The Columbia Space Shuttle Disaster

Dogged Engineer’s Effort To Assess Shuttle Damage

Over and over, a projector at one end of a long, pale-blue conference room in Building 13 of the Johnson Space Center showed a piece of whitish foam breaking away from the space shuttle Columbia’s fuel tank and bursting like fireworks as it struck the left wing.

In twos and threes, engineers at the other end of the cluttered room drifted away from their meeting and watched the repetitive, almost hypnotic images with deep puzzlement: because of the camera angle, no one could tell exactly where the foam had hit.

It was Tuesday, Jan. 21, five days after the foam had broken loose during liftoff, and some 30 engineers from the National Aeronautics and Space Administration and its aerospace contractors were having the first formal meeting to assess potential damage when it struck the wing.

Virtually every one of the participants — those in the room and some linked by teleconference — agreed that the space agency should immediately get images of the impact area, perhaps by requesting them from American spy satellites or powerful telescopes on the ground.

They elected one of their number, a soft-spoken NASA engineer, Rodney Rocha, to convey the idea to the shuttle mission managers.

Mr. Rocha said he tried at least half a dozen times to get the space agency to make the requests. There were two similar efforts by other engineers. All were turned aside. Mr. Rocha (pronounced ROE-cha) said a manager told him that he refused to be a ”Chicken Little.”

The Columbia’s flight director, LeRoy Cain, wrote a curt e-mail message that concluded, ”I consider it to be a dead issue.”

New interviews and newly revealed e-mail sent during the fatal Columbia mission show that the engineers’ desire for outside help in getting a look at the shuttle’s wing was more intense and widespread than what was described in the Aug. 26 final report of the board investigating the Feb. 1 accident, which killed all seven astronauts aboard.

The new information makes it clear that the failure to follow up on the request for outside imagery, the first step in discovering the damage and perhaps mounting a rescue effort, did not simply fall through bureaucratic cracks but was actively, even hotly resisted by mission managers.

The report did not seek to lay blame on individual managers but focused on physical causes of the accident and the ”broken safety culture” within NASA that allowed risks to be underplayed. But Congress has opened several lines of inquiry into the mission, and holding individuals accountable is part of the agenda.

In interviews with numerous engineers, most of whom have not spoken publicly until now, the discord between NASA’s engineers and managers stands out in stark relief.

Mr. Rocha, who has emerged as a central figure in the 16 days of the Columbia’s flight, was a natural choice of his fellow engineers as a go-between on the initial picture request. He had already sent an e-mail message to the shuttle engineering office asking if the astronauts could visually inspect the impact area through a small window on the side of the craft. And as Mr. Rocha was chief engineer in Johnson Space Center’s structural engineering division and a man with a reputation for precision and integrity, his words were likely to carry great weight.

”I said, ‘Yes, I’ll give it a try,’ ” he recalled in mid-September, in the course of five hours of recent interviews at a hotel near the space center.

In its report, the independent Columbia Accident Investigation Board spoke of Mr. Rocha, 52, as a kind of NASA Everyman — a typical engineer who suspected that all was not well with the Columbia but could not save it.

”He’s an average guy as far as personality, but as far as his engineering skills, he’s a very, very detail-oriented guy,” said Dan Diggins, who did many of the interviews for the report’s chapter on the space agency’s decision-making during the flight and wrote that chapter’s first draft before it was reworked and approved by the board. Never in hours of interviews did Mr. Diggins find a contradiction between Mr. Rocha’s statements and facts established by other means, he said.

Mr. Rocha’s experience provides perhaps the clearest and most harrowing view of a NASA safety culture that, the board says, must be fixed if the remaining shuttles are to continue flying…

Mr. Rocha said that when he learned of the foam strike in a phone call on Friday afternoon, he gasped. All weekend he watched the video loop showing the strike, and at 11:24 p.m. on Sunday, he sent an e-mail message to the manager of the shuttle engineering office, Paul Shack, suggesting that the astronauts simply take a look at the impact area.

Mr. Shack never responded. But by Tuesday afternoon, Mr. Rocha was showing the loop to the so-called debris assessment team at the meeting in Building 13, where he had his own office. As arresting as the images were, the team agreed, they were too sketchy to draw conclusions without new images.

To engineers familiar with the situation, the request was an easy call. ”We all had an intense interest in getting photos,” said Steven Rickman, a NASA engineer whose staff members served on the assessment team. ”As engineers, they’re always going to want more information.”

In his second e-mail appeal for satellite imagery, Mr. Rocha wrote in boldface to Mr. Shack and other managers, ”Can we petition (beg) for outside agency assistance?”

But Mr. Rocha did not know that the strange politics of the NASA culture had already been set in motion. Calvin Schomburg, a veteran engineer who was regarded as an expert on the shuttle’s thermal protection system — though his expertise was in heat-resisting tiles, not the reinforced carbon-carbon that protected the wings’ leading edges — had been reassuring shuttle managers, Mr. Diggins said. Mr. Schomburg either ”sought them out or the managers sought him out to ask his opinion,” Mr. Diggins said.

Whether because of Mr. Schomburg’s influence or because managers simply had no intention of taking the extraordinary step of asking another agency to obtain images, Mr. Rocha’s request soon found its way into a bureaucratic dead end.

On Wednesday, an official Mr. Schomburg had spoken to — Ms. Ham, the chairwoman of the mission management team — canceled Mr. Rocha’s request and two similar requests from other engineers associated with the mission, according to the investigation board. Late that day, Mr. Shack informed Mr. Rocha of management’s decision not to seek images.

Astonished, Mr. Rocha sent an e-mail message asking why. Receiving no answer, he phoned Mr. Shack, who said, ”I’m not going to be Chicken Little about this,” Mr. Rocha recalled.

”Chicken Little?” Mr. Rocha said he shouted back. ”The program is acting like an ostrich with its head in the sand.”

Mr. Shack, Mr. Schomburg and Ms. Ham declined to comment for this article or did not respond to detailed requests for interviews relayed through the space agency’s public affairs office.

On the day he talked with Mr. Shack, Mr. Rocha wrote an anguished e-mail message that began, ”In my humble technical opinion, this is the wrong (and bordering on irresponsible) answer.” He said his finger hovered over the ”send” key, but he did not push the button. Instead, he showed the draft message to a colleague, Carlisle Campbell, an engineer.

”I said, ‘Rodney, that’s a significant document,’ ” Mr. Campbell said in an interview. ”I probably got more concerned or angry than he did at the time. We could not believe what was going on.”

But Mr. Rocha still decided he should push his concerns through official channels. Engineers were often told not to send messages much higher than their own rung in the ladder, he said.

Taking the Issue Higher

The next day, Mr. Rocha spoke with Barbara Conte, a worker in mission operations, about spy telescopes. In a written response to reporters’ questions, Ms. Conte said her colleague ”was more keyed-up and troubled than I had ever previously encountered him.”

That day, she and another NASA employee, Gregory Oliver, took the issue to Mr. Cain, the Columbia’s flight director for landing, at an unrelated meeting.

”We informed LeRoy of the concern from Rodney” and offered to help arrange an observation by military satellites, Mr. Oliver wrote on March 6 — a month after the accident — in a previously unreleased e-mail chronology of shuttle events. The message continued, ”LeRoy said he would go talk to Linda Ham and get back to us.”

About two hours later, at 12:07 p.m. that day, Mr. Cain sent out his own e-mail message saying he had spoken with management officials, who had no interest in obtaining the images. Therefore, Mr. Cain wrote, ”I consider it to be a dead issue.”

It was not over for Mr. Rocha, though. On Thursday afternoon, Jan. 23, he encountered Mr. Schomburg, the expert on the heat-resisting tiles, on the sixth floor of Building 1, where most of the managers had offices. They sat down in the anteroom of an office and began arguing about the need for imaging, said Mr. Rocha and the investigative board’s report.

Mr. Schomburg insisted that because smaller pieces of foam had broken off and struck shuttles on previous flights without dire consequences, the latest strike would require nothing more than a refurbishment after the Columbia landed. Mr. Rocha maintained that the damage could be severe enough to allow hot gases to burn through the wing on re-entry and threaten the craft.

As their voices rose, Mr. Rocha recalled, Mr. Schomburg thrust out an index finger and said, ”Well, if it’s that bad, there’s not a damn thing we can do about it.”

On Jan. 24, eight days into the mission, engineers and managers held a series of meetings in which the debris strike was discussed. At a 7 a.m. meeting, Boeing engineers presented their analysis, which they said showed that the shuttle probably took the hit without experiencing fatal damage.

Those results were hastily carried into the 8 a.m. meeting of the mission management team, led by Ms. Ham. When a NASA engineer presented the results of the Boeing analysis and then began to discuss the lingering areas of uncertainty, Ms. Ham cut him off and the meeting moved along. The wing discussion does not even appear in the official minutes.




On the morning of January 21, 2003, the Mission Management Team (MMT) for NASA mission STS-107 – the twenty-eighth flight of the space shuttle Columbia – held a teleconference, its second since the Columbia’s launch on January  16.  An hour before the meeting Don McCormack had been briefed by members of the Debris Assessment Team (DAT), a group of engineers from NASA, Boeing, and Lockheed Martin  who had spent much of the previous five days evaluating the possible consequences of a large debris strike on the Columbia.  During the shuttle’s  ascent into the atmosphere, a large  piece of foam had broken off the left bipod area of the shuttle’s external fuel tank and had smashed into the ship’s left wing. None of the cameras that were tracking the shuttle’s launch had provided a clear picture of the impact, so it was difficult to tell how much damage the foam might have caused. And although by January 21  a request had been made for  on-orbit pictures of the Columbia, they had not been approved. So the DAT had done what it could with the information it had, first estimating the size of the foam and the speed at which it had struck the Columbia, and then using an algorithm called Cater to predict how deep a piece of debris that size and traveling at that speed would  penetrate into the thermal-protection tiles that covered the shuttle’s wings.

The DAT had reached no conclusions, but they made it clear to McCormack that there was reason to be concerned. McCormack did not transmit that sense of concern to the MMT during its teleconference. The foam strike was not mentioned until two-thirds of the way through the meeting, and was brought up only after discussions of, among other things, a jammed camera, the scientific experiments on the shuttle, and a leaky water separator. Then Linda Ham, who was the MMT leader, asked McCormack for an update. He simply said that people were investigating the possible damage and what could potentially be done to fix it, and added that when the Columbia had been hit by a  similar strike during STS-87, five years earlier, it had suffered “fairly significant damage.”  This is how Ham answered: “And I really don’t think there is much we can do so its not really a factor during the flight because there is not much we can do about it.”

Ham, in other words, had already decided that the foam strike was inconsequential.  More important, she decided for every one else in the meeting that it was inconsequential, too. This was the first time the MMT had heard any details about the foam strike.It would have been logical for McCormack to outline the possible consequences and talk about what the evidence from past shuttles that had been struck with debris showed. But instead the meeting moved on.

Hindsight is, of course, twenty-twenty, and just as with the critiques of the U.S. intelligence community after September 11, its perhaps too easy to fault the MMT at NASA for its failure to see what would happen to the Columbia when it reentered the Earth’s atmosphere on February 1.  Even those who have been exceptionally critical of NASA have suggested that focusing on this one team is a mistake because it obscures the deep institutional and cultural problems that plague the agency.

…sifting through the evidence collected by the Columbia Accident Investigation Board (CIBO), there is no way to evade the conclusion that the team had an opportunity to make different choices that could have dramatically improved the chances of the crew surviving.  The team members were urged on many different occasions to collect the information they needed to make a reasonable estimate of the shuttle’s safety. They were advised that the foam might, in fact, have inflicted enough damage to cause “burn-through” – heat burning through the protective tiles and into the shuttle’s fuselage- when the shuttle reentered the Earth’s atmosphere. ….

The Performance of the MMT is an object lesson in how not to run a small group, and a powerful demonstration of the way in which, instead of making people wiser, being in a group can actually make them dumber.

as found in James Surowiecki: The Wisdom of Crowds, p.p 173-174.

NASA Report on Columbia Disaster