Cascadia High Speed Rail in Bellingham: thoughts on station locations
Within the next decade or two, construction will begin on a new High Speed Rail (HSR) system connecting the major cities of Cascadia (the cross-border region that spans British Columbia, Washington State, and Oregon). Currently, only the very most rudimentary studies have been completed (see the 2018 Feasibility Study, the 2019 Business Case Analysis and the 2020 Framework for the Future report), but it is already clear that this new HSR system will repay its investment many times over with increased economic opportunities and dramatic reductions in our region’s climate impacts. High Speed Rail in Cascadia will create new levels of interconnection between our region’s three major metro areas — Vancouver B.C., Seattle, and Portland — but it will also be transformative for those of us who live in the smaller cities in the region, like Bellingham. It will connect us to job centers in the big cities, but also encourage more jobs to be located in our own towns. It will also help spur more sustainable land use in our smaller cities, making our growth more environmentally-friendly and discouraging sprawl that is the number one threat to our farmlands and natural spaces.
At this early stage, the official maps of the proposed route are necessarily vague. All of the official studies so far don’t even show any hypothetical track alignments, just various configurations of which cities might be served along the route.
Within the next year or two we will probably learn more information about the business case for various station locations, along with better cost estimates for different route options. But even before we see the results of the next round of study, we can start to make some educated guesses about which route options might make sense, and which station locations will be more desirable than others.
In this post I’ll focus on where a potential HSR station might be located in Bellingham, based on what we know about the constraints of the HSR geometry and reviewing what unofficial analyses are already out there.
Assumptions:
First, let’s quickly review our assumptions about what HSR needs, both on a technical level (specifically constraints on the geometry of the track necessary to support high speeds) and with respect to land use planning.
Speed
The current HSR studies led by WSDOT (with supporting funding from Oregon, BC, and Microsoft) propose speeds of 250 mph (402 km/h) or greater. We should note that these proposed speeds are higher than almost all operating high-speed surface transit systems in the world, hence why these studies are technically called “Ultra High Speed Ground Transportation” or UHSGT. The current studies compare steel-wheel High Speed Rail with more exotic technologies like MagLev — Magnetic Levitation trains that hover slightly above their tracks to attain even higher speeds — and Hyperloop — a hypothetical system using small pods that travel inside vacuum tubes to avoid any air resistance. MagLev is only in use in a few places around the world, and Hyperloop is still many years away from entering regular service anywhere. These latter two technologies could attain higher speeds than HSR, but likely at much greater cost.
For the purposes of this post I’m going to assume that we will end up using tried-and-tested High Speed Rail technology, not MagLev or Hyperloop. Some of the observations below would still mostly apply to MagLev, and somewhat to Hyperloop, to varying degrees.
Is 250mph an attainable goal? Very few HSR systems can reach these speeds, and it is even on the high end of speeds for new lines that are currently being designed around the world. The new HS2 line under development in England between London and Birmingham has a design speed of 250mph, but California’s HSR project under construction is “only” designed for 220mph maximum. For now, aiming to build 250mph rails in Cascadia is probably a worthy goal, but I expect that once detailed planning begins, there will likely be some compromises made in order to keep costs manageable.
The reason we care about proposed speed when we’re discussing route planning is because higher speeds require straighter tracks. To achieve 250mph our tracks can have a maximum curvature of 7000m (22,966 ft). If the tracks have to be curvier (to go around a hill instead of tunneling through it, for example) then the train has to go slower on those segments. 250mph is certainly achievable, but the tracks would be very expensive to construct, relying on a significant amount of tunnels. That being said, the economic benefits would also be greater the faster the train can go.
It’s also important to note that brand-new tracks will be need to be constructed, and in most areas we will need to purchase all new right-of-way. There will be some areas where we can build new tracks alongside existing ones, ideally staying within the right-of-way already owned by the railroad company, saving the costs of acquiring new land. The alignments of existing railroad tracks are always nice and flat, which is important, but they’re only useful for HSR if they’re also very straight. We can also save money by using publicly-owned right of way alongside or in the median of I-5, or along other roadways.
Land use & station location
In addition to the tradeoffs related to the track alignment, we also have to consider a number of tradeoffs when selecting a station location.
- Existing ridership density vs. potential redevelopment
Placing the station as close as possible to existing high density areas such as Bellingham’s downtown core means we instantly tap into the largest possible ridership, and we have the smallest impact on the city’s built environment. Downtown already has strong infrastructure that can be expanded on, and there is still lots of room for new construction. But on the other hand, a station further away from downtown would offer the opportunity to build a whole new dense neighborhood around the new station. But this would take time to construct, and if done poorly it could easily trigger more sprawl not less, especially if the station is surrounded by parking lots instead of homes and businesses.
2. Walkable station areas vs. locations at a transportation hub such as an airport
In larger cities such as Seattle and Vancouver, it might make sense to locate a station at a mass-transit hub, even if the location is less ideal in other ways. In Bellingham, we don’t have much of an existing mass transit system. We have a few small bus stations, but these could easily be relocated to be near the eventual HSR station, rather than building the HSR station near them. The only transit hub that it would make sense to build near would be our airport. We could be sacrificing some opportunity to build more housing and offices near the station, but in exchange we would become an even more valuable node in the regional network, and it would bring many more nonstop flights to new destinations to our airport, which would benefit Whatcom County in its own ways.
3. cost cost cost
As with the track alignment dilemmas, similarly with station placement we might have to forgo an ideal location if it is simply too costly to build or to acquire the necessary land.
In general, however, the UHSGT Business Case Analysis found that in high speed rail projects around the world, these cheaper peripheral station locations usually do not live up to their expected potential.
“Greenfield station areas, which are undeveloped land on the urban periphery or in rural areas, offer an opportunity to spur economic activity at a lower cost than central city locations. Nevertheless, research on tech hub locations indicates that the most desirable locations are in the central city.”
— Exhibit C: Memorandum on “Green Field” Stations on page 55, (page 219 in the PDF)
Cities served
We know the route needs to connect somewhere in the Seattle metro area on one end and somewhere in greater Vancouver on the other end. Any route between those metro areas probably would pass through Bellingham (although not necessarily, as we will see below). Almost certainly, there would be a HSR station in Bellingham (as assumed in all the studies so far) with many daily trains stopping here. But there would also likely be many express trains that pass through Bellingham on their way from Seattle to Vancouver without stopping. For example, most of the scenarios examined in the 2019 Business Plan analysis included 12 daily trains stopping in Bellingham, with an additional 9 express trains from Seattle to Vancouver that would pass through Bellingham without stopping.
In general, new High Speed Rail lines provide economic benefits to smaller cities along the route, not just the major cities at either end of the line. While HSR trains can’t stop too many places along the way without losing their speed advantages, after the massive capital investment to build the lines in the first place, it can still be viable to provide a healthy amount of high speed service to smaller cities along the line.
Why is Bellingham such a challenge?
Building the HSR line through Bellingham will likely be some of the most difficult and expensive sections of track to build because of our uniquely challenging topography.
Bellingham sits at the meeting of two very different terrains. To the north of Bellingham is flat, wide-open farmland that is physiographically an extension of British Columbia’s fertile and densely-populated Fraser Valley. But to the south of Bellingham (including the southern half of the city itself) there’s roughly ten miles of rugged terrain (the Chuckanut Mountains) before reaching the next flat area in the Skagit Valley around the city of Mt. Vernon. These Chuckanut Mountains are part of the foothills of the Cascade Range to the east, and they abut the waters of the Salish Sea to the west. Basically, between Bellingham and Mt. Vernon there is no flat terrain suitable for easy construction of High Speed Rail tracks. Undoubtedly some expensive tunnels and bridges will be necessary in this area.
Possible alignments and station locations in Bellingham
Now we’ve come to the meat of this article. Here I’m going to review some of the station locations proposed in various other blogs and reports, plus some of my own exploration of locations that I haven’t seen discussed yet. The earlier work I’m referring to consists of three main sources:
- The theoretical alignment maps published by Rudy Niederer (from Cascadia HSR, an organization unrelated to Cascadia Rail) in 2015 and revised in 2021
- The excellent “Seattle-Vancouver High Speed Rail” series by Alon Levy on the Seattle Transit Blog in 2017
- The official HSR study and business case analysis from WSDOT published in 2018 and 2019
Looking at each of these sources, plus some of my own analysis, I can see about six possible station locations, some of which have multiple alignments that could access them. Let’s look at each location in detail:
Option 1: Fairhaven
The current Amtrak station in Bellingham is located in the Fairhaven neighborhood. Fairhaven was originally a separate city that merged with other cities on Bellingham Bay to form the city of Bellingham in 1903. Its small historic district is a tourist destination in Bellingham, but it is several miles away from Bellingham’s larger Central Business District. The Bellingham Amtrak station was moved from downtown to Fairhaven in 1995 to be closer to the Alaska Ferry terminal, which has weekly sailings to Alaska.
The 2017–2018 “Ultra High‐Speed Ground Transportation Study” and the 2019 “Business Case Analysis” prepared for WSDOT assumed that the only station location in Bellingham would be located at the site of the current Fairhaven station. However, exploring alternative station locations in Bellingham seems to have been outside the scope of their analysis. Similarly, potential track alignments were also out of scope for these studies, so it’s understandable that they did not go far enough to observe that Fairhaven is not an ideal nor feasible location for Bellingham’s HSR station.
Advantages:
- Fairhaven is already one of Bellingham’s urban villages.
- There are some available parcels for Transit Oriented Development (TOD).
- Fairhaven is already a transit hub, in theory, although Greyhound and the infrequent Alaska Ferry don’t really count.
Disadvantages:
- TOD is relatively constrained, and most of the developable land is susceptible to sea level rise.
- Fairhaven is 3 miles from Bellingham’s Central Business District (CBD).
- Bellingham’s more heavily-used transit hubs are in downtown and in the Cordata neighborhood on the north side of town, far from Fairhaven.
- This location is also the furthest from the rest of the population in the county (such as other growing cites like Ferndale and Lynden) and also far from the planned growth within Bellingham itself, which is largely expected in the north side of town.
Finally, and perhaps most challenging, is that the track geometry approaching Fairhaven is nearly impossible to bring up to HSR standards. Just south of Fairhaven Station the current tracks hug the curves of the shoreline, with little room to add additional tracks and no way to easily straighten any of the sharp curves. These coastal tracks are also susceptible to frequent mudslides which regularly cause the closure of the current Amtrak Cascades service.
There appears to be no realistic way for a new HSR line to access Fairhaven station. In his proposed maps from 2015, Rudy Neiderer proposed that Fairhaven could possibly be served with either a massive 14.9-mile-long tunnel under the Chuckanut Mountains and under Bellingham Bay, or a combination of mountain tunnels along with a 4-mile-long trestle bridge crossing the bay. However, both of these options were removed in Neiderer’s 2021 revised plan.
Of these options, the long bridge seems unlikely to be approved: it would block any shipping traffic to the Port of Bellingham, and would also likely be struck down as a violation of treaty obligations with the local Lummi Nation. A much smaller over-water pedestrian walkway proposed over Bellingham Bay was recently abandoned because of its impact on treaty rights.
The long tunnel would likely be possible, engineering-wise, but very expensive. And if we are going to pay for a tunnel long enough to cross under the entire length of the Chuckanut Mountains and the city of Bellingham, then we would have the flexibility to place the station anywhere in the city, not necessarily at the compromised location in Fairhaven.
Option 2: Airport
The other station location Neiderer proposes is at Bellingham International Airport. In general, connecting airports with a HSR station is a great idea that leverages the strength of both transportation modes. Bellingham’s airport has plenty of spare capacity and could become a kind of overflow airport for Seattle and Vancouver. There’s also a lot of room for TOD near Bellingham’s airport, especially if some additional bridges are built across I-5 to access the buildable land that sits directly across the freeway.
The only major downside to placing the HSR station at the airport is that it’s 4.5 miles from the CBD (even further than the Fairhaven station), making this the most distant station option that we’re looking at, relative to its proximity to downtown. If the airport becomes the location for Bellingham’s HSR station, I can imagine a very strong case for building a Bus Rapid Transit line between the airport and downtown, or even a short streetcar line.
Alignment 2a: Airport via tunnel
Similar to his proposal for the Fairhaven station, Niederer suggests a 14.9 tunnel under the Chuckanuts and Bellingham itself, with an underground station at the airport, and the northern mouth of the tunnel somewhere northwest of the runways near Slater Road. From there it would connect with the existing relatively flat BNSF railroad right-of-way and onward north to the border. Again, this long tunnel would be the most expensive option, but it would be by far the fastest. However, there may be some other options.
Alignment 2b: Airport via Bakerview Road and connecting to an inland route along Highway 9
I heard this proposal from a gentleman at the Cascadia Rail Bellingham Chapter launch party in 2018, and unfortunately I forgot his name!
In this scenario, instead of any major tunneling, the southbound tracks would head east from the airport roughly following Bakerview Road on the north side of Bellingham. East of Hannegan Road the tracks would begin to follow the abandoned “Bellingham Bay and British Columbia Railroad” (OpenStreetMap link) heading northeast out of town to a point where they’d cross the Nooksack River and join up with the existing Northern Pacific Railroad’s Sumas Subdivision (OpenStreetMap link) to head south paralleling State Highway 9.
The advantages of this route are that it follows a relatively flat course, eliminating the need for many tunnels and bridges (with the exception of a new bridge over the Nooksack River). By following Highway 9, this inland route avoids going directly through the Chuckanut Mountains, passing through a narrow valley cutting through the Cascade foothills.
The major downside to this alignment is that it is not very direct. On a train heading from Vancouver to Seattle, after stopping in Bellingham you have to head north again and quite a bit east to join up with the Highway 9 alignment. Compared to a tunnel plowing straight through the Chuckanut mountains, this inland route would inevitably add travel time. But on the other hand if budget constraints make a long tunnel impossible, then the inland route (traveling at 250mph) might be time-competitive with a slower route through the Chuckanuts that requires more twists and turns using fewer tunnels.
So, despite the longer route, the flatter terrain (and potentially lower property acquisition costs by avoiding the heart of Bellingham) makes this probably the cheapest of all the routes we are considering.
Alignment 2c: Airport via I-5
A third option for connecting to the airport would be to use the I-5 right-of-way through Bellingham. Luckily, the airport is right next to the freeway, so we could connect to it quite easily without any tunneling under the runways. In the next two options (Iowa Street and Lakeway) we will look at the I-5 alignment in a bit more detail.
Options 3 and 4: Iowa Street or Lakeway via I-5
In their “Seattle-Vancouver High Speed Rail” blog post, Alon Levy takes a more pragmatic view than Niederer, focusing instead on the most cost-effective route through Bellingham along I-5. This route would require trains to slow down as they pass through Bellingham, but the construction costs would be much lower than Niederer’s plan. Assuming that trains will follow I-5, Levy finds two possible station locations that would be as close as possible to Downtown Bellingham, at the two existing freeway exits that serve downtown: Iowa Street and Lakeway.
Any alignment using I-5 still has the problem that some of the curves in the city of Bellingham (especially those just south of Lakeway, around Exit 252 Samish Way) are still too tight. Levy thinks these can be eased to about at 3,600 foot radius (1100 m) which would only allow speeds of 105 mph.
Comparing the two station locations identified by Levy, Iowa Street and Lakeway are broadly similar, but each has its own strengths and weaknesses.
The area around a potential Iowa Street station is mostly low-rise commercial and light industrial landuse, with a large number of sprawling car dealerships just to the east of the freeway. This would be a good location for rezoning into a dense new mixed-use neighborhood without many negative impacts on nearby residential neighborhoods. One point of concern about the Iowa Street location is that it sits on the banks of Whatcom Creek, a small salmon-bearing stream. Despite the industrial landuse abutting the creek in this area, there have been some efforts to restore the habitat along the creek in recent decades, and any future HSR station and associated redevelopment would need to be careful not to impair this rehabilitation.
Lakeway Drive has the advantage of being slightly closer to the Central Business District, and currently has a mix of residential and commercial landuse. However, on the west side of the freeway sits the York Neighborhood, an area of mostly single-family detached homes that is listed on the National Register of Historic Places. It is unlikely that this neighborhood would be able to take the necessary increase in housing density around the station. If we built a station next to this neighborhood while leaving the landuse unchanged, it would be incredibly disruptive to the housing market in this small area, making those few homes within walking distance of the station extremely desirable and even more unaffordable for Bellingham residents.
Option 5: Bellis Fair Mall / Golf Course redevelopment
Finally, I’d like to propose two other station locations that I haven’t seen discussed anywhere else.
First, we should consider where we could locate the largest possible redevelopment, if we decide that our goal is to get the biggest investment in new housing with as little disruption as possible to existing neighborhoods. (Whether this is necessarily the optimal solution for our climate goals is debatable, but it seems likely that it would be the optimal political and economic solution to get the high speed rail project to succeed).
Conveniently, two of Bellingham’s largest parcels are located right across from each other on I-5, and they are both prime examples of old-school landuse planning that is anathema to any possible success in our fight against climate change. These two sites are: Bellis Fair Mall, and the Bellingham Golf & Country Club.
The Bellis Fair Mall site is approximately 63 acres, and the Bellingham Golf & Country Club is 142 acres. In these roughly 200 acres, at the population density of Barcelona (145 people per acre) we could house 29,000 people, or roughly a third of Bellingham’s current population. We could even save nearly all of the trees on the golf course by building only on the fairways.
For a similar proposal in Seattle, see “Let’s Tee Off for Housing” by Ryan DiRaimo in The Urbanist. One difficulty in Bellingham is that this is a privately-owned golf course, so it’s not quite as “easy” politically as proposals to convert publicly owned golf-courses into housing in places like Seattle.
Even looking at the assessed value of the golf course, we can see that the Whatcom Assessor determined that the market value of the land was $30 million in 2021, but was only appraised at $7.9 million. This is clear evidence that a private golf course is hardly the best use for this prime, centrally-located piece of land. Despite not being easily accessible to the public, the country club has been granted special “open space” tax status, meaning that the county is passing up hundreds of thousands of dollars in additional property tax revenue it could be getting if this land were put to better use with a new mixed-use neighborhood with thousands of new residents.
This problem is not unique to Bellingham. To read more about golf course valuations and taxes, see this article from the Seattle Times: “Are Seattle’s exclusive private golf courses getting a huge break on property taxes? Critics say it’s time to recalculate.”
Also, for any readers who might be fans of the sport of golf who think it’s sacrilege to build a new neighborhood on top of a golf course, I’ll point out that in the Birchwood Neighborhood Plan, the country club is already zoned as “Multifamily Residential, High Density.” It not a question of if we build housing on the grounds of the golf course, it’s a question of how and when.
I see two possible alignments that could serve a station at a redeveloped Bellis Fair Mall and golf course: Alignment 5a via I-5, and Alignment 5b via Bakerview Road and the inland route. Both of these alignments are the same routes we discussed earlier as options for an airport station, with the only difference that the station would be near Meridian Street instead of at the airport.
Option 6: Downtown/Waterfront
Finally, the last station location that we ought to consider is perhaps the most obvious, but also the most challenging: Bellingham’s Central Business District.
Interestingly, neither Niederer nor Levy tried to place a station in Downtown Bellingham, and it wasn’t mentioned in any of the WSDOT studies published so far.
In general, as the 2019 Business Case Analysis observed, city centers are the most desirable station locations, and Bellingham’s downtown is no exception. The city center neighborhood is already has the highest population density of all of Bellingham’s neighborhoods, and it is also the primary commercial and retail hub of all of Whatcom County. Furthermore, it’s adjacent to Bellingham’s large but slow-moving waterfront redevelopment project on the 237-acre site of the old Georgia-Pacific pulp and tissue mill.
Further reading on the waterfront redevelopment:
- “The future arrives on the Bellingham Bay waterfront” by Alex Meacham
- “Is that all there is? Bellingham’s Attempt to Reclaim its Waterfront” by Mark Hinshaw
- “Why Bellingham’s Waterfront Took So Long, and Why It’s a Good Idea” by Floyd McKay
However, in addition to the usual downsides of building a station in a city center (higher costs of property acquisition and difficulty squeezing a station into crowded existing infrastructure), the topography of Bellingham’s downtown is also particularly challenging. The existing BNSF railroad that passes through downtown hugs the bluff along the edge of Bellingham Bay, and has a sharp curve right where it passes downtown, making it too curvy for use as a high speed rail alignment. Additionally the presence of the bluff and the hills immediately to the south of the CBD will likely require some expensive tunneling to manage the abrupt changes in elevation. To the northwest of downtown the existing tracks are relatively straight, and possibly usable as the HSR alignment.
Despite these challenges, the upside is so promising that we should still give a city center station location some serious consideration. A modest one or two-mile-long tunnel under downtown could connect to affordable I-5 alignment discussed above. The northern portal of the tunnel would at the bluff beneath Bay Street between Holly and Chestnut Streets, with the tracks then connecting in a straight line to the existing surface tracks paralleling Roeder Avenue at Central Ave.
Given this track alignment, there seem to be two possible station locations:
One option would be an underground station beneath the three city blocks bordered by Holly, Bay, Chestnut, and Railroad Ave. This location would be the most desirable in terms of proximity to the existing downtown and the new waterfront redevelopment, but it would also be expensive to construct, not least because it needs a station oriented in the NW-SE direction in an area where the city blocks run from the SW to the NE. There are a few historic buildings in these three blocks, but there are also several surface parking lots that could be used for excavation and for the new station entrances. Ideally, most or all of the historic buildings in these blocks could be preserved.
The other option would be a surface station just west of the mouth of Whatcom Creek in Old Town, right at the location of the historic BNSF station which was used by Amtrak up until the construction of Fairhaven Station in 1995. This small historic station could be incorporated as part of the much larger new structure that HSR would require.
Summary/Conclusion
I have to reiterate that I am not a civil engineer, and all of these alignments would need far more research before we know if they’re even feasible, let alone how expensive they would be to construct. But we can at least begin to start to see the differences and potential trade-offs between the various options.
There are a few key takeaways from this analysis:
- A High Speed Rail station at Bellingham’s current Amtrak station in Fairhaven is simply not possible. If and when High Speed Rail comes to Bellingham, the station will be somewhere — anywhere — else.
- Across the whole Cascadia corridor from Eugene, Oregon to Vancouver BC, the Bellingham area is where there is the most uncertainty about alignment, cost, and speed (perhaps with the exception of the approaches to the region’s three major cities of Portland, Seattle, and Vancouver). Whether the eventual HSR project ends up being the most ambitious and transformative plan, or the most modest and frugal version of itself, the range of choices will result in huge differences in what alignment will be chosen in Bellingham. Perhaps there will be a massive, expensive tunnel where trains can travel at their peak speed, or perhaps the route will wind its way along the curvy, hilly right-of-way of I-5, with the resulting time penalty for all trains traveling between Seattle and Vancouver, whether or not they are also making stops in Bellingham.
- The location that is eventually chosen in Bellingham will say a lot about what kind of future we expect for our city, and what kind of future we hope to manifest for ourselves. There are three main ways our future could play out, depending on where the station gets built:
- Do we focus on Bellingham as commercial and transportation hub by building a station at the airport?
- Do we build for a green future while looking to the past, by building a station near downtown and leveraging our existing infrastructure? This would cause the least amount of sprawl, but would also be the most politically fraught, with NIMBYs fighting any new density tooth-and-nail.
- Or do we start with a clean slate and build a new neighborhood from scratch? We might achieve the highest density this way, but this path contains the greatest unknowns. This might be seen as a “defensive” move for Bellinghamsters who want to protect their current neighborhoods from change and to prevent Bellingham from becoming an unaffordable bedroom community.
In a follow up post, I hope to explore some of these future dynamics in more detail, in particular the idea of building a new neighborhood, or even a whole new “hub city” in Whatcom County, as proposed by former governor Chris Gregoire and the Cascadia Innovation Corridor.
What can we do next?
This is an excellent time to get involved in the early stages of the High Speed Rail planning process, here in the Bellingham area but also in cities throughout the Cascadia region.
- Contact your state representatives and senators and ask them to support High Speed Rail.
- Get in touch with local government officials (county and city government, local port commissions, etc.) and ask them to include High Speed Rail in their long term plans.
- Get involved with volunteer advocacy groups like Cascadia Rail!
- Talk to your friends and family and ask them to imagine how High Speed Rail would provide the most benefit to their community.
Do you have thoughts about any of the station locations I explored above, or are there any possible route alignments you think I missed? Please let me know!
