Three Half-Assed Freeways That Nobody Will Miss


The world’s longest on-ramp, on the west edge of downtown Minneapolis.

We’ve been down this road before. Urban freeways impact their neighborhoods through noise, pollution, and by creating barriers to biking, walking, and local trips. No matter how many sound walls you erect, freeways erode quality of life for a quarter- or half-mile in any direction. Fumes, noise, asthma, speeding traffic… nobody wants to live next to a freeway.


The 4th Street viaduct where it crashes into the downtown street grid.

Another big problem is what to do when the freeway ends. Anytime a high-speed road dead-ends into a low-speed street, we’re going to have problems. On-ramps are the most dangerous areas for anyone on foot or bicycle, because drivers’ perceptions have become acclimatized to fast roads. A good example is the (now-removed) Washington Avenue “freeway” that ran directly into the heart of the University of Minnesota campus, sending cars at 40+ mph into throngs of students. Very bad idea! Freeways are only as good as their network connections. Anytime you try to stop one, you undermine the whole point of a limited-access road in the first place.

Finally, and most paradoxically, freeways create their own demand. If you build a large new road, people will drive more. Take the road away, and people will drive less. Freeway removal projects in other cities have shown that much of the traffic simply disappears, some of it migrating to other roads or to other times of the day, some of it evaporating altogether.

Urban freeways are a mixed bag, to say the least, and we should take seriously the idea that we’d be better off without some of them. Here are my favorite three candidates for creative destruction.

Freeway: Hiawatha / MN-55

hiawatha-aveHistory: This is a state highway, one of the first freeways built in the Twin Cities. You can tell, too. Hiawatha was designed before we knew how to properly build a limited-access road. For years it was a south-east shortcut from downtown to the airport. Ten years ago, we built an expensive LRT along it. Today it has a medicore bike path, wide sidewalks, and carries 20,000 cars each day through a key park, many neighborhoods, and into downtown.

Main Problem: Everybody hates this so-called freeway. Waiting at the lights along Hiawatha is an exercise in frustration. But the main problem is the way Hiawatha divides the neighborhoods on either side. Whether in car, on foot, or on a bicycle, crossing Hiawatha is lengthy, unsafe, and unpleasant. Many times, I’ve watched people dash across the 6-lanes of traffic, running across this quasi-freeway to or from the LRT stations. Devoting all the space around transit stations to a half-assed freeway is no way to capitalize on an expensive investment.

Magic Wand: Waving my magic wand, I’d turn Hiawatha into an everyday 3-lane road. I’d take the Western half of the right-of-way and develop the land. Some of it could be a linear park, some of it could be mixed-use urbanism. Crossing this street would become easy and comfortable. The neighborhoods on either side would re-connect. Streets like Lake, 38th, and 46th would be continuous again, and see large increases in foot-traffic. Quality of life in a large part of South Minneapolis would be improved.


Freeway: Ayd Mill Road

History: This is a long- controversial railroad trench running through Saint Paul. For years it was a little-used shortcut from the South-East suburbs to the western half of the city. In 2002, Mayor Randy Kelly, weilding Saint Paul’s unchecked mayoral powers, connected this so-called freeway to Interstate 35-E over the objections of the City Council. Today it remains a political football, carrying 15-20K cars each day.

Main Problem: Because it’s below grade, Ayd Mill is less of a barrier for the neighborhoods. Rather, its main problem is the traffic it generates and redistributes through the city. The corner of Selby and Snelling is ground zero. Every day for hours, traffic pours off the “freeway” to take a shortcut to I-94, making this key corner highly unpleasant. You see similar effects along Hamline and Lexington.

Magic Wand: There are a lot of ideas for how to fix this road. Some want to cut the road in half, creating a two-lane version with a trail alonside similar to the river road. Others want to build an expensive connection on the North end to I-94, turning this into a more proper “freeway.” If I had a magic wand, I’d choose ‘none of the above.’ Instead, remove all the pavement from this trench and turn it into Saint Paul’s “midtown greenway.” Biking and walking paths, parks, community gardens, all set into a lovely valley. Instead of detracting from the neighborhood, this area would become an asset.


Freeway: 4th Street Viaduct Onramps

4th-st-viaductHistory: I-394 is the last proper freeway ever constructed in the Twin Cities (knocking on wood right now). When they built it, they included a viaduct to build two on-ramps running along 4th Street on the Western edge of downtown. At the time, most of the land around the extension was mix of LULUs: light industrial space, dilapidated buildings, vacant lots, the gargabe burner. Today, this land is prime space for development.

Main Problem: There are really two problems. First, having onramps lead directly onto city streets encourages cars to speed through what should be a walkable city. This is the spot where a man was killed earlier this year.

Worse than that, the 394 elevated causeway depresses everything in its shadow. The backside of the North Loop neighborhood is devalued by the road.

Magic Wand: Just take it down and replace it with a surface boulevard. All the land alongside the road would double in value. If you think the North Loop is amazing today, just imagine it without a freeway causeway hanging over it like a rusty guillotine. “HERC Burner Lofts” anyone?


Letting go is hard to do. When freeway teardowns have been tried in other cities, they’ve been surprisingly successful. All three of these so-called freeways were bad ideas. With a little bit of leadership and creativity, we could correct our mistakes.

45 thoughts on “Three Half-Assed Freeways That Nobody Will Miss

  1. Froggie

    FYI, the 3rd/4th St viaduct has *NOTHING* to do with I-394. I-394 in that area is below-grade. That said, I concur that 3rd/4th St access to I-94 could be converted into an at-grade street.

    I think the residents along Lexington Ave would be against your Ayd Mill proposal. IIRC, they believe it’s taking traffic off of Lexington Ave.

    Hiawatha will still need to be 4 lanes (2 each way). Also, you didn’t mention a *HUGE* context item here with Hiawatha…in that it borders the strip of largely industrial land along the railroad spur that parallels it south of Lake St (and was much more industrial with the Hiawatha Ave EIS was approved in 1985 and the road was rebuilt between 31st and the creek in 1988). It’s not just Hiawatha that separates the neighborhood but this industrial land.

    About the only place you could realistically narrow/reconfigure Hiawatha is north of Lake St.

    1. Froggie

      Nope…about 8-10 years apart. The 3rd/4th St ramps were built ca. 1981. While a few of the bridges and the 5th St Parking Ramp over 394 were built in the mid-80s, I-394 proper wasn’t completed until 1992.

  2. Matt Steele

    I agree for the most part, but my gut says Ayd Mill helps adjacent neighborhoods rather than hurting them. It takes traffic off city streets, although Selby/Snelling is definitely a mess that needs work. I’m more of the opinion that we should give people choices where any outcome is better for the neighborhood. Connecting Ayd Mill to 35E (or keeping it open) should be tied to traffic calming and corridor beautification on Snelling, Selby and Randolph.

    1. Tim Santiago

      Agreed. Selby and Snelling stands to become an even better commercial center should the traffic be rerouted differently off of Ayd Mill. I would add Hamline Avenue to the list as well, as many people take this road from 94 to Ayd Mill. I hate seeing Concordia students running for their lives to and from practice each day. A three lane option with pedestrian crossings is what I envision everytime I drive down this road.

    2. Nathaniel

      Also – we need to get rid of the silly rule that semi-trucks can’t drive on 35E. They drive on Snelling as a result, through neighborhoods and take Ayd Mill.

      1. Xan

        We could make a rule that businesses that rely on semi-trucks should not exist inside the 494/694 loop. That will take them off Snelling. They are only there because of obsolete industrial zoning.

        1. Bill LindekeBill Lindeke

          that’s not a good idea; there’s lots of great industrial property mixed in with the Mpls and StP urban fabric that is crucial to keeping tax revenue & jobs in the city.

  3. Allen

    While road design often splits neighborhoods I wouldn’t place that blame simply on MN 55. the busy LRT line there would do the same just as the highway does. Plus, as others have eluded too, the industrial strip that includes the grain elevators, along the freight rail line was a natural barrier before MN 55 grew and before LRT doubled down on the pedestrian block.

    1. Alex CecchiniAlex Cecchini

      I don’t know that I’d call ~7 minute headways each direction for the LRT “busy” the way that a 4+ lane highway is busy. For context, the number of people traveling on 55 in cars (assume an average of 1.59 passengers per vehicle) is 47,000 – 57,000 per day (depending on location and direction, vehicle counts from MnDOT are 30-36k/day). While this is >2x the passenger count of the LRT, it represents many times the number of vehicles passing each intersection (with more congestion, pollution, noise, etc emitted on the whole). I won’t argue that the LRT line has a good sized ROW that adds a degree of separation from one side to the other (regardless of how busy it is), but the actual tracks and gravel are only 30′ of the ~190′ of Hiawatha’s total ROW (while car infrastructure gets ~100′ at intersections).

      1. Bill LindekeBill Lindeke

        Just an observation, but vehicle counts are a bit more varied than that, more like 20K on the southern portion. A lot of that traffic seems to originate in Mpls, and could be re-routed or transitioned into other modes.

        1. Alex CecchiniAlex Cecchini

          No doubt, I was using the larger northern number for context. The lower number would certainly point to a less “busy” road in total to cross (while still being far more vehicles than a LRT every 7 minutes each direction) but also makes you question the need for the ROW of 55 itself.

  4. Tim Santiago

    I live near Ayd Mill and I’ve been obsessed with its history and its future since I moved here 10 years ago. Personally, I think the corridor should be better integrated with the community. Auto access should remain, but it could be more like the Mississippi River parkways (narrow, slow, no shoulders). It could also be like Midtown with ramp access for bikes and peds alike at each of the ramps. Filling in ramp/loop spaces with park/picnic space would be great for the neighborhood, since many parts of Mac-Grove (like my place near St Clair and Hamline) are far from open space. Additionally, like the Greenway, there is potential for redevelopment opportunities to better integrate the corridor with the surrounding neighborhood. The closed nature of Ayd Mill now, and its proximity to businesses and homes, seem to make it an escape route for criminals, as evidenced by the recent armed robbery of Kowalskis. Creating better accessibility for everyone would undermine this IMO. Great post!

  5. cl

    Matt is right on in my opinion. Ayd Mill being below grade acts like more of a tunnel for cars to get around off the regular grid. The issue is that it doesn’t connect to 94 ( AND that there isn’t a bike lane ). If I had a magic wand I’d connect it to 94 and put in a bike path- which of course is by far the most expensive option. The reality is Highland, Mac-Grove, and parts of Merriam Park have very little freeway access so Ayd Mill does serve a purpose for automobiles getting places, and does so wo/ clogging the streets. The headaches it causes are specifically due to the lack of a 94 connection. That said, while this isn’t the exact neighborhood that brought us the 45mph 35E, it is of a very similar mindset so anything will be controversial. And Bill, don’t hate on those “mayoral powers” in St. Paul, sometimes leadership needs to happen as much as I like to think that the Democratic process of urban development should always yield win/wins it doesn’t always work that way. I’ve either served on the boards of, or shown up as a concerned citizen with several neighborhood groups around St. Paul, and I swear most of the residents that show up (participate) come armed with their militant particularisms and objectives- which usually doesn’t give a rat’s a## about what is best for the city. Which isn’t always terrible, but certainly can make easy things tough without real reason to do so. And don’t touch Hiawatha, as was pointed out it was born from being an industrial “wall” that happened to fall on a line between Downtown and the airport so it’s functional though ugly, bad urbanism. But don’t touch it because when I have my old architect, planner college friends come to town from New York that’s the main part of my Mpls tour. Which is a St. Paul guy’s way of showing outsiders a particular perspective when I have to take them to Mpls. Ha.

    1. Bill LindekeBill Lindeke

      1) Bike path + freeway for cars = FAIL

      2) Live by the mayor, die by the mayor, I suppose. (In this case, the latter.)

      3) Nice use of “militant particularism!”

      4) Have no fear, there are lots of other tour possibilities.

      1. cl

        1. Fine- Connect it to 94, make it 40 M.P.H. and make the regular grid more accomodating to bikes.
        2. I’ll take my chances on having the powers to implement a bold vision vs a potential “rogue” mayor. Besides I like it when mayors don’t need to always defer to the city council (Especially with some of St. Paul’s city councils through the years) This is what elections are for in my opinion.
        3. Thanks. If one get’s through several David Harvey books as an undergrad then one has the right to appropriate some of his great terms as one see’s fit. You’d love how I’ve used that term with audiences that I know haven’t read David Harvey. It’s a brilliant term-very high level of “transferability of meaning(s)”.
        4. I don’t like to change things that worked well in the past. When I find something that works for my intended purpose I like to know it’ll be there over time. St. Paul guy after all. Like good jokes, I get more miles from them than a set of Michelin’s
        Thanks for another good article.

        1. Bill LindekeBill Lindeke

          2. then why is nobody legitimate running against Coleman? part of the problem with the strong mayor system is that nobody is willing to put their neck out to run against an incumbent. that’s what’s happening this year in St Paul.

    2. Xan

      Doesn’t that rail line in Ayd Mill go right into the newly remodeled Union Depot? What a wasted opportunity. And does it not, going the other direction end up in the Greenway leading to Uptown? And we want to connect this to 94 why?

  6. Alex

    Thanks for harping on these mistakes Bill. I believe the World’s Largest On/Off Ramps were built as part of the Northside I-94 segment. Seems like this is by far the most obvious candidate for removal and I’m dismayed that none of the Mayoral candidates are demonizing it.

    Not sure how developable the Developable Land would be, considering it would only be around 50-60′ and of course between an arterial and an LRT line. I think Hiawatha would work fine as a parkway-like space anyway, just make the mediocre bike path into a nice bike path, buffer it and the sidewalk on both sides with trees, and make a slip/parking lane on the east side. By the time anyone is willing to spend money to reconstruct this fairly new roadway much of the industrial land will have been redeveloped anyway (too bad so many of our mistakes are recent).

    I really don’t think the industrial land uses were as much of a barrier as some commenters here have stated – they actually intersperse with the Snelling Ave residential. Why would people have fought the freeway as hard as they did if there hadn’t been continuity to these neighborhoods?

    1. Matt Steele

      Imagine that we could develop the western half of Hiawatha Ave, and the clear zone as Bill suggested: I think it would also be possible to develop on air rights over the existing LRT. This mixed use would a) allow for bigger floor plates creating more incentive for developers, b) buffer the western residential neighborhood from train/car noise. And since this is along a rail corridor, we could skip the parking minimums.

    2. Froggie

      Most of the Hiawatha freeway fight wasn’t in those neighborhoods…it was down along the creek. Some of the plans there either had the creek rerouted or rerouted Hiawatha along the edge of the neighborhood immediately west of where Hiawatha crosses the creek. THAT’S where the freeway fight happened. North of 46th wasn’t as much of a fight…because Hiawatha was already seen as a buffer between the neighborhoods to the west and what was at the time much heavier industrial use along the east side.

      1. Sam NewbergSam Newberg

        My understanding is there were two fights – the one you mention by/south of the creek in the 1990s when the roadway was “improved” along the eastern edge of the neighborhood. The other fight was decades earlier but after homes were taken to “improve” Hiawatha in to a freeway that never happened. The resulting roadway in 1990 or so is neither a freeway nor an urban street, and a terrible compromise that makes nobody happy. The Blue Line was chosen in this corridor because there was extra right-of way along the west side of Hiawatha. It’s just too bad planning for Hiawatha Avenue itself and light rail couldn’t have happened at the same time and with a similar mindset of helping pedestrians and cyclists first. Now we’re faced with retroactive pedestrian crosswalk improvements but so far no fundamental changes to Hiawatha as a roadway. Sigh.

        1. Froggie

          Sam, I was referring to the fights of the ’70s and early ’80s, which focused in the area between 46th and 50th, where the bulk of the “rerouting” was proposed. Well aware of “Camp Coldwater”, but that was long after the FREEWAY plans were dropped.

      2. Bill LindekeBill Lindeke

        I’m not convinced that industrial land per se must be a buffer. Sure, some of it is. But that doesn’t mean that key streets (36, 46, Lake, Franklin, 24th…) can’t run relatively continuously through the area. There are lots of examples of industrial land co-existing with residential neighborhoods. Those are some of my favorite parts of the Twin Cities (e.g. The real barrier isn’t industrial land, but the 4+2 lanes of speeding turning traffic on a not-quite-necessary non-freeway.

  7. David LevinsonDavid Levinson

    I agree with most of the commenters that Ayd Mill is a useful facility, and would be improved by a connection to I-94. I would make it parkway like, with roundabouts instead of traffic signals, so that it would be continuous flow, but appropriate traffic calming so that it would be slow.

    The world’s longest entrance ramps could be at grade, but keep in mind there will still be a lot of motor vehicle traffic that would be at grade and have to be contended with, so the road will still act as a barrier. This may be a minor problem. The other thing to keep in mind about the “shadow of the North Loop” is that the land uses that go there have to go somewhere. So if not here, where? I am perfectly willing to acknowledge that is not the highest and best use given the proximity to downtown, but before we gentrify, we need to figure out what neighborhood is going to take the HERC, bus depot, and the remaining LULUs.

    Hiawatha at 3 lanes is implausible given that it is a state highway, connectivity to the airport, and the level of traffic it actually carries. I don’t think with the LRT on one side and the commercial rail tracks on the other side that a highway is bad use, given state highways will be somewhere. I don’t see demand for land in the next 20 years that requires highway right-of-way, given the existing vacant parcels, land at the stations and the tracks themselves all have plenty of potentially developable air rights that are unexploited.

    An interesting case could be made if you said you wanted Hiawatha LRT to be grade separated to increase speeds and improve safety. This would require elevating (or depressing) the cross-routes (or depressing or elevating the LRT), which could make the Hwy 55 a full-assed freeway instead of half-assed. Sometime you just need to go all-in. Then build air rights development across Hiawatha Avenue at the stations to exploit the new higher accessibility levels.

    1. Matt Steele

      I’ll use Hiawatha as a Minneapolis example of a proposed facility trade. I propose that MnDOT implement the 35W Access Project from 46th St. to Downtown, and widen 62 from the Crosstown Commons to Hwy 77. And in exchange, Hiawatha sees reduction in capacity such as three-laning it as Bill suggests.

      We can de-stroad our communities by improving roads in exchange for improving streets. After all, Hiawatha as it exists is a poor facility for everyone. It’s hostile to the neighborhoods and it’s a worthless way to get from Downtown to the Airport. It’s already faster to take 35W and 62, so let’s focus on keeping those as good roads, and Hiawatha as a good street to encourage local development.

      1. Froggie

        Unless you also plan on some massive rezoning and industrial removal between Lake and 46th, you’re not going to get the full impact you’re expecting.

        1. Matt Steele

          Massive rezoning is absolutely fine with me, since this is along a light rail line it actually seems fitting. Also industrial removal has already been happening, as it’s the natural outcome of land uses moving to their highest and best use.

  8. Walker AngellWalker Angell

    What role does induced demand play in these? Most of us humans tend to take the path of least resistance in some, most, or all areas of our lives. The easier we make driving individual motor vehicles, the more people will drive individual motor vehicles. And this is not, I believe, a viable in the long-term.

    Will connecting Ayd Mill to 94 take traffic off of local streets or will it, in time, just induce more driving by more people? From an auto perspective it makes huge sense to connect it when you look at our motorway system, but on the other hand, we’ve gotten along OK this long without it. Would connecting it create more traffic on 35E south of Ayd Mill? More traffic on 94 west of it? Now we need to enlarge both of these?

    How realistic is traffic calming? I’m a big fan of roundabouts, but will this be like so many higher speed roundabouts in Europe where cars still fly around them at fairly high speed? What kind of noise and pollution would this provide the adjoining neighborhoods? Partial-tunneling might help this aspect, but would still leave future congestion issues.

    How much of the induced demand is longer distance travel replacing what would otherwise be local trips (that could be done on feet, bike, or transit)? Suddenly it’s a lot easier to drive to a restaurant on the west bank or pick up a few things at Target instead of a local store.

    1. Bill LindekeBill Lindeke

      Really good questions Walker. That’s why I’m kind of sanguine about my not-gonna-happen Hiawatha plan. Lots of that traffic seems to be from South Mpls to downtown, induced demand that could be fairly easily shifted to other modes. 20 years from now, this could be on the table. (I’d make it happen sooner, but I don’t really have a magic wand.)

        1. Froggie

          However, Robert Moses Pkwy was A) effectively a spur, and B) had a small amount of traffic compared to what Hiawatha Ave has, so downgrading was more easily done.

      1. Matt Steele

        Part of the reason why I propose capacity exchanges is to reduce the effect of induced demand. If we were to improve Ayd Mill connectivity while simultaneously imposing significant calming on Randolph, Selby, and especially Snelling…. the end result would be roughly the same amount of traffic (since it’s not a net increase in mobility) but better facilities for cars (roads) and better facilities for neighborhoods (streets).

  9. Xan

    3rd and 4th could just as easily turn onto Olsen/55 where a few blocks later they would cross 94 anyway. Those things as they are are absolutely ridiculous. There is no reason to cut through that area, elevated or not.

    1. Froggie

      Historical context: you may or may not have been around ca. 1980, but that area was nothing but dilapidated warehouses when the 3rd/4th St connectors to I-94 were built.

      1. Xan

        I realize that. All those things do is allow people who are leaving downtown to avoid 2 or 3 traffic lights. That’s a pretty ridiculous (and expensive) solution to what is essentially a non-problem.

  10. Xan

    Also, I vote 394 as a runner up. Without it there could be a massive urban park stretching from Loring, through the Parade, to Bryn Mawr to the north and Cedar Lake to the east, and people might learn there is a lake down there too, under the freeway. What a mess it makes crossing 94, and right at the Basilica no less. What a shame.


    I also harbor some disdain for the 3rd/4th Street viaduct through the North Loop. I started a topic over at UrbanMSP about it before learning that the NL master plan recommended (really? wut?) keeping the viaduct. I practiced a little willful ignorance and went ahead and mocked up a plan for the streets in the area that would not only benefit the NL, but also near North by better connecting the two neighborhoods across the 94 trench. I have no illusions that this would ever happen, but I can’t help but dream.

    Here’s the topic:

    Here’s the existing street layout:

    And here’s my dream redo:

  12. Morgan

    The 94 access road needs to be brought down as soon as possible. It is totally unnecessary and a huge barrier to better reclaiming one of the most exciting areas of the city.

    While I don’t have any loved lost for Hiawatha, I have greater issues with Cedar Avenue. I would much rather see a severally calmed Cedar Avenue with about half of the vehicular traffic. If that traffic goes to Hiawatha I would be OK with that. Cedar is over-used and very dangerous.

    1. Matt Steele

      The easiest and most cost effective solution is to get rid of the bridge at Lake Nokomis and do some major calming of the freeway-to-freeway interchange at 62 (or get rid of it).

      Here’s an idea I made last year which would keep neighborhood to neighborhood connectivity but would significantly reduce the appeal of Cedar Avenue as a replacement to 35W:


        A roundabout where Cedar, Old Cedar, Nokomis Pkwy, Edgewater (and Woodlawn?) is a great idea. It’s a perfect way of indicating to drivers that “you’re not on a freeway any more” and would greatly simplify the mess of intersections and restricted movements that plague that area and confuse drivers.

        I’d go a step further and include a second roundabout at the north end as well, where Cedar, 52nd, and Nokomis Pkwy come together. That light is really only necessary because of the current Cedar situation.

        I don’t have an opinion on connecting Nokomis Pkwy to Bloomington at 54th.

Comments are closed.