Building Out Washington’s Passenger Rail Network with ‘Cascades Local Rail’

A diesel multiple unit at Zagreb Central Station. (Dorianbezak from Wikipedia)
A diesel multiple unit at Zagreb Central Station. (Photo by Dorianbezak from Wikipedia)

Last year, when Washington’s east-west passenger rail study came out, the results were mostly uninspiring for a full corridor initially, but one particular section on demand in the Kittitas and Yakima Valleys stood out to me:

Snippet of corridor demand in the Kittitas and Yakima Valleys. (WSDOT)
Snippet of corridor demand in the Kittitas and Yakima Valleys. (WSDOT)

In short, local transit demand exists in communities along where rail lines run that are not currently served sufficiently by local transit options. The current Amtrak Cascades program cannot address unmet needs like these, but perhaps a slightly different program might help. To this problem, I pose the following idea: Cascades Local Rail.

The concept is relatively simple: the Washington State Department of Transportation (WSDOT) could extend the Cascades program in partnership with local transit agencies to provide local intercity service along existing rail lines or rail corridors. As this would be a local service, smaller diesel multiple unit (DMU) trains could be used instead of the full Cascades trainsets. Construction costs could be kept to a minimum by using existing stops or building very simple shelters with right-sized platforms. New railyards to store these trains would also be needed. On corridors already used by the Amtrak Cascades service, rail improvements could be shared between WSDOT and the local transit agency since both services would benefit.

Funding mechanisms could include grants, funds from local transit agencies (replacing some intercity routes), or from tax districts similar to the Sound Transit taxing district. Transit funds from the carbon fee and cap-and-trade legislation that the state legislature is working on would also be a potential funding source.

Any of these changes would require new legislation from the state legislature, but if demand exists and could benefit multiple areas, it’s worth both investing in and giving local communities the opportunity to join in on the kind of transit programs that are mostly focused in the Everett-Seattle-Tacoma corridor.

Below, I will examine a few sample corridors where this kind of service might be useful.

Cle Elum – Ellensburg – Yakima – Toppenish – Pasco

Highlighted potential DMU route
Map of East-West Study Area. Potential Cross Valley route highlighted in Blue.

First, the line east of the Cascades that brought this idea to light is worth considering. A quick Google Maps search of transit options between these cities finds that while adjacent city pairs do not have unreasonable connection times, the overall connections between towns over longer distances soon gets out of hand.

BusRail Estimate
Cle Elum to Ellensburg60 minutes*33 minutes
Ellensburg to Yakima61 minutes50 minutes
Yakima to Toppenish39 minutes16 minutes
Toppenish to Pasco180 minutes66 minutes
*As per Hope connector schedule, not listed on Google Maps.

The comparisons become even more dramatic though, if you compare through-routing (assuming three minutes dwell time per station):

GreyhoundRail Estimate
Ellensburg to Pasco233 Minutes138 Minutes
Yakima to Pasco123 Minutes85 Minutes
Dramatic reductions for intercity travel versus direct Greyhound bus routes.

Given that many of the Greyhound bus routes don’t operate every day, running multiple trains per day at a competitive price could provide a superior service to what is currently offered. In addition, long-term improvements on this route could both build ridership for an eventual cross-state Amtrak Cascades service and over time build enough demand to address the big problem with a cross-state route: the slow mountain route.

GreyhoundRail Estimate
Seattle to Ellensburg180 Minutes242 Minutes
Note that the 176 minutes of the rail estimate is the segment between Auburn and Cle Elum alone.

North Lakewood – Marysville – Everett

The BNSF/Amtrak route runs almost entirely parallel to SR-99 and right through the middle of Marysville.
The BNSF/Amtrak route runs almost entirely parallel to SR-99 and right through the middle of Marysville.

In the Marysville area, a roughly 12-mile long stretch of the Burlington Northern Santa Fe (BNSF) mainline could connect the new Lakewood apartment and shopping complex with Downtown Marysville and Downtown Everett. Traffic issues have often been cited in the past as reasons that Marysville could benefit in being added to the North Sounder line between Everett and Seattle. However, Marysville is currently not part of the Sound Transit taxing and service district.

From Everett, there are currently several Sound Transit buses running south that use carpool lanes, and in the not too-distant-future light rail will reach Everett Station which would make a new rail connection between Everett and cities further north much more attractive.

The Marysville-North Lakewood rail segment not only runs through the middle of town, but adjacent to SR-99 and through all of the major activity centers in town. Marysville could likely support at least two to three DMU-style stations with a single terminus station in Lakewood which has space to build a railyard to store the DMU cars and maintenance bay.

There is probably at least enough ridership to support trains every 15 to 20 minutes as evidenced by the current frequency of the adjacently running Community Transit Routes 202 and 201. To make this kind of frequency reliably possible, the main line through Marysville wound need to be double-tracked and grade-separated for major intersections (e.g., 4th Street and 88th Street NE). This would be a major safety improvement and also benefit Amtrak Cascades heading to Vancouver, British Columbia as well.

A unique benefit of this Cascades Local Rail service is that it would bring passenger rail storage further north and leave trains unaffected by the routine temporary winter closures from heavy rain events and landslides on the waterside segment between Everett and Seattle. Given that multiple agencies would benefit from this arrangement (e.g., Amtrak Cascades, Community Transit, and City of Marysville), funding could be shared among multiple agencies as well. Local officials would likely strongly support this plan on the idea of grade-separating major intersections alone, as freight trains frequently shut down all east-west traffic through the town.

Tacoma – Lacey – Olympia

Potential rail alignment to Downtown Olympia. 1) Sound Transit right-of-way, 2) BNSF mainline and St. Clair Division, and 3) Woodland Trail.
Potential rail alignment to Downtown Olympia. 1) Sound Transit right-of-way, 2) BNSF mainline and St. Clair Division, and 3) Woodland Trail.

The lack of frequent rail service between Olympia and Tacoma seems a bit unfortunate when one considers that publicly-owned rails already extend from Tacoma Dome to DuPont. This trackway is already shared with Sounder trains and eventually Amtrak lines that travel to Portland, Eugene, and Los Angeles. Adding more daily service between Tacoma and the end of the publicly-owned tracks could make a lot of sense, but it would make even more sense if a new segment were built as far as downtown Olympia.

The issue with the current Centennial Station serving Olympia and Lacy is that it’s nowhere near any urban area. It’s not even within the urban growth boundary of Lacey, so it’s not an ideal station for frequent transit service. The least expensive-looking extension to reach Downtown Olympia appears to be using St. Clair Junction and reactivating rail on the Woodland Trail. It could be a somewhat politically-complicated proposition since it’s currently a trail and also crosses a number of main roads on the way in, but this path provides a decent station location for Lacey and much better access to Downtown Olympia. Even if the station remained adjacent to I-5 in Olympia, it would provide far superior access than the current station.

Possible Downtown Olympia alignment highlighted in yellow. Genesee and Wyoming Line shown in brown and Woodland Trail in pink.
Possible Downtown Olympia alignment highlighted in yellow. Genesee and Wyoming Line shown in brown and Woodland Trail in pink.

Most of the Woodland Trail is separated from connecting roads, so it could easily be re-tracked and (with proper fencing) could even still share space with a walking and biking trail. The trouble spots are in Downtown Lacey where the track way would likely need to be elevated and in the crossing of I-5 to reach Downtown Olympia. Agreements with BNSF for rail use between Nisqually and the end of St. Clair Junction and potential use of the Genesee and Wyoming rail in Downtown Olympia to reach the downtown area could also be somewhat expensive, though the rail segments are fairly short. Additionally, land would need to be purchased in Olympia for a station and small railyard storing DMUs. There are several areas where costs could be shared though.

Additionally, rail capacity through the DuPont-to-Tacoma section would benefit Amtrak, Sound Transit, and this Olympia service, so those costs could be shared. DMUs could share Sounder platforms reducing station-building costs. And this program could provide an incentive to the state legislature to replace the infamous Mounts Road rail bridge across I-5 with one that can take more trains and allow faster, safer travel.

Electrification and expansion

For routes that prove successful, switching from diesel-powered DMUs to electric might be a prudent choice. It would reduce the carbon footprint of the services further and the old diesel vehicles could be sent to other services in the state not ready to convert or to corridors ready to experiment with a new DMU route. Multiple electrified small routes along the Amtrak Cascades corridor could help build the case for eventually electrifying the whole corridor which would reduce the carbon footprint not only for Amtrak but for statewide freight services as well.

Previous studies

If rail lines like this seem like a good idea that ought to have been studied before, that’s because they have been. Back in 2010, a WSDOT study was commissioned to analyze the possibility of running a DMU commuter train through Southeast King County from Auburn through Maple Valley and Black Diamond out to Ravensdale. The study found that a line through the area would be feasible with either 30- or 15-minute frequencies. Despite a seemingly positive study, it was never funded. Part of this might be due to a lack of synergy with any other planned use. Perhaps if a long-term plan to restore service through Stampede Pass Tunnel were aligned with this project, it would make more financial sense.

The snub of passenger rail happened even as the 2015 “Connecting Washington” package invested $16 billion in state transportation– just funneling money overwhelmingly to highway expansion. Instead, we get periodic studies, including another in 2020 that reiterated the viability of east-west passenger rail services and a state rail plan that continues to kick around the idea without much of an action plan.

In summary

The above ideas are not meant to be an exhaustive list of all potential local routes that might be worth considering, and the ones listed here may or may not be the most viable options. Given the desire to expand rail to more parts of the state and the lack of good intercity transit options in many areas, it might be worth adding this kind of study in parallel with future high-speed rail studies (in addition to fully funding the planned WSDOT build out of Amtrak Cascades service). These services together would provide a lot of intercity transit connections that simply do not exist today and make reducing car trips in more parts of our state more feasible as we seek to reduce the carbon footprint of Washington State.

We hope you loved this article. If so, please consider subscribing or donating. The Urbanist is a 501(c)(4) nonprofit that depends on donations from readers like you.

Charles is an avid cyclist that uses his bike as his primary mode of transportation. He grew up in the Puget Sound, but is currently overseas living in Japan. He covers a range of topics like cycling, transit, and land use. His time in Tokyo really opened his eyes to what urbanism offers people and has a strong desire to see growth happen in Seattle.

Inline Feedbacks
View all comments
Lael White

These are great ideas and not new. Thomas White has been designing similar for 20 years. He was instrumental on the team that developed the Amtrak Cascades Long Range Plan in response to RCW 47.79.020 (which also includes East/West expansion in its final phases). Readers might want to check out his E/W proposal draft.

Chris B. brings up a point that none of it works unless you have tons of EV rentals at the stations. That’s really the only possible argument against rail expansion that might make any sense but only if considered separate from a well-designed integrated transportation system – that we already have built out massive urban sprawl so how is rail going connect people in outlying areas? People go as far as to say rail just won’t work in the US because of this. The answer is we must build on-demand EV shuttle service as part of rail stations and transit centers. And/or rental service, but that has downsides if it’s private – in reliability and in cost. None of it will work if it’s profit driven – the best solutions will be in response to public need. A public EV shuttle service as part of transit/multimodal transportation could be a huge jobs creator – good-paying union jobs program in training, standards, protocols, vehicle maintenance, servicing, customer care, and driving, development of apps to suit the purpose – van pools for busy commutes or small EV’s for individual trips. Consistent service, all with reduced fare options that are built in to the system like with an Orca card or senior bus pass… apply online or via a station attendant. In other countries with efficient rail systems, people obviously don’t just ride the train and then walk around the stations. They have rural and small town communities too. The many and frequent and reliable connections from the station hubs are key to their success. Electric trolleys like all the ones we tore out, shorter light rail/commuter lines, and EV shuttle need to be firmly embedded in our vision for a well-functioning and climate-friendly transportation system. The models exist.

Here’s a model closer to home:

Mark Foutch

In Olympia there already is an active rail line coming from the BNSF mainline to the south. It passes through Tumwater then downtown where it would join the E-W Woodland Trail route and pass through the 7th Ave. tunnel, emerging on Capitol Lake just below the main Capitol campus. A glass covered escalator could take passengers up to the campus. The line continues up Percival Creek to near South Puget Sound Community College and an industrial park. It used to continue on west to Aberdeen.


I like this idea, but I wonder about the use of Greyhound as your comparison, especially now that FlixBus runs in Washington. They’re often faster and cheaper than Greyhound and essentially run as express buses between moderately-sized cities, which could greatly reduce these estimated bus travel times and make bus service (albeit with increased frequency) the more financially feasible option.

Chris Burke

Some of your time comparisons are cherry picking against truly abysmal bus times. If, as your last paragraph states, the goal is to reduce car trips, then you should be comparing train travel times to car travel times. This makes the train look a lot less desirable. For example, Seattle to Ellensburg takes 242 minutes on the train, but only 102 minutes by car. Furthermore, unless all you plan to do is walk around the station area in Ellensburg, there is no infrastructure in Ellensburg to help you get around. If you drive, you have your car with you. This is generally true of all the routes you propose.

I think a lot more needs to happen before regular passenger rail service could become reality in most of these areas. Like widespread adoption of electric rental cars in every town, say. There are very strong economic reasons passenger rail hasn’t leapt into existence along these routes [remainder omitted due to inaccurate information].

Stephen Fesler

To clarify, Sound Transit commuter rail annual operations is around $60 million and systemwide operations for all modes is $380 million, which does not account for farebox recovery. Other costs like administration, leases, state of good repair brings the system costs to under $600 million per year.

Chris Burke

I agree, it will cost less than a billion per year to operate the trains. However, note that ST spent plenty to prepare the Sounder lines for passenger travel. I can’t find that number, but I’m going to say it was probably on the order of a billion dollars, just like ST’s annual operations in total cost on the order of a billion dollars (2019 was $545 million). Hey, I like rail travel as much as the next fanboy, but outside of major metropolitan areas it’s going to remain a niche form of travel, unless a lot of other changes to society happen concurrently.

Jacob Padgett

For the Lakewood/Marysville/Everett section, there is another thing to consider as well. 

There is a spur line that goes into downtown Arlington splitting off from the BNSF main line in north Marysville that could possibly be turned/used for passenger service (I am fairly certain it is only used twice a week for freight) that would have the future benefit of providing service to the Cascade Industrial Center Arlington and Marysville are developing jointly. 

The spur could join the mainline and continue in to Everett. Both lines could service connections in Marysville, as the spur joins just north of 116th.

This, combined with a dedicated bus on 172nd that ran east/west from North Lakewood to a small station at the spur crossing near 67th (you could put it south of the crossing, nothing there now) would vastly increase service in the already congested area and provide connections to the centennial trail as well. 


Good concept. For Cle Elum to Pasco, I’d recommend calling it “Yakima River Rail,” for the obvious reason that a big portion of the segment is not in the Cascades.

In terms of tying it into a state-funded program, besides Amtrak Cascades, the state also runs several branded intercity bus routes through the “Travel Washington” program.

Statewide Intercity Bus Network.png

Also, for the Yakima River Rail alignment, I’d recommend adding a few summer/fall weekend trips from Seattle to Walla Walla to capture tourists visiting wine country. It’s be exceptionally nice to bring back the spirit of the Spirit of Washington train. Growing up, our farm was along the rail line and I thought it was dreamy to see the Spirit show-up … the possibilities of travel across the state to see all its beauty.


At least in the case Cle Elem->Ellensburg, the 60 minute-33 minute time advantage of rail is an apples to oranges comparison. Considering that the area between the two cities is 100% rural, a train would run without stops. So, a fair comparison would be an express bus, not a milk run that spends half an hour each trip going door to door within Ellensburg.

A bus travels the same freeway that cars do, at the same speeds, so would take the same amount of time, which, according to Google, is 28 minutes, not 60. (

The corridor here doesn’t need rail. What it needs is simply better bus service, particularly more frequent bus service. Ideally, Ellensburg/Yakima/Tri-Cities would be connected by something resembling hourly service. The current service levels is not even close. Focusing on rail or electrified rail is a mere distraction, when the area hasn’t even gotten its bus service in order.