Bust open the Bottleneck with buses

Last Thursday I threw a post up on streets.mn that proved using mathematics! that the Hennepin-Lyndale Bottleneck is overbuilt as a roadway and could be reduced by a lane in most segments without risking more than moderate congestion. This result fits with my experience as a frequent user of the Bottleneck at rush hour over a period of many years – the freeway-scale design makes it feel like you’re inching through the facility at slug speed, but in reality traffic moves through the Bottleneck no slower than on other nearby urban streets. On top of the space that could be recaptured from the extraneous lanes, the existing lanes are mostly far too wide, so a lot of pavement can be reduced just by rebuilding the lanes at a more appropriate width.

Scroll down for pics!

So what to do with all this extra space? Anyone who glances over the last year or so of posts here, if still awake, can guess what I say: bus lanes. The thousands of bus riders that travel through the Bottleneck could be sped through daily and be insulated from occasional congestion, and those sitting in cars would watch as the buses made it through faster on this highly visible facility, encouraging them to think of transit as a better option. One of the great transportation weaknesses in Minneapolis is that the Devil’s Backbone (the ridge that comprises Lowry and Loring hills) creates a wall that makes it difficult to travel between Downtown and Uptown. It’s in the city’s best interest to encourage as much of that traffic as possible to take place on space-efficient travel modes such as mass transit.

Here is a map-like graphic I used on the streets post to show which segments could lose a lane without risking much congestion:


Most of the route along Hennepin is green, indicating that a lane can be removed, and the one segment that isn’t is just barely over the 75% threshold. This segment has an enormous amount of queuing space (650′ for the lanes coming from the Bottleneck and over 2000′ for the lane coming from I-94), so I’d suggest that here too a lane be removed and replaced with a bus lane. Here is a lane diagram of the Bottleneck with extraneous lanes removed and replaced with bus lanes where needed:


This provides a bus lane through the entire facility for buses coming to or from South Hennepin, and for much of the facility for Lyndale buses. The bus lanes would also be used for right turns indicated by standard lane symbols, which simply involves a switch from a solid while line to a dashed line shortly before the intersection, along with a right-turn symbol or two. I’ll reproduce an example from British Columbia here, because a lot of Minneapolitans seem to have difficulty picturing this:

The northbound segment just north of Franklin is more tricky, due to traffic exiting to I-94. The bus should still have priority, so I’d recommend striping a short lane for general traffic north of the intersection that is required to merge across the bus lane (yielding to buses in the process) before exiting:

HennepinAtFranklinNorthLegBusLanesThis may seem tight, but there is about 330′ or the length of a downtown block in which to accomplish this, which shouldn’t be a problem for traffic moving at urban speeds (20-25mph). I’ve depicted it within the existing curb-to-curb width, but as the northbound segment is being rebuilt as part of The Project, I’d suggest that the general traffic lanes be reduced to 10.5′ a pop, with the bus lane at 12′, so that it can be reduced 2′ overall and adjacent sidewalk/boulevard made a bit less pathetic/dismal.

As long as we’re discussing the above image, I’ll mention that it depicts the existing southbound roadway, the 34.5′ of which will not be touched as part of The Project. In this space I’ve ruthlessly slashed one of the general traffic lanes and replaced it with an offset bus lane. Additionally I portray the corner with a striped (and bollarded) curb extension, which should be added to every corner on a street with parking as the city’s adopted policy recommends.

It’s possible that something similar could be done on Lyndale at the north leg of the intersection with Franklin. The roadway there happens to be exactly the same width as Hennepin, and the traffic patterns seem to be mostly similar. I think that the city’s policy to reduce VMT is enough to justify replacing a lane in each direction with a bus lane, and additionally this area has been screaming for an extension of the Lyndale bikeway to Franklin (not to mention more pedestrian space, the lack of which forces an awkward dance at the southbound bus stop and taking turns in front of Rudolph’s). If this were to occur, here’s a suggested cross section:


However, there is a much higher volume of vehicles per lane on Lyndale than on the corresponding segment of Hennepin, and anyway this segment of the Bottleneck isn’t going to be rebuilt as part of The Project. It should still be restriped, though, to improve the currently awkward required movements and outrageously overwide lanes. Additionally, the Lyndale bikeway should be extended south by replacing the existing parking lane with a bollard-separated two-way cycle track. Here’s an idea for how this would look:

lyndale-ave-s-restriped-north-leg-at-franklinThis could also improve traffic flow by removing the scary merge of southbound traffic just north of the intersection (technically the traffic from I-94 is supposed to yield but they often don’t). It would do this by replacing one of the lanes of Bottleneck traffic with a bus lane, then giving each stream their own lane at the intersection and banning Bottleneck traffic from turning left (they have had plenty of opportunities to travel in that direction already). Here is a diagram:

SouthLyndaleRestripeLaneDiagramThis plan reduces capacity as measured by square feet of pavement, but I think it will actually improve traffic flow by reducing conflict points and increasing clarity about where to go (note how the northbound lanes now have one clear lane to get in that will take them either to Hennepin or to Lyndale & the freeways). I’ll point out that here too there is enormous queuing space, so dozens of cars could pile up (er, behind, not on top of each other) before impacting an intersection.

If you look back to my overall lane diagram above, there are a few other places where reducing a lane actually has the potential to aid the flow of traffic by making destination more clear. The southbound lanes are an example; currently the four lanes are ambiguous about which will go where, but if you cut a lane it’ll be one destination per lane, from left to right: 15th/I-94, Lyndale, Hennepin. The other spot improved by a lane reduction is the northbound lanes where it splits into Hennepin vs. Lyndale/freeways. Currently one of the lanes splits into both destinations, making the signage confusing. Since Hennepin downtown has been reduced to two lanes inbound, there’s absolutely no need for three lanes to split off the Bottleneck here, so I’d say do one general traffic lane to Hennepin that can be flared out to two at the intersection if necessary.

As you may have noticed, my plan mostly doesn’t actually reduce the number of lanes, but rather replaces some lanes with bus lanes. But there are many other changes needed along the Bottleneck, such as improving the space for bikes and peds and general greening, all of which requires space to do. Amazingly, the current lanes are so overbuilt that even assuming one of the current lanes in each direction is replaced with a 12′ bus lane, by reducing the remaining lanes to 10.5′ a substantial amount of space can be captured (zoom your eyes to the 4th & 5th columns from the right):

BottleneckLaneWidthsNote that each line of the chart indicates only one direction, so that for most of the segments, at least 10′ can be converted to bike/ped space or green buffers. Also, my analysis hasn’t even touched on the pointless “access” road between Groveland and Douglas, which could be eliminated altogether but at least could be substantially narrowed for a significant aesthetic improvement and public gathering space.

Public works has promised to release their latest design for the reconstruction at a public meeting on August 4th. I’m hopeful that they will use the space gained from narrowing lanes to separate the bike & ped streams along the Loring Bikeway, and my best case scenario is that one of the obviously superfluous lanes heading to Hennepin downtown will be dropped. But aside from that I don’t expect any substantial changes, in part because of Public Works’ continued auto-orientation but also because of the rumor I mentioned in the streets post that got so much attention in the comments. Rebuilding the Bottleneck substantially the same would be a tragedy for Minneapolis, not only because it dooms the city to decades more of unpleasant, auto-centric commutes but because it is a huge opportunity lost for a great central public space. If this occurs, rest assured I will expound on this rumor and call to account those responsible for the tragedy.

Postlude, because I really haven’t written enough yet: I of course think that the freeway ramp overpasses should be torn down and replaced with surface facilities like a roundabout or traffic circle, thus freeing up developable space as well as providing room for much greater pedestrian connectivity and cycletracks throughout. The plans shown here operate only within the constraints of the current reconstruction project, which do not allow us to ponder changing the freeway interchange, possibly with the ulterior motive of requiring their continued presence.

A Better Nicollet Mall Survey

I just took the Nicollet Mall survey that I guess they borrowed some interns from Fallon to write, and I have to admit, I had a little bit of trouble understanding the questions. They just didn’t speak my language I guess, they really had trouble seizing the urbane zootgeist that emanates our primer throughfare. So I came up with my own survey, hope you like it:

If Washington Ave doesn’t deserve bus lanes, what does?

According to Hennepin County, around 7,500 bus riders will travel on Washington Ave at peak hour (4:30-5:30 PM) between Hennepin and 35W on an average weekday in the year 2035. For some perspective, that’s about the same amount of cyclists estimated to ride the Washington Ave Bridge on a typical day, which is the busiest location for cyclists in Minneapolis. To be honest, I’m not really sure where Hennepin County got that number, but they mention something about Metro Transit estimating 30 passengers on an average peak hour bus, and if that’s true, that means around 5,000 riders are commuting by bus on this segment of Washington at peak hour today, which would seem to rival the number of cars.

These numbers are fuzzy, obviously, but it seems clear that a large number of people are riding transit on Washington Ave. So why isn’t Hennepin County proposing a layout that would benefit that mode? In fact the four proposed layouts actually make things worse for transit by moving most bus stops to right-turn lanes, where they face the delay of having to pull in and out of general traffic, and where riders face the safety threat of vehicles turning right around the bus.  Besides the sheer number of existing transit trips, there are other reasons that a responsible analysis of options for Washington Ave would include dedicated bus lanes, which I’ll detail below.

Preparing for battle

Preparing for battle

The Gateway Ramp is a major bus layover facility. Part of the fuzziness of the bus rider numbers above, I think, is that they assume average occupancy for the buses running on Washington, about half of which actually pick up and drop off most of their passengers on Marquette or 2nd, so run mostly empty on Washington as they access the Gateway Ramp to lay over. Even if they’re not carrying passengers on Washington, though, it is important to the passengers they pick up later that they not encounter congestion, so their eventual passengers will benefit from dedicated facilities that allow them to be picked up reliably. In addition, the Gateway Ramp has been apparently been designated as a layover facility for an unspecifiedly enormous number more buses so that the City can do what it wants with the Nicollet Hotel block. That likely means that 30-60 additional buses will be soon be traveling on Washington between the Gateway Ramp and Hennepin Ave, relying on a congestion-free route to deliver timely service. (The Gateway Ramp is also a convenient place for the up to 6,000 employees in Ryan’s recently proposed development to catch an express bus.)

Clustering transit and providing dedicated lanes on Washington will maximize the impact of transit investment, create a more legible system, and improve route spacing. Hennepin County’s analysis provides a depiction of the bird’s nest of transit routes on Washington:

Page 13 from DRAFT Traffic Operation Analysis - Apr2013This diagram should set off alarms at Metro Transit. If transportation engineers need to create a diagram like this to understand the network structure, what chance does a lifelong suburbanite retiree who just bought a condo on Washington have? Bus lanes would offer reassurance to confused riders that yes, they can catch a bus on this street. If Metro Transit were to use the bus lanes for its various archaically routed local services that use Washington for a portion of their trip already, it would be able to focus shelter improvement money on this one street instead of spreading it between several (not that there is any apparent shelter improvement on the downtown segments of these routes currently). This would also have the effect of maximizing frequency (a rider traveling between 7 Corners and Hennepin could catch any of 3 routes), adding legibility (riders would not have to memorize where the 7 & 22 turn off of Washington), and spacing (the thousands of new housing units being added to the Mill District face a long walk to convenient transit service).

These advantages are recognized and supported by the City of Minneapolis, which recommends reorganizing downtown transit to cluster along three corridors they call spines (a biological metaphor that becomes less apt the more spines you have). The buses running closest to the riverine edge of downtown are left as they lay, probably out of inertia. Yet these services would benefit from “spining” too, and perhaps more, since lower-frequency services will gain more from higher effective frequencies due to clustering. I have made a table of the number of buses at the peak hour on Washington Ave by segment and direction, based on data from Hennepin County, but adding a spine scenario, which assumes the 3 and the 7 proceed along the length of the corridor and the 22 travels on Washington east of Hennepin (it also adds the 14 west of Hennepin as it travels today but was not included in the Hennepin County data for some reason; I’d add that it may make sense to add the 14 to this spine west of Chicago or 11th Ave S).

pm peak bus load avg headway washingtonIn the segment where reconstruction is imminent (outlined on the table), average headways are expected to be three minutes or less at peak hour in 2035, and are currently under one minute for all but one block in the westbound direction. The spine scenario brings average headways in each direction to under 3 minutes, and by 2035 both directions of Washington will carry a bus  less than every 2 minutes. These are really substantial bus volumes, unlikely to be exceeded by any Nicollet Mall, Hennepin, or the main E-W bus spine. So why are those streets candidates for bus facilities (even if they’re half-assed ones), but not Washington?

Of course, most of this service could cluster on 3rd or 4th Sts instead of Washington, but those seem to have fewer advantages and more disadvantages. Briefly, Washington connects better to the remainder of the routes on the east and west ends, which means less delay caused by turning. 4th St is an awkward distance from the LRT stations on 5th St, too far for first-time users to see the transfer stop from the station, and also too far to really work as combined effective frequency, yet not spread enough for the larger portion of downtown to benefit. Washington is convenient to the two fastest-growing neighborhoods in the state, and with this effective frequency could provide easy access for the residents of these new dense buildings to regional transit (LRT or Highway BRT). Finally, in order to fit (ideally two) bus lanes on 3rd or 4th, you need a curb-t0-curb width that leaves too little space for sidewalks. Currently the sidewalks are reduced to 10-12′ on these streets, whereas the wider right-of-way on Washington would allow for ample sidewalks in addition to the bus facilities.

But assuming we continue our practice of ignoring the huge current use and future potential of bus transit, why should we prioritize transit rather than bikes or cars? Well, Washington is actually not as connective for cars & bikes. OK, there are a pair of big freeways on the each side of Downtown that make it a convenient route for cars, but even those are duplicated by other exits a few blocks away (or will be soon). In terms of surface connections, it’s also not very useful for cars. As I’ve argued before, and as residents tend to agree, Cedar is inappropriate as an auto commuting route. North Washington has some destinations, but is superseded by 2nd St by the time it gets to Plymouth Ave (certainly North Loop destinations don’t generate enough car trips to justify 3 lanes).

For bikes, too, Washington is not ideal as a through route. Of course the U of M is a big destination, but to reach it from Washington you need to turn at least twice and/or carry your bike up the stairs behind Willey Hall. A better U of M connection to Downtown is CPED’s (possibly abandoned) proposal for a path in the trench that would connect to the LRT trail at Curry Park, which would maximize connectivity and have the greatest separation. Even if you could somehow create a surface route between Washington and the U of M, it would likely be slower than a trench route and the LRT trail because of the left turn and all the stoplights. Anyway, the LRT trail is likely to be at least as important a source of bike trips into downtown as the U of M (or at least that’s the goal), and Washington both connects poorly to it and is out of the way for people trying to access the core (requiring two left turns).

3rd St would work best for a regional bike facility that goes through downtown (unlike West River Parkway, which bypasses it), especially because 3rd St offers connections to the Northside that Washington doesn’t. As noted above, Washington itself kind of peters out as a frontage road to I-94 north of Plymouth Ave, but even the parts that are there will be difficult to retrofit for bike facilities – certainly it wouldn’t be able to do any better than duplicate the lanes that exist on 2nd St N. 3rd St, on the other hand, connects directly to the LRT trail on the east, and with some additional cantilevering of the sidewalk along the 4th St Viaduct could connect directly to the Cedar Lake Trail and be extended across the Cut and through the Interchange to the bike lanes on 7th St N, basically the main bike route between Downtown and the Northside (it could also connect to the off-street trail that could logically be placed along Olson Hwy, but doesn’t seem to be in anyone’s plan for some reason).

washington or 3rd bike routes

Of course people will still want to use bikes and cars to access destinations on Washington Ave. Bus lanes actually work really well for this since they are used heavily primarily at the peak hours, and at other times they can be flexed for other uses, including parking. A bus lane works much better for bikes than a general traffic lane because there are typically far more gaps between buses than cars. At rush hour on Washington you wouldn’t want to bike the length of the street, but the minute gap between buses will allow you to bike on one of the ample adjacent facilities on 1st, 2nd, 3rd, or 4th, then up one of the north-south bike routes (for example 1st, Hennepin, Nicollet, 4th, 5th, Portland, Park, or 11th), and then the one or two blocks remaining to your destination. I would suggest 16′ shared bus-bike lanes, separated by a solid white line except for the 150′ or so before right turn intersections, and symbolized by a diamond. 5-6” advisory bike lanes could be striped to guide cyclists toward the left side of the lane to minimize the amount of leap frog, and a 1-2′ mountable curb could be placed between the Shared Bus-Bike Lanes (SBBLs) and general traffic lanes to provide a buffer for cyclists and to discourage the spread of congestion by stupid or greedy motorists.

Would all this fit? For the most part, yes:



SBBL configuration

SBBL configuration

You can add SBBLs and fit within the right-of-way and have sidewalk space at least as wide or wider than most of Hennepin County’s proposed layouts and what is there now. SBBLs are an ideal compromise solution that provide for the existing and future demand of cars and transit, but also provide a more comfortable space for bikes and opportunities for parking. It is a shame that Hennepin County only does planning for transportation by car instead of transportation for all, or there may have been a possibility for a holistic solution that would be appealing to a larger group rather than their special-interest focused layouts.

If a street that carries 15,000 transit passengers in a typical day – as many as some light rail lines in the US – doesn’t deserve dedicated bus lanes, what street does? Is it realistic to expect that the maybe 50 miles of light rail being developed in the Twin Cities will be able to shift the millions of daily trips here to a lower-emission mode? Buses are crucial to our current transit system and will continue to be crucial to our future transit system, which represents our best hope for achieving environmental and equity goals through transportation policy. If one of the cycle track options is built, I will certainly enjoy riding it to Grumpy’s every once in a while. But if the Washington Ave process means that the Twin Cities is just shifting from focusing all transportation planning on making it nice to drive to focusing all transportation planning on making it nice to bike, I’m taking the first bus out of here to someplace that plans transportation comprehensively, without mode bias, and with an eye towards societal goals.



Pedestrians prohibited on streets.mn

Today on streets.mn I examine Bloomington’s latest bungle. Bloomington is the suburb I lived in the longest, it has some really nice topography for the Twin Cities that great parks like Hyland, Moir, and the Minnesota River recreational area takes full advantage of, and it has some decent bones in terms of the integrity of their street grid. But man do they fuck up a lot for pedestrians. See for yourself on my post (or I suppose you could go to Bloomington and walk around, but then you might die and I’d feel bad).

Oh yeah and I tacked on this graphic for some reason:

Because all of these signs are bad for pedestrians except the No Turn on Red

Traffic Control Device for Non-Vehicular Traffic Vehicles

Encountered by a traveler to the Champs de Targets

How exactly does a pedestrian comply with a stop sign?  Do both feet need to be firmly planted, approximately parallel so as not to suggest movement, in order to come to a complete stop?  Is a pedestrian at a stop sign required to turn his or her head in each direction, or does a nonchalant scan of the field of vision suffice?

Pedestrian activity has been compared to such graceful movement as ballet, clouds, and animal migration.  Why wasn’t a simple LOOK OUT FOR THE TRAIN YOU IDIOT sign good enough?  Why do we need to be subject to the same confining rules as our twitchy vehicular brethren?

Drunk Driving & Other Delights

George Smith… A name that will live in infamy…

1897:  Officers make first drunk driving arrest

On this day in 1897, a twenty-five-year-old London taxi driver named George Smith became the first person ever arrested for drunk driving after slamming his cab into a building.  Smith later pled guilty and was fined twenty-five shillings.  In the United States, the first laws against operating a motor vehicle while under the influence of alcohol went into effect in New York in 1910.  In 1953 Robert Borkenstein, a former Indiana state police captain and university professor, invented the Breathalyzer.  Easier to use and more accurate than earlier devices, the Breathalyzer was the first practical device and scientific test available to police officers to establish whether someone had too much to drink.

From my History Channel page-a-day calendar today.  Btw, apparently 25 shillings in 1897 would be £114.27 in 2011, or $182.87 at current exchange rates.  Luckily the penalty is a bit steeper these days.

When is National Year Out?

Tonight is National Night Out, the one night a year where we block cars from certain streets so they can be used in a way that actually enriches the community.  The City has a list of all the “official” NNO events, and it’s fun to look at the column that lists the planned activities to see what people would use their streets for if they didn’t have to fear for their lives every time they set foot on it:

  • Grilling/kids games
  • sit in street, watch planes
  • kids riding bikes in the street
  • Johnny Cash tribute band
  • Welcom[ing] new neighbors
  • hanging out for adults, sidewalk chalk/colors for kids
  • Plant/book exchange
  • a lot of laughter
  • Gospel rap
  • Barbecue, pinata, water balloon toss
  • kids running around
  • discuss cute dogs
  • Basketball Tournament
  • Share how things are in the n’hood
  • Hopefully visit from fire engine !
  • Kids “own” the street
  • Zumba
  • Beers, Brats, Buddies
  • self defense demo
  • celebrate life of a long-time neighbor who passed away recently
  • schmoozing
  • chicks-on-sticks
  • Chili cook-off
  • Chicken Wing Contest
  • kids bike decorating
  • possibly tours of gardens and/or guitar playing
  • the kids like to ride their bikes/play games in the street

As you browse the 33 page list, it becomes almost overwhelming how many of the activities that people have to wait till this one time a year to use their street for are just plain everyday activities.  When I was growing up in the suburbs, we played in the street all the time.  City kids I guess can only do that once a year, and only if you jump through enough bureaucratic hoops, and only if your street is deemed “inessential for traffic flow.”

By far the most common activity listed is “socializing” or a variation of it.  Of course socialization happens on these blocks on other nights, too, but only in people’s yards, or squeezed onto a narrow sidewalk.  Since most neighbors drive, random socialization can only happen if no one’s listening to music, or no one is stopped behind your car.

Sure, driving is an easy and comfortable way to get around, but is it worth it?

Yes, I am sitting in my apartment on the computer instead of at NNO event.  Leaving now…