No easy fixes for Lake Street traffic

Arlington’s Board of Selectmen will consider a plan February 5 to ease rush-hour traffic on congested Lake Street by putting a traffic signal on the Minuteman Commuter Bicycle Path.

The Minuteman and Lake St. in a less-congested hour. View is north. Photo: Phil Goff.

The Minuteman at Lake St. in an uncongested hour. View is north. Photo: Phil Goff.

The Transportation Advisory Committee, which has been studying the problem for more than a year, recommends the project with an estimated cost of $150,000. If approved, the Town would establish a design committee.

At the risk of oversimplifying a very complicated problem, I warn: Be careful what you wish for.

A signal has the potential to improve trip time on Lake Street, dramatically during the evening rush hour, but at the cost of more cut-through traffic and at the expense of path users.

Paradoxically, the signal has little likelihood of shortening traffic queues, especially inbound in the evening commute. Traffic would flow more quickly but there would be more of it. Lake Street would probably not be less congested at rush hour.

Unlike the Mass. Ave. Project design, this decision involves real tradeoffs between users: cars vs bicycles and pedestrians. It thus has the potential to fuel one of Arlington’s favorite feuds.

The issue illustrates how perplexing and counter-intuitive traffic issues can be. It also suggests that there are limits to what traditional traffic-engineering techniques can tell us and what they can accomplish.

Lake Street Traffic and the Minuteman Path

Lake Street traffic, already very bad at rush hour, is almost certainly made worse by interactions with bicycle and foot traffic on the Minuteman Path. The path crosses the street 200 feet west of the traffic signal at Brooks St. near the Hardy School.

Traffic backs up at Brooks St. outbound (towards Route 2) in the morning and inbound (towards Mass. Ave.) both morning and evening, though the real crunch is the evening commute, when the traffic queue can reach as far back as Cross St. in Belmont.

The case in favor of a traffic signal can be summed up with two observations.

First, traffic modeling done for the TAC finds that a signal can improve inbound evening trip time for motor vehicles by nearly 7 minutes (though this may be overstated).

Second, there are more drivers than path users at rush hour.

The signal in question would be a standard red-green-yellow signal, timed with the nearby signal at Brooks, with walk lights and perhaps special bicycle signals as well. (It’s nothing like the nonstandard signals where the path crosses Mill St.)

The new lights would be timed with the signal at nearby Brooks Street, which might also be altered.

The traffic counts on which TAC relies are as follows:

Morning Commute

  • Lake Street motor vehicles: 956
  • Minuteman commuters: 256

Evening Commute

  • Lake Street motor vehicles: 1,011
  • Minuteman commuters: 316

Counts are for a single peak hour, not for the whole commute.

The impact of a Minuteman signal on the evening commute (currently measured at 601 vehicles on one April evening peak hour) would be so profound that the TAC believes the improvement will increase traffic by a third:

Existing traffic volumes show Lake Street has approximately 800 vehicles westbound in the AM peak hour and 610 eastbound during the PM peak hour. Because of the substantial eastbound queuing on Lake Street during the PM peak hour, the 610 observed vehicles understate the actual demand.

Therefore, 800 vehicles was used to represent the eastbound actual existing demand during the PM peak hour. (TAC report to Selectmen, unnumbered 5–6, emphasis added)

(There’s actually an easy way to test that assumption, which is to count traffic in the winter when ice, cold, and darkness reduce Minuteman traffic, and its effect on Lake St., to near zero.)

You’ve probably noticed that the above does not try to assign moral responsibility for the situation, as in, Cyclists behaving badly, or, Motorists foolishly yielding. Rather it focuses on questions of what can be done (though it confines the answer to engineering solutions).

In my opinion both motorists and cyclists behave badly there, but that’s probably better as a topic for another post.

The difficult part

The foregoing makes, I think, a pretty strong case for a new traffic signal at the path and Lake.

So, what’s not to like? Why isn’t this a slam dunk?

I’m going to have to write another post about that, but here are the problems in brief.

  • There is tremendous, and growing, demand for a way to and from Route 2 that avoids the super-congested Alewife area (Thank you, Cambridge!).
  • Consequently, improvements in trip time will inevitably draw new traffic seeking a way through. Some of these new vehicles will be cut-though drivers seeking a way to Cambridge, Somerville, and beyond. Expect more traffic on Marathon Street, Broadway, and Mass. Ave., at least.
  • Easing the horrible Brooks Street bottleneck will (alas) shift the bottleneck to the T intersection with Mass. Ave. To avoid this, some drivers will likely seek their Northeast Passage through the Hardy School neighborhood.
  • The new traffic will also slow things down, so that the time savings for the inbound evening commute (based on 800 vehicles) will be less. Probably not 7 minutes after all.
  • Meanwhile a signal will affect the path all day, not just at rush hour (when drivers outnumber path users).
  • So it’s not just about balancing competing traffic needs, but about a change to Arlington’s most popular park to benefit commuters.

There are also some some unresolved safety issues about this change that involve small children walking to and from school.

A lot more explanation is appropriate, but these are not simple issues.

I do not envy the Selectmen on this one.

Advertisements

14 comments so far

  1. Thouis Jones on

    This seems like overkill. I agree that it’s very doubtful they’ll see anywhere near 7 minutes improvement. New development will increase the congestion around Alewife, and both backup and cut-through will increase.

    If they’re trying to address safety at the path intersection, they might as well just make it a 4-way stop. It’s close to being one, anyway, during rush hour, anyway, given how many cars end up yielding to pedestrians and bikes (even if they don’t have to, to the latter).

  2. Ryan Silva on

    Why don’t they just change the bike path signal to a flashing yellow during non-peak hours? I don’t mind a signal during rush hour when there’s lots of traffic on both lanes, but it would be silly to wait to cross Lake during off-peak hours if no cars are coming.

    • Adam Auster on

      Ryan, you mean as a way to minimize the impact of the signal on path users? (It would probably be flashing red for the path, I should think.)

      But then what is the status of pedestrians? I think that once the crosswalk is signalized they lose their right of way and must press the button and wait for a walk signal every time.

      So does that mean pedestrians must wait, but cyclists can just go (after stop)?

  3. nynexman4464 on

    I think a light here makes some sense for safety. It’s always a gamble crossing here for drivers and pedestrians – a game of who is going to stop first.
    I don’t think a four way stop works here. The minuteman crossing is so close to brooks ave it means that drivers coming from mass ave with a green can’t clear the intersection at brooks because there are cars in front of them stopping at the path. A light makes it so drivers have a synchronized green allowing the intersection to clear. It won’t solve all the problems, but it might help.
    In terms of making it less of a pain for path users – why not make the light flashing yellow on lake st and flashing red on the path during off peak times? That would basically make the intersection operate the way it does now, with a stop sign for path users.
    I don’t see the proposal on the transportation board page (http://www.arlingtonma.gov/town-governance/all-boards-and-committees/transportation-advisory-committee/reports-recommendations-guidelines) but maybe if the crossing was on demand, or used a sensor similar to Mill St might also help.

  4. Peter Fuller on

    Traffic light as proposed at bikepath and Lake St, slaved to existing light at Brooks and Lake, would improve safety margin there, which is reason enough to go forward with this plan. Any vehicle traffic improvements on Lake St will be icing on the cake.
    The estimated project cost of $150,000 raises my eyebrows a bit: seems a lot for a traffic light. There must be more to this than meets the eye. I look forward to TAC posting their full report on the Town website so we can all see the detailed proposal.

  5. David Markun on

    The newly reworked traffic signal at the intersection of Bedford St. with the Minuteman Trail in Lexington offers a chance to experience something like this could be. It has fun talking pushbuttons for pedestrians and riders to press (set back from the curb so that riders can press them without dismounting and without riding into the road) and it has whizzy red/amber/green bicycle-logo lights facing the path. My observation there is that the signal does a good job of expediting the auto traffic at busy times, and making the trail traffic safer. It is now much rarer than it used to be for an auto driver to stop when they have the green light, probably because the set-back position of the pushbutton puts waiting cyclists into a less threatening position. I think there is no need to worry about delaying trail users at times of no auto traffic; at such times the on-demand light comes almost instantly, and beyond that, at Bedford St. at such times many trail users do not even press the button, they just cross against the red bike.

  6. markk02474 on

    With the intersection of Mass Ave and Lake Street remaining as a bottleneck, problems that can be reduced are danger and global warming. A traffic light coordinated with the one at Brooks makes the Minuteman crossing safer and reduces the starting and stopping motorists experience.

    To reduce the bottleneck at Mass Ave, Selectmen should have adopted what MassDOT encouraged, a longer right turn lane for Lake Street motorists turning eastbound on Mass Ave. Selectmen never discussed their bad decision in public, nor were meetings with MassDOT on the subject open to the public, nor detailed in meeting minutes.

    Finally, Selectmen need to work with our state Senator and Representatives to widen route 16 from 4 to 6 lanes between Mass Ave. and Alewife before Cambridge builds any more condo and apartment complexes. Furthermore, Mass Ave westbound in Cambridge needs a second left turn lane to route 16 towards route 2. Likewise, two left turn lanes from route 16 to Mass Ave westbound to Arlington are needed. These will provide much needed relief to Lake Street. This is 30 years overdue. Its critical now that Cambridge has accelerated development in the area.

  7. Susan Stamps on

    Safety of path users crossing Lake St is my concern, not convenience of Lake St traffic. The lights at Bedford St and Hartwell on the bike path are the right model in my opinion. In both locations, the light is green for the traffic and only changes to red when the path user activates the walk light. There is a big gate on the path at the crossing visible from a far distance, alerting the path user to slow down and assess the situation and then usually hit the walk light before crossing. I have never seen problems at either of those crossings but every day there are scary incidents at the Lake St crossing.

  8. Susan Stamps on

    I think the safety improvements, lights, etc. might be eligible for Community Preservation Act funds as infrastructure improvements in an outdoor recreational resource and Arlington’s CPA fund will likely be up and running by the time a final decision is made.

  9. Rob Mulligan on

    Yes, “If you build it, they will come” needs to be considered. I would like to see that light in Lexington mentioned – although the proximity to Brooks Ave. really complicates things. In the morning, there are actually a decent number of cars exiting Brooks on both sides onto Lake St. towards Rt 2… but if the path light was red because the Lake lights were red, they will then be jammed up at path and seems as though nothing will ever get through! It seems prudent to ‘prototype’ any changes with human-based traffic control for a week and get a real sense of what the impacts are. Also, good point on the Mass Ave light. Is the light at that T intersection 100% utilized when green at rush hour? If and when that gets fully used, it seems the benefit of the path control light is diminished.

  10. dr2chase on

    I’d vote for the synchronized red-yellow-green at rush hour, including a DO NOT BLOCK CROSSWALK sign for the cars, and something else for off-peak. The problem with flashing-yellow for the cars is that it might cause drivers to treat the crosswalk as not being a fully fledged crosswalk, thinking that there is a button that needs to be pressed, etc, and thus not yielding to pedestrians and cyclists.

    An alternative might be a red-yellow-green at all times, but with a beg button for the bikes/peds that behaves differently at off-peak — this would be less ambiguous.

    But I would also advise a hard look at the throughput at other intersections. Put plainly, if the current combo is a true bottleneck, then there should be minimal backups and delays at Mass Ave (on one end) and at Route 2 (on the other end). If there’s backups there already, then this would be money wasted.

  11. Charlie on

    The goal of a signal here should be to increase safety of the path users, if that’s something that’s desired, not to increase traffic capacity on Lake St. You’re absolutely right that traffic looking for a way around the congested Route 2/16 intersection would easily fill up any “new” capacity on Lake St. If the signal is coordinated with the signal at Brooks St, that also means that for significant portions of the day when traffic on Lake St is much lighter, much of the time path users would often be forced to wait for no obvious reason to them and would be very tempted to ignore the signal altogether.

    This is a rare case where I actually think status quo is the right answer. I see no reason to believe that that path users need more protection when crossing, and increasing the traffic capacity would simply invite more traffic.

  12. Adam Auster on

    These are very interesting and thoughtful comments. Here’s what I think.

    A number of people—nynexman, Peter Fuller, David Markun—suggest that a traffic signal will address safety issues, and in fact the TAC report makes a similar claim.

    However, the data from TAC do not support this idea, because however it may feel the intersection is pretty safe already. There are few reported collisions, given traffic volume, and those collisions produce no serious injuries and often no injuries at all.

    That’s probably because the current layout promotes those annoying feelings of uncertainty, which leads to slow cautious behavior. Speed things up, as we are (ironically) trying to do, and things could actually get worse, not better.

    The counter argument is that traffic signals will finally make everyone stop when they are supposed to stop and go when they are supposed to go. Consequently this regulated traffic will have zero collisions and it won’t matter how fast everyone is moving.

    David Markun points to crossings in Lexington that are signalized and that seem to work well (TAC does too).

    However, these intersections do not have the same kind of volume and are not right next to an elementary school and to dense residential neighborhoods. Furthermore, TAC finds only 75% compliance even there.

    Other Lexington crossings are not signalized and also seem to work.

    I think the honest thing to do would be to drop the safety claims and admit that this is about making life better for commuters, especially in the evening. This seems likely to me based on the modeling that TAC has done, although I think the committee may have overstated the benefits in that respect.

    Commuters do have a rough deal, and fixing the problem is hardly out of bounds. But I’d like a better handle on the actual benefits and costs, including how much new cut-through traffic.

    • dr2chase on

      One thing I notice is that some people tend to gravitate towards rules by habit, and others (usually not by habit, usually by hard experience) learn to look for results and data. The current intersection is not terribly friendly to the rule-followers — the cyclists often act like rolling pedestrians, which is “against the rules” (note that some cyclists, for example me, can smoothly dismount and walk across the intersection, and it is slower and causes more delay, not less).

      So I think that one appeal of a signal is that will do a better job of making rules explicit and removing this troublesome ambiguity, even though it appears that (as you note) the ambiguity works okay for safety.


What do you think?

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: