Metro-North Railroad crews at work repairing a damaged section of track near the Spuyten-Duyvil station on Tuesday. Photo: Marc A. Hermann / MTA New York City Transit
There’s something very dramatic and unsettling about seeing a passenger rail train, once filled with people, lying on its side scattered about its tracks and the woods nearby. It’s wrong for a train to be off its track, and it’s newsworthy when one jumps the rails. This past Sunday’s Metro-North derailment provided us with a tragic reminder of the worst that can happen when a train derails, particularly one traveling at excessive speeds.
In the aftermath of the incident, safety takes second stage. Politicians throughout the region issued calls for comprehensive studies and sounded alarm bells. Connecticut Gov. Dan Malloy penned a letter to the MTA. “I am asking MTA/Metro-North to develop an action plan that addresses communication, safety reporting, inspection and maintenance programs, remedial short term action plans, and longer term capital investment programs to upgrade the infrastructure,” he wrote. Change needs to happen now.
New York’s junior senator Kirsten Gillibrand issued a similar call with particularly strident language. She wrote a letter to Federal Railroad Administrator Joseph Szabo expressing “deep concern over the recent derailment of a Metro-North commuter train in the Bronx, and for the safety of New Yorkers and others who use the Metro-North railroad every day.”
“Yesterday’s accident is the latest in a long list of accidents on MTA’s system, and comes on the heels of a freight train derailment near the same turn in July… This is simply unacceptable,” she wrote. “I renew my call for an immediate comprehensive safety evaluation of the MTA system and procedures to ensure that we do not experience a similar tragedy in the future. Additionally, I request that you provide my office with an overview of any steps that have been taken by the FRA to address MTA commuter rail safety.”
What Gillibrand and Malloy are saying has some truth to it, but there’s also some kneejerk fearmongering. Meanwhile, the incident has created the perception of safety problems. One rider said to The Times on Sunday, “You think you’re safe on the train. I know I’m going to be taking a car for a while.”
It’s that reaction that the coverage over the last few days and the statements made by politicians has fed. Sunday’s derailment is a terrible story with a tragic ending for four riders and horrific injuries to many others. These four fatalities though were the first passenger deaths in Metro-North’s 31-year history. WNYC crunched the numbers and found that, since 1993, for every 1 billion train passengers, seven have died. In 2012 alone, 33,561 Americans died in traffic incidents. The comparable motor vehicle death rate is 108,000 for every 1 billion drivers.
Now, I’m not going to further minimize what happened Sunday. Fatalities or not, Metro-North’s safety record, as the FRA noted on Tuesday, has been abysmal of late, and the technology exists to ensure that no one — zero people — dies on in a crash derailment due to excessive speeds. The MTA though hasn’t fully funded the positive train control program and may not have it ready until 2019. That we can build a $4.5 billion subway stop but can’t scrounge up a quarter of that to save lives speaks volumes about our priorities in non-emergency situations. But I digress. (In an excellent post, Patrick at The LIRR Today delves into this issue and more.)
I’d like to know from politicians where the general outrage is when seven pedestrians die in car crashes as they have over the last week in New York City. I’d like to know why it’s a struggle to fund mass transit until something calamitous happens and dramatic photos — of flooded stations, of derailed trains — are splashed across front pages. Investing in transit is a commitment, but it’s well worth it in added mobility and, yes, saved lives. Metro-North needs to improve its safety record, and it likely has to overcome a brain drain. But it needs support from start to finish and not just at the end.
33 comments
How is the MTA’s fraudulent statement that PTC is “still under development and is untested and unproven for commuter railroads the size and complexity of Metro-North and LIRR” not a knee-jerk reaction?
MTA is toooooooooooooooooooooooooooo slow on implanted PTC since they awarded the contract just this year. Poor planning and poor management on both MTA railroads.
The MTA also resides in the USA USA USA bubble. If it’s not here, it isn’t real! Forget that other civilized countries have it for commuter lines; we don’t so it’s not possible.
Based on passenger miles, rail is still far safer than driving a motor vehicle, but that safety record could be even better with PTC, which would help avoid the kind of mass casualty event like the one from this past weekend.
The MTA must do more to improve safety for the public, and claiming that the PTC isn’t proven for commuter rail is asinine and a cop-out to avoid obligations under federal law to install these safety measures.
Gov. Cuomo and state legislators must appropriate the funds necessary to get this done. But they wont – or wont do so with sufficient funds to address the problem in a timely fashion. They’ll force the MTA to take money from other programs to get PTC installed.
Now that we’ve proven trains are totally unsafe, can we turn MNRR into a BRT network? Pretty please?
Better yet, I think the rail corridor should be turned into a park.
Maybe a bike trail over better yet let turn into PARKWAYS!!! It will cut lots of people commute times on I-95 or I-87.
“We have to protect our phoney baloney jobs here, gentlemen! We must do something about this immediately! Immediately! Immediately! Harrumph! Harrumph!
— Gov. William J. Lepetomane (as recorded in “Blazing Saddles”)
Give Chuck Schumer a paddle ball racquet and I’ll be he could do a mean Mel Brooks imitation. Not sore about Sen. Gillibrand, though.
Politicians always rush forward when highly unusual tragic events occur, demanding investigations and changes, even if their lack of concern beforehand is one of the reasons why preventive measures weren’t in place, because the pols thought $$$ could be better spent elsewhere.
We’re still about a day or two before theses same pols figure out the LIRR has had a pre-PTC warning system in place for years, which will lead to a new round of outrageous outrage by the same politicians, and likely very much slower mandated speeds by the MTA between Yonkers and Spuyten Duyvil for the next several years to satisfy them, at least until PTC arrives on Metro North.
Sigh. These slow zones are a knee-jerk reaction to everything. In its dying years, the Milwaukee Railroad didn’t have money to repair a bridge that had burned, so instead it slapped a 10 mph slow restriction on it; when a train carrying copper ore crossed it, it collapsed, and the Milwaukee had to pay an expensive settlement for environmental remediation (link).
Hell, the curve in question could support slightly higher speeds. I’m not sure how superelevated the curve is, but at today’s cant deficiency (5″, =125 mm) and 75 mm freight-friendly cant it would be 63 km/h, and at full 150 mm cant and at the cant deficiency the FRA is about to permit (~6″, =150 mm) it would be 77 km/h. Allowing these somewhat higher speeds isn’t going to make the curve any less safe: even a heavy diesel loco with a high center of gravity can run the curve safely at 150 mm cant deficiency. At 82 mph, or 132 km/h, it wouldn’t become any less survivable.
In fact, pushing speeds closer to the comfort limit – still very far from the safety limit – improves finances slightly by reducing labor costs slightly and increasing ridership on the margins, so it gives the MTA more working money to spend on PTC.
I think it’s safe to say a lot of dumb transit decisions are less about doing what makes sense and more about creating work.
I would go out on a limb and say we can afford a bit less train safety if it means more overall safety. Cars are always going to be the most dangerous mode, and reducing dependence on them will increase safety the most.
For once I agree with Bolwerk.
A train crash because the TO was a known drunk and the union covered from him (union square), no. One train crash every 30 instead of every 40 years because the safety factor is 1.7 instead of 2.1, or the slow zone before a curve is 1/2 mile not 2 miles, whatever. I’d be more worried about the huge car to platform gap and exposed 3rd rails outside the city. Those are obvious, deadly, and prohibited in all sane places.
NYTC slowing the entire system so as to allow/cover themselves for another drugged T/O with a conductor covering for him as s/he misses multiple stops, openes the door over the track not the platform etc etc, that’s criminal.
It wouldn’t hurt to bank that track just a little more, and order freights to never stop right there.
That’s not always the case, though. Private railroad managers who slap slow restrictions aren’t doing so because they’re trying to increase crew staffing cost – they’re trying to defer maintenance.
As for safety, consider how many people are afraid of flying because of crashes and terrorism, and how many are afraid of driving because of same. Passengers don’t react the same way to voluntary risks (what Eric F says downthread can be mitigated with safer driving) as to involuntary ones (trusting the pilot/train driver/bus driver). If railroads aren’t as proactive in anticipating safety problems and eliminating them as the airlines are required to be, passengers are going to remember those major accidents and not the per-p-km safety rates and drive.
I’d love to see the mta refuse to provide safety upgrades unless their funding source is 100% transparent and doesn’t cut the current budget whatsoever.
Mass transit IS dangerous.
Because as you are walking from your transit stop to your destination, you could get clipped by a car.
Or even be shoved by a padestrian causing you to fall & be killed by a motorist who was texting while driving!
Beware of apples to oranges statistics comparisons.
In this case, it looks like fatalities per rail trip are being compared to fatalities per motorist over the course of a year’s worth of trips.
Based on a per mile traveled basis, I found these stats that still show commuter rail is 17x safer than driving:
CARS/LIGHT TRUCKS: 7.28 fatalities per billion passenger miles
COMMUTER/LONG-HAUL TRAINS: 0.43 fatalities per billion passenger miles
BUSES: 0.11 deaths per billion passenger miles
AVIATION: 0.07 deaths per billion passenger miles
Thanks for pointing this out.
Thanks for doing the math – I noticed that but didn’t have time to look up the actual stats.
But really there’s no excuse for people being killed at all on a modern rail system, short of suicides. Yes it is absolutely a different standard than we hold our roads too, but that’s because rail transit is better than road transit.
I’m with you that train travel is very, very safe.
That said, the comparison of auto deaths to train/plane deaths is not apples to apples. A person can quite easily drive in such a way as to greatly reduce his collision potential. A person who does not drive recklessly, avoids driving in highly adverse weather and does not drive under the influence of alcohol or drugs is basically subject only to potential bad drivers around him and acts of God, etc. That will by no means get you down to zero probability of death, but you are much closer to that area than a general average would get you.
The other question is whether PTC would actually cause MORE deaths. Money is finite, and dollars diverted to these systems could instead go into system maintenance, incremental improvements to dicey areas, etc. Metro North’s safety record would not lead one to believe that it needs to suddenly mass-automate safety protocols.
The PTC mandate opens some pretty cool doors. It might become operationally legal to do something like run a subway train along LIRR trackage.
Studies have shown that drivers tend to dramatically overstate their own capabilities compared to those of the general population. This would lead one to believe that the mean is actually more of a true representative mean than you would think. Even in a car, the chance of death each time you leave the house is infinitesimal, but it’s still an order of magnitude higher than rail or plane travel, period.
I would wager that if you ONLY looked at innocent victims of DUI or talking/texting drivers, you would still have a higher death rate than travel by plane or train.
“Even in a car, the chance of death each time you leave the house is infinitesimal, but it’s still an order of magnitude higher than rail or plane travel, period.”
Agreed 100%
“Studies have shown that drivers tend to dramatically overstate their own capabilities compared to those of the general population.”
I’m not talking about capabilities. I’m talking about maintaining your car, not driving drunk or impaired, not going 90 in a 45, not texting, etc. For bonus points, don’t tailgate. That will not drive down your risk to zero but it will substantially lower your collision risk. Personally, I choose routes that avoid roads I believe are dangerous and minimize left turns. When I can’t avoid certain roads, I drive more cautiously on them. I won’t drive in blinding rain, and will pull off a highway exit when possible when there is horizontal rain. I’m sure I’m not alone in this. My number could still come up, but I very much believe that my odds are better than any general average.
I’m not sure it would be hard to put a number even on that.
Let’s take BoerumHillScott’s numbers at face value and assume you can 100% eliminate your personal risk of human error. If an act of God risk of death while driving is still 5% and the risk of someone else doing something to you is 10%, maybe you lower your odds to 1.092 (= 0.15 * 7.28) per billion passenger miles. This also would have to assume you always drive alone, and aren’t driven by someone who might be less competent than you.
Even being that generous, your odds are more than double the passenger-train odds, and I think BoerumHillScott’s numbers are skewed by mainline passenger rail being a popular suicide choice and the fact that cars are often hit at grade crossings. The risk to the passengers in the train is probably lower.
No matter what, you are going to be much safer on a train. You certainly can’t get your own human error down to zero either. I would just take the average rates with a big grain of salt.
There was a big spate of regional airline crashes back in the mid or early 90s. Thankfully, things have been much improved since then. At the time,the talking heads on tv would always intone that flying was safer than driving by an enormous margin. I always found the comparisons to be way to cute using the big averages because they were conflating main carrier cross country flying with all types of driving. Again, the averages do not account for a safety minded person driving from say Birmingham to Little Rock on all interstates vs. a turbo prop regional making the same trip in a fog bank. It’s a narrow point, but I just don’t think averages tell you the whole story.
I agree, but really I don’t see any mode as a neat substitute for another. They have overlapping purposes on the margins, but even that’s limited. When MNRR stops working, local transit and highways can’t pick up the slack. Generally speaking, rail stops making sense as the best option for a long-distance trip over 5 hours.
Also, you raise a funny point: TV talking heads are probably much more likely to own airline stocks than passenger rail stocks. Or maybe I’m just missing people defending MNRR.
Also, Talking Heads and Congresspeople fly all the time, willingly. They don’t train much, and when they do, they’d rather not be. Acela being the exception, for the most part. And WMATA, but in DC they only care about WMATA when it crashes. Then everyone cares.
“COMMUTER/LONG-HAUL TRAINS: 0.43 fatalities per billion passenger miles”
Does this include people driving, walking or jumping in front of the train? We’re considering only passengers on the train, or passengers on the platform who may on rare occasions fall or (NYC) be pushed in front if the train.
I suspect passenger fatalities on a train are dramatically less than in cars, by a factor of over ten to one, simply because train systems are run very conservatively now.
Mixed passenger and freight operations are always going to have problems though, nature of the thing.
The Times finally did an article on the existing safety systems instead of just focusing on the lack of PTC:
http://www.nytimes.com/2013/12.....utine.html (this URL currently works but doesn’t make much sense given the article title is “Doomed Metro-North Train Had Warning System, Just Not in Operator’s Cab”. Google that if you have issues.)
General FYI for those who have trouble with Times articles: set your browser to private browsing to evade the paywall. That’s Ctrl+Shift+P on Firefox in Linux and WIndows. I think the command is the same for IE and Chrome. On Macs, it might be Apple-Shift-P.
Also, on Firefox, you can right-click a URL and select “Open in New Private Window.”
The counter on articles is reset to zero within the private browser. After 10 articles, just restart the private browser.
A note to the ptc knee-jerkers, especially the fed pols who mandated it: it is NOT failsafe; just look the dc metro, which essentially has it, and yet has had multiple collisions, with fatalities.
And these numbers are not legit: “since 1993, for every 1 billion train passengers, seven have died. In 2012 alone, 33,561 Americans died in traffic incidents. The comparable motor vehicle death rate is 108,000 for every 1 billion drivers.”
You are probably comparing trips [on trains] and drivers [who make thousands of trips each year] – a good example of apples and oranges.
So you’re saying trains aren’t a few orders of magnitude safer than cars?
If by “essentially has PTC” you mean “has a system that doesn’t even know what the trains’ correct deceleration rates are,” and “has a train supervision system with unreliable communication signals between the trackside signals and the trains,” then sure, the DC Metro has PTC. This demonstrates the advantage of using a debugged international standard with a good track record rather than inventing a bespoke system.