Ft Worth TX: 133-Vehicle Accident on Icy Bridge in I-35W Toll Lanes

[abqtraveler]

So I do not understand how a overpass can cause a pile up.

Because northern drivers are used to snowy and icy conditions.

That and my dad taught me & my brother that if an overpass could have ice on it, when you come up to it, let off your accelerator and coast over the bridge.  Once you're clear, then you can use your accelerator once again.  Of course, this doesn't work for those super long ones, but does for the majority of the super short ones on Interstates.

Yes, and I've experienced it firsthand several times.  The worst part of I-135 going north here in Wichita is actually after it stops being the Canal Route and has become ground-level again.  It's just north of that point, where it goes over the railroad and Hydraulic/29th Street.  The reason it's worse is because, while the Canal Route itself is long and straight and level, that bridge has a slight curve and a slight incline.  Any ice there, and your drive wheels do funny things that they don't do on the Canal Route itself.

I also remember the time I was driving OK at 50-55 mph coming south down I-135, with some snow and ice but nothing out of the ordinary.  Bridges, curves, no problem, just slow down a little bit and all is well.  Then I came around the bend at 17th Street, where it becomes the Canal Route, and there were two wrecks on the viaduct, and my car spun out to 90 degrees before I could regain control.  Sometimes one bridge is fine but another is not.  That's just the way winter weather driving is.  People in this part of the country get a lot more experience learning these things than people in Dallas.

And that's why folks up north have studded snow tires, particularly on rear-wheel drive vehicles. A set of chains also comes in handy, particularly for deep snow.

[bwana39]

This is black ice. Snow melts quickly in Texas then refreezes. It is then black ice. What happens in the overpasses are humps. You come from a longish clear section where you are at clear road speed, you go over one hump (that has no visability over) and there is the traffic that has crashed going up the next one. CRASH. CRASH , CRASH

Even northern transplants struggle here. Driving in icy conditions in Texas is way different than where i am today in Colorado.

[kphoger]

Even northern transplants struggle here. Driving in icy conditions in Texas is way different than where i am today in Colorado.

Yep.  Even driving here in Wichita is more challenging than where I grew up in northwestern Kansas, even though we get less snow here.  Ice is a whole different beast.

[jakeroot]

Yes everyone mentioning a lower speed limit would have help obviously hasn't driven this exact stretch of highway south of Ft. Worth where the speed limit is 60-65 mph and everyone is easily doing 90.  It's no joke at all, they are literally making you feel like you are sitting still when you are also speeding at 70.  I am not talking about 1 or 2 cars, but everyone on I-35W.

I would hope such speeds wouldn't be common sights on mornings like those where the pileup occurred. Below freezing temperatures and the chance for black ice mean that drivers should have been driving perhaps much more cautiously than they were. I can understand northerners being a bit less cautious, but southerners aren't used to this crap, so why was everyone hauling ass like nothing was going on? You'd think they'd be a bit more timid.

The point of lower limits during certain occasions (remember: variable limits, not a permanent lowering of the limit) would be to warn drivers of an upcoming incident. Drivers had very little warning along those express lanes of the upcoming crash. If signs like these were in place, drivers could have been alerted to the crash. The per-lane signs could have shown a very low limit (25 or 35?) or even a red "X", and then the accompanying VMS could have displayed "CRASH AHEAD PREPARE TO STOP". This wouldn't have necessarily prevented any more collisions, but it certainly would have cut down on the number of drivers going full speed into each other.

[ethanhopkin14]

Yes everyone mentioning a lower speed limit would have help obviously hasn't driven this exact stretch of highway south of Ft. Worth where the speed limit is 60-65 mph and everyone is easily doing 90.  It's no joke at all, they are literally making you feel like you are sitting still when you are also speeding at 70.  I am not talking about 1 or 2 cars, but everyone on I-35W.

I would hope such speeds wouldn't be common sights on mornings like those where the pileup occurred. Below freezing temperatures and the chance for black ice mean that drivers should have been driving perhaps much more cautiously than they were. I can understand northerners being a bit less cautious, but southerners aren't used to this crap, so why was everyone hauling ass like nothing was going on? You'd think they'd be a bit more timid.

The point of lower limits during certain occasions (remember: variable limits, not a permanent lowering of the limit) would be to warn drivers of an upcoming incident. Drivers had very little warning along those express lanes of the upcoming crash. If signs like these were in place, drivers could have been alerted to the crash. The per-lane signs could have shown a very low limit (25 or 35?) or even a red "X", and then the accompanying VMS could have displayed "CRASH AHEAD PREPARE TO STOP". This wouldn't have necessarily prevented any more collisions, but it certainly would have cut down on the number of drivers going full speed into each other.

Texas has those systems.  Outside of San Antonio, its a lot of fancy signs to tell you not to drink because it's Super Bowl weekend.  I hope this gets them to use the system the way it's supposed to be used.

[Bobby5280]

Wouldn't the lack of Jersey barriers only have created the conditions for head-on collisions instead?

I don't think anyone would suggest removing the concrete barrier separating the North and South directions of traffic. But there are pro and con reasons for and against removing a concrete barrier separating Southbound toll/express lanes and Southbound general purpose lanes. I've seen some configurations where the only thing "separating" premium express lanes from general purpose lanes is a double set of road stripes.

IMHO, TX DOT and various other toll road agencies should shut down express lanes when icy road conditions are fixing to develop. I agree with others about having dynamic speed limits, at least in major cities. I also think most of the current traffic advisory signs installed overhead absolutely suck. Most of them look like they date back to the early 1990's. They could install full matrix LED boards capable of showing critical information in a far more effective manner.

[bwana39]

Yes everyone mentioning a lower speed limit would have help obviously hasn't driven this exact stretch of highway south of Ft. Worth where the speed limit is 60-65 mph and everyone is easily doing 90.  It's no joke at all, they are literally making you feel like you are sitting still when you are also speeding at 70.  I am not talking about 1 or 2 cars, but everyone on I-35W.

I would hope such speeds wouldn't be common sights on mornings like those where the pileup occurred. Below freezing temperatures and the chance for black ice mean that drivers should have been driving perhaps much more cautiously than they were. I can understand northerners being a bit less cautious, but southerners aren't used to this crap, so why was everyone hauling ass like nothing was going on? You'd think they'd be a bit more timid.

The point of lower limits during certain occasions (remember: variable limits, not a permanent lowering of the limit) would be to warn drivers of an upcoming incident. Drivers had very little warning along those express lanes of the upcoming crash. If signs like these were in place, drivers could have been alerted to the crash. The per-lane signs could have shown a very low limit (25 or 35?) or even a red "X", and then the accompanying VMS could have displayed "CRASH AHEAD PREPARE TO STOP". This wouldn't have necessarily prevented any more collisions, but it certainly would have cut down on the number of drivers going full speed into each other.

Texas has LOTS of the VMS. They tend to have generic messages. There are location specific data on them at times, but outside of metro Houston, I don't recall ever seeing real-time data. The wreck ahead signs are in use around Houston.

The dynamic  speed limits might  be helpful, but probably not within tyhe purview of TXDOT and perhaps would complicate the rules of the (emissions) attainment zone agreements. 

The Jersey barriers may have complicated this wreck, but any type of barrier would have created a similar outcome. The Jersey barriers save more lives in a given year than they create,

[kphoger]

The Jersey barriers save more lives in a given year than they create

[jakeroot]

Texas has LOTS of the VMS. They tend to have generic messages. There are location specific data on them at times, but outside of metro Houston, I don't recall ever seeing real-time data. The wreck ahead signs are in use around Houston.

Texas has those systems.  Outside of San Antonio, its a lot of fancy signs to tell you not to drink because it's Super Bowl weekend.  I hope this gets them to use the system the way it's supposed to be used.

The VMS signs are just one part of how variable limits work. The variable per-lane limits would show which lanes are closed, open, suggested speeds; the VMS sign would tell drivers why, even if something more or less generic like "reduced speed zone" -- it's better than lowering the limits but without any signage to explain why.

Example from Seattle showing how they work in tandem. There's also these mini displays that display shorter helpful messages or speed limit signs. They can also displays arrows and X's to close lanes or warn of caution.

Here's a page where they show all the symbols and messages that you can see in use around the Seattle region.

[edwaleni]

Easy response. Pre-emptive salting and sand.

If the temp is going to get within a few degrees of freezing in the forecast, your pour salt.

If it is going to get below 15 degrees, you use calcium chloride with sand, not sodium chloride.

Do it proactively in the event of, not wait until. If you wait, then its too late.

[kphoger]

Brine is utilized to assist at the beginning of snow and ice. Once it gets wet it quickly loses its effectiveness. If it rained before the freezing drizzle occurred, it does no good as its washed away. If a freezing drizzle lasted for a period of time, the brine helped at first, then its effectiveness disminished quickly.

Salting roads before a storm generally doesn't help. Salt is like trash on a roadway. As traffic goes by it it gets swept off a dry Road. Salting a sriveway or parking lot may be okay, because traffic is moving at very slow speeds. Traffic on highways, at highway speeds, just blows it off.

Easy response. Pre-emptive salting and sand.

If the temp is going to get within a few degrees of freezing in the forecast, your pour salt.

If it is going to get below 15 degrees, you use calcium chloride with sand, not sodium chloride.

Do it proactively in the event of, not wait until. If you wait, then its too late.

jeffandnicole, who actually works in snow plowing, seems to disagree with you.

[ethanhopkin14]

Texas has LOTS of the VMS. They tend to have generic messages. There are location specific data on them at times, but outside of metro Houston, I don't recall ever seeing real-time data. The wreck ahead signs are in use around Houston.

Texas has those systems.  Outside of San Antonio, its a lot of fancy signs to tell you not to drink because it's Super Bowl weekend.  I hope this gets them to use the system the way it's supposed to be used.

The VMS signs are just one part of how variable limits work. The variable per-lane limits would show which lanes are closed, open, suggested speeds; the VMS sign would tell drivers why, even if something more or less generic like "reduced speed zone" -- it's better than lowering the limits but without any signage to explain why.

Example from Seattle showing how they work in tandem. There's also these mini displays that display shorter helpful messages or speed limit signs. They can also displays arrows and X's to close lanes or warn of caution.

Here's a page where they show all the symbols and messages that you can see in use around the Seattle region.

Yes, but the key element here would be for them to actually work instead of informing us how many people have died that year on the roads in Texas.  The system is in place, but it is never put into motion the way it was designed.  It actually annoys me there are so many VMSes in the state with either nothing on them or some stupid generic jingle about slowing down on them, not because of traffic, but how to avoid a ticket. Also, those digital arrows above lanes are all over the state and are just off.  I don't understand why you install a system and never use it.

A side note, this all happened within minutes.  How would information get relayed to the system?  Does it have to be called in?  Is it somehow automated by scanning roadways for slowdowns?

[ethanhopkin14]

Easy response. Pre-emptive salting and sand.

If the temp is going to get within a few degrees of freezing in the forecast, your pour salt.

If it is going to get below 15 degrees, you use calcium chloride with sand, not sodium chloride.

Do it proactively in the event of, not wait until. If you wait, then its too late.

We don't have these resources down here.  It is seen as a waste of money for the event it happens maybe once every five years.  Same reason our houses down here don't have insulated or heated piping.  There is no point putting that expensive system in the houses here just so we can use them for a day and a half every five years. 

[jakeroot]

Yes, but the key element here would be for them to actually work instead of informing us how many people have died that year on the roads in Texas.  The system is in place, but it is never put into motion the way it was designed.  It actually annoys me there are so many VMSes in the state with either nothing on them or some stupid generic jingle about slowing down on them, not because of traffic, but how to avoid a ticket. Also, those digital arrows above lanes are all over the state and are just off.  I don't understand why you install a system and never use it.

A side note, this all happened within minutes.  How would information get relayed to the system?  Does it have to be called in?  Is it somehow automated by scanning roadways for slowdowns?

Well yes, they need to be active and working. There's really no point to them if they just sit off half the time. The regular VMS displays should be off when not in use (those generic jingles are a waste of electricity, and drivers ultimately just ignore them), but the speed limit signs really should always be on, unless the traffic is really light.

The system in WA is monitored 24/7 by a traffic management center, but it's largely automated using loop detectors. It's true that this did happen quite fast, but I think it would have been caught in some way. Or at the very least, a lowered limit could have been put in place with "ice" warnings on the VMS displays.

[rte66man]

Brine is utilized to assist at the beginning of snow and ice. Once it gets wet it quickly loses its effectiveness. If it rained before the freezing drizzle occurred, it does no good as its washed away. If a freezing drizzle lasted for a period of time, the brine helped at first, then its effectiveness disminished quickly.

Salting roads before a storm generally doesn't help. Salt is like trash on a roadway. As traffic goes by it it gets swept off a dry Road. Salting a sriveway or parking lot may be okay, because traffic is moving at very slow speeds. Traffic on highways, at highway speeds, just blows it off.

Easy response. Pre-emptive salting and sand.

If the temp is going to get within a few degrees of freezing in the forecast, your pour salt.

If it is going to get below 15 degrees, you use calcium chloride with sand, not sodium chloride.

Do it proactively in the event of, not wait until. If you wait, then its too late.

jeffandnicole, who actually works in snow plowing, seems to disagree with you.

Oklahoma has been experimenting with spraying on a brine mixture before bad weather. The theory is it should stay in place and keep the roads clearer longer. I'll have to look and see if they have published results.  Oklahoma State University has also experimented with a self deicing bridge deck. I believe it has heating elements in the concrete.
https://hvac.okstate.edu/sites/default/files/HPCC_Bridge_Deck_Deicing.pdf

[kphoger]

They spray with brine before bad weather in Wichita too.  But not if it's going to be raining first.

[stevashe]

*  Because we in the US are a weird, exceptionalist country, we use the same set of values for horizontal curve radii at a given design speed regardless of the maximum superelevation (e_max) used in a particular area.  8% is a pretty common value for e_max, but it can go as low as 4% in areas susceptible to frequent icing.  It helps to have at least an intuitive understanding of the superelevation provided in your area so you can choose your speed accordingly--lower maximum superelevation values translate to reduced tolerance for speeds above the hands-off speed.

I have no idea where you got this idea from, but it's absolutely not true. The AASHTO design guidelines definitely have separate tables of minimum radii for different values of max. superelevation, see below screenshot from the table of contents of "A Policy on Geometric Design of Highways and Streets, 7th Edition" (not going to post the actual tables since AASHTO seems to be fairly protective of their copyright).



If there is some state, county, city, or other agency not using these standards, that's on them. And I'd be quite surprised given the potential liability.

[hotdogPi]

I think what he's saying that if you're using the maximum, which you usually are, it doesn't matter what the maximum is. If you're using a constant value, the maximum does matter, but it shouldn't (6% superelevation has the same physical properties whether the maximum you're allowed to use is 6% or 8%, but the two tables give different values, while 6% with 6% maximum corresponds to 8% with 8% maximum).

[J N Winkler]

*  Because we in the US are a weird, exceptionalist country, we use the same set of values for horizontal curve radii at a given design speed regardless of the maximum superelevation (e_max) used in a particular area.  8% is a pretty common value for e_max, but it can go as low as 4% in areas susceptible to frequent icing.  It helps to have at least an intuitive understanding of the superelevation provided in your area so you can choose your speed accordingly--lower maximum superelevation values translate to reduced tolerance for speeds above the hands-off speed.

I have no idea where you got this idea from, but it's absolutely not true. The AASHTO design guidelines definitely have separate tables of minimum radii for different values of max. superelevation, see below screenshot from the table of contents of "A Policy on Geometric Design of Highways and Streets, 7th Edition" (not going to post the actual tables since AASHTO seems to be fairly protective of their copyright).

It is true, as I think you can verify for yourself by looking at the nomographs for different values of e_max.  For an 80 MPH design speed, for example, a 6000 ft curve would receive superelevation of 4.3% for e_max = 8% and 4% for e_max = 6%.  (I'm working with the sixth edition of the Green Book, but I assume yours is similar.)

I wasn't actually talking about minimum radii, which--as you say--are different for different values of e_max at the same design speed.

Another way of expressing this point is that the same combination of curve radius and superelevation can correspond to different design speeds depending on the e_max chosen for design.  This creates issues with design consistency, as discussed in a 1999 FHWA report prepared as part of the research program that led to the Interactive Highway Safety Design Model (which has been used by a number of agencies to assess proposed alignments for new highways).

If there is some state, county, city, or other agency not using these standards, that's on them. And I'd be quite surprised given the potential liability.

I am personally not aware of any cases (or at least any that have been litigated in any recent decade) that hinged entirely on design criteria for horizontal alignment.

There are some states, such as California, that have their own design standards for horizontal and vertical alignment.  From past conversations with practitioners, I think the concern that comes into play is ensuring the state-specific standards stay above a floor defined by whatever edition of the Green Book FHWA has adopted for federal-aid highways.  (There was a mini-controversy in 2004 when AASHTO dropped the "dead dog" criterion for forward sight distance, resulting in lower k-values being allowed for summit curves.  There was talk of amending state DOT highway design manuals to mandate continued use of the older, more conservative values.)



As regards the Fort Worth pileup, the Green Book has interesting things to say about friction coefficients on ice.  They can range from 20% down to 5%, the lower limit corresponding to black ice forming on wet roadways as they drop just under the freezing point--exactly the conditions that came into play in this incident.