In my last post, we took a look at what it might take to get a high speed rail corridor roughly along the old Oregon Electric (OE) corridor stood up. As a reminder, for ~$4.7 billion we could have both high speed regional service from Portland to Wilsonville, as well as half-hourly intercity service from Portland to Eugene with stops en route in Wilsonville, Salem, Albany, and Eugene. These cities have a population of just over 1 million, and represent 36.2% of the state’s population that lives in incorporated areas.

Spending nearly $5 billion to connect 36.2% of the state’s urban population is still a good investment, but we should take a deeper look as to what it would take to integrate more of the Willamette Valley into our high speed passenger rail network. Because our speculative spine for high speed service is along an existing rail corridor, we can spend relatively small amounts of money to connect a few branches to the main line. But first, we need a brief overview of what a branch line is and which ones could be integrated into our high speed service.

Branch Line Overview

In railroading, a branch line serves a lower traffic area and consist of a short spur off of a mainline. Branch lines used to be fairly common in the early 20th century, but are little used for passenger service currently. Short line freight railroads have typically staked their entire existence on branch line service, usually transloading to a Class I at a small yard where the branch and main lines converge.

Anyways, the best example of a branch line that currently operates passenger service is in the great state of New Jersey. NJ Transit operates the 2.7 mile long Princeton Dinky which runs from the Princeton Junction station on the Northeast Corridor to Princeton Station on the university campus. It’s scheduled to arrive 5 or 6 minutes before departing trains, and leaves about 5 or 6 minutes after incoming trains providing easy transfers between the school at the most important rail corridor in the country.

Without the Dinky, a bus could connect the two – but the Dinky is faster, more reliable, and obviously way cooler. Princeton doesn’t generate enough traffic to justify trains originating and terminating at the school itself, but it does generate enough traffic to justify a 5 minute shuttle for connections. It’s a good service, and something we can incorporate into our service plans.

Branch #1 – Corvallis

The tenth largest city in the state and the home to the largest university (Oregon State), Corvallis is currently very poorly served by intercity transit. While the old OE branch has long been abandoned, a former Southern Pacific (SP) branch connects to the main line at Albany. At 11.6 miles long, it’s not the most direct route – but with a modest top speed of 150 km/h (93 mph), the trip would be just 8 and a half minutes (it’s 17 minutes by car, or 38 minutes by bus today).

The project would cost in the neighborhood of $52 million, with $20 million dedicated for a new station in Corvallis. Lower speeds and single track make it extremely affordable, and with the addition of Corvallis’ 60,000 residents we are now at 38.2% of Oregon’s city dwellers being connected by this HSR project.

Branch #2 – Yamhill County

Out in the western stretches of the valley is wine-rich Yamhill County. Served by an SP branch, the principal cities of Newberg and McMinnville are the 24th and 17th largest in the state. Unfortunately for us, our original plan does not include provision for an intercity station where the branch meets our trunk line (near the Tualatin River in Tualatin), so additional construction at the hypothetical Tualatin Junction station would be required and some of the trips between Portland and Eugene would have to be amended to stop here, rather than Wilsonville. But it would still connect to all the fast commuter services, and given the nature of the Yamhill County line and its continued growth as a Portland/Washington County exurb, this would likely be its primary use.

At 28.4 miles long, it’s quite long for branch-style services and it’s likely that significant construction would have to be undertaken around the Chehalem Mountains where the line snakes through a narrow gorge. A 3 mile cutoff tunnel (single track) would cost about $130 million, and brings the total cost to just under $300 million ($170 million with no tunnel). With provision for five stations (McMinnville, Lafayette, Dundee, Newberg, and Sherwood), we add another 87,000 city residents to our service area and now serve 41.1% of the state’s city population.

Branch #3 – Milwaukie

In what would probably function as an extension of the Yamhill branch, we can serve Milwaukie and Lake Oswego on the other end of the old SP branch. While these cities are firmly in Portland’s orbit, the addition of passenger service on this line would make trips between Milwaukie and Lake Oswego unfathomably faster – as the existing railroad bridge is the only crossing of the Willamette between Sellwood and Oregon City. It also would allow for somewhat shorter trips between southern Portland and the trunk line, since Union Station is not always convenient to travel to.

Lake Oswego and Milwaukie account for about 60,000 more residents, bringing our running total to 43.2% of the state’s city population. The cost of this project would be about $50 million, with $30 million being for stations – which in Milwaukie could integrate a cross platform transfer with the MAX Orange line. The entire run at 150 km/h would take about eight minutes (it’s a 20 minute drive in traffic).

Branch #4 – Lebanon

Lebanon is a small city of just 18,000 but given its proximity to Albany and an already existing right of way to use means the line could be built out for just $45 million. But the real reason for choosing to include this branch relates to future service to Eastern Oregon, as Santiam Pass is an attractive choice for getting over the mountains. In any case, adding Lebanon brings us to 43.8% of the state’s city population and the total travel time would be about 10 minutes as well.

Branch #5 – Cottage Grove

At the south end of the line, arguably outside the Willamette Valley proper a branch extension to Cottage Grove is also worth considering in light of a future southerly extension to the Umpqua and Rogue river valley population centers. While Creswell (about halfway down the line) and Cottage Grove combine for just 15,000 people, the line is owned and operated by a shortline freight (Central Oregon Pacific) which took over operations on the original SP line between California and Oregon, which should make things easier for passenger service improvements.

For a cost of $85 million, and a travel time of about 17 minutes we’re now at 44.4% of the state’s city population being directly connected to our high speed line.

Washington County

And now we have arrived at the most important part of the “branch line build out” section of our high speed rail project. While we could just extend existing commuter rail service to Hillsboro and call it a day, as TriMet’s WES service connects to our service at Wilsonville, there is enough excess capacity on the line (and potential demand) to run trains through from Hillsboro to Salem. The six cities served have a combined population of about 320,000 (plus we could serve 50k additional people in Aloha, which is unincorporated).

Unfortunately, the connections with existing TriMet service outside the 57-TV Highway bus are not great. The Beaverton WES station would have to be moved away from the MAX station, and the alignment through Hillsboro is also a bit away from the existing MAX. This is just a result of the specific locations of the two old rail lines that are being used and is largely unavoidable without incurring astronomical costs.

But in any case, with 200 km/h service on this line and more intensive regional style service patterns the line ought to be fully double tracked which brings costs up to the $250 million range. End-to-end travel times of about 30 minutes blows the current 2+ hour transit time and 50 minute drive time out of the water, and like our SW Corridor High Speed Regional service this would induce a serious amount of demand. Plus, running trains through reduces the need for more station and yard space near Wilsonville (where there is room, but where land values should skyrocket and be dedicated to higher and better uses).

And with the addition of our Washington County branch, we now serve 8 out of the 10 largest cities in the Willamette Valley – the other two of which (Gresham, Springfield) are connected fairly well to local transit in the primary city (Portland, Eugene) which will have a high speed station. And a grand total of 55.3% of Oregonians who live in incorporated places are now served by either the trunk line or a branch (39% of the state’s population).

The Power Good Branches

A tree is only as strong as its branches, and our high speed rail service needs more than just a trunk. With just six minor projects totaling about $800 million ($650 to $770 million for trackage, plus another $150 million or so for trains), a further half-million Oregonians would have high speed train stations in their cities. Sure, many of them would operate well under high speed rail speeds – but the timed branch transfers to the proper high speed rail trunk would make travel vastly easier within the valley.

Corvallis to Portland would become a 45 minute transit trip, down from 3+ hours today. Forest Grove to Eugene would become a 75 minute trip, down from 5+ hours today. So many more trips would become not just possible, but practical and competitive on public transportation. In particular, trips that involve two branches in this schematic are almost all miserably inefficient and thus “driver only” in practice.

Take our Newberg to Corvallis trip above, involving what has to be the most insane deviation I’ve ever seen in a “shortest transit route” option. On the Oregon High Speed Rail System, this trip could be done in about an hour – faster than the quoted drive time despite still involving two transfers. And this gets exactly to the fundamental problem with our existing state-run intercity services: the utter failure to serve any parts of the state not already served by freeways.

Let’s consider POINT, Oregon’s state run bus service. It may provide a great connection between Portland and Eugene in the mornings, and provide much needed service in southern and eastern Oregon but the network is far too sporadic to even really be considered a network. In the valley, it only serves cities which are already connected to each other via Amtrak (and also Woodburn). If you want to get to Corvallis, McMinnville, Lebanon, or Cottage Grove you’ll be forced onto a transit service that can only be described as bare bones and that is likely not well integrated with POINT – meaning long waits. The Eastern service is excellent, and I don’t have any real complaints about it – I just wish it were extended from Bend to the valley. And the Southwestern service is very strange, with the stops in red actually being served by Josephine County Transit (a service that the website says is suspended). It’s only four routes, and two have no useful connections

Putting It All Together

Ultimately, while these various branch lines serve areas with lower ridership demands than the central spine they still are destinations in their own right. Dundee has a booming wine industry, connecting it to the largest market in the state with a 30 minute train ride would be great for business. Lebanon has a medical college, connecting it to Portland, Salem, and Eugene via rail would give students and employees a far better option than driving to get around the state. And of course, Corvallis has Oregon State – arguably the most important single destination in the state to serve well by public transportation given its large student population and need for peak service around things like OSU football games.

High speed rail is about more than just a fast travel time between one end and the other – it’s about the communities that are served. If Oregon wants to be more than just the state where Portland happens to be, our transportation investments have to reflect that. On a personal note, I love the Willamette Valley and having only three cities outside greater Portland that are practical for me to travel to without a car deeply affects my experience of it.

Unfortunately for small city living and loving Oregonians, our department of transportation is unwilling or unable to provide reasonable accommodation. Preferring to widen Main Street turned state highways to allow Jimmy McPortland to get to the coast 5 minutes faster, or putting the entire state transportation planning apparatus in service of two or three freeway mega projects in Portland creates an environment where rural areas and smaller cities get continually less investment.

A high speed rail program would not be cheap, but it’s unclear to me if Oregon can afford to not invest in a new way forward. Every dollar spent now on freeway infrastructure is a dollar investing in a future with continued fossil fuel dependence and sprawl and we need political leaders who are willing to champion bold plans that can actually be achieved. If you want a future in Oregon where transportation is fast, electric, and affordable then start advocating. Emphasize building a high speed spine that allows for easy branch line connections, and most of all educate your representatives on what high speed rail actually means.

What Are We Doing Now?

Currently, plans are slowly progressing for the Cascadia Ultra High Speed Ground Transportation project, but given that this project only includes Portland and is certain to be overengineered to the point of not being feasible to construct. This is a bit of a tangent, but estimates for just Vancouver BC to Portland are about $40 billion for about 450 km of track – that’s $88 million/km (over $140 million/mile). I think this would be money well spent, but it’s also a bit ridiculous. Do we really need to plan for 250 mph (400 km/h) top speeds?

A quick search will tell you that there are a grand total of zero high speed railways that operate 400 km/h service – with a minor exception for the Shanghai Maglev which operates at 431 km/h (268 mph). The Shanghai Maglev is only 18 miles long, and is an express train from the airport – not an intercity service. The current highest top speed on a commercially operated high speed train is the Beijing – Shanghai line in China, with top speeds of 350 km/h (217 mph). As an aside, this line cost about $35 billion (2011 $, $47 billion in 2023 USD), near the top end of the Cascadia plan, makes $1 billion/year in profit, and serves cities with a total population of nearly 50 million. It’s also fully elevated and almost 50% longer than the likely alignment for Vancouver to Portland.

It’s pure hubris to imagine that somehow Vancouver BC, Seattle, and Portland can generate anything approaching Beijing – Nanjing- Shanghai traffic, so construction costs that are 50% higher per mile seem difficult to justify. Maybe we don’t need this service to literally be the fastest in the world, and conventional high speed service at 250 to 300 km/h (155 mph to 188 mph) is what we should be after. Or maybe our state departments of transportation have no clue what other countries do for high speed rail, and just throw out random numbers. Or maybe they don’t realize the difference between a mile and kilometer and figure we can just plan for impossibly fast trains and hope someone will be able to manufacture them.

What I hope you all take away from this series is that high speed rail is a tangible, achievable goal if our elected officials can manage to get things right. Find your legislators email addresses (mine is Rob Nosse) and get them to pay attention to what our state DOTs are actually doing as it pertains to passenger service. It should be possible for almost every city in the Willamette Valley to be connected to each other by a high speed rail service, and all for less than $7 billion. Or we can rebuild a total of 5 miles of freeway in Portland. Agitate, organize, get informed and tell a friend.