By Joe Stephens and Lena H. Sun
Washington Post Staff Writers
Sunday, September 27, 2009
One of the first moves Metro officials made after a subway crash killed nine people this summer was to sandwich older rail cars, similar to one crushed in the accident, between newer, sturdier cars. While repeatedly portraying the move as one that might improve safety, interviews and newly obtained documents show Metro conducted no engineering analysis before launching the initiative.
Metro took the action "as a means to address public perception," Metro safety chief Alexa Dupigny-Samuels told a safety panel in July in a previously undisclosed letter.
Shifting the cars "was not undertaken because these railcars are 'unsafe' or [because they] pose a hazard," she wrote in the letter to the Tri-State Oversight Committee, a regional safety panel. Officials repositioned the cars, she said, "to provide an added level of reassurance." Dupigny-Samuels did not cite any scientific support for the move.
Committee members concluded that Dupigny-Samuels viewed the measure as "purely a public relations effort," records show.
Meanwhile, Metro General Manager John B. Catoe Jr. was striking a different tone in public appearances.
At a June 30 news conference, Catoe said, "To ensure as much as we can the crashworthiness of our cars and of our trains, we're moving these cars to the belly of the train configuration." Five days later in an op-ed column in The Washington Post, Catoe wrote, "We have placed the 1000-series rail cars at the centers of our trains, hoping that this will make them less vulnerable."
And a week after that, in testimony to a congressional subcommittee, Catoe said that Metro was working "to enhance safety" and that "the cars are specifically placed based upon . . . crashworthiness." In the news conference and in his testimony, Catoe stressed that he considered the cars safe to operate.
Metro spokeswoman Lisa Farbstein confirmed last week that the agency conducted no analysis showing that the measure improved safety. She said she was unaware of any other transit agency that had taken similar steps and was unaware of any outside entity that had sanctioned the move.
"We have no engineering analysis that says it will make it safer, that is correct," Farbstein told The Post on Wednesday. "We do not have any engineering analysis that says one way or the other. We certainly don't believe it's going to make any train set less safe. . . . So as a precautionary measure, we thought it would be best to belly them."
She added that Metro considers the 1000 series cars safe even when on the ends of trains. The cars remain in the center of the trains.
On Friday, Farbstein supplied The Post with a Web link to an 11-year-old study, which she said showed that the first car in a striking train suffers the greatest amount of crush. The study, however, said that its findings applied to a "generic, high speed" train and that it relied on the "probably unrealistic" assumption that the cars would not buckle or override each other, as 1000 series cars have done in past accidents. The study did not examine what happens when dissimilar cars are mixed in the same train.
Metro board member Jeff McKay said establishing scientific support was never a concern. When the board met in an emergency session the day after the crash, some Metro executives were reluctant to belly the older cars because of manpower costs and logistical concerns, he said. But the board insisted.
"In reality, we knew it meant nothing, but the perception was that even if we don't think it matters, we need to do it because it gives people a sense of security," McKay said.
In following weeks, regional safety officials struggled to understand the reason for the maneuver.
Kenneth Korach, a consultant for the Tri-State Committee, said in a July 22 e-mail to members: "As of this point, we still do not know if WMATA [the Washington Metropolitan Area Transit Authority] has increased or decreased the risks associated with the 1000 series cars by placing them in the middle of trains. This leads me to the conclusion that given WMATA's reluctance or inability to provide us with this information, that no analysis was conducted."
Public perception is of increasing importance to Metro, which has been beset by concerns about safety and complaints about service. Simultaneously, Metro has been pursuing hundreds of millions of dollars from Congress to modernize its fleet and pay for other capital improvements.
After the June 22 accident, Metro retained two crisis communications consultants, one paid $4,000 a day and the other $275 an hour, to augment its five-person media office. Metro has moved aggressively to counter news articles, posting long rebuttals on the agency's Web site that it labels "Corrections and Clarifications."
Metro has declined in recent weeks to release documents in response to a number of open records requests, saying its staff is busy with the crash investigation. The Post obtained records related to the repositioning of rail cars through an open records request to the Virginia Department of Rail and Public Transportation. Metro did not respond to a similar request.'It Cost a Lot of Money'
The 1000 series cars, built by Rohr Industries, are more than 30 years old and lack new technology for absorbing an impact. They have long been a source of concern.
After 1000 series cars crumpled during a 2004 crash at Woodley Park, the National Transportation Safety Board recommended that Metro accelerate retirement of the cars or strengthen their frames. The 1000 series, the board said, was "vulnerable to catastrophic telescoping damage and complete loss of occupant survival space."
Metro officials said they could not afford either option but promised to replace the cars over a number of years, a response the NTSB described at the time as "unacceptable."
On June 22, a Red Line train slammed into a train stopped near the Fort Totten Station, and lead car No. 1079 -- a 1000 series car -- compressed to a third of its original length as it rode up and over the preceding train. NTSB Chairman Deborah A.P. Hersman again publicly questioned Metro's failure to replace the cars and, seven days after the crash, Metro announced plans to reposition them between newer cars.
Metro has almost 300 of the 1000 series cars, which comprise a quarter of its fleet. The reconfiguration proved costly and time-consuming because operators had to move hundreds of cars among rail yards at night. At a news conference, Catoe compared the complex logistics to sorting tiles on a Rubik's Cube. Cars still need to be reshuffled nightly.
"It cost a lot of money," Farbstein said, adding that she was unable to give a specific figure. "It was a significant amount of overtime, I know, to make that change."
The initiative surprised the Tri-State Oversight Committee, which monitors Metro's safety but has no authority to force the agency to make changes or share information. "We were notified through the press release" that announced the move June 30, said committee chairman Eric Madison.
On July 17, Madison wrote to Dupigny-Samuels, the Metro safety chief, saying that for eight months Metro had failed to respond to the committee's request for a comprehensive strategy to reduce the hazard posed by the cars. He said Metro had yet to provide the committee with any engineering analyses, crash-test data "or even anecdotal evidence" to support the move.
Dupigny-Samuels responded July 21, saying that a 1997 study determined that reinforcing the cars would be impractical and costly, so Metro had instead decided to replace them over time with newer models. She did not refer to any documentation showing that the new configuration was less dangerous but asserted that "the 1000 series railcars are safe."
"The re-configuration of these railcars is being undertaken as a precautionary measure," she wrote.
On Sept. 16, Madison wrote back, saying Dupigny-Samuels's letter appeared to indicate that the bellying strategy was "not intended to mitigate any hazard nor make the system in any way safer." He said the committee assumed that Metro had no professional analyses or other evidence to validate the move. And Dupigny-Samuels's letter appeared to be "directly contradictory" to Metro releases saying that in the new configuration "the newer cars may absorb the majority of impact," Madison wrote.
He questioned whether moving the cars might create "unforeseen hazards." And he questioned whether distributing 1000 series cars among several maintenance shops, which tend to specialize in particular models, could create risks of its own.
The letter asked Metro to tell the committee by Sept. 30 "whether 'bellying' these railcars provides a real safety enhancement or, as you note in your July 21st letter, this measure is purely a public relations effort."
The NTSB declined to comment on Metro's actions because it is investigating the crash. But on July 14, in little-noticed testimony before the House, the board's chairman declined to endorse Metro's strategy.
"The safety board has not taken a position on whether or not putting the cars in the center was the right thing to do," Hersman testified. "We did ask them to look at the evaluation of these cars in a scientific way. . . . We don't have the engineering data to necessarily support the placement."