Crash prone 'modern roundabouts'

[MNHighwayMan]

No problem Jake.  I used (fumbled) with Photoshop to make the overlay design above but I'm just not very good with that program.  I'll take a look at Adobe Illustrator.   

Or if you're unwilling to pay Adobe the unreasonable prices they charge to use their software, Inkscape is the free and open-source alternative to Adobe Illustrator. It's what I use, and is more than good enough to create sign graphics (or things like roundabout layouts), even if it lacks a lot of the fancier features Illustrator has.

Adobe's CC programs are expensive if you're a non-vocational user, but they're dirt cheap for those who use them professionally. That doesn't necessarily describe me, but $50/month for Adobe's entire suite is a bargain if you want my opinion.

Or $0 forever for CS6 if you're willing to put in a bit of effort to visit sites of slight ill-repute.

(Not that I've done that, no... Totally not.)

[jakeroot]

No problem Jake.  I used (fumbled) with Photoshop to make the overlay design above but I'm just not very good with that program.  I'll take a look at Adobe Illustrator.   

Or if you're unwilling to pay Adobe the unreasonable prices they charge to use their software, Inkscape is the free and open-source alternative to Adobe Illustrator. It's what I use, and is more than good enough to create sign graphics (or things like roundabout layouts), even if it lacks a lot of the fancier features Illustrator has.

Adobe's CC programs are expensive if you're a non-vocational user, but they're dirt cheap for those who use them professionally. That doesn't necessarily describe me, but $50/month for Adobe's entire suite is a bargain if you want my opinion.

Or $0 forever for CS6 if you're willing to put in a bit of effort to visit sites of slight ill-repute.

(Not that I've done that, no... Totally not.)

Oh I did when I was younger! But the subscription model is much easier to manage (financially), and I always have the latest updates so I'm happy to being doing it legally now.

[tradephoric]

Another fatal Summerlin roundabout crash just happened.  This comes 7 months after a fatal crash at the Village Center Dr and Meadow Hills Drive roundabout in Summerlin back in November.  The very first modern roundabouts built in America haven't done a very good job at preventing fatalities lately.

Motorcyclist killed in crash at roundabout in west Las Vegas
https://www.reviewjournal.com/local/summerlin/motorcyclist-killed-in-crash-at-roundabout-in-west-las-vegas/

[jakeroot]

Hmm. He hit the curb going right? I'm guessing that means he misjudged the curb between the right turn slip lane and the circulating lanes? Hard to say, but I know those slip lanes haven't always been there.

[roadfro]

Another fatal Summerlin roundabout crash just happened.  This comes 7 months after a fatal crash at the Village Center Dr and Meadow Hills Drive roundabout in Summerlin back in November.  The very first modern roundabouts built in America haven't done a very good job at preventing fatalities lately.

Motorcyclist killed in crash at roundabout in west Las Vegas
https://www.reviewjournal.com/local/summerlin/motorcyclist-killed-in-crash-at-roundabout-in-west-las-vegas/

Hmm. He hit the curb going right? I'm guessing that means he misjudged the curb between the right turn slip lane and the circulating lanes? Hard to say, but I know those slip lanes haven't always been there.

The "first modern roundabouts" built in America are the two north of Summerlin Pkwy, on Village Center Circle at Town Center Dr and at Hills Center Dr/Meadow Hills Dr, both constructed circa 1990.

This particular crash happened at Town Center Dr & Hualapai Way, constructed circa 1995 and modified to include the right turn slip lane islands between 2000 and 2004 (according to Historic Aerials imagery).

[tradephoric]

They just can't seem to fix the notorious Britannia roundabout in Adelaide.  Even after a major $3.2 million reconfiguration of the roundabout in 2013 that converted it from a single roundabout to a double roundabout, casualty crashes are the exact same as before the reconfiguration (at the time experts said the reconfiguration would 'reduce casualty crashes by up to 23 per cent').   The troubled intersection is going to be discussed at an upcoming Adelaide's City Counsel Meeting and the new option on the table is grade-separation. 

Traffic experts float new Britannia roundabout overhaul
https://indaily.com.au/news/2018/06/25/traffic-experts-float-new-britannia-roundabout-overhaul/

[MNHighwayMan]

I mean this in the sincerest way possible, but: do you search for news about roundabouts every morning, or something? I'm curious how it is you find all these news articles all the time.

[tradephoric]

I mean this in the sincerest way possible, but: do you search for news about roundabouts every morning, or something? I'm curious how it is you find all these news articles all the time.

The very first reply on this thread was made after I linked a news article about an Ann Arbor roundabout that experienced 170 crashes in 2013, which was the most crash prone intersection in Michigan.

Speaks more to the (in)flexibility of Michigan drivers than it does any inherent safety risk with roundabouts.  New roundabouts elsewhere have seen major drops in the number of crashes.

I wanted to see if froggie was right.  While some roundabouts have certainly seen major drops in the number of crashes, froggie failed to mention all the examples of roundabouts that have seen major increases in the number of crashes.  I think this thread has done a good job highlighting which types of roundabouts are likely to see big increases in crashes, but that does take a lot of research.  It's not enough just to show that there have been big increases in crashes at 2x2 and 2x3 roundabouts in Michigan, but when you can show that this is a pattern nationwide I think it's more effective.  To answer your question no I don't search for news stories about roundabouts every morning, but it's not that hard to keep track of the known problem roundabouts if you really want to.  I'd argue that many supposed roundabout experts have an affinity for roundabouts, and don't want to highlight the problematic ones as this could erode public support for their beloved roundabouts!  I don't care about that.

[jeffandnicole]

...To answer your question no I don't search for news stories about roundabouts every morning, but it's not that hard to keep track of the known problem roundabouts if you really want to...

Being that you have a tendency to write posts about roundabouts that have no known history of problems and happen to have a single fatal accident, and your posts are filled with links to newspaper and TV news websites, your answer seems to be a non-answer.  Or are you saying you look for stories in the afternoon or evening, so "I don't search every morning" is technically correct?

[tradephoric]

Being that you have a tendency to write posts about roundabouts that have no known history of problems and happen to have a single fatal accident...

A single fatal crash is a significant event.  According to published data, signalized intersections in America average a fatal crash about every 100 years and drivers have only been driving through modern roundabouts in this country for 27 years.    Considering there have been 2 fatal crashes at Summerlin Nevada roundabouts over the past several months - the first community in America to build modern roundabouts — they instantly become fatality prone intersections.  Supposedly roundabouts reduce fatal crashes by 90% when compared to signalized intersections.  If that's the case, I'd expect a fatal crash at a roundabout once every 862 years or something insane like that.. not 2 fatal Summerlin roundabout crashes (the birthplace of the modern roundabout) in just 27 years. 

[jakeroot]

Supposedly roundabouts reduce fatal crashes by 90% when compared to signalized intersections.  If that's the case, I'd expect a fatal crash at a roundabout once every 862 years or something insane like that

Yes, if that 90% was compared to any old signal. But to the best of my knowledge, that data was based on roundabouts which replaced crash-prone signals. I don't think there's any expectation that all roundabouts reduce fatal crashes by 90%, since not every intersection experiences that many fatalities.

[kalvado]

Supposedly roundabouts reduce fatal crashes by 90% when compared to signalized intersections.  If that's the case, I'd expect a fatal crash at a roundabout once every 862 years or something insane like that

Yes, if that 90% was compared to any old signal. But to the best of my knowledge, that data was based on roundabouts which replaced crash-prone signals. I don't think there's any expectation that all roundabouts reduce fatal crashes by 90%, since not every intersection experiences that many fatalities.

Another common narration for roundabouts is that due to lower traffic speeds and more favorable collision angles (no high speed t-bones), fatal accident rate drops way way down, as well as serious injury ones - even if property-only go up. 90% seems very conservative number with such a narration... 

[jeffandnicole]

I mean this in the sincerest way possible, but: do you search for news about roundabouts every morning, or something? I'm curious how it is you find all these news articles all the time.

Being that you have a tendency to write posts about roundabouts that have no known history of problems and happen to have a single fatal accident...

A single fatal crash is a significant event.  According to published data, signalized intersections in America average a fatal crash about every 100 years and drivers have only been driving through modern roundabouts in this country for 27 years.    Considering there have been 2 fatal crashes at Summerlin Nevada roundabouts over the past several months - the first community in America to build modern roundabouts — they instantly become fatality prone intersections.  Supposedly roundabouts reduce fatal crashes by 90% when compared to signalized intersections.  If that's the case, I'd expect a fatal crash at a roundabout once every 862 years or something insane like that.. not 2 fatal Summerlin roundabout crashes (the birthplace of the modern roundabout) in just 27 years. 

Followup question: Are you a politician?  Because you never actually answer the questions posted.

[tradephoric]

I love it.  La Crosse's first roundabout replaced a signalized intersection that didn't have any crashes the year before the roundabout was built.  It's really hard to improve on ZERO crashes.  What would a 35% reduction of zero crashes be?  The real question is why did the city pick an already safe intersection to construct a roundabout?  Admittedly the previous intersection was an odd staggered four-way intersection, so that probably had something to do with it, but it doesn't appear to have been picked based on its safety record.  Days before the official opening of the roundabout in 2017, a motorcylist nearly killed himself after plowing through the roundabout at high speed.  But... but... but... we were told roundabouts reduce total crashes by 35%!  Tell that to the people of Ann Arbor, where a double-lane roundabout there saw a 10 fold increase in crashes.  But how gullible do you have to be to believe that ZERO crashes will be reduced by 35%.  Silly headline indeed!  According to the article there were 7 crashes last year at the roundabout (defies expectations.. shocking that there would be more than ZERO crashes!). 

First year roundabout accident totals on Cass Street defy expectations
http://www.1410wizm.com/index.php/home-m/item/34063-first-year-roundabout-accident-totals-on-cass-street-defy-expectations

[jakeroot]

Here's hoping those seven collisions are just a first year thing.

I just drove through a brand new (two week old-ish) roundabout near Anacortes, WA yesterday, and there were plenty of horns. No accidents, though. I don't know if there were any accidents before the roundabout, but I drove through this roundabout (the Sharpes Corner Roundabout) and it's brother to the south (the Howards Corner Roundabout) seven times in a very short amount of time. So I give it a thumbs up. The old signalized intersection was just overloaded. No traffic issues when I approached any of the seven times.

The only issue I see is the path overlap along the southern edge of the roundabout (left entry in the photo below). Drivers were negotiating it fine, since both lanes had drivers entering simultaneously, preventing drivers from just ignoring the lane lines. But some drivers do let their guard down, and might just go from the right lane to the inner circulating lane if they're not paying attention.

https://twitter.com/wsdot_north/status/1012706908422062088

Video from WSDOT:

https://twitter.com/wsdot_north/status/1011630333807112192

Note the link in the tweet; there are live cameras of both roundabouts online: http://www.wsdot.com/traffic/cameras/OakHarbor.aspx?cam=9498

[kalvado]

I just drove through a brand new (two week old-ish) roundabout near Anacortes, WA yesterday, and there were plenty of horns. No accidents, though. I don't know if there were any accidents before the roundabout, but I drove through this roundabout (the Sharpes Corner Roundabout) and it's brother to the south (the Howards Corner Roundabout) seven times in a very short amount of time. So I give it a thumbs up. The old signalized intersection was just overloaded. No traffic issues when I approached any of the seven times.

The only issue I see is the path overlap along the southern edge of the roundabout (left entry in the photo below). Drivers were negotiating it fine, since both lanes had drivers entering simultaneously, preventing drivers from just ignoring the lane lines. But some drivers do let their guard down, and might just go from the right lane to the inner circulating lane if they're not paying attention.

Looking at old Google maps images, conversion of southbound exit from the intersection to 2 lanes is the trick. 

[tradephoric]

Here's hoping those seven collisions are just a first year thing.

I just drove through a brand new (two week old-ish) roundabout near Anacortes, WA yesterday, and there were plenty of horns. No accidents, though. I don't know if there were any accidents before the roundabout, but I drove through this roundabout (the Sharpes Corner Roundabout) and it's brother to the south (the Howards Corner Roundabout) seven times in a very short amount of time. So I give it a thumbs up. The old signalized intersection was just overloaded. No traffic issues when I approached any of the seven times.

The only issue I see is the path overlap along the southern edge of the roundabout (left entry in the photo below). Drivers were negotiating it fine, since both lanes had drivers entering simultaneously, preventing drivers from just ignoring the lane lines. But some drivers do let their guard down, and might just go from the right lane to the inner circulating lane if they're not paying attention.

Looking at the aerial of the old intersection it does appear vehicles traveling WB turning left onto SR20 were nearly backing up onto the through lanes of traffic.  The increased capacity of the roundabout with the double left turns now should clear out those left turners, which is a great thing.

My concern is traffic traveling the other direction.  Here is a streetview of the road about 1,000 feet from the roundabout.  Traffic approaching the roundabout is traveling downhill through a curve in what resembles a limited access highway.  Now the same arguments can be made when the intersection was a traffic signal, but in the scenario where an out of control truck loses his brakes, the trucker approaching a traffic signal at least has the chance of blowing through a green light (and if he does blow through a red light, an observant left-turner who has a green may yield to the multi-ton truck, because they don't want to die).  Now as the out of control truck with no brakes approaches the sharp curves of a roundabout... well good luck slowing down to 25 mph so you don't tip your load over.  This roundabout just seems tailor made to be a truck rollover hotspot (and I do see a fair share of 18-wheelers in the google earth aerial).   Will I be at all surprised if I hear about a truck rolling over at this roundabout within 6 months?  Not at all.



EDIT: I may be exaggerating how big a downhill grade it is approaching the roundabout.  It looks like it's going downhill, but i'm sure you have a better idea about that.  Even still, a truck just traveling along what appears to be a limited access highway isn't necessarily going to be expecting to see a roundabout pop up in the middle of the road.  Hopefully there are a LOT of advanced warning signs for this roundabout in particular. 

[jakeroot]

Here's hoping those seven collisions are just a first year thing.

I just drove through a brand new (two week old-ish) roundabout near Anacortes, WA yesterday, and there were plenty of horns. No accidents, though. I don't know if there were any accidents before the roundabout, but I drove through this roundabout (the Sharpes Corner Roundabout) and it's brother to the south (the Howards Corner Roundabout) seven times in a very short amount of time. So I give it a thumbs up. The old signalized intersection was just overloaded. No traffic issues when I approached any of the seven times.

The only issue I see is the path overlap along the southern edge of the roundabout (left entry in the photo below). Drivers were negotiating it fine, since both lanes had drivers entering simultaneously, preventing drivers from just ignoring the lane lines. But some drivers do let their guard down, and might just go from the right lane to the inner circulating lane if they're not paying attention.

Looking at the aerial of the old intersection it does appear vehicles traveling WB turning left onto SR20 were nearly backing up onto the through lanes of traffic.  The increased capacity of the roundabout with the double left turns now should clear out those left turners, which is a great thing.

My concern is traffic traveling the other direction.  Here is a streetview of the road about 1,000 feet from the roundabout.  Traffic approaching the roundabout is traveling downhill through a curve in what resembles a limited access highway.  Now the same arguments can be made when the intersection was a traffic signal, but in the scenario where an out of control truck loses his brakes, the trucker approaching a traffic signal at least has the chance of blowing through a green light (and if he does blow through a red light, an observant left-turner who has a green may yield to the multi-ton truck, because they don't want to die).  Now as the out of control truck with no brakes approaches the sharp curves of a roundabout... well good luck slowing down to 25 mph so you don't tip your load over.  This roundabout just seems tailor made to be a truck rollover hotspot (and I do see a fair share of 18-wheelers in the google earth aerial).   Will I be at all surprised if I hear about a truck rolling over at this roundabout within 6 months?  Not at all.

https://i.imgur.com/GtNpUu5.jpg

EDIT: I may be exaggerating how big a downhill grade it is approaching the roundabout.  It looks like it's going downhill, but i'm sure you have a better idea about that.  Even still, a truck just traveling along what appears to be a limited access highway isn't necessarily going to be expecting to see a roundabout pop up in the middle of the road.  Hopefully there are a LOT of advanced warning signs for this roundabout in particular. 

In my limited experience, a truck without any brakes going down that hill (not a steep hill, but a long hill nevertheless) would have crashed long before reaching the roundabout, either into the accompanying forest or a barrier. There's too many curves for something that heavy to not have already tipped over. The roundabout is basically in the middle of a curve. If it did make it to the roundabout, certainly that would be quite a show, but I doubt it would. A good trucker would try and bail before that, I hope.

Regardless, a truck without any brakes could slam into anything in its path, be it other cars or barriers. Sure, they could make it through that green light, assuming other waiting cars in either lane could somehow react to a truck approaching them at 80+, but that's not like a final checkpoint. There's a constant merge on the right they'd have to deal with, plus a pretty strong curve after the signal. Then another curve not long after that, with a signal, and then another two signals with curves beyond that. If they somehow made it through all that without crashing or stopping, there's a very steep bridge over the Swinomish that would have stopped them.

With that said, might there be a tip over at this new roundabout? It's possible, but only because a driver missed the signs, not because it's out of control.

FWIW, a truck would tip over in a roundabout for the same reason that one might tip over on a freeway clover: inattentive driving. But sometimes, it's just the best option for traffic flow. This area is rural with mostly random arrivals. It's brilliant.

[kalvado]

Truck without brakes is an elephant in the room, and it is pretty much a given that someone will have a bad day over that...

Brilliant.. Would just expanding left turn from east to south to 2 lanes be an adequate solution? Or circular motion (reduced wait time at low traffic, reduced maximum throughput compared to light controlled motion) is a crucial part of improvement?   

[jakeroot]

Brilliant.. Would just expanding left turn from east to south to 2 lanes be an adequate solution? Or circular motion (reduced wait time at low traffic, reduced maximum throughput compared to light controlled motion) is a crucial part of improvement?

The key modification was making SR 20 between Sharpes Corner and Howards Corner right-in right-out only, which required implementing some sort of U-turn situation at both intersections. Roundabouts were the obvious solution.

FWIW, here's an image uploaded to WSDOT Flickr ten years ago (two years after intersection improvements were considered here) that shows the types of collisions that occurred at the signal:



The original design actually included a southern underpass for eastbound traffic, but this was scrapped at some point:

[kalvado]

The key modification was making SR 20 between Sharpes Corner and Howards Corner right-in right-out only, which required implementing some sort of U-turn situation at both intersections. Roundabouts were the obvious solution.

That explains a few things, although dumping turning traffic into a historically problematic intersection is an interesting solution.
I wonder if there was a significant development to the south of that area and corresponding traffic increase?

[jakeroot]

The key modification was making SR 20 between Sharpes Corner and Howards Corner right-in right-out only, which required implementing some sort of U-turn situation at both intersections. Roundabouts were the obvious solution.

That explains a few things, although dumping turning traffic into a historically problematic intersection is an interesting solution.

Well, you could say the same thing about freeways. They localize most of the conflict points at interchanges. If you reduce the number of cars going through a specific intersection, you might reduce the chance of collisions at said intersection. But the benefits of doing so may be outweighed by the negatives of allowing added freedom of movement in other areas.

WSDOT typically studies these things to death, so I'm sure they considered the increased traffic at the roundabouts from all the U-turning traffic. If anything, it just stops traffic from entering from one of the sides that might flood the roundabout, allowing traffic to enter freely on the other side for a few moments. This is not really a major issue, but the U-turning traffic does provide a small benefit in this way.

I wonder if there was a significant development to the south of that area and corresponding traffic increase?

SR-20 to the south is the only road access to Whidbey Island, in addition to a couple ferries. 80k people live on Whidbey, and there's major shopping in Burlington, Arlington, and Mount Vernon to the north and east. It's very likely that congestion has gotten much worse in the last 10 or 20 years.

According to Historic Aerials, the intersection was changed from a Y-intersection to a T-intersection in the 70s when SR-20 and SR-20 Spur were widened to dual carriageway. No idea how long the Seagull intersection had been in place, but since at least the 90s (though probably since it was built).

[kalvado]

That explains a few things, although dumping turning traffic into a historically problematic intersection is an interesting solution.

Well, you could say the same thing about freeways. They localize most of the conflict points at interchanges. If you reduce the number of cars going through a specific intersection, you might reduce the chance of collisions at said intersection. But the benefits of doing so may be outweighed by the negatives of allowing added freedom of movement in other areas.

WSDOT typically studies these things to death, so I'm sure they considered the increased traffic at the roundabouts from all the U-turning traffic. If anything, it just stops traffic from entering from one of the sides that might flood the roundabout, allowing traffic to enter freely on the other side for a few moments. This is not really a major issue, but the U-turning traffic does provide a small benefit in this way.

Sure, there is such a thing as free lunch - but often in a form of cheese in a mousetrap. I wonder if right of way was an issue in setting up 2+1+2 road with a turning lane. Given it was 2 lanes to begin with - I can easily see that being a no-go.. A lot of local variables, as always..

[jakeroot]

That explains a few things, although dumping turning traffic into a historically problematic intersection is an interesting solution.

Well, you could say the same thing about freeways. They localize most of the conflict points at interchanges. If you reduce the number of cars going through a specific intersection, you might reduce the chance of collisions at said intersection. But the benefits of doing so may be outweighed by the negatives of allowing added freedom of movement in other areas.

WSDOT typically studies these things to death, so I'm sure they considered the increased traffic at the roundabouts from all the U-turning traffic. If anything, it just stops traffic from entering from one of the sides that might flood the roundabout, allowing traffic to enter freely on the other side for a few moments. This is not really a major issue, but the U-turning traffic does provide a small benefit in this way.

Sure, there is such a thing as free lunch - but often in a form of cheese in a mousetrap. I wonder if right of way was an issue in setting up 2+1+2 road with a turning lane. Given it was 2 lanes to begin with - I can easily see that being a no-go.. A lot of local variables, as always..

For sure. WSDOT is not quick to acquire ROW unless it's absolutely necessary. Land prices are very high in the Pacific Northwest, even out in the boonies. As the project was already delayed almost ten years, they decided to take the money they had and do what they could. Almost certainly the reason a southern under-crossing wasn't included.

A five lane road with a center turn lane might have worked (would have been tight), but you'd still have traffic turning left into the center turn lane to merge. Which is fine, but there's still that whole "turning across traffic" thing. This is obviously standard stuff on virtually all roads, but WSDOT had the opportunity to eliminate (or minimize) those left turn conflicts without harming the overall flow.

This is how the road looks now. Classic 2+1 road with a C-curb down the middle. There's room to expand on either side (if ROW were purchased), but I don't know if it would have helped the situation here. Based on the design of the southern roundabout, it looks like WSDOT has designed the project to accommodate an additional northbound lane in the future.





No idea what's going on in that second photo, BTW.

[kalvado]

No idea what's going on in that second photo, BTW.

Traffic wise? Truck in a roundabout, nothing to see. Tandem can probably go 3 MPH on such a roundabout - and possibly up to 5 MPH if driver is in a rush and willing to take some risk...