500ft, <0.25mi, <1mi, >1mi, or NONE). Built on the US DOT National Transportation Atlas Database (NTAD) natural gas pipeline layer (sourced from EIA).

Who this is for

• Pipeline ROW / encroachment teams monitoring third-party construction, grading, or new structures near a pipeline corridor.
• Energy and industrial site-selection teams screening candidate parcels for pipeline proximity before a land option or lease.
• Excavation, HDD (horizontal directional drilling), and utility-locate contractors doing a first-pass desktop check before dispatching a locate crew.
• Phase I Environmental Site Assessment (ESA) consultants who need a documented pipeline-proximity finding for a report.
• Real estate / title / insurance due diligence teams screening a property portfolio for gas-transmission-line exposure.

What it does

For each input site, the actor:

1. Builds a bounding-box (envelope) query around the site at your chosen radius (default 5 miles).
2. Queries the NTAD/EIA natural gas pipeline ArcGIS FeatureServer for polyline segments intersecting that envelope.
3. Computes the true nearest-point-on-line distance (not just centroid distance) for every candidate segment, filters to those actually within the radius, and sorts nearest-first.
4. Emits one row per nearby pipeline segment (up to maxPipelinesPerAsset), each carrying the segment's own distance/operator/type/status plus per-asset summary fields (nearest distance/operator overall, total pipeline count within radius, and the encroachment band).
5. If nothing is found within radius, emits a single explicit "clear" row (clear_flag: true) so a zero-result site is never silently missing from your dataset.

Example input

{
  "assets": [
    { "lat": 29.76, "lon": -95.37, "label": "Houston TX site" },
    { "lat": 38.48, "lon": -98.02, "label": "Rural KS control" }
  ],
  "radiusMiles": 5,
  "maxPipelinesPerAsset": 25
}

Each asset accepts { "lat": ..., "lon": ..., "label": "..." }, a "lat,lon" string, or a [lat, lon] array.

Output fields

What happens when the upstream is down

Every row this actor emits is billable, so it will not charge you for a run that answered nothing.

• Some sites screened, some failed — you get every row, including an ERROR row for each site that could not be screened. Those rows carry null (never false) in clear_flag, nearest_distance_miles and the rest: null means not checked, and must never be read as "no pipeline nearby".
• Every site failed — the run FAILS, emits nothing and bills nothing. A run in which no site could be screened has no answer in it, and returning a dataset full of ERROR rows would both charge you for nothing and hide the outage behind a green run. The failure message names the upstream error.

Two independent copies of the same layer

The primary source is the US DOT NTAD layer on geo.dot.gov. Since 2026-09-16 that host's entire FeatureServer tier has been answering

HTTP 200 {"code":404,"message":"Service not found"}

So each run now gates its source before screening anything:

1. A live integrity check runs first, on whichever copy is being considered — a national feature-count floor, a positive canary (a Houston point that must return at least one segment), and a negative control (a mid-Pacific point that must return zero, which is what catches a spatial filter that has silently stopped being applied).
2. If the primary is unreachable, the run falls over to an independently hosted copy of the same EIA layer published by Esri U.S. Federal Datasets, which must pass that same check before it may serve a single row. Every emitted row then carries backend_failover: true and names the backend, so a failover is never silent.
3. If a copy answers but its integrity check does not hold, the run FAILS rather than falling over. Both copies carry the same EIA layer, so a fallback would launder the same wrong answer.
4. If no copy is reachable, the run FAILS, emits nothing, bills nothing.

The fallback is not a perfect copy, and that is disclosed rather than hidden. EIA's canonical file (last published 2020-04-27) has 32,961 segments and the primary matches it exactly. The Esri copy has 32,892. They agree on every interstate and intrastate line; the fallback is missing 69 gathering segments in 24 systems, mostly offshore Gulf of Mexico plus Wyoming gathering headers such as Jonah and Lost Creek. At a point on the Jonah line the primary reports <500ft and the fallback alone would report 0.7 mi. So when a run is served from the fallback and a site's search area touches one of those lines, the row carries failover_coverage_gap: true, names them in failover_gap_systems, and a no-pipeline row gets clear_flag: null instead of true. Both fields are null when the primary served the run.

Content equivalence between the two copies was measured, not assumed: replayed against this actor's own pre-outage output, the fallback reproduces it exactly (Houston 29.76,-95.37 at 5 mi → 2 segments, nearest 4.62 mi to Kinder Morgan Texas Pipeline Co, Intrastate, Operating; the published 5-site demo → the same 14 rows).

Use as an MCP tool

This actor is Pay-Per-Event (Result-only billing) and discoverable to AI agents via mcp.apify.com — Claude, Cursor, and other MCP clients can call it directly as a "find the nearest gas pipeline to this address/coordinate" tool, chaining it with geocoders or other siting/proximity actors (electric grid, wetlands, floodplain, critical habitat) for a combined due-diligence workflow.

FAQ

How close is the nearest gas pipeline to an address? Geocode the address to lat/lon, pass it in assets, and read nearest_distance_miles and encroachment_band from the result.

Is there a pipeline near my property? Yes — set a radiusMiles that matches your ROW/setback concern (e.g. 0.25–1 mile) and check pipelines_within_radius and clear_flag.

Is there a gas transmission line map API? This actor wraps the US DOT NTAD natural gas pipeline layer (sourced from EIA) behind a simple lat/lon-in, distance-out API — no GIS software required.

What does the encroachment band mean? It's an editorial proximity classification (<500ft / <0.25mi / <1mi / >1mi) based on the distance to the nearest in-radius pipeline segment — useful for flagging sites that need a closer ROW/encroachment review or an operator call.

Data source & disclaimer

Source: US DOT National Transportation Atlas Database (NTAD), "Natural Gas Pipelines" layer (sourced from EIA), a public ArcGIS FeatureServer, with the Esri U.S. Federal Datasets copy of the same EIA layer as an automatic fallback when NTAD is unavailable (see source_backend on every row). This is an informational proximity screen only — not a locate ticket, survey, or official right-of-way determination. Pipeline geometry is a compiled national reference and can lag as-built conditions. Before excavation, siting, or a Phase I ESA conclusion, call 811 / your state one-call center and confirm with the pipeline operator of record.

Pricing

Pay-Per-Event: $8 per 1,000 result rows (Apify default-dataset-item event). Empty/clear results still cost the buyer only the tiny actor-start.