Dozens of cyclists have crashed on a slice of railroad tracks in Knoxville. What lessons can be gleaned from this infrastructural pitfall?
“She had to wear this second-skin bandage for months afterward, just because it was kind of a deep gouge,” says Cherry, an associate professor of civil engineering at the University of Tennessee.
Shortly after, Cherry was contacted by his department head, who happened to see a bad accident from his office window at the same crossing. In this case, the rider was carted away by EMTs. The department head asked him why he wasn’t doing anything about it. “I said, ‘Well, nobody’s done any research on it, and I don’t have funding at the moment for that,’” Cherry says. “But he shamed me into it. He said, ‘So somebody’s injured now because you didn’t have the time?’ I said, ‘OK, if you put it that way, I’ll do something about it—if you give me your window with a view.”
The result of that exchange is a recent study in the Journal of Transport & Health, coauthored by the university’s Ziwen Ling and Nirbesh Dhakal, and perhaps the most brutal compilation of bike crashes you’ll ever see outside of Tour de France wrecks. From a window-mounted camera, Cherry and his colleagues logged more than 50* accidents at the railroad crossing, involving paths on both sides of the street, in just two months in 2014.
In many cases, riders fly off cycles and land on their arms or heads, winding up clutching their bodies in pain. Some are pitched into the roadway and narrowly avoid being run over by cars. If you’ve ever had a nasty spill, you might not want to watch this footage—for me, it revived painful memories of when I split my chin open like a ripe plum on a deep Seattle pothole.
When Cherry uploaded the video last month on YouTube, commenters shared their own stories about this tricky crossing. “I broke my pelvis and was in a wheelchair for three months,” said Annabel H. “I had a black eye and a whole lot of bruises. I had no idea it was such a treacherous crossing,” reported Gina O.
As many riders know from painful experience, crossing rails embedded in the street is a treacherous undertaking on a bike. There are at least 100,000 at-grade rail crossings in the U.S., not counting city trams and streetcars (which are also notorious for taking down cyclists). But it’s tough to gather data on how many crashes they cause because so few are communicated to the authorities. “The work I looked at, we saw people getting hauled off on ambulances and other things, but very, very few police crash reports,” says Cherry. “There’s a lot of rail infrastructure throughout Tennessee, and I can only imagine how many unreported crashes are occurring statewide or even nationwide.”
That’s part of what motivated Cherry and company to conduct what they call the nation’s first “empirical analysis of rail-grade crossings and single-bicycle crashes.” To them, the problem wasn’t with the cyclists. It was with the roadway design and the fact nobody knows, scientifically speaking, the best way to bike over railroad tracks.
“There’s no guidance on really what is the minimum angle—that’s what I found when I started looking into the literature,” says Cherry.
Most experienced riders know the ideal way to do it: As the folks at Bicycling say, cross at a 90-degree angle. That’s the “gold standard” many infrastructure designers strive for. But in cases when the crossing has gaps running in different directions, it might be best to pedal through at 45 degrees. Of course, all this is more complicated when metal tracks are wet, a situation that can turn even a savvy cyclist into a hollering missile directed fast into the pavement.
The Knoxville crossing is particularly tricky, due to the angle that riders must approach the tracks. The researchers often filmed riders trying to cross at a perilous 10 degrees. If cyclists swooped and sharply turned, they could cross at a safer 45 degrees. If they rode straight through they had a bigger risk of hitting the dust, especially if they were female or riding in a group. Cherry suspects the latter cyclists might be pedaling side-by-side, thus limiting their ability to maneuver across the track, or perhaps chatting among themselves to distraction. As for the gender factor, it could be related to experience, he guesses, or a “torque that occurs in the handlebars that maybe a guy could overcome with more upper-body strength.”
So what’s the magic angle to bike across railroad tracks? After reviewing both successful and unsuccessful crossings in their bone-bruising footage, the researchers concluded the famed 90 degrees isn’t necessary—a lesser “bronze standard” of 60 degrees is still extremely effective.
“The main thing we learned is that after a 60-degree angle crossing—that is, the bike crosses the tracks at 60 degrees—we didn’t see any crashes,” says Cherry. “And really, we only saw a couple of minor crashes between the 30-degree and 60-degree mark. So if any agency can squeeze in a 30-degree crossing, they can probably do pretty well to solve a lot of the problems.”
That’s what Knoxville eventually did. After pondering a 90-degree crossing that would cost $200,000, partly due to the route being near a river and needing retaining walls, the city and the railroad company settled on a cheaper, roughly 60-degree “jughandle” detour on the side of the street where people were tumbling into traffic. “The total cost was $5,000 for all of that, which is unbelievable, really,” Cherry says. “This has been years in the making, with probably hundreds of crashes there, and it took $5,000 worth of in-house crew time and materials.” (The city later made the path on the other side, located on a greenway, angled to about 60 degrees.)
Of course, some over-confident cyclists ignore the new paths and barrel right over the tracks, often to their bloody detriment. That makes Cherry think other solutions should be tested in the future, like bollards to corral riders the correct way or fillers to stick into railroad gaps so bike wheels don’t get stuck.
“I frankly think we need to make infrastructure that is, as the snobby cyclist says, idiot-proof,” he asserts. “Most people say, ‘Oh, it’s common sense.’ But you look at our video and it seems at least a third of the people who ride are experienced cyclists. Even the most experienced cyclists can get blindsided by one of these.”
* Clarification: The study focused on 32 crashes on the side of the street closest to traffic because the other side had issues with its “restricted view and geometric characteristics”; however, if you include the other side, there were at least 53 crashes in the two-month study period.