Dr. Gridlock

Jan 13, 2015

The Washington Post's Dr. Gridlock, Robert Thomson, will be online to take all your questions about Metro, traffic throughout the region and other transportation issues.

Welcome, travelers, to a special edition of the chat to discuss Monday's tragedy on the Yellow Line outside L'Enfant Plaza. Please submit your questions and comments now.

Has Metro explained why they couldn't back the train up to L'Enfant?

No, and Metro officials say they won't be discussing any aspect of the incident publicly. They say the investigation is now in the hands of the National Transportation Safety Board.

So we don't yet know the answer to that very important question: Did the power fail on the train, or was there another issue that prevented the operator from backing the train into L'Enfant Plaza to get the passengers out of the smoke in the tunnel.

I saw the article on safe evacuation. In your opinion, is there a time when riders should go against the instructions of the operator and leave the train?

I rode downtown on the Red Line this morning and was looking around the crowded car imagining what we'd all do if it filled with smoke.

Therefore, I don't blame anyone for any action they took on that Yellow Line train. There, but for the grace of god.

I know there are problems with what people call "self-evacuation." You put yourself in danger, because you are attempting to walk along a darkened track bed without knowing if the third rail power is on and withing knowing if another train might be approaching.

Also, it would be impossible for the train operator to back the train into the station once the operator knows that there are people out there on the track bed.

That's why so many people in this case and in previous cases complain about what they perceive as the lack of information coming from the operator or rescue personnel.

You've got to be talking to the riders constantly, or they will feel abandoned -- and legitimately so.

Is there any word on how long it actually took firefighters to get to the train?

In this and any other similar incident, expect a lot of the initial information to be modified. I remember a very early moment after the 2009 Red Line crash when officials were puzzling over how two trains going in opposite directions could have wound up on the same track. Of course, that wasn't the case, as we learned shortly afterward.

With the Yellow Line incident, we don't have riders saying, "Oh, there's smoke coming into the car. I wonder what time it is?" So from the riders' side, we just have a few best guesses on how long they were stuck.

We're working on refining that, as well as the official estimates from the rescue personnel.

NPR reported that self-evacuations caused power to be shut off to the train and prevented it from returning to the station. Is that true?

That's possible, but we don't have confirmation of that. It's early days. Their are conflicting, or at least overlapping, explanations of what happened and in what order it happened.

We don't know if the power went out because of the electrical problem, or if the power was deliberately turned off -- and if so, who turned it off. It could be turned off from the Metro operations center or on the scene. It could have been turned off when responders realized there were people on the tracks, or it could have been turned off when the rescuers arrived to walk down the tracks.

How can it be that each time there is a major Metro disaster it is as though it was unforeseen? Is there no playbook for includes how Metro will communicate with emergency responders?

Yes, there's absolutely a playbook on this. The most recent time I'm aware of that the playbook was updated was after the Green Line train got stalled between Navy Yard and Anacostia in January 2013.

Don't be thinking this kind of problem is unforeseen, or really weird. It's something that transit agencies and emergency responders plan for. Focus on that.

While I understand firefighters waiting until they were sure that the third rail was off before evacuating people, it seems that they were concerned because trains were still running on other tracks in L'Enfant so they weren't sure it was off. If that was the case, why not just shut down the station completely?

This is what we're talking about, from The Post story:

"Eugene A. Jones, the interim chief of D.C. Fire and Emergency Medical Services, said firefighters did not immediately enter the tunnel to help the riders because they were not sure whether the subway’s electrified third rail had been deactivated.

"But Jones said the delay was “nothing like” the length of time described by passengers.

“While the power may have been turned off to the track bed where the firefighters would go in to make rescues, they heard trains running on the lower level” of the station, Jones said. “So in their mind, they wanted to make sure that before they put people on the track bed that the power was actually off."

To rescue the train, it would not have been necessary to turn off the third rail power on both levels at L'Enfant Plaza. Only the Green/Yellow tunnel south of L'Enfant Plaza was affected.

With rush hour building, train operators would rightly have wanted to preserve as much of the system as possible. They would not have wanted to shut a critical segment of the Orange/Silver/Blue Line if they could avoid it.

Was it the entire train that filled with smoke or just one car / other cars? Also, how often does Metro perform mock fire drills? This needs to happen at least quarterly IMHO..

I'm not sure how often fire drills are performed. I know they are performed. That's one thing I wondered about when I saw that quote about the reluctance to enter the tunnel because of the other train noises. Is that not something emergency responders are familiar with at the very busy L'Enfant station?

On the smoke issue: We think the smoke spread throughout the train. In fact, it got as far back as the L'Enfant Plaza platforms, which I think were about 800 feet away.

Was Metro following a standard protocol in having left passengers inside a smoke-filled train car for 40 minutes?

I'm sure that was not Metro's goal. It's more a choice of lesser evils. Would it be less worse for the riders to remain on the train, or to get out on the tracks?

The operator apparently perceived that the problem was outside the train, and told people to stay put. That seems like good advice, as long as the rescue is conducted promptly, and the passengers are constantly informed.

We have talked about the need to inform passengers every time a train stops, and to repeat that information periodically, even if there's nothing new. That's under normal circumstances. Imagine if you're in a rail car filling with smoke.

Talk to me on a speaker? I'd want someone holding my hand.

Is Baltimore Metro safer than DC Metro?

I have no reason to doubt the safety of the Baltimore system. It's much smaller than Metrorail, which is one of the very biggest in the nation. (Biggest is NYC.)

Can you clarify whether the train operator ever expressly told the passengers to stay on the train or to evacuate?

Couple of snippets from The Post story:

A rider reported that the train operator said, "there’s a problem, nobody move," then the car quickly began to fill with smoke.

And several others reported that the train operator announced to passengers that they should stay low to escape the smoke while they waited to be evacuated.

That conforms with our understanding of Metro policy in such situations. We have no report that the train operator advised anyone to evacuate before rescuers arrived to help them.

Brilliant PR move. It will take NTSB months if not a year to answer the question, keeping metro off the hook. Please continue pushing metro on this point. Riders deserve answers, and frankly metro employees who let the train sit there and resulted in the death of someone need to be fired.

Yes, on getting answers. We don't know that "metro employees let the train sit there." Reporters are working on understanding what happened in what sequence.

How was the Fire Department not sure that the third rail was turned off? Was it a miscommunication by WMATA? A fundamental misunderstanding of how the metro system works? The article this mornings mentions that since they heard trains running on the lower level, they weren't confident that the 3rd rail had been turned off. This is the biggest failure in my view.

There's only one way to know for sure that the third rail is turned off -- and you don't want to perform that experiment.

I certainly wouldn't blame firefighters for wanting confirmation from Metro officials that the power was off. But my understanding is that there are relatively straightforward ways of checking with Metro officials on that. (There's a phone by the tracks.)

Was is just the first rail car ?

It was a six-car train, and we have no reason to believe the smoke was confined to just one car. In fact, it looks much more like the smoke was entering all the cars.

Also, the problem, as we understand it now, was not with the cars themselves. The smoke was coming from outside the cars.

Why does Metro tend towards shifting blame ("those passengers never waited as long as they said") instead of taking responsibility and promising to make sure this never happens again?

It's a pretty common thing in reporting on emergencies to have responders note that the people to be rescued overestimate the time it took. In fact, that would be perfectly understandable in this situation. One look at those pictures riders took inside the train tells you that every moment must have seemed like an eternity.

But that doesn't mean their estimates were wildly off.

By the way, I haven't seen any Metro official quoted as saying that the passengers were off on their time estimate. I do see interim D.C. fire chief Jones saying that the delay was “nothing like” the length of time described by passengers.

Why wasn't there clearer communication between the fire department and WMATA regarding whether the third rail was turned off? Is this not practiced regularly in drills?

For reasons you can imagine, this is a key part emergency response planning. There are well-established protocols for getting the third rail turned off and communicating that information.

One issue that came up in at least one previous incident is that once you turn off the power, you can't move the train out of the tunnel, or send in another train to fetch it. But that's different from the communication issue, which should be much more clear-cut.

How long, precisely, did it take for the various emergency responders to a) be notified; b) reach the train; c) begin evacuations and; d) complete the evacuations? A timeline would help put things in perspective.

You're asking the right questions and reporters are working on a timeline. And trying to resolve conflicts that always come up when different people relate their versions of the timeline.

Since the District fire-EMS has a $700,000 mock-up Metrorail tunnel at their training facility and presumably has trained for this situation, how do they explain the long delay in getting the passengers out of the train yesterday?

In the early going, officials have not characterized this as a major delay. We also don't have a credible timeline yet.

(I'm pretty sure every single passenger in that smoke would characterize the response as delayed.)

Having lived in the this area since 1985, I am completely baffled as to why Metro still has such problems communicating among its employees and with passengers. That would seem to be the easiest thing for Metro to address. Yet, time and again, we hear that Metro cannot clearly communicate, either during routine service disruptions and also during emergencies, such as yesterday. I would have thought that the NTSB's report post-June 2009 (the derailment) would have address this issue. But clearly that's not the case. Does the NTSB have any enforcement power if it determines that Metro did not follow its recommendations with respect to communications?

The NTSB makes "recommendations," which transit agencies take very seriously but don't always follow. For example, the NTSB recommended many years ago that Metro replace the oldest cars in the rail fleet, the 1000 series cars. But you're still riding on them.

On communications in emergencies: This is indeed a chronic problem. Metro officials have promised for years that this would be fixed.

And as you say, this has not been fixed.

During the administration of General Manager Richard Sarles, we've had a few incidents in which trains were stalled for a long time awaiting rescue. In those incidents, passengers said they had little information to go on. That's how we got the term "self-evacuation."

I was on what may have been the last train to stop at L'Enfant yesterday on the Orange/Blue/Silver line after the accident. The lower level had some smoke. What pulled my nose out of my book was the Metro staff member outside informing people to either get on the train or evacuate the station immediately. He then pushed a (possibly homeless) man into our car, and advised riders to stay on the train. He was brisk and professional, and helped maintain some calm (the shouting coming from the upper level and the smoke were making people very nervous, of course). So good job, at least by the staff within the station at maintaining order and calm.

If anyone has eyewitness information about what happened at L'Enfant Plaza -- on the train or in the station -- please contact me via e-mail at Robert.Thomson@washpost.com.


Could you please clarify who and what has responsibility for Metro oversight? What role is played by new Mayor Bowser? The City Council and its members? Congress? Federal agencies? Thanks.

All kinds of agencies have an oversight piece of Metro. For example, there's the Tri-State Oversight Committee, which reviews Metro safety plans. The D.C. Council periodically holds oversight hearings with Metro officials. Other governments that contribute to Metro funding do the same. Congress has held hearings on Metro safety .

You know the NTSB is involved.

Dr. G, Two comments: There were no useful announcements from Metro about what what happening at L'Enfant Plaza when i entered Foggy Bottom Metro about 4:30. It was only random choice that i chose not to try for a Silver/Blue/Orange train to L'Enfant to switch to the Silver line. Second, there were no intelligible announcements from any of the train operators on the Tokyo-subway packed Blue Line trains to Franconia about the problem. Second comment: really stupid, stupid, selfish people packing and blocking the doors of the trains rather than moving to the center (where there was space) just so they wouldn't miss their exit -- six stops down the line. A question: how long will the Yellow line be shut down?

We don't have a time estimate on when the Yellow Line will be back in business. The problem is the tunnel segment south of L'Enfant Plaza where the smoke incident occurred. It that's out, then the Yellow Line Bridge over the Potomac is out.

Talking to riders and offering reassurance and direction is great. Of course this assumes that the speakers in the train cars are functional. What role, if any, did the sound quality and clarity of the speakers play in this? I cannot imagine that WMATA's "wait until the 7000 series" dodge can be taken seriously at this point.

I haven't seen a specific complaint about the quality of the speakers aboard the train. It's been more like, The speakers weren't used enough.

You're certainly right that there are often problems with the speakers. On the Red Line this morning, I know the operator was telling us about the disruption in Yellow Line service, but the only words I could make out were "Yellow Line."

Does metro have diesel rescue engines in the system that are able to pull a train into a station in the event of electricity loss? I imagine there were quite a few trains stuck behind this disabled one, but really how hard is it to tell all of the operators to put it in reverse and head back (surely the trains can operate in reverse?). If the station area could be cleared, sending a resuce engine to pull the train into the station shouldn't have taken more than a few minutes (unless metro doesn't have a plan in place for such a thing, and instead is just winging it). What would they have done if the train was actually on fire? Tell people to stay in place and wait?

An evacuation is not something that riders train for. How are they supposed to know what they're supposed to do and when they're supposed to do it without instruction from the train operator or emergency personnel.

Has anyone memorized those evacuation placards in the trains? People do what they think is best to save their lives. And they might not be right, which leads to tragedy.

No, Metro doesn't have diesel rescue engines. The standard procedure is to off-load the nearest train and have that train move the disabled train, or at least recover the passengers.

That's problematic if the third rail has been turned off or somehow disabled.

Metro can't seem to do anything right. What do you see as the problem? Why can't the system run as efficiently as say the NYC subway? Or is than an unfair comparison?

I spent a few decades riding the New York City subways. If they're more efficient than Metro, then a lot has changed since I lived there.

But here's what I think is the key question riders should be asking: Are we better off now than we were in 2009? We're more than halfway through the massive and disruptive rebuilding program launched while Sarles was the general manager.

If they've fixed so much, how come things don't seem better. How can a tunnel get filled with smoke, and a train get stalled long enough that scores of people need to go to the hospital and one rider dies?

Last week, officials and civic leaders were praising Sarles for restoring Metro's "safety culture." Are we really there yet?

"Metro officials say they won't be discussing any aspect of the incident publicly." Why? Many concerned riders would like to know.

When the NTSB takes over an investigation, the NTSB expects to be in charge of releasing information about the investigation. That seems reasonable.

And if you're a Metro official, very convenient.

Does Metro have a policy and system to notify customers system-wide should there be a major incident such as this one?

Sure. It has the e-mail alerts and Twitter alerts. There was plenty of information going out Monday afternoon.

But in this case, as in many other less serious incidents, the magnitude of the problem doesn't become clear in those messages. We say that just last week with the cold weather delays, equipment failures and cracked rail.

Riders want to know, OK, you tell me trains are single-tracking, but I'm six-miles away. What does that mean to my trip?

Metro has made improvements, but still has difficulty communicating useful information to riders immediately after a disruption.

What was the direct cause of the fatality?

We don't know the cause of death yet.

The head of Metro's board has talked in the past about trying to arrest people who self-evacuate. Is it now time to eliminate that discussion so that people who worry about their own personal safety in an emergency don't have the specter of arrest hanging over them?

I can't imagine transit police arresting people as they stagger out of a tunnel from a disabled train.

Is there any evidence people were avoiding Metro today? I personally drove, to avoid the delays, and also avoided 395, expecting it would be especially crowded.

No. The Post had reporters out watching the trains this morning, but there's no evidence that ridership was way down. I'm sure some people avoided the trains, just as they did after the 2009 Red Line crash.

I can tell you that my Red Line train was packed, late in the morning rush.

Was there a reason the train couldn't go forward? Would breaking a window open decreased the smoke in the train car?

We're not sure why the train stopped. The operator may have seen trouble ahead and stopped the train. Or the train may have lost power. We don't know.

Aside from the difficulty of smashing a train window (it's not like they stash bricks under the seats -- it sounds to me like the equivalent of drilling a hole in the bottom of your boat to let the water out. The smoke problem was outside the train.

"Don't be thinking this kind of problem is unforeseen, or really weird. It's something that transit agencies and emergency responders plan for. Focus on that." Okay, I'll focus -- if this wasn't unforeseen why did they do such a poor job responding?

We don't know that the rescue squad did a "poor job" responding. We do know that the response took too long for the passengers aboard the train and that a passenger died while scores of others had to be hospitalized.

Without question, there needs to be improvement in the emergency response system. (I feel like I've probably written that sentence several times in the past few years.)

I've read that metro has been trying to replace most of it's "arc insulators" during the major trackwork they've been conducting. Do we know if this section of track has had such major trackwork as of late? If so do we think it would have affected the chances of such electrical arcing occuring?

We're not even sure it was an insulator, or some other piece of electrical equipment that generated the electrical arc.

Am I hearing that first responders who are trained to go into a burning building cannot safely walk along a track bed with a hot third rail? That's ludicrous - just don't get near the third rail - it's a wide track! The train was only 150 feet away from the station - it could have been evacuated quickly and the passengers could have walked back safely even with a hot third rail..

I've walked down a darkened Metro tunnel, grateful to have one track walker ahead and another behind. I don't envy anyone who does that for a living.

Should something similar happen again, what should riders do? Obviously not evacuate. But cover our mouths, lay on the floor? How can I prepare?

I would read those safety instructions on the placard by the middle doors. But I think the main responsibility for the future is on the shoulders of Metro officials. They're the ones who need to prepare.

Thanks for joining me today. We've had a very long chat on a very important topic, but I still couldn't get to all of your questions and comments. If you'd like to follow up, please send an e-mail to me. I have two addresses: Robert.Thomson@washpost.com and dgridlock@washpost.com.

While we've been chatting, Post reporters have been working on answering your questions, so watch for updates on washingtonpost.com this afternoon and in tomorrow's newspaper.

In This Chat
Robert Thomson
Robert Thomson is The Washington Post's Dr. Gridlock. He offers therapy for that most intimate relationship: the one between you and your commute. You can read his work on the Get There blog, as well as in the Metro section of The Washington Post.
Recent Chats
  • Next: