Bollards: A Barrier to Safety?

Bollards are, in theory, intended to keep people safe. So why do people keep getting hurt by them?

Bollards. You know, those posts they put in the middle of the trails, usually at entrances?

In theory, they’re there to keep vehicles from entering the trails and endangering pedestrians and cyclists on the trails.

But in practice? It seems, lately, that they’re there to create one more hazard for the users they’re supposed to protect.

Arlington cycling advocates have been pushing Arlington County to remove a number of useless and/or dangerous bollards on Arlington’s trails for a while, but there seems to have been an uptick of bollard-related injuries this year. It’s time for the County to formulate a coherent policy on where bollards should be used. Then, it needs to act to remove the bollards that pose an undue danger to trail users.

If the purpose of bollards is to deter motor vehicles from accessing pedestrian and bicycle facilities, they should only be located at points where trails and streets intersect. This makes particular sense at points where a driver may mistake a trail for a street. There aren’t too many of these situations in Arlington, but the W&OD near Bluemont Park is no stranger to SUVs driving to the picnic pavilion. And I can imagine how bored teenagers might find it fun to try to drive up the access bridge from the Roosevelt Island parking lot. So yes, bollards do have their place.

However, there is no reason bollards should be placed in trail/trail intersections or along the middle of trails. Yet Arlington is not only full of these instances, but they seem to be proliferating.

Some of them are historical artifacts – the bollard at the W&OD trail intersection just east of Lee Highway is there because a road used to lead up to that point. But despite all the work that was done to turn that area into a park with a trail, the original bollard was left there, needlessly endangering trail users. It, and other similarly redundant bollards, should be removed.

Then there are the new bollards that have been installed because . . . well, honestly, it’s something of a mystery to me. For example, it’s still not entirely clear why the forest of bollards on the new bridge near East Falls Church were installed. The bridge was already inside of a protected perimeter (i.e., there are no cars anywhere near here). Yet the County installed three bollards on one end and one on the other (just to keep things even?). After being pressed at a public hearing, the County seems to have relented, removing some of the bollards.

But not all of the bollards were removed from the new bridge. It appears – and now I should emphasize that I’m speaking from second-hand information – that some parts of Arlington County government think that bollards are a great way to calm trail traffic. This would explain, in part, the bollards posted like ski-gates on the bridge. But this over-application is plainly out-of-step with the best practices of transportation professionals.

The Federal Highway Administration provides guidance specifically for the use of bollards and other barriers. It notes

Even "properly" installed bollards constitute a serious and potentially fatal safety hazard to unwary trail users. [ . . .] For these reasons, bollards should never be a default treatment, and should not be used unless there is a documented history of intrusion by unauthorized cars, trucks, or other unauthorized vehicles.

Bollards are often a hazard to trail users, who can crash into them, possibly resulting in serious injury or death. Poorly installed bollards can lead to head-on collisions. Bollards are involved in "second user" crashes, where the first user hides the bollard until it is too late to avoid it, even if the first user has adequate sight distance. These crashes can produce serious or incapacitating injuries. This can happen to pedestrians as well as bicyclists or other higher speed users. 

There have been several serious crashes directly related to bollards in Arlington County. One, earlier this week, appears to have happened at the west entrance to the bike/pedestrian bridge crossing the George Washington Parkway to the Mount Vernon Trail. This bollard is utterly superfluous, and is just one of many the County should have removed by now.

When the County finally does take the danger of bollards seriously, and starts removing the unnecessary ones, it should be sure to remove the bollard collars left behind. This is the part of the bollard base that sticks out of the ground. Short, metal and usually poorly marked, it’s a menace to pedestrians and cyclists alike. Concerns about empty bollard collars have been raised before, but their removal is not a priority for the County. That needs to change.

Arlington County needs to develop a consistent policy regarding bollards that takes into account the FHWA’s guidance and the actual dangers posed by vehicle traffic at proposed bollard locations. Any future bollard installation anywhere in Arlington should respect this policy. Finally, the County should take an inventory of existing bollards, and develop a schedule for removal of those that don’t comply with this policy.


Where are all these bollards? Check out this map, which has been maintained by citizens concerned about this problem.  You, too, can contribute.

Mark Blacknell is chairman of the Arlington Bicycle Advisory Committee, president of the Washington Area Bicyclist Association, and a League Cycling Instructor.


About this column: A regular examination of cycling in Arlington and what its growth means to our community.

Michael Doan June 21, 2012 at 01:29 PM
I've seen some plastic bollards that bounce back after impact. Wouldn't those solve the problem?
Mark Blacknell June 21, 2012 at 01:57 PM
Michael - those bollards still represent a hazard (albeit with a lower possibility of severe injury). They're still subject to being hidden by riders/pedestrians in front of you, and while clipping the top (with your handlebars) is something most people can recover from, hitting the base isn't. And if you check out the bases of most of those kinds of bollards in the area? You'll see rubber from tires. So yeah, they're somewhat better than the concrete/steel bollards, but nothing is probably better than them.
Yoeun Pen June 21, 2012 at 02:10 PM
As a lycra-wearing cyclist with several thousand miles on my road bike, I want MORE bollards. They're not for keeping cars off the trail. They're to protect cars and people from reckless cyclists speeding down the trail, ignoring stop signs and blasting through intersections. I've ridden long enough to know that these types of riders make up a significant portion of the cycling demographic. They complain about sharing the road and wanting dedicated European bike lanes, but at least cyclists in Europe stop at red lights!
Mark Blacknell June 21, 2012 at 09:16 PM
I'm curious - how do you think that these bollards actually fix the problem you're ID'ing, Yoeun?
Yoeun Pen June 21, 2012 at 10:05 PM
It forces cyclists to slow down as they approach the bollard. They're like speed bumps, but safer.
Peter June 21, 2012 at 10:36 PM
Meanwhile, Mommies pushing strollers and skateboarders are on the dedicated 'bicycle' lanes.
dave June 22, 2012 at 10:53 PM
based on FHA guidelines and documented crashes I feel confident a large settlement is in the making. too bad county representatives can be personally sued for negligence.
Eric Wagner June 23, 2012 at 01:05 AM
Good observations, Mark. Consider that this is happening in a county that is generally bike friendly! Alexandria has its own critical issues with bollards at the Woodrow Wilson Bridge, as you well know.
Geof Gee June 23, 2012 at 01:13 PM
Interesting ... so are you for putting bollards on roads where car drivers often ignore the rules of the road. Consider the areas around Memorial Circle with pedestrian crossings where drivers are often way over the 25 mph speed limit. http://goo.gl/maps/9ptJ Now that I think about it more, perhaps we should rip out seat belts and put a spike in the middle of every steering wheel pointed at the driver. Six inches will be enough. Mind you, I concur with the underlying premise that changes in risk result in changes in behavior. But if you believe that bollards on Washington Blvd and the spike in the steering wheel are bad ideas even though is far more mortality and serious injuries associated with driving, then we might want to re-think the position on bollards.
Yoeun Pen June 23, 2012 at 01:50 PM
Real mature, Geof. Your analogy with seat belts and spikes on steering wheels is ridiculous because bollards aren't inherently dangerous. They're simply not. A brick wall is dangerous if I ride right into it, but not if I drive around it. These cyclists should be paying attention to the road whether they're watching out for patches of gravel, potholes, negligent drivers, speed bumps, pedestrians, old ladies or bollards. If you're not barreling down the trail heads down at 30 mph, you will see all these objects on the road with plenty of time to respond. I'm from Seattle and I assume there are bollards everywhere on the trail, because in Seattle there are. Sometimes even gnarly mazes that force cyclists to stop and walk their bikes through it right in the middle of the trail. I've never heard of these problems from the giant biking community there.
Yoeun Pen June 23, 2012 at 01:52 PM
Trails are not dedicated bike lanes. They're shared use trails. Cyclists tend to forget that.
Geof Gee June 23, 2012 at 03:37 PM
There is a whole literature on what I mentioned. Peltzman is the guy who wrote and did an early analysis with seat belts. You can follow ~30 years of literature after him. A spike in a steering wheel is only dangerous if the driver crashes at a "high" speed. So drivers would drive more carefully/slowly/alertly and there should be far less collisions. Isn't that your precise argument with bollards ... that people can mitigate their risks by adjusting their behavior? Think about how safe our roads would be such that all those speeding cyclists would ride in the roads too! Moreover, with cars becoming safer conditioned on being in a collision, there is an obvious externality. If drivers go faster or act more carelessly on the margin after the introduction of safety devices, their personal risk may change little but the risk to pedestrians, cyclists, motorcyclists, and other road users increases. If we share the same belief that getting yourself killed/injured is pretty bad but getting someone else killed/injured is far worse, I'd think you'd be even more motivated to place those bollards in the middle of Washington Blvd.
Yoeun Pen June 23, 2012 at 04:10 PM
But a bollard is not a six inch spike in the steering wheel. That's my point. It's not inherently dangerous like a spike is.
Geof Gee June 23, 2012 at 04:59 PM
Both are "dangerous" conditioned on a certain level of care and attention. If drivers change their behavior enough -- similarly with cyclists and bollards -- the spike won't be "dangerous". I'm arguing that "inherent" is poorly defined here. Anyway, a spike in the center of the driving wheel is a simple rhetorical device. It would be a mistake, for everyone, to be fixated on it. If we want to encourage safer speeds and/or behavior on MUPs/roads we should be thinking about what the right tradeoff between risk and "return" -- the convenient, safe, fun travel -- and what are the best ways to encourage the best behavior.
Geof Gee June 23, 2012 at 05:01 PM
It seems to me that MUPs are a place where there is a wide range of ability among users. We have adults, seniors, and kids using the MUP with their feet, bikes, whatever. Consequently, some model about how these different groups will react to different environments is needed. Now the general argument for bollards is about preventing unwanted motorized vehicles from using the MUPs. Yoeun adds that they are also a "safety" device. Mark addresses the first argument. The second argument could be true theoretically. But I'm thinking that you'd have to carefully look at the data and have a pretty good working model of how those MUP users react to these devices. Moreover, you'd also want to consider how bollards perform as safety devices relative to alternatives. These are clearly empirical questions and I'm inclined to think that the engineers at FHA are basing their opinion on whatever evidence is out there. Just speculating based on anecdotes and what I know about the development of risk behavior, I suspect that bollards have some serious negative outcomes -- particularly since they tend to be located near streets -- and they perform poorly as safety devices for younger MUP users. Perhaps I missed it, but as someone who does a lot of utility riding and riding with my kids with trailers, the bollards are a royal PITA since they often leave too little space.
Allen Muchnick June 27, 2012 at 06:33 PM
Yes, trail bollards are a bicycling hazard, but the new flexible bollards are a trivial hazard compared to the County's plan to install streetcar tracks in both curb lanes of Columbia Pike. Arlington bicyclists need to get their priorities straight.
Kenneth Parrott May 09, 2013 at 08:00 PM
I agree that the flexible bollards are much less of a hazard, but they won't prevent a car from going down the trail.
Carmen Sandiego February 10, 2014 at 10:19 AM
I'm actually a huge fan of bollards. They have all kinds of uses and practical applications. One reason I like them is because you can put them in your driveway and keep cars from parking or getting too close to your yard. I'm thinking about getting them from <a href='http://www.singhfab.com.au/bollards.html' >bollards rocklea</a>.
bryan flake April 14, 2014 at 06:05 PM
When my brother and I were in high school, we went to a twenty-four hour taco stand at two a.m. We were pulling through the drive thru, when my brother got a bit to close to the safety bollard and dented the whole driver's side of the car. When I see safety bollards, I think about our taco excursion that night. bryanflake1984| <a href='http://www.lighthouse-bollards.com/Removable.html' >http://www.lighthouse-bollards.com/Removable.html</a>


More »
Got a question? Something on your mind? Talk to your community, directly.
Note Article
Just a short thought to get the word out quickly about anything in your neighborhood.
Share something with your neighbors.What's on your mind?What's on your mind?Make an announcement, speak your mind, or sell somethingPost something