I already wrote about the MLK Jr Way S bike lane concepts, but SDOT gave a few more details about the project during the May Bicycle Advisory Board meeting (PDF) that are worth sharing.
First, some background. SDOT is conducting some early planning for bike lanes on MLK between S Judkins St and Rainier Ave S, so essentially from Mount Baker Station to the I-90 Trail. The project team is going to develop the 30% design in the summer and fall, but the project isn’t scheduled for construction until 2023.
They presented three options, with option three being by far the most popular based on comments to my original story and SDOT’s presentation.
Not only is option three the “community preference to date,” according to SDOT’s presentation, but it is also the most affordable. As I wrote in my original story:
In general, two one-way bike lanes is the best practice unless there is a good reason to do otherwise. So I guess without any such reason, the one-way bike lanes would be best.
The concept for a two-way bike lane on the east side of the street dates back to a 2008 Southeast Transportation Study, which noted that a an east-side bike lane would connect better to the parks along the street. The two-way bike lane on the west side of the street was featured in the great 2015 Accessible Mount Baker concept and would have the fewest driveway crossing:
But the one-way bike lanes in option three should also have quality access to either side of the street so long as the project creates a complete street with safe crosswalks at every intersection. And that should be the goal, though perhaps it needs to be more explicitly stated in the project description. Installing bike lanes should make installing safe crosswalks much easier, anyway. And with safe crossings and bike lane protection through intersections, any benefits of the two-way options pretty much evaporate.
Two-way vs one-way bike lanes
This is part of why I decided to write this follow-up. I wanted to share this info, but I also wanted to talk about how to decide between one-way and two-way bike lanes when developing any street design plan.
One-way bike lanes are typically the best practice, but that doesn’t mean two-way bike lanes are bad. Nothing is that simple. There are many situations where they are a great solution or maybe even a necessary compromise. To name a few examples, two-way bike lanes are often a good option when making a connection between, say, two trails with access points on the same side of the street like Gilman Ave W. Or if one side of the street is next to a large park or waterway and therefore has far fewer intersections than the other side like Green Lake Way or the Seattle waterfront trail. Two-way lanes can also be useful on dense one-way streets like 2nd Ave downtown, where there is strong demand for bike trips in both directions and very limited space to work with.
It is often true that beyond a certain amount of bike traffic, the effectiveness of two-way bike lanes breaks down. This is why cities with huge bike-riding mode share typically avoid two-way bike lanes. But to be honest, that will be a great problem to solve some day in the future. We have a long way to go before we get there. And there’s no reason why future SDOT couldn’t turn the two-way lanes we build today into one-way lanes and build another bike lane on the other side of the street. And that will be easier to do when there are so many people biking that they don’t fit anymore. But those are decisions for another day.
This is why I don’t find the need for future-proofing to be a convincing reason to resist two-way bike lanes in today’s Seattle. Right now, we need to use whatever tools we have to make basic connections so more people can get where they need to go and ridership can keep growing.
To put this all another way, one-way bike lanes should be the default option unless some special circumstances suggest a two-way would be better. This stretch of MLK Way S doesn’t seem to have a convincing reason to go with two-way bike lanes.
Two way protected bike lanes are the spawn of satan. They become difficult and dangerous if you need to make a turn. They’re also vulnerable to drivers pulling out of driveways and failing to notice bicyclists coming from an unexpected direction. I dislike pretty much all of the two way PBLs in Seattle.
I actually think two-way PBLs work on one-way streets as cyclists riding the opposite way are more segregated from traffic.
Agreed! completely. (Same is mostly true for 1 way bike lanes…)
Another problem with many of the new bike lanes is lack of passing room. With an unprotected bike lane, you can pass in the car when there’s no cars. With a protected lane, you could be stuck behind someone going 8 mph for miles.
This makes me wonder if the sweet spot is bike lanes buffered by a couple feet of diagonal stripes, but no physical barrier, such as posts or curbs.
Two way bike lanes, at least you can use the oncoming lane to pass when there isn’t much traffic.
Protected lanes are really for those not comfortable with a standard lane. You can pass on a one-way bike lane with bollards riding through the gaps.
Often, the gaps are positioned so close together, you have to almost come to a complete stop to go through them, which makes passing difficult.
Of course, one option for fast cyclists is to just ignore the protected lane altogether and ride in the car lane. But, then you’re holding up the cars, and there’s a lot of drivers out there who think that since there is a bike lane, you’re required to use it, and will honk at you (or worse) if you don’t.
In Copenhagen, the protected bikes are generally wide enough that you can pass somebody if you need to, so this isn’t a problem.
2 way bike lanes are an acquiescence by the cycling community to accept our place as second class citizens on the roads. Much of our rights and safety have already been traded away, so that even the infrastructure that we do have is often compromised by construction projects, clueless ride share drivers, and all manner of drivers in general who are just oblivious to cyclists and cycling infrastructure. There is little dignity in two way bike lanes. Just my opinion.
So I strongly opposed the 2-way in the previous story, but wow – it seems like everyone else has even more fundamental dislike of 2-way lanes than I do :)
I guess for me I don’t actually dislike 2-way lanes at all. Not fundamentally — I just oppose them *here*, because SDOT has not shown any real willingness to design good, functional transitions that work with them.
One way is better, except perhaps on one way streets:
1. You go with the traffic flow
2. You have less to watch – driveways on one side, turning cars on the other, but no bikes in the other direction.
3. You can exit or enter the traffic lanes
4. You follow normal traffic rules when going through an intersection
5. You don’t have to jig-jag across the street to enter or exit.
6. It’s easier to pass another cyclist
Two way had better have a really good justification and the lanes definitely need to be wider to avoid head-on collisions and allow passing.
One way each side is usually preferred, but not always. Two-way can work better on streets with frequent bus service and bus stops on the other side. Two-way will be better on East Marginal Way S where there will be only one intersection on the bike lane side, but there are port terminal entries on the opposite side with lots of turning semi-truck traffic. Width of bike lanes makes a difference, too.
How to choose between one-way and two-way bike lanes:
Do you care more about perception of safety than actual safety? Slap a two-way bike lane down!
There isn’t a single two-way bike lane in Seattle that is safer than the alternative that existed before it was put in place. Accepting them is as a reasonable design is admitting that we just don’t care enough to actual take the street space required to install safe AND comfortable cycling infrastructure, and that we value perceived comfort more than safety.
Andy B, that seems like a purely ideological statement, or just a personal perception. Crash and injury data shows that the opposite is true.
Hi Don. Thanks for sharing that link – I hadn’t seen that report posted.
I’m confused why you say that the Seattle-specific data shows the opposite to be true, when the report itself says there isn’t enough actual crash data to draw statistically significant conclusions.
From the video analysis data, the only reported component of it that distinguishes between one-way and two-way data states:
“Two-way PBLs have had more crashes than one-way PBLs on a per mile basis. It should be noted, however, that most two-way PBLs are located in dense areas with generally more bicycling activity and motor vehicle traffic.”
I’d love to see disaggregated data, but from what’s available in that report I don’t think it’s possible to draw any positive conclusions about two-way PBLs.
You’re spot on that my claim is based on personal perception, though – I don’t think it’s impossible to make a safe two-way PBL, just that it hasn’t been done in Seattle, largely for lack of political will.
Andy, I think we’re all pretty much in agreement at this point. I’ll just add that, yes, you can make a completely safe 2-way PBL but it would require more space and more protected intersections. The space generally isn’t therre and the extra signal time for intersections would probably get huge pushback.
Actually, some numbers on this would be interesting. Both of you are making claims and I’ve never seen data on this. Could both of you post references ? Thanks.
Here is Seattle’s most recent bike & ped safety analysis report. See page 17 and following for PBL’s.
Don, that link is broken. I went to the vision zero site but can’t seem to find this reference.
Don’s link works if you select the entire text – it looks like the link itself got broken at the .pdf part.
I don’t know of any Seattle-specific data to either support or refute my claim, it’s based on personal perception and observation.
Not Seattle-specific (NY, DC, and Portland) but this recent IIHS study (https://www.iihs.org/news/detail/some-protected-bike-lanes-leave-cyclists-vulnerable-to-injury) notes:
“Compared with a major road with no bike infrastructure, the risk of a crash or fall was much lower on two-way protected bike lanes on bridges or raised from the roadway — for example, within greenways. In contrast, the risk of a crash or fall on a two-way protected bike lane at street level was much higher than that of a major road.” I can’t link the full journal article but it cites the crash/fall risk as 11.4 times higher on a “two-way protected bike way, light separation” facility versus an unimproved major road.
The “light separation” is consistent with all our on-street two-way PBLs in Seattle.
Ah, I hadn’t tried copy & past, just clicking on it. Thanks.
Anyway, the report states they found “Two-way PBLs have had more crashes than one-way PBLs on a per mile basis.” They also pretty much disqualified their findings by saying there weren’t enough crashes to be statistically significant. They enhanced their findings by examining video of near misses.
So, hard to say how to take that data. I’d say it’s inconclusive.
For now, I’ll stick with one-way PBLs since I think they function better from a practical standpoint.
What is relevant about Seattle’s data is that it is from the actual Seattle’s PBL’s we are talking about, not other cities, and that there is no data showing any PBL’s being more dangerous that the streets before the PBL was put in place. There is only data showing higher crash rate on 2-way vs 1-way including bike/bike and bike/ped, which just makes the case that 1-way is preferable to 2-way usually but not always, and that PBL is preferable to door zone bike lane or no bike lane on arterial streets that are desirable bike routes. PBL’s are new, so data is short term, the treatment is an evolving work in progress. They are working, but also are experiments that will sometimes fail and need revision, as was done in the changes from 2nd Ave pilot project to 2nd Ave second phase. The data to date shows no fatalities. On unimproved Dexter and on 2nd there were fatal crashes killing people on bikes in years just before the 1-way and 2-way PBL’ s were installed.
Sounds like we agree, Don:
There’s no statistically significant data that shows that 2-way PBLs in Seattle are any safer than what they replaced, and the data from other cities where it has been studied shows they are much worse.
One-way PBLs in Seattle can be an improvement, which means that if the City wants to take a data-informed approach to design instead of one based on wishful thinking and experimentation with human lives, it would be better off spending what little it does on facilities that are actually shown to work, instead of gambling on a treatment that has been proven to be more dangerous elsewhere, and remains unproven here.
Even if the cost (in dollars and political capital) to get those one-way PBLs is dramatically higher than two-way, it’s worth getting less total miles of improvement to have it done right.
Andy B, I don’t think we are in agreement on safety for 2-way PBL’s. Your assertion that ,”there isn’t a single two-way bike lane in Seattle that is safer than the alternative that existed before it was put in place”, is not supported by crash data. The Feb 2020 SDOT City of Seattle Bicycle and Pedestrian Safety Analysis does not say that two-way PBL’s are less safe than what existed prior to the PBL installation.
On the contrary, it says:
“PBLs are among the bikeways identified in the Plan that have demonstrated safety benefits…”
“There have been few crashes on the PBL network relative to crashes citywide.”
“Key findings from the crash analysis include:Most streets with PBLs saw a reduction in bicycle crashes when comparing before and after crash frequencies.“
After several years now of data collection and analysis on the very streets we are talking about (not Copenhagen or Portland or NYC) it finds a high increase in ridership and lower rate of injuries and fatalities to cyclists in Seattle. This trend correlates to installation of Seattle’s first significant PBL’s, both one-way and two-way.
This data does show that “Two-way PBLs have had more crashes than one-way PBLs on a per mile basis. It should be noted, however, that most two-way PBLs are located in dense areas with generally more bicycling activity and motor vehicle traffic “ It does not show that two-way PBL’s are worse than the pre-existing condition.
There were fatal bike/vehicle crashes on streets like Dexter and Second Avenue in years just prior to installation of PBL’s, with none since. That doesn’t mean the risk is zero, or that bike riders should be blissfully unaware of danger, but it’s a pretty good sign that PBL’s are not “the spawn of Satan”.
We live in a built up City with many physical constraints on roadway and street re-configuration. The city’s street use manual sets standards and provides guidance, but gives flexibility to adapt to conditions. It seems unwise to throw options out of the toolbox that might work well in specific locations. We have a wide variety of conditions, not amenable to a one-size-fits-all approach. It also seems unwise to demand that only optimal solutions be implemented. When it’s a choice between “all or nothing”, the result is “nothing”.
Although one-way PBL’s are preferable, there are some pretty nice two-way PBL’s. My favorite is the one crossing the Spokane Street Bridge at street level, separated by a concrete barrier, with plenty of width and good pavement. I am really looking forward to the one planned for East Marginal Way S from S Horton to S Atlantic, providing a 2-way PBL with only one intersection instead of a southbound 1-way that would cross several port terminal entries and exits with heavy semi-truck traffic and poor sightlines.
Thanks for listening.
Hey Don, my mistake, apologies.
I feel like you might be conflating the findings in that report a bit in your post.
I agree that the report does not support my claim that the two-way PBLs are worse than nothing, but it also does not refute it. Because it based on such a limited dataset of two-way PBLs (only data through 2017, and excluding any street that has since been reconfigured), it simply isn’t reasonable to use it to draw conclusions specifically about two-way PBLs from that data.
You’re correct that there hasn’t been a fatality on Second Ave yet, but we’re talking about two very different time periods for a facility to be in service (gosh, how long was that awful left-side paint-only bike lane there?), which is why it’s important to rely on statistically significant data rather than anecdote.
I do agree with you that facilities like the Spokane Street Bridge trail are great – it may be a bit of terminology difference but I really consider those kind of facilities in a whole different class from on-street two-way PBLs. That IIHS report I linked to classifies them as “Protected bike lane, heavy separation”, and the data they report suggests that they are indeed dramatically safer – over 100 times safer than an on-street two-way PBL, and about 10 times safer than either an unimproved road or an on-street one-way PBL. Those are best in class!
Thanks for the interesting conversation.
SDOT’s own guidelines for arterial lane width (11 ft for through traffic adjacent to opposite direction travel or adjacent to curb lane, 8 ft flex zone, 10 ft turn lanes) appears to be violated in every option but Option 2. There really should be a requirement that a project’s departures from the street design manual be itemized and explained for every project, because they’re based on best practices. At least when you violate building codes, you have to specifically note and ask for exemptions. https://streetsillustrated.seattle.gov/design-standards/roadway-construction/roadway-width/
One more voice against 2-way bike lanes. The disaster that is the entries for the Gilman bike lane is a daily problem for me–I don’t want to connect the two bike paths, I want to get to my home in Magnolia. There’s no Magnolia entrance from the north (Discovery Park) end, and there’s no way to get to Thorndike from the south end–to prevent cars from entering the bike lane, there is curb, not just bollards. Virtually all of the 2-way bike lanes I can think of have the same issue of only connecting specific destinations, not truly enabling cycling as transportation.
I really can’t stand two-way bike lanes. I think they are dangerous and inadequate. they are typically not wide enough for what they intend to accomplish, become dangerous for possible encounters between cyclists and pedestrians planning to cross at an intersection easily get confused about looking both ways or cycle traffic when car traffic is anticipated from one direction only. I believe that one-way lanes on both sides, with the flow of traffic are much safer.
Another vote here for no more two lanes. “very limited space to work with” on 2nd Ave downtown is due to what? standard car lanes, that is what. Get rid of the car lanes. Get rid of the cars.
Unlike many Seattle streets, there’s actually plenty of width to work with on 2nd Ave. The issue on 2nd is the frequent headway of buses stopping at long bus stops on the west side, Getting rid of cars is not only a fantasy, it would require lots more buses. It makes more sense to organize this street by putting buses on one side and bikes on the other, with cars and trucks in between. The bike lanes are too narrow, but they are in a good location.