Rail transportation data, from network geometry to live departures. · Head-to-head
NTAD North American Rail Network Lines vs Data.gov US Rail Catalog
Which rail transportation data, from network geometry to live departures. data fits your job: NTAD - North American Rail Network Lines, or Data.gov - US Rail Catalog. API, files, or your warehouse. Daily, weekly, or hourly.
NTAD - North American Rail Network Lines
Data.gov - US Rail Catalog
Coverage, side by side
| NTAD - North American Rail Network Lines | Data.gov - US Rail Catalog | |
|---|---|---|
| Geographic | `STFIPS`, `CNTYFIPS`, `STCNTYFIPS`, `STATEAB`, `COUNTRY` and `FRADISTRCT` on every segment | `has_spatial` flag plus location keywords and per-record spatial extents |
| Temporal | Snapshot vintage carried at release level; no date per segment | `modified`, `issued`, `accrualPeriodicity` and `temporal` intervals per record |
What each contains
Pick by fit, not by loyalty.
| NTAD - North American Rail Network Lines | Data.gov - US Rail Catalog | |
|---|---|---|
| Record identity | `FRAARCID` link ID plus `OBJECTID` and `FRFRANODE`/`TOFRANODE` topology node numbers | Publisher-attributed record `title` with issuing-agency provenance |
| Named subject | `DIVISION`, `SUBDIV`, `BRANCH` and `YARDNAME` railroad-defined naming | `title` and `description` supplied by the publishing agency |
| Attribution | `RROWNER1`-`RROWNER3` reporting marks for operational control, maintenance and ownership | `publisher.name`, e.g. Federal Railroad Administration or Bureau of Transportation Statistics |
| Geography keys | `STFIPS`, `CNTYFIPS`, `STCNTYFIPS`, `STATEAB`, `COUNTRY` and `FRADISTRCT` on every segment | `has_spatial` flag plus location keywords and per-record spatial extents |
| Temporal key | Snapshot vintage carried at release level; no date per segment | `modified`, `issued`, `accrualPeriodicity` and `temporal` intervals per record |
| Classification | `NET` network class, `STRACNET` corridor flag, eleven-value `PASSNGR` passenger taxonomy | `distribution[].mediaType` with `has_download` and `has_spatial` filters |
| Measures | `TRACKS` main-line count, `MILES` and `KM` segment lengths | None - quantities sit inside the datasets each record describes |
What each does better
NTAD North American Rail Network Lines
Ownership carved to the segment. RROWNER1 through RROWNER3 carry the reporting marks of whoever holds dispatching control, maintenance responsibility or property ownership on each link, and TRKRGHTS1 through TRKRGHTS9 stack up to nine other railroads holding rights over the same steel. Sampled rows read like operating paperwork: KPR over the Kelowna subdivision in British Columbia, CSAO on successive Pennsylvania fragments measured in tenths of a mile.
Corridor and passenger intelligence. Every segment carries an S or A flag under the 2018 STRACNET Strategic Rail Corridor Network designation, and the eleven-value PASSNGR code separates Amtrak from commuter operations, Via Rail, Ontario Northland, rapid transit and tourist lines.
Physical texture no index can substitute. NET distinguishes main sub-network from yard tracks, out-of-service lines, passing sidings over 4,000 feet, abandoned grades, trails on former right-of-way, rail ferry connections and industrial leads; TRACKS counts mains; TIMEZONE assigns the corridor clock. All 302,771 features arrive in one flat table keyed by FRAARCID, drawn at 1:24,000 scale or better and measuring roughly 164 MB as raw geometry.
Data.gov US Rail Catalog
Breadth beyond geometry. The roughly 550 rail matches cover ground the linework never touches: the Federal Railroad Administration contributes more than 100 records across its Highway-Rail Crossing Inventory, highway-rail accident series reaching back to 1975, railroad equipment safety and rail equipment accident files; states and cities add grade-crossing, delay and incident tables - 76 records from New York, 47 from Maryland, 28 from Chicago.
Metadata discipline on every entry. Each record carries a publisher name, abstract, issued and modified dates, a declared publication rhythm in accrualPeriodicity (sampled FRA accident files read R/P1M) and explicit temporal intervals such as 1975 onward. Distribution entries are typed by media type, and has_download plus has_spatial flags narrow a search to directly usable or mappable records. See DCAT metadata record.
It indexes the other side of this page. Among its 50-odd BTS records sits the North American Rail Network family itself, alongside rail nodes and intermodal terminals - so the catalog doubles as a finding aid for the very geometry it competes with here.
The verdict
Verdict: sample both, pick by fit. Let the unit of analysis decide. If your question names a piece of steel - who owns the line into this yard, which corridors carry the STRACNET flag, how many mains run past this junction? - NTAD - North American Rail Network Lines already contains the answer as rows. If your question names a landscape - which agencies publish grade-crossing inventories, accident tallies or delay statistics worth joining to that geometry? - Data.gov - US Rail Catalog is the instrument shaped for it, because finding the right twenty records among 550 is the whole job.
On craft the scores sit close: 10/10 against 9/10 on the Datadory rubric. Neither wins by default; the returned samples make the call.
Sample both, pick by fit. See NTAD - North American Rail Network Lines · See Data.gov - US Rail Catalog
Fair questions
Is NTAD North American Rail Network Lines better than Data.gov US Rail Catalog?
Different instruments for different questions. The linework wins whenever the question is spatial: 302,771 segments with owner reporting marks, trackage-rights holders, STRACNET flags and passenger codes attached to each. The catalog wins whenever the question is discovery: roughly 550 rail-matching records from federal, state and city publishers. Sample both, pick by fit.
Which dataset covers more geography?
The linework, deliberately. It spans all 50 US states, the District of Columbia, Canada and Mexico within longitude -156.69 to -59.93 and latitude 14.67 to 64.93. The catalog stays US-centric, though several BTS and FRA series it indexes extend across the northern and southern borders.
Do the two datasets overlap anywhere?
Literally. Among the catalog's 50-odd Bureau of Transportation Statistics records sits the North American Rail Network family itself, alongside rail nodes and intermodal terminals. One side is the described geometry; the other is the description pointing back at it.
Which one should a rail-network analyst pick?
Start with NTAD - North American Rail Network Lines when the unit of analysis is a segment: ownership chains across RROWNER1-3, up to nine trackage-rights holders, STRACNET corridor screening, track counts and mileages. Then widen with Data.gov - US Rail Catalog to pull crossing inventories, accident series and state delay tables that join onto that geometry by county, state or corridor.
Can Datadory deliver both datasets together?
Yes. Request them separately or merged onto one calendar, delivered daily, weekly, or hourly - your call. Every delivery arrives as clean, documented rows with the field dictionary included, so your team evaluates real records before committing to either source. Or take both in one feed.