A rail identity crisis in Minneapolis

Map of the proposed Hiawatha, Central, Southwest, and Bottineau lines and stations planned for the city of Minneapolis.

Map of the proposed Hiawatha, Central, Southwest, and Bottineau lines and stations planned for the city of Minneapolis.

This is what the region’s planned light-rail system looks like if you cut it off at the city limits of Minneapolis. It includes the Hiawatha, Central, Southwest, and Bottineau lines. While the first two largely stick to arterial corridors, Southwest (the planned Central/Green Line extension) and Bottineau (the Hiawatha/Blue Line extension) go as quickly as possible to under-populated freight railroad trenches. Again, this is in Minneapolis, where the vast majority of the region’s current transit trips begin and end. Something is wrong with this picture.

As of 2013, there are 11 active stations: Ten of them opened in June 2004 when the Hiawatha Line (now Blue Line) started operation, and the eleventh—at Target Field–opened in November 2009 to coincide with the start of the Northstar commuter service. What’s really surprising is that, despite the addition of three more routes, the number of stations within the city is only expected to double from 11 to 22: Four have been constructed in Minneapolis for the Green Line (Central Corridor), with probably another five expected to be added for the Southwest extension to Eden Prairie. The Bottineau extension of the Blue Line is only planned to add two stations within the city limits, almost completely bypassing the north side. Looking at these future routes, it almost seems like the Hiawatha Line would have been designed to run out of the city even faster, if not for that pesky Mississippi River in the way.

There are a couple of stops that will be just across the border, of course: Many people would be surprised to discover that the existing VA Medical Center station is just barely outside the city, in the unincorporated territory of Fort Snelling (along with the Fort Snelling stop and the two MSP Airport stops). On the Green Line, the Westgate station is barely past the border—one of the two platforms even extends into Minneapolis by just a few feet. On the planned Bottineau extension, stops at Plymouth Avenue and Golden Valley Boulevard are also just barely outside the city.

It’s really remarkable how Minneapolis only gets token access in the Southwest and Bottineau projects. With locally-preferred alignments primarily along freight rail corridors. They are no longer urban transit projects, but are instead commuter or regional services in disguise. And once that fact is acknowledged, it’s worth looking at how commuter and regional rail lines are set up elsewhere. In particular, why do we have to with building a double-track light-rail line right next to a freight line when they both have the same 4-foot, 8½-inch gauge? It would arguably be better for the existing freight lines to be upgraded to double-track and run passenger service over that instead of building something totally separate.

There has been some precedent for running light rail on freight tracks: The NCTD Sprinter in San Diego and New Jersey Transit’s River Line between Camden and Trenton are two examples. Austin’s Capital MetroRail uses essentially the same vehicles as NJT’s service, though it is classified as “commuter rail”. Those are mostly low-frequency services using significant stretches of single-tracked, non-electrified routes—because it was cheaper to build that way, of course. A closer analog to what’s planned for the Twin Cities is likely the UTA TRAX Blue Line in Salt Lake City, which is double-tracked and uses overhead catenary for power.

Frustratingly, the Federal Railroad Administration and Federal Transit Administration have required those lines to use “temporal separation” so that passenger service has exclusive access to tracks during the day, while freight services have exclusive access to track in the overnight hours. The diesel-powered vehicles on the Sprinter, River Line, and MetroRail services are used in Europe on mixed-traffic routes, but somehow we haven’t figured out how to do that in the United States because of overblown fears that freight and passenger trains would crash into each other. Even though some services like the River Line have implemented active signaling systems to automatically stop trains before a crash could even happen, the federal government has balked at the idea of having freight service share track with lightweight passenger rail vehicles during daylight hours.

At the other end of the scale, there are some very busy bus corridors in Minneapolis which deserve to be upgraded. Over time, they’ve been cut out of being potential light-rail lines, though the city is currently pursuing adding streetcar services—something that has caused tension between the city and the Metropolitan Council. Riding the Blue Line down Hiawatha Avenue, it’s easy to see why it was derided as a “train to nowhere” ahead of its opening. There are few obvious destinations along the bulk of the route, at least in comparison to historic streetcar corridors like Hennepin Avenue, Nicollet Avenue, Chicago Avenue, University Avenue, and others. The Blue Line does have a massive traffic generator at its southern end with the Mall of America (think of the daily attendance at the Minnesota State Fair, and multiply that by every day of the year), but it’s still hard to imagine that rail service along any of those streets would carry any fewer people per mile as the Hiawatha corridor. So are streetcars really the right way to improve service along a busy commercial street which likely has higher inherent transit demand?

The Hiawatha Line blew past ridership expectations when it opened, and now ranks as the 5th-busiest light rail system in the country in terms of boardings per mile. Rather than going “lighter” than “light rail” with streetcars, it seems clear that the true urban corridors in Minneapolis should be getting re-examined for fully grade-separated service like a subway (the “heavy rail” that light rail is “light” in comparison to). That concept was largely discarded in the region many years ago, the argument usually being that “Minneapolis isn’t dense enough”. Yet the idea hasn’t really been revisited even with Hiawatha surpassing ridership levels that weren’t expected for another 15 years at least.

A comparison of the sizes of 40- and 60-foot buses, a typical streetcar, and 2- and 3-car Blue Line trains. 3-car trains are current standard weekday service.

A comparison of the sizes of 40- and 60-foot buses, a typical streetcar, and 2- and 3-car Blue Line trains. 3-car trains are current standard weekday service.

Back in the streetcar era, the frequency of service on Nicollet Avenue was every five minutes off-peak and three minutes or less during the busiest parts of the day—it should have been possible to look down the street and see two or three coming your way and a couple of others heading in the opposite direction. Of course, the good thing about sub-3-minute frequencies is that they’re showing up all the time, which makes riders happy. But cramming that many rail vehicles onto a surface line mingling with regular traffic means that any disruption can propagate down the line pretty quickly. Each streetcar also needs its own operator, while a longer train can get by with fewer employees per number of passengers.

I’m not sure transit demand is quite high enough that streetcars would go back to historic levels if they were implemented on Nicollet Avenue again, it would probably get pretty close. The Blue Line is running 3-car trains all day long (though they’re probably only needed at peak times), so streetcars along Nicollet would have to run three or more times as frequently as Hiawatha’s current 10-minute schedule at certain times of day. The frequency can be reduced if multiple streetcars are chained together, but that means larger boarding platforms are needed.

Proper subway lines have higher capacity, faster and more frequent service, and a minimal visual impact on the land above. Longer trains can be used since they don’t have to fit between street-level intersections, and the vehicle types can be switched up if the entire route is grade-separated (traditional subway lines allow passengers to walk between cars because the operator cab can be reduced in size—there isn’t a need for all-around visibility and there’s no risk of crashing into automobiles). Underground trains can operate very frequently and at higher speeds since there isn’t cross traffic to worry about—even curves can be eased to allow faster trains, since they don’t have to fit past buildings. With rights-of-way reduced or even eliminated, more land is available for developers on the surface, meaning that costs can be offset by very intense transit-oriented development.

Washington, D.C.’s Metro system has really been a poster child for that aspect of building underground. They’ve also discovered that underground lines stand up better over time since they aren’t exposed to weather—a major thing to consider in Minnesota with our heavy use of road salt in the winter. Going underground may also allow the deadlock to be broken between the idea of “railstitution” of busy bus routes with streetcars and the alternate concept of “Arterial BRT” (not technically bus rapid transit since it lacks exclusive lanes, but pulling in every other aspect including better stop spacing, level boarding, and off-board payment).

While the momentum behind existing transit projects in the Twin Cities shouldn’t be totally disrupted, it is clear that something has gone wrong with our list of priorities when bridges and tunnels abound on the suburban Southwest LRT corridor, while the far more crowded urban segment along Washington and University Avenues saw tunnels discarded over cost concerns. This isn’t just a problem for the Twin Cities region either—many other light-rail corridors across the country have taken the path of least resistance, and ridership has typically suffered because of it. That’s not acceptable in a future where people will be asked to live more multi-modal lifestyles.


Streets.mn is a non-profit and is volunteer run. We rely on your support to keep the servers running. If you value what you read, please consider becoming a member.

15 Responses to A rail identity crisis in Minneapolis

  1. Froggie August 3, 2013 at 12:42 pm #

    Maybe we can take the term “BRT Creep” and apply it to LRT: “LRT creep”: when tunnels make sense but are cut because they “cost too much”…

  2. Joe August 4, 2013 at 11:20 pm #

    The Green Line’s Extension has more stations in Minneapolis than the CCLRT does… Can we either say St. Paul doesn’t matter or can we extend this to include the sister city as well?

    • Matt Steele August 5, 2013 at 8:22 am #

      I think the commentary is centered around the built environment–what type of land use can support rapid transit–and St. Paul is a much more compatible land use than low-density industrial in Eden Prairie.

      WRT Green Line stations in Mpls, pick any single Green Line station east of downtown and it would likely have higher ridership than Royalston/Van White/Penn/21st St combined.

  3. helsinki August 5, 2013 at 6:07 am #

    A prime candidate for the first subway in Minneapolis should be Lake Street.

    Specifically, a fully underground line with at least seven but not more than 9 stations seems to make most sense. These would be: (1) West Lake [connect to SWLRT], (2) Uptown [Hennepin], (3) Lyndale, (4) Nicollet, (5) Chicago, (6) Bloomington, and (7) Hiawatha LRT.

    This route is approximately 4.3 miles.

    Arguably, the cost would be far lower than that of the recent LRT lines. Costs are of course hugely variable (See: http://blogs.crikey.com.au/theurbanist/2012/02/15/why-do-subways-cost-so-much-to-build-here-than-elsewhere/). Tunnels would likely be built with boring machines (cut and cover construction is cheaper, but very disruptive above ground and therefore usually politically untenable). Our MSP experience with tunnel boring machines (or “TBM”) during the Hiawatha Line construction under the airport cost $110 million for 7 300 ft, or 1.38258 miles (See: http://enr.construction.com/features/transportation/archives/021125.asp). At $79.56 million per mile (in 2002 dollars), this was not terribly pricey.

    Not only would a Lake Street Subway connect the SWLRT and Hiawatha LRT (sorry, Green line extension and Blue Line respectively) at both ends, it would also tie together the main south MInneapolis thoroughfares. Currently major bus routes and candidates for future streetcars (ideal, in my opinion, given that the built environment of Minneapolis was designed around this mode), these routes would perform better with a subway. Not only would a subway facilitate super-fast intra-Minneapolis transportation and massively stimulate development on and around Lake Street, it would also open up city-suburban connections that are now concentrated downtown.

    Sounds preposterously fanciful? I hope not. Far more serious, in my opinion, than the streetcar-in-the-Greenway-trench idea that’s been around for a while.

    • Alex Cecchini
      Alex Cecchini August 5, 2013 at 12:06 pm #

      Why not use a Lake St subway to either continue connecting over to provide single-seat service from Uptown to downtown St Paul (with connections along the way along Selby or turning south and connecting commercial nodes along Grand/Summit, Macalester, and hitting the hospital area before reaching downtown)?

      The Lake-St + Selby/Grand Subway could also interline with a circle line as David proposed http://transportationist.org/2010/04/29/minneapolis_circle_line/ that completes connections to the CCLRT, future Bottineau LRT and Broadway Streetcar, and swings back to hit the SWLRT. In this way, suburban commuters/travelers have options to reach destinations all around Minneapolis in much shorter times (cutting down tangential distances while also running in grade-separation with fewer stops).

      Make real investment in surface arterial connections with dedicated lane aBRT/Streetcar (don’t want to get in to technology discussion..) along corridors like Hennepin, Chicago-Freemont, Snelling, etc.

      I could easily be convinced that a cut/cover Nicollet-Central LRT (underground from ~Lake St up to Central/Broadway with at-grade service north/south of this area) would be a valuable link for city mobility as well as bringing in first-ring suburban needs as well.

      Disregarding costs, what would this look like from a system perspective? What are the missing links? It seems like a good chunk of NE Mpls and St Paul north of University would be cut off from this mobility unless they lived close to Central or Snelling… North Minneapolis would be well-served by Bottineau/Broadway/Circle lines (and even the addition of Van White/Royalston in the SWLRT extension). Thoughts?

      • Alex Cecchini
        Alex Cecchini August 5, 2013 at 3:48 pm #

        Something like this:

        http://goo.gl/maps/ibg8l

        Challenges with this (beyond cost, obviously..)

  4. Boris August 5, 2013 at 10:41 am #

    Rather than an underground rail line on Lake Street, why not go elevated? If we make it classy like the 7 Train in Sunnyside, Queens, it could be a much more affordable option. Likewise, we could use driverless technology like the Skytrain in Vancouver, and make it relatively affordable to operate.

  5. Bill Lindeke
    Bill Lindeke August 5, 2013 at 12:38 pm #

    Frustrating that the political necessities of expediency and geographic / regional “balance” mean that the most ideal places for transit are left by the wayside.

    St Paul’s University Ave will be the TC’s first real test of what actual urban rail transit can look like. Let’s hope we get it right! (So far, results inconclusive.)

  6. Matt Brillhart August 6, 2013 at 3:21 pm #

    Mike, this is an excellent post. I’m disappointed the comment thread thus far has been derailed by a pointless discussion about a tunnel Lake Street (I mean, WHAT?!)

    Before leaving the realm of fantasy entirely, I am kind of intrigued by David’s “Circle Line” concept, once our anti-urban LRT expansion (the next two line extensions anyways) is complete.

    In other news, the Star Tribune just editorialized in favor of re-examining an urban routing of SWLRT and now three mayoral candidates in Minneapolis have called for slowing things down and re-evaluating the situation.

    http://www.startribune.com/opinion/editorials/218173352.html
    http://minnesota.publicradio.org/display/web/2013/07/25/mpls-mayoral-candidate-winton-wants-to-redirect-sw-light-rail-to-nicollet
    http://www.minnpost.com/two-cities/2013/08/minneapolis-mayoral-candidate-hodges-wants-slow-southwest-lrt-plan

    • helsinki August 7, 2013 at 12:03 am #

      Perhaps you missed the end bit (“Proper subway lines” , “Washington D.C.’s Metro”).

      What I took away was that heavy rail on highly used urban corridors close in to the urban core makes more sense than heavy rail flying over and tunneling under highways in Eden Prairie chasing the fickle residents of the metro fringe. Couldn’t agree more.

  7. Ian Bicking August 7, 2013 at 12:04 am #

    Would it be at all possible to build a cut/cover system on streets that are less essential to the current transportation system, and so wouldn’t cause as much disruption? Where the train runs isn’t important, only where the entrances are and how hard it is to get from the entrance to the train (depth also causing problems). Would the tunnels be the same depth regardless of construction, or would they be shallower with cut/cover?

    • Matt Brillhart August 7, 2013 at 9:04 am #

      Ian, I think that is a very good question. For example, why tunnel Nicollet when you can tunnel Blaisdell? You could even still have access portals to the stations on Nicollet, or on the cross-streets between Blaisdell & Nicollet. I wish this option would have been explored more in depth when we were discussing 3C options. Instead, the threat of “open cut tunnels” and years of business disruption on Nicollet were used to sway public opinion away from our only shot at an urban alignment of SWLRT.

      • Froggie August 8, 2013 at 1:39 am #

        That’s not completely fair. The lack of interlining with Central/Hiawatha was also a negative factor against a Nicollet alignment.

        • Alex Cecchini
          Alex Cecchini August 8, 2013 at 9:53 am #

          From a near-term operations perspective, yes. The AA gave too little credit to the number of people who would transfer from SW to CC (or vice versa) given the short walk to the station. If you view the SW routing as up through Uptown and then continuing north across the river and up Central, this may be a better option. You could interline in this tunnel with a line that continues down Nicollet and does something different on the north-side (up University parallel to Central or heading east along University/4th to serve Dinkytown and Stadium Village).

          All sorts of options when not thinking near-term on the operations side. Of course, this could be done with the Kenilworth routing as well as long as we don’t build overly expensive tunnels that lock us in to keeping train service between West Lake and Van White for the foreseeable future.

          • Froggie August 12, 2013 at 7:56 am #

            From a larger perspective, I don’t disagree. But that operates under the assumption that it’s feasible to extend an LRT line across the river into Northeast. No such analysis has been done since the late 1980s, and a good bit has changed since then.

            A very good “what if”. But that’s all it is. A “what if”. Meanwhile, SWLRT is a bit more firmly etched in stone (well, at least in papyrus) at this point.