What Other Cities Say About Cleveland’s Unusual Bike Lane Buffer

Cleveland’s seemingly backward buffered bike lane on Lorain Avenue. Photo: Satinder Puri.

pfb logo 100x22Michael Andersen blogs for The Green Lane Project, a PeopleForBikes program that helps U.S. cities build better bike lanes to create low-stress streets.

For all their benefits, protected bike lanes can be complicated. Between maintaining barriers, keeping them clear of snow and preserving intersection visibility, it’s understandable that cities opt not to include them on every street project.

Buffered bike lanes, though, are pretty simple: if you’ve got at least two feet of roadway to spare, you lay down some hash marks between car and bike lanes and double the comfort of biking on a street.

Except in Cleveland, apparently.

When the above image started circulating online this summer, many people assumed some sort of miscommunication was afoot in Cleveland. The main point of a buffered bike lane, as made clear by everyone from AASHTO to NACTO, is to separate bikes from moving cars and/or the doors of parked cars, not to protect bikes from curbs.

But as more information emerged and it began to seem as if Cleveland was not only doing this intentionally but might be planning to repeat the design elsewhere in town, we wondered whether this might be a new trend in street design.

So we emailed cities around the country and asked their bikeway designers to say whether they’d ever want to use this setup. Here’s what they said.

Mark Zwoyer, PE, assistant administrator of the department of public works in Indianapolis:

We probably would not use this in Indianapolis. We use cross hatch buffers to separate the bikes and the motor traffic, but not to offset bikes from the curb.

Dongho Chang, PE, chief road engineer in Seattle:

Buffer seems to be in the wrong place. We’d want to buffer the bikes from travelled vehicle lane.

Jonathan Lewis, AICP, assistant director of planning in Atlanta’s transportation department:

I feel like I’m missing something? Why is the buffer next to the curb? I don’t see an upside. In Atlanta, during torrential rains, the curbside bike lanes flood. I guess that’s an upside to this design, but it doesn’t seem worth the trade-off. It is not a configuration we would consider in Atlanta.

Nathan Wilkes, PE, of the transportation department of Austin, Texas:

I think the photo shows well where the rider wants to be. In Austin, TX we always put our buffer to traffic where there is the most value in creating separation between fast moving traffic and the person on the bicycle to create a bicycle facility that a larger portion of the population is comfortable in. The only use I have seen as shown in the photo is if there were parked cars to the right of the bicycle lane where there was more concern about the door zone than the moving traffic. … In short it looks to be a waste of buffer material applied incorrectly.

Rachael Bronson, associate city planner for the Denver Department of Public Works:

Without parallel parking, I’d prefer this to be on the car side. Or maybe buffer on both sides if there are a lot of access points where bikes may need a buffer on the curb side.

She cc’d Michael Koslow, PE, a senior engineer at Denver DPW, who added:

I agree that car-side buffering would be better; however if there’s additional room it’d be good to keep bicycles toward the middle of a wider lane for conspicuity by drivers.

Matthew Dyrdahl, AICP, bicycle and pedestrian coordinator in Minneapolis:

We typically buffer the other side, meaning put the buffer between the travel lane and the bicycle lane.

He cc’d Simon Blenski, bicycle plannner in the Minneapolis Public Works Department, who added:

We typically install a buffer on the travel lane side. We have installed buffers on both sides of the bike lane where there is excess space and where parking is adjacent to the bike lane.

I can only think of one example where we only have a buffer on the curb side and not the travel lane side. It is an exceptional case on Hennepin Ave S between 12th St and 13th St, where there is a wide parking bay next to the bike lane. The buffer was placed in the excess parking bay width.

We have never done a buffer directly adjacent to a curb with no parking present.

OK, so we were zero for eight on finding another city that would consider this sort of buffer. But we also reached out to a consulting engineer who works with many cities: Rock Miller, PE, of Stantec Engineering.

Miller, who’s based in Orange County, California, said there is one possible reason to stripe a buffer this way: to ensure that the bike lane is close to fast-moving trucks. If it’s close enough, the wind from those trucks will whip away leaves, seeds and bark.

Another rationale: to reduce the amount of pavement that snowplow operators feel the need to keep clear in winter.

Vehicular cyclists push for minimal buffer, as shown, because wind sweep from passing trucks tends to reduce debris. … There could also be a significant snow storage and removal factor. These are perhaps valid concerns if maintenance will have limits, but new riders will want the buffer on the traffic side, and to also have thorough sweeping.

As for Cleveland road engineer Andy Cross’s stated concern that putting the buffer between bike and auto traffic would increase the risk of cars turning right in front of bikes, Miller said he’s not aware of evidence that either supports this or denies it.

Because we have so little exposure (volume) data, we really don’t know if one treatment is more risky than another or if a simple design adjustment greatly changes the results.

Ultimately, Miller said, “it is up to the designer, with input from users, to decide how to use the space.”

You can follow The Green Lane Project on LinkedIn, Twitter and Facebook or sign up for its weekly news digest about protected bike lanes.

44 thoughts on What Other Cities Say About Cleveland’s Unusual Bike Lane Buffer

  1. As a resident of Austin, I think this is out of context.

    Our buffered bike lanes never (I can’t think of one anyways) have this huge of a buffer. This is a typical one near my office in suburban hell:


    If we ever contemplated putting in one with as large a buffer as the one in Cleveland on a street with non-trivial curb cuts, I’d be advocating for the buffer to be on the curb side as well.

  2. I design these things for a living, and certainly all current guidance would tell you the buffer is on the wrong side in this photo.

    On the other hand, giving the designer the benefit of the doubt, I can actually see one good reason to have the buffer on the curb side for this particular stretch of roadway. There appear to be many, many driveways, and many, many utility poles, and the utility poles are only two feet or so from the curbline. This means that a driver exiting from one of those driveways has very poor sight distance up the road and will need to nose their car into the roadway in order to see up the street. If the bicycle lane were right at the curb this would mean a lot of drivers (a) would not have good sight distance up the bike lane, and (b) will be nosing their cars into the bike lane when they exit the driveways.

    Splitting the buffer to have two narrow buffers on either side of the bike lane may have been a better approach.

  3. > Splitting the buffer to have two narrow buffers on either side of
    > the bike lane may have been a better approach.
    Or just make the lane extra wide and leave it up to the cyclist to choose their line. Ultimately hash marks on the road are to let cars know where they’re not allowed. Cyclists will ignore them and just ride where they feel safest.

  4. In the photo above, there’s a silver car pulling out of a driveway. It’s fairly obscured by the poles and the driver seems to be nosing out into the buffer zone to see. So maybe this isn’t as crazy as it seems.

  5. This comes down to the conflict between proven international best practices vs. failed vehicular cyclist ideology. Andy Cross clearly adheres to the vehicular cyclist ideology, which says if you can’t force bicyclists into the general purpose lanes (which is the goal of VCs) then you should put them as close to traffic as possible. It’s almost as if VCs want to actively discourage cycling by anyone who isn’t a fit male.

    It’s a great example of why it’s important for the bicycling community to organize and force out any left-over VCs that are still making decisions regarding bike infrastructure. The biking community in Houston was able to force out their long-standing VC Bicycle Coordinator who outright refused to stripe any bike lanes, let alone safe protected infrastructure. Bicyclists in Houston are now much better off for it.

    Also, prepare for the inevitable onslaught of VC commenters once this gets posted on one of their online refuges. A note to everyone else: rational arguments don’t work with irrational people (i.e. those whose entire ideology is premised around pretending the Netherlands doesn’t exist). You might as well be arguing with astrologists. Please just ignore them.

  6. Ideally, there would be two buffers — one on each side. As somebody whose only serious cycling injury occurred when an oblivious pedestrian darted into the bike lane and took out my handle bars, I’m all for some space between the sidewalk and the bike lane. Throw in things like storm drains that are typically right up against the sidewalk, and it makes even more sense. Enough sense to use in place of a buffer between cyclists and cars? Probably not, but it’s not super crazy.

  7. Not just storm drains, but all the gravel, sand, dead leaves, and litter that collect in the gutter, making it dangerous to bike (or skate) close to the curb.

  8. I keep forgetting how unfamiliar some of you Rust Belt folks are with Sunbelt exurbia. This road is theoretically 45 mph (signed), obviously design speed is quite a bit higher (but not 70). But this bike lane is pretty typical for the exurbs (i.e. you’re not going to get better than this in our lifetime).

  9. Yes! We definitely do. SO many bike lanes are 5′ wide next to 50mph traffic. Terrifying. It’d need to be a video contest, though, to show the full extent of the terror.

  10. I would not want to be the city attorney for Cleveland. If you’re going against so much consensus, prevailing best practices, etc, you need really good analysis to support doing something different. It’s why so many cities don’t want to innovate, but then I don’t practice in Ohio.

  11. Agree. I don’t think hash marks are needed all over the place. Those thermoplastic markings are actually quite bumpy and annoying to ride on.

  12. I agree, hugging the curb in this location is asking for trouble. If I were riding it, I’d probably go closer to the curbe where there are no driveways, but then swing away at every driveway.

  13. The vertical posts in your picture are in a traffic-adjacent buffer. Streetsblog is saying that this Cleveland bike lane should have a traffic-adjacent buffer rather than a curb-side buffer. There’s nothing inconsistent about holding that postion while also praising the design of the Queens Blvd cycletrack: both have traffic-adjacent buffers (and just fyi, the beige area isn’t a curb-side buffer, it’s a pedestrian area).

    A better analogy to the Cleveland situation would be if Andy Cross put the vertical posts on the right side of the lane, thus rendering them useless. Just like his curb-side buffer.

  14. Kenneth,
    I think you’ve made a great case for the design in the context of this particular part of the street; however, it doesn’t seem as though that was the true rationale behind Cleveland’s decision. If it were, answers like “we don’t have enough data” wouldn’t have been given. This post also indicates that the city plans to replicate this design throughout other streetscape projects, not all of which I imagine have the same sight distance issues with driveways, etc.

    From an ideological perspective, I’m also not sure how I feel about shoving bicyclists out next to traffic/unsafe conditions due to the 1) poor design of public infrastructure, and 2) ignorant drivers who fail to consider the consequences of pulling out into the road blindly. The obvious solution to this is exactly what you noted as the best compromise- splitting the space and placing a buffer on each side.

  15. None of those things change the layout or proximity to cars that is the basis of the complaint. If Cross made the outside line a bit thicker, put plastic sticks every 40 feet, and called the hatched area a sidewalk expansion, this would be fine?

    Why would anyone do anything but a real, high (by American standards) quality cycletrack if they have 9 feet of curbside width to work with? It boggles the mind.

  16. If Cross took 2-3 feet from the curb-side buffer and put in a 2-3 foot traffic-adjacent buffer (just like the picture you posted; NATCO recommends at minimum 18″ inside the buffer lines if you are going to put in cross-hatching) then yes, it would be far better than the current implentation and I’m positive Streetsblog writers would agree.

  17. Most of the response express the attitude of everyone else does it this way… But where is the engineering, research, data that shows one method IS better than the other? I think a better response is needed especially if towns and cities are to take bike infrastructure more seriously!

    I do find it very telling that the cyclist in the photo is riding in the buffer.

  18. My point was “why not go for the best given a similar budget and the same space“. US cities can’t afford the best.

  19. We have a lot of those out here in the Golden State. The worst are the ones on roads that are (over)built for “future demand” but leave the outside lanes unstriped, resulting in an expanse that is literally in some cases, nearly 30 feet wide. In some places, they have the audacity to paint a “generous” six foot curbhanger for bicyclists to use. They’re almost never usable and hardly what is needed, especially in the context of the surrounding areas.

    Here’s one in my region where a cyclist got mowed over last year (you can see the ghost bike):


  20. The issue is, VC/BD advocates do not miss the mark. While largely a coping mechanism more than something to actually promote over other alternatives, that style of riding is the safest in much of the current road environment in America. Plopping paint doesn’t fix many of the underlying problems and issues still exist that need to be mitigated. Painting bikeways to promote safer riding practices is part of that.

    This design does not force anyone to ride closer to the “car” lanes as they are free to do as the man in the picture is doing: ride in the buffer. However, the intent is to encourage riding farther from the curb which will generally make a bicyclist more visible. Visibility is important and is something that is obviously sorely needed in this instance as evidenced by the car in the picture nosing into the lane to be able to see beyond the poles. Additionally, it allows bicyclists to be more predictable and ride in a straighter line that doesn’t dodge people leaving driveways, broken glass, gravel, storm drains, and other obstructions that frequently accumulate closer to curbs. These are facts that need to be acknowledged.

    Furthermore, as already mentioned above, VC is a viable coping mechanism due to the current environment. No doubt, the environment does need to change. However, paint is not that change. Regardless of which side of the bikeway has the hash marks, the fact of the matter remains that these bike lanes are far from best practice and something like this shouldn’t even exist in at all. (Recall that most bike lanes are striped on what is considered the paved shoulder of a highway. A shoulder, paved or not, is intended for use by runaway vehicles for their recovery. Putting vulnerable users directly in their path is asinine, especially on multilane roads. Dutch guidelines, attached, would never allow a painted bikeway on a roadway with two or more lanes per direction and the Danish are similar.)

    Arguably, the best part of the implementation here is that it could (and should) be used to concurrently narrow the adjacent travel lanes and thus lower speeds, which would be a far bigger improvement in safety than switching the side of the buffer. There are more important things to take care of like clearing that line of power poles and closing driveways throughout the corridor, which would in turn reduce the visibility issues leading to this buffer choice in the first place.

  21. Yes, we could afford Dutch level cycling infrastructure in the United States.

    Could Cross find it in his existing budget for cycling? Nope.

  22. Making the leap from “coping mechanism” to “preferred alternative” is pretty nutso, though. I bike with very bright lights in order to be able to see the road in neighborhoods with few or no street lights. However, I don’t then go around arguing against street lights, pretending that my bike lights are a better and safer option.

    It is important to both acknowledge via street design people’s rational fears of biking in traffic while also building facilities that maximize safety/visibility between road users. One can build all the bike lanes they want, but if they are designed such that few people are willing to use them then it is a pretty useless effort, and the safety of the few cyclists who still brave those streets will hit a ceiling regardless of their individual behavior, since they will be rare and less expected by drivers. I agree that much more than paint is needed, but if more people can be encouraged to bike in the short term using just cheap and temporary materials then everyone benefits as a result, and the more significant improvements will then have more support from people who see it as benefitting them directly.

  23. How much money do you think Cross has his hands on for this project? Hint: its not remotely enough to construct a raised cyclepath.

  24. “to ensure that the bike lane is close to fast-moving trucks.”

    I had to read that twice. What sane person wants to be riding CLOSER to fast-moving trucks?

  25. I tend to agree, but it really also depends on the project too. If they’re already repaving, adding some concrete to create the raised curbing and paving the bikeway a little higher isn’t the most expensive proposition in the world. However, I am aware that it’s not necessarily that easy and there are certainly things that can easily complicate things (i.e. wastewater management) and drive up the cost.

  26. Painted bike lanes are almost always a coping mechanism, not a preferred alternative. There are already plenty of studies out showing that most people still aren’t going to ride with painted bike lanes and I watch people calmly pedal along on the sidewalks right next to bike lanes almost daily. In other words, there is a rather limited audience for paint and I doubt that switching the buffer to the traditional side would result in a substantial increase in ridership here. However, it does continue to encourage a safer riding position for those who do venture into the street. That’s an indisputable fact.

  27. Having a debate about paint-only versus physically protected bike infrastructure is a much more nuanced conversation, though, as opposed a debate about bike infra versus no (or very minimal) bike infra.

    As for paint-only bikeways, I agree that they could also be considered a coping mechanism, but only in terms of coping with the lack of adequate funding, design, and public support for physically protected bikeways, which would indeed be preferable in many situations but require resources that aren’t often available, at least where I operate.

    I would indeed be interested to see some studies on the effects or ridership as well as crash rates/severity with regard to left versus right side painted buffers. My guess is that the differences in both categories would be pretty minimal.

  28. The curbside buffer? Should be exactly the width of the storm drain grate. That’s, like, less than ONE FOOT WIDE on most streets.

Leave a Reply

Your email address will not be published. Required fields are marked *


Cleveland Traffic Engineer Puts Buffer on the Wrong Side of the Bike Lane

Cleveland is finally installing buffered bike lanes along some major streets, but with the buffer between the bike lane and the curb, not between the bike lane and traffic. At first, many people thought this design was a mistake. But it has now been painted on two streets at the behest of Cleveland’s traffic engineer, Andy Cross. Local blog GreenCityBlueLake reports that […]

San Diego Claims Space for Biking With Its First Road Diet

San Diego certainly has the right kind of climate to be a great biking city. But its streets — wide, fast, scary — have been holding it back. That’s beginning to change. On Tuesday, the city striped two buffered bike lanes on major streets. The writers at Bike SD say it’s an important precedent for the city: […]

Take a Look at Tampa’s First Protected Bike lane

Tampa is starting to make progress on safe bike infrastructure. Last weekend, Tampa Mayor Bob Buckhorn led a celebratory ride to mark the opening of the city’s first protected bike lane — a curb-protected two-way lane on Cass Street downtown. The Cass Street project is one of the first protected bike lanes in Florida — a notoriously dangerous […]

Tampa Installs Its First Green Bike Lane

There are splashes of green appearing in downtown Tampa, as the city installs its first buffered bike lanes on Platt, Cleveland, and Brorein streets, complete with green intersections. The Platt Street redesign trimmed the one-way, three-lane road down to a two car lanes plus a buffered bike lane. (The bike lane on Cleveland will offer a travel option for cyclists heading in […]