There is a solution to puddles at corners

All pedestrians experience the frustration of puddles at corners (these are in addition to the puddles that form at most driveways). There is a solution, it is simple, and it calms traffic too.

Don’t dip the sidewalk when crossing side streets. Simple, eh? Now this is different from Ottawa’s too-gentle experiments with “intersection tables” and other timidities.

Here’s some close up pictures of what a pedestrian-first crossing of a sidestreet could look like. And there won’t be slush or puddles designed in from the beginning, like in Ottawa.



Isn’t that simple? And workable? And it will succeed in slowing down car traffic as it leaves a busy street in enters a slower residential street.

Here’s a close up of the corner detail:



Note there is a catch basin right at the traditional spot in the corner to catch water running along the gutter. The yellow dot pattern helps direct pedestrians to the crossing. In Ottawa, we use slots or groves cut into the sidewalk to convey that tactile message, so it’s not like everyone is going to be falling off the curb. With all our snow plowing, we might prefer to make the road slope be concrete too.

I suggest that raised crossings only because we have so much already-built current road infrastructure. These crossings could be added without tearing up all the road.

Of course, the better solution is to stop treating city streets as if they were country roads that slope water into an adjacent ditch. Drain the water away from the sidewalk, to the centre line of the road. Salt and car traffic will keep the snow mushed and washing down the catchbasins.

It might even make citizens want to walk somewhere, or even to walk to the bus stop again. But we have a century of anti-pedestrian design to overcome. And we shall.

16 thoughts on “There is a solution to puddles at corners

  1. Is not the road sloping away from the center also a safety feature, i.e. the sloping makes cars tend to drift away from each other as opposed to toward each other?

    I don’t know if this is true or, even if it is, how much of an impact it actually has, but it’s the only argument I can think of against inward-sloping roadways.

      1. The convex shape was definitely developed for drainage in rural environments from Roman times, but I’m sure there are those who would argue it is a motor vehicle “safety” feature as well, and I imagine that would be the principal objection to concave cross sections with centreline drainage.

        But I would turn that argument on its head: perhaps if roads are sloped to the centre, and therefore appear more dangerous (whether they are or not), then motorists might drive more carefully, i.e. slowly.

        There actually are some streets in Ottawa with centreline drainage. There’s a number of quasi-private streets west of Centrepointe, such as Stonebriar Drive, that drain to the centre. I haven’t visited them in person, but looking at them on Streetview they do seem to have far fewer catchbasins than regular city streets.

        A lot of parking lot driving aisles also function with centreline drainage, and since parking lots seem to have (require?) zany slopes, if there is a safety issue, it should become apparent there.

  2. I don’t drive enough to have a strong opinion on road slope and drifting into the curb. But all divided streets eg Carling, slope to the centre line for two to four lanes, and at large intersections like on hunt club or woodroffe where there are multilanes, the road slopes both to the outside curb and the median curb. Do cars often drift out of their lane into the curb? Is that due to the slope or the motor pulling the car to where the sleepy motorist directed it?

    I suspect someone somewhere (probably a Councillor) would object to cars sliding on ice in the centre of the road and hitting the opposing car, thinking this is less hazardous than driving onto the sidewalk … but I think there would be much less ice in the centre as all the salt and tire friction would be higher there than the outside lane edge. Snow would continue to be plowed to the edge, making it a soft receptacle for errant voitures.

  3. What do you think of your raised crosswalk solution as far as it concerns cyclists? Wouldn’t this, given current road cross-sections, turn every crosswalk into a super speed bump for cyclists?

    1. David – That might not be a bad thing. I am a cyclist and get completely ticked off by other cyclists blasting through intersections against the light as they look for cars, do not see them, and then it is full speed ahead. Getting T-boned by a cyclist at 20kph is enough velocity to kill.

      1. fjf: “Getting T-boned by a cyclist at 20kph is enough velocity to kill.”

        Umm, check your physics. The chance of death when a pedestrian is hit by a car at 32km/h is just 5% (source:, a car will impart much more energy then a bike at the same speed. Pedestrian fatalities caused by cyclists are so rare as to be statistically insignificant. Not to say they never happen, but seriously.
        More relevant to David’s original query: “…turn every crosswalk into a super speed bump for cyclists?”. No, because cyclists are not impacted by speed bumps the same way cars are. These table intersections will not be noticed by even the fastest of cyclists.

        1. Thanks for the URL. But where in that mass of information did you get the precise fatality rate for a 32 km h impact? I cannot find it and I looked.

          Without being able to consult the source I am skeptical. In the 1970s at UBC a cyclist was following the roadway centre line in fog and travelling slowly due to near 0 visibility. He died when he encountered a second cyclist travelling in the opposite direction using the same navigation tactic. A second instance is drawn from a local sports field. Young woman playing field hockey on a turf surface died in July?? of last year due to an impact injury. It is difficult to sustain speeds much above 12 km h in human locomotion. Stand on a concrete slab, loose consciousness and fall the wrong way and you can sustain a fatal injury. Concrete is seriously hard stuff!!

          Granted a 2 ton mass at 32 km h will impart more energy than a 200 lb cycle and rider at the same speed. But the nature of the impact counts a great deal. And then we can get into the trope on lies, damned lies, and statistics. Statistics shows that most statistics are sheer bunkum. I have no wish to be the pedestrian outlier. Or the cyclist either.

          1. Here is the quote:
            Car Speed % fatally injured pedestrians
            32 km/h 5
            48 km/h 45
            64 km/h 85
            And as a graph the probability of fatal injury for a pedestrian colliding with a vehicle looks like this (source: Pasanen, 1991)”
            It is the last paragraph of a short document just above a graph that plots out the above table. I don’t believe you followed the link.
            Your two anecdotes are meaningless, and separated in time by over 40 years? Come on get a grip.
            This is what saddens me about this conversation: A freak accident in the 70’s and a sports injury, play a larger role in your mind then the very real likely hood that a car will recklessly run you down in an intersection. A scenario that plays out every day in North America. Yet, the danger is bikes. Sad.

  4. This approach would have other applications and might save the city money. I am thinking of kirkwood which has a series of speedbumps placed at the mid-point of the street segment between intersections. So cars slow for the speedbump and then race forward through the unsignalled intersection toward the next speedbump. In other words cars are paying attention to the speedbumps not to the pedestrian traffic supposedly being protected by those speedbumps.

    Place the speedbump at the intersection and cars would slow before entering the intersection. This should improve pedestrian safety as les voitures are slowing at exactly the point they are likely to encounter pedestrian traffic rather than accelerating in order to make up time lost to slowing for the last speedbump.

    To make this work only one side of the intersection would likely be raised but this still should be effective for opposing traffic. The plows already get through existing speedbumps without taking them out so that should not be a problem.

    And thanks for the post. Not in 100 years would I have thought that there was an alternate set of arrangements for intersections. Always thought the way Ottawa does it must be the way of the rest of the world.

  5. We all know that catch basins overflow or become clogged with snow and ice… That’s why we get these puddles. With centreline drainage you could end up with a wide puddle down the centre of the street, obstructing the travelled lanes. Although puddles are inconvenient to us pedestrians, they could be downright dangerous to motorists and certainly hamper emergency response times. There’s also to obvious fact that snow would get plowed to the sides and then melt towards to middle, creating a wet road succeptable to freezing.

    The real engineering solution is to overcome catch basin blockages from snow/ice, but it impracticle for them to be designed to accommodate extreme rainfall events.

    1. How would we end up with wide puddles down the centre of the street with centreline drainage?

      The reason that edge-of-road catch basins become clogged with snow and ice is because the snow plows plow the snow right on top of them in great huge banks. Short of having high-wattage heaters running on the catch basins themselves, I don’t see how there’s an real engineering solution to address snow-and-ice clogging of edge-of-road catch basins. The problem might be mitigated somewhat by plowing to the centreline, but then all the things you’re worried about would present themselves again.

      With centreline drainage the catch basins would generally remain clear due to the fact that the roads get plowed.

      I think I’d take roads susceptible to freezing over actual freezing, which is what we get with the enormous puddles right now. Those puddles can sometimes hide ice, amongst other things (particularly if the puddle forms in a shaded area, it can remain frozen while water from sun melt flows over it – eventually the ice below will melt, but for a few hours it can remain frozen). In reality, roads sloped to the centreline are only going to freeze over during a flash freeze event – and it’s not like our sloped-to-the-side roads were immune to that anyway.

  6. you could end up with a wide puddle down the centre of the street, obstructing the travelled lanes. …. I don’t think so. There are puddles on the sides of the roads because our engineers design in all sorts of things to slow the flow of water, like depressions at every driveway, where water can collect due to the gutter being used to store snow and ice. If the gutters were clear and the sidewalks left elevated at every driveway, the gutters would be free to drain to the catchbasins. This is exactly what the centreline of the street is: well plowed, laden with salt, and free flowing to the catch basins which would be twice as frequently located (ie, half the draining distance) if the same number as are at the curbs were moved to the centreline. Ergo, puddles would be rare, As for the emergency vehicles canard, I see this trotted out whenever someone can’t think of convincing reason to oppose a bulb out, or speed bump, or a bike lane, the Big Bad Boggeyman of delayed emergency vehicle times comes up. That’s why, of course, the Main Street complete street cannot work, because of the ambulances, see ??

  7. The city has installed structures similar to these along the eastern end of the Byron Linear – at Clarendon and Harmer Ave’s. I’d love to see similar installations along the entire length of the BLP (Granville, IPD, Hilson, Kirkwood, Tweedsmuir, Athlone) and the Transitway path from Bayview to Churchill.

    I’m not sure about drain placement, but this offers a local case study…and an opportunity for Eric to venture to the ‘far west side’ with camera in tow.

  8. Anything that knocks someone to the ground is potentially fatal. There are been a number of fatal cases in Ottawa in the last decade where people were punched in the head outside a bar, lost consciousness, fell and struck their heads,

    I knew someone who fell from the second step on a ladder. He didn’t make it. Falling is bad. Anything that knocks you down on a hard surface is potentially fatal.

    Your comment below, which basically calls another poster a liar is over the line in my opinion.

    Finally, you seem very sensitive about the issue of pedestrian and bike interaction. I can say from personal experience that I have felt threatened by cars and bikes, but drivers are more likely to seem surprised and apologetic, whereas the cyclists who bike aggressively enough to be dangerous seem offended that you got in their way. They feel they own their speed, and the pedestrian just stole it from them. I think those cyclists spend so much time optimizing their interactions with dangerous cars like mice avoiding elephants that they forget they aren’t the smallest and slowest thing on the road.

    Bad cyclists make cyclists look bad, just like bad drivers make drivers look bad.

    1. Peter Drake – I assume you were responding to me and my back and forth with fjf.
      You have spectacularly missed the point. The point that David raised, the potential inconvenience to cyclists.
      David – “What do you think of your raised crosswalk solution as far as it concerns cyclists?”
      My answer, based on experience, is raised intersections will not slow down cyclists. I should have put this first in my response to David, ahh hindsight being 20/20.
      Still, fjf and now you do deserve a response. As I initially said, “Pedestrian fatalities caused by cyclists are so rare as to be statistically insignificant. NOT TO SAY THEY NEVER HAPPEN, but seriously.” (emphases added). I acknowledge the frailties of human beings! Stop with your meaningless anecdotes, and stop inflating them. This conflating the possible with probable is infuriating. We have city streets where normal rate of travel is >70kph (Main, Carling, Holland etc.) yet we tie ourselves in knots over a few cyclists travelling <30kph, madness.
      Why is basically calling another poster a liar, over the line? fjf claimed to have "looked". Further he used his inability to find the precise fatality rate for an impact at 32kph to discredit my source. "Without being able to consult the source I am skeptical." All fjf had to do was -f and key in 32 and he would have found it. Still, the information was not hidden in a “mass of information”. it is clear, easy to find. The first entry on a table headed: “Car Speed % fatally injured pedestrians”. Really I feel I was quite restrained. Obviously fjf did not look, did you?
      If only “…bad drivers make [made] drivers look bad.” perhaps I’d be less sensitive to the lack balance in the bike/car vs. pedestrian comments.
      Here is the bottom line, Bike/pedestrian improvements when they come together improve the lot of EVERYONE, including drivers. Focusing on the rare or imaginary conflict between pedestrians/cyclists serves nobody, but the status quo.
      We should be putting our energy into pedestrian areas that do not flood in winter thaws or spring rains. Intersections that privilege active transport, not cars. Main street areas that are welcoming to pedestrians and cyclists who ultimately are the consumers who will keep these areas vibrant. Lowering speed limits to 30k as is being advocated in GB with the “20s Plenty” campaign. Falls are dangerous!,Lets figure out ways to make the sidewalks less slippy, rather then obsess over a few misbehaving cyclists (who in the scheme of things present very little threat).

Comments are closed.