Portland Trimet Portolan Data Catalog (Mirror)

This is a portolan catalog version of https://developer.trimet.org/gis/ - taking the static data layers there and making them fully cloud-native and interactive.

Contents

root

route-stops

Details

Visibility: Public
Owner: Chris Holmes
Created: 3 Aug 2026
Last Updated: 3 Aug 2026

Analytics

28 days

Downloads

3,249

Data served

114.3 MB

Countries

28-day downloads

11 Jul 2026

7 Aug 2026

Analytics

README

TriMet Route Stops

Public transit stops for all bus and rail lines. For stops served by multiple lines there are multiple records in this dataset. 8,314 Point features covering the TriMet service district in the Portland, Oregon metropolitan area, published by TriMet as tm_route_stops and last updated at the source on July 30, 2026.

The same stops as stops, but exploded by the service that calls at them: one row per stop per route-direction, 8,314 rows over 6,316 distinct stops. A stop served by four routes appears four times.

This is the only layer in the catalog that connects a stop to a route, and it carries stop_seq, the position of the stop along its route-direction, which is what lets you reconstruct the order of stops along a line.

Agents: see AGENTS.md for join keys, verified query recipes and the caveats that will otherwise cost you a wrong answer.

🗺️ Open this collection in the data browser →

Preview it on a map, inspect the schema and license, and download the GeoParquet or PMTiles directly.

Quick start

Read it straight from object storage — DuckDB fetches only the byte ranges it needs, so there is nothing to download first:

Or with GeoPandas:

Suggested uses

Building the stop sequence for a route — the ordered list a timetable or a routing engine needs.
Finding transfer points: stops where many distinct rte values meet.
Answering "which routes serve this stop" and "which stops does this route serve", neither of which any other layer here can do.

Limitations and inappropriate uses

Every count is inflated unless you deduplicate. count(*) is 8,314, which is not a number of stops. Use count(DISTINCT stop_id) for stops, or work from the stops collection.
stop_seq restarts at every route-direction. It is meaningful only within a (rte, dir) group. Ordering the whole table by stop_seq is meaningless.
No times, no frequencies. Sequence is order, not schedule. Nothing says how long it takes to get from stop 4 to stop 5.
Directions are not mirror images. The stop list for dir = 0 and dir = 1 of the same route often differs in length, because of one-way streets and stops served in one direction only.

Schema

Column	Type	Description
`rte`	int32	Route number.
`dir`	int32	Direction of line serving this stop. Values: `0` Direction 0; `1` Direction 1.
`rte_desc`	string	Route name.
`dir_desc`	string	Description of route direction.
`type`	string	Type of service. Values: `AT` Aerial Tram; `BUS` Bus; `CR` Commuter rail; `MAX` Light rail; `SC` Streetcar.
`stop_seq`	int32	Stop sequence number.
`stop_id`	int32	Unique identifier.
`stop_name`	string	Intersection or street address of stop.

Column descriptions are TriMet's own, taken verbatim from meta_tm_route_stops.shtml. The same text is carried in table:columns in collection.json.

Visualization

Style	What it shows
`default.json`	One mark per stop per route-direction, colored by TYPE. Because a stop served by several routes appears several times, marks are drawn semi-transparent — the busiest transfer points show up as the darkest spots on the map.
`by-direction.json`	Splits stops on DIR. Along a two-way corridor the inbound and outbound stops sit on opposite sides of the street, so this style shows the paired-stop structure of the network directly.
`by-sequence.json`	Ramps color across STOP_SEQ, the position of a stop along its route-direction. Following the ramp traces the direction of travel, and the gradient makes route ends easy to pick out. Sequence numbers restart at every route-direction, so the ramp is only meaningful when the view is filtered to one route.
`frequent-service.json`	Stops on route-directions flagged FREQUENT, drawn over the rest. This is the stop-level companion to the routes Frequent Service style, and it answers a question the routes layer cannot: which stops a rider can use without consulting a timetable.

The PMTiles layer is named route-stops. Styles reference it as ../route-stops.pmtiles, so they load unmodified against this directory.

Files

File	Size	What it is
`route-stops.parquet`	343.6 KB	GeoParquet 1.1, 8,314 rows in 1 row group(s), zstd, Hilbert-ordered, bbox covering column
`route-stops.pmtiles`	2.3 MB	Vector tiles for display, every feature kept at every zoom
`thumbnail.webp`	48.5 KB	Rendered from `styles/default.json` over a light basemap
`collection.json`	—	STAC Collection metadata

Provenance

Produced by TriMet GIS (4012 SE 17th Ave, GIS3, Portland, OR 97202, gis@trimet.org) and distributed at developer.trimet.org/gis as tm_route_stops.

The originals are linked as assets and are the authoritative copy: Shapefile · KML · Metadata

This collection was produced by converting TriMet's Shapefile to GeoParquet, keeping its native EPSG:2913 (NAD83(HARN) / Oregon North (ft)), and to PMTiles in WGS84 for display. No features were added, removed or edited. The exact commands are in tools/convert.py.

Contributing

This catalog is generated, and both halves are open source: the catalog at cholmes/portolan-catalog-trimet and the data browser at cholmes/trimet-data-browser. If something here is wrong or could be better — a description, a column definition, a query, a style — open an issue or send a pull request. Corrections to the underlying data go to TriMet at gis@trimet.org.

License, and a warning

TriMet distributes these datasets free of charge, but the terms of use do not grant redistribution rights. Section 4 of TriMet's Terms of Use reserves reproduction, modification, distribution and republication of site Content without written consent; the redistribution license in section 5 covers the Web Services API, not these GIS downloads. The collections therefore declare "license": "other" with a link to those terms, rather than claiming an open license the source does not offer.

Practically: use the data, and contact gis@trimet.org before redistributing it or building a product on it. If you need transit data under clear open terms, TriMet's GTFS feed is the better starting point.