TGS US Pipelines ArcGIS Feature Layer

TGS - the subsurface and well-data company behind decades of US exploration mapping - maintains a national pipeline layer titled US Pipelines (SPG), and it runs deeper than mapping data usually bothers to. A server-side count taken during catalog research on 2026-08-21 returned 731,388 line segments, spanning the lower 48 plus Alaska, each with roughly 44 attribute fields riding alongside its polyline geometry.

A CATEGORY field sorts every feature into one of three classes - Pipelines-Gas, Pipelines-Oil or Pipelines-Product - and beneath that sits attribute depth most pipeline maps skip: current operator and owner names, operating status, nominal diameter in inches, liquids capacity in thousands of barrels per day, gas capacity in million cubic feet per day, flow direction, interstate designation, system and subsystem naming, county-and-FIPS geocoding, and a per-record last-modified stamp. Even small abandoned gathering lines retain operator, system name, diameter and modification timestamp, so retired systems stay traceable instead of vanishing off the map.

For oil and gas storage and transportation teams this is the physical network itself: whose steel runs where, how wide, carrying what, and in which direction. Get a sample of this dataset and we return real rows cut to the corridors, commodities and operators you name.

Every observation arrives flat: one polyline segment with its attributes denormalized onto the row, so knowing who operates a line, what it carries and how wide it is takes no joins. Two captured Texas records, exactly as delivered:

# US Pipelines (SPG) -- one row per pipeline segment
CATEGORY           Pipelines-Oil
CMDTY / CMDTY_DES  CRD / CRUDE OIL
DIA_IN             5.0
OPR_NM             LASER MIDSTREAM COMPANY, LLC.
SYS_NM             HASKELL GATHERING
STATUS_NM          ABANDONED
STATE_NM           Texas
MOD_DATE           20221201173417

CATEGORY           Pipelines-Oil
CMDTY / CMDTY_DES  CRD / CRUDE OIL
DIA_IN             5.0
OPR_NM             LASER MIDSTREAM COMPANY, LLC.
SYS_NM             HASKELL GATHERING
STATUS_NM          ABANDONED
STATE_NM           Texas
MOD_DATE           20221201174045

Two habits surface immediately. Retired infrastructure stays named - both rows are ABANDONED 5-inch crude gathering lines on the HASKELL GATHERING system, yet the operator LASER MIDSTREAM COMPANY, LLC., the commodity and the diameter all survive - and every record self-stamps with MOD_DATE down to the second: 20221201173417 decodes to 1 December 2022, 17:34:17, six minutes before its twin at 17:40:45. Lineage at that resolution is what makes change detection possible rather than guesswork.

Roughly 44 fields ship with the layer; eleven verified groups carry most of the analysis. OPR_NM is the load-bearing column - the current operator name behind every by-operator footprint map - while CATEGORY is the three-way commodity splitter and DIA_IN plus the capacity families are the sizing filters. Definitions below were confirmed against delivered records rather than guessed from column headers.

Additional fields on request: the remaining structural columns travel alongside the core set - commodity grouping (CMDTY_TYPE), reported length (LENGTH), the gas-capacity twins (CAP_MMCFD/CAPE_MMCFD), interstate-versus-intrastate designation, BOE region rollup, source-reliability codes (QUALITY/QUALITY_NM), internal identifiers (USPIP_ID/FIDPNET), the major-trunk flag (MAJOR_LINE), creation/launch/online dates and a carried-content description (CONTENT). Their populated behavior is confirmed against live rows when we prepare your sample - the same pass locks column naming for your pipeline.

• Geography: continental United States including Alaska - Web Mercator bounds run roughly -158 to -67 degrees longitude and 25.7 to 71.4 degrees latitude - with per-segment state, county and combined FIPS coding so tabular cuts reproduce the map.
• Temporal window: history carried on the rows themselves - DATE_CREAT, DATE_OF_LA and ONLINE_DTE preserve creation, launch and in-service dates, and MOD_DATE stamps each edit, sampled values decoding to December 2022.
• Granularity: one record per polyline segment - 731,388 of them - with operator, commodity, diameter and capacity attributes held at segment level rather than rolled up per system.

Against the wider Datadory catalog - average quality score 7.81 across all 1,744 datasets - this slice scores 7/10, docked for undocumented code semantics and capacity populated on only part of the network rather than for schema depth.

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

You pick the channel and the cadence; the field dictionary above travels unchanged across all three. Bulk files suit overnight warehouse loads and GIS imports, structured payloads suit event-driven pipelines, and the segment table lands in spreadsheet- and SQL-friendly shapes for analysts who live in a BI tool. Because the grain never changes - one row per pipeline segment - switching cadence later is a settings conversation, not a re-integration project.

• Operator footprint mapping - OPR_NM and OWNER roll 731,388 segments up into per-company networks; the sampled operator, Laser Midstream, still appears on abandoned 5-inch gathering steel that most commercial maps have already dropped.
• Corridor capacity modeling - DIA_IN with CAP_MBBLD (thousands of barrels per day) and CAP_MMCFD (million cubic feet per day) turns the map into a throughput model, and FLOW_DIR values such as BIDIRECTIONAL flag which corridors can reverse.
• Commodity-lane studies - CATEGORY's three-way split plus CMDTY_DES descriptions separates crude, natural-gas and refined-product systems without parsing free text.
• Right-of-way awareness - STATUS_NM keeps ABANDONED lines visible and located, which is exactly what siting, surveying and encroachment checks near old gathering systems need.
• Feature engineering - a labeled national polyline corpus with reliability codes and per-segment attributes drops into routing, risk and infrastructure-screening models unreshaped.

Data scientists & ML engineers rank first in our persona research for this industry: 731k labeled segments form a training-ready spatial corpus for routing and screening models (data scientists use cases). Market researchers & consultants size midstream footprints by operator, system or state straight off the attribute table (market researchers use cases). Developers & data-product builders stand logistics dashboards and mapping products on the segment table (developers builders use cases). Investors & quant researchers read per-operator network reach as an asset-quality screen. Journalists, academics & students cite operator-attributed pipeline geography in infrastructure work without filing a single data request.

• Oil & Gas Storage & Transportation data hub - the pooled industry view, from terminal directories to storage inventories (hub).
• OGIM - Oil and Gas Infrastructure Mapping database - EDF's global GeoPackage: 6.7 million features including 1.2M+ km of pipelines, for when the question stops being US-only (global infrastructure set).
• HIFLD Petroleum Terminals (POL Terminals) - operable US bulk terminals of 50,000+ barrels shell storage or receipt capability; the endpoints where these pipeline segments deliver (terminal layer).
• Global Energy Monitor - Global Oil Infrastructure Tracker - asset-level crude and NGL transmission routes worldwide with ownership chains and status (GOIT).
• Head-to-head: vs HIFLD Petroleum Terminals (POL Terminals) - line geography against terminal geography, compared field by field (comparison).

Full inventory sits on the oil gas storage transportation data hub.