Updated

A former NASA official who led a study three years ago that faulted the way the agency dealt with safety risks told the Columbia investigation board Thursday that the same problem appears to have played a role in the shuttle disaster.

Henry McDonald, an engineering professor, appeared as a witness as the board held its first public hearing on what caused the shuttle to break up over Texas on Feb. 1, killing all seven astronauts.

McDonald said he was disappointed the space agency did not adopt more of his team's recommendations. He noted that the same type of communication breakdown he warned about seems to have hindered engineers who evaluated damage to Columbia's left wing by launch debris and concluded the shuttle and its astronauts were safe.

"It's a replay," McDonald told reporters after addressing the board.

McDonald said that the system used by NASA to assess shuttle flight risk was flawed and that there seemed to be a perception among workers that "if I've flown 20 times, the risk is less than if I've just flown once."

He said he tried to convince engineers and managers that this was not true.

"If you have a 1-in-100 chance of risk of an event occurring, the event can occur on the first or the last" time, he said. "We were continually attempting to inform them unless they change the risk, positively, they still have the same issue even after 50 flights or 60 flights."

Investigators believe that foam, ice or other debris that fell off the shuttle's big external tank during liftoff Jan. 16 damaged the left wing and allowed hot gases to burn through the wing and destroy the shuttle.

Foam had broken off during previous launches without serious incident. And while Columbia was still in orbit, a group of Boeing engineers evaluated the possible damage to the shuttle's thermal tiles and concluded the spacecraft was in no danger.

The chairman of the Columbia investigation board, retired Navy Adm. Harold Gehman Jr., said of McDonald's 2000 report: "It's eerily prescient."

Eight members of the board, dressed mostly in black, took part in the sparsely attended hearing, with fewer than 100 observers in the black-curtained 500-seat auditorium at the Clear Lake campus of the University of Houston. The board plans public hearings twice a week for two out of every three weeks until further notice. The next hearing is March 17 in Houston.

McDonald's shuttle safety study was commissioned after a pair of close calls on a 1999 Columbia launch: one involved a short circuit, the other a leaking main engine. At the time, McDonald was director of NASA's Ames Research Center in Mountain View, Calif.

He found that NASA depended on archaic database systems for logging shuttle problems, and so it was difficult if not impossible for top-level managers to quickly know the history of any particular issue. He said shuttle program manager Ron Dittemore, who also appeared before the board Thursday, could have benefited greatly from a more modern, consolidated system.

"I have no concern at all that people like Ron Dittemore, presented with the facts, will make the right decision, no concern at all," McDonald said. "The concern is presenting him with the facts and many of them are very deep."

Some improvements have been made since then, he said, but it still is an inefficient system and does not provide decision-makers with the failure probability for any particular shuttle part.

McDonald suggested not many of his panel's 120 recommendations were fully adopted by NASA. He eventually left the space agency, and is now teaching at the University of Tennessee at Chattanooga.

Earlier in the 4-hour hearing, Gehman asked another witness, the director of Johnson Space Center, about critics who contend that NASA, while repairing one nitty-gritty shuttle problem after the other, sometimes misses "the big, big safety issues."

"Would you just give me your views of ... how you satisfy yourself that you've got an eye open not only to the snake that's right at your ankle, but what's over the hill?" Gehman asked Jefferson Davis Howell Jr., the center director.

Howell agreed that the space agency has a tendency of "taking care of business today." But he noted that "some of our best talent" is working to improve the shuttle and keep it flying.

Gehman and other board members were also tough in their questioning of a Boeing engineer who, back in the early 1990s, was part of a team that looked at external tank issues.

The engineer, Keith Chong, acknowledged that no one on the team studied the shedding of insulating foam from the tank, even though it had been a problem since the beginning of the program.

"To be honest with you, that was not discussed," he said.

Chong said the primary focus was on finding an environmentally friendly replacement for Freon in the foam; the substitute was not as good and caused a "popcorning" of the foam that NASA eventually learned how to deal with, he said.

In Washington, NASA Administrator Sean O'Keefe told lawmakers at a Senate subcommittee hearing that NASA faces dangerous staff shortages because of looming retirements and fewer college graduates with the skills it needs.

"We lost some individuals with skills we couldn't afford to lose" during the past decade, O'Keefe said. "Through downsizing and the normal attrition process, we lost key areas of our institutional knowledge base."

The space agency has been lobbying for employment changes, including higher pay and bonuses, for more than a year.