Space Shuttle Disaster

PBS Airdate: October 14, 2008
Go to the companion Web site

NARRATOR: The disintegration of the space shuttle Columbia, in 2003, was a blow to the nation; for NASA it was unthinkable.

JON CLARK (Flight Surgeon, NASA): We were at the landing site, and then all the cell phones started going off, with the astronauts that escort us, the family escorts. And then, boom, all of a sudden we're rushed to vehicles.

NARRATOR: Columbia was America's first space shuttle, and it redefined what a spacecraft could do: it took off like a rocket and landed like an airplane; it was reusable.

ROGER LAUNIUS (National Air and Space Museum): NASA really believed that space flight was going to become routine. And they really believed that it was their responsibility to build a vehicle that would make space flight something that happened all the time, with relative ease, comfort and inexpensiveness.

NARRATOR: But the loss of Columbia forced NASA to reevaluate the very idea of a space shuttle, just as the loss of Challenger, 17 years earlier, killed the dream of safe and routine space travel.

HOWARD MCCURDY (American University): They all kind of had conflicting design requirements, as a result of trying to get too many people agreeing on a single vehicle. The accident, in effect, was planned from the moment of the conception.

JOHN LOGSDON (Space Policy Institute): This is a tragedy: that it took the loss of seven lives to force the realization that there had not been a compelling, strong rationale for sending humans into space.

NARRATOR: The tragedy highlights deep questions now facing NASA, as it designs the next generation of spacecraft and determines the future of America's role in space. Up next on NOVA: Space Shuttle Disaster.

Major funding for NOVA is provided by the following:

Well, it might surprise a lot of people that ExxonMobil would be interested in lithium ion battery technology applied to hybrid electric vehicles. Our new battery separator film is a true breakthrough that's going to enable the deployment of more hybrid vehicles, faster. This means a tremendous reduction in greenhouse gases, the equivalent of removing millions of cars from the road. I think this is the most important project that I've worked on in my career.

And David H. Koch. And...

Discover new knowledge: HHMI.

And by the Corporation for Public Broadcasting, and by contributions to your PBS station from viewers like you. Thank you.

STORY MUSGRAVE (Astronaut): Launch is...it's acceptance of the risk. Now, the first time's okay, because you don't know what they're going to do to you. It's the subsequent times that are hard. The vibrations and the noise are such that you cannot believe the shuttle and its systems are going to tolerate this abuse.

NARRATOR: On January 16, 2003, Mission STS-107 took off from Cape Canaveral aboard the space shuttle Columbia.

STORY MUSGRAVE: It is shaking so bad you are afraid you're going to lose your teeth. I mean, it's hard.

NARRATOR: The commander of the crew was Rick Husband, veteran of one previous space mission. His pilot, Willie McCool, was flying for the first time, as were mission specialist Dave Brown and Ilan Ramon, the first Israeli in space. Payload commander Michael Anderson, flight engineer Kalpana Chawla, and Navy officer Dr. Laurel Clark, completed the crew.

JON CLARK: I had been concerned about the erosion of safety culture, and that's based on my observations in the involvement I had taking care of the crew. And I can remember a discussion I had with my wife before the mission. When I talked to her about how risky space flight was, and I went through some examples, her response was very interesting. She goes, "Well, if it was so risky, why didn't NASA tell me that?" At the time, she thought I was just being negative.

NARRATOR: Since the shuttle Challenger accident in 1986, 17 years earlier, NASA had clocked 87 successful missions. But the spacecraft the crew was flying that morning was still considered experimental.

The shuttle is the most complex spacecraft ever designed. And liftoff is especially unforgiving. Once the rockets fire, there's no turning back.

JON CLARK: The particular phase of the mission that's always been the most challenging is the launch. You've got a lot of energy there that's being unleashed. The ability to do anything during that first two minutes with the solid rocket boosters is non-existent.

NARRATOR: Tracking cameras followed every second of the launch, but until the images were analyzed the next day, no one would notice anything wrong. It seemed a perfect launch. Eight minutes later, Columbia was in orbit, and the crew began a mission dedicated totally to science.

Columbia was the oldest shuttle in the fleet, and the only one not designed to dock with the international space station. Ironically, this shuttle was originally designed with a space station in mind, but not the one we have today. The original plan was far more ambitious.

In fact, the story of the shuttle really begins during NASA's glory days, as Neil Armstrong was preparing to land Apollo 11 on the Moon.

NEIL ARMSTRONG (Archival Audio): Houston, Tranquility Base here. The Eagle has landed.

JOHN LOGSDON: NASA did a lot of planning, in the late '60s and early '70s, of what its initiative after Apollo should be.

NEIL ARMSTRONG (Archival Audio): That's one small step for man...

JOHN LOGSDON: That got grandiose after the success of Apollo...

NEIL ARMSTRONG (Archival Audio):...one giant leap for mankind.

JOHN LOGSDON: ...twelve-person space station by 1975, a 50-person space station by 1980 and a 100-person space station a few years later. And they realized that throwing the rockets away every time would kill the economics. And so, you needed something reusable, you needed something to shuttle from the Earth to the space station.

NARRATOR: But as NASA was planning the future, the powers in Washington were planning to slash NASA's budget, starting with the remaining Apollo missions.

MICHAEL GRIFFIN (NASA Administrator): Nixon took us out of lunar exploration. That administration cancelled the last three Apollo missions—for which the hardware had already been bought and paid for—and did not fund any significant successor program.

HARRISON "JACK" SCHMITT (Astronaut, Geologist): The work left undone was very clear, even before the last missions were flown.

NARRATOR: As a trained geologist, Jack Schmitt was the first scientist, and the last, to walk on the Moon.

JACK SCHMITT: The Apollo 17 mission took us to a deep mountain valley on the Moon, deeper than the Grand Canyon. The mountains on either side were 6- and 7,000 feet above us, 2,100 meters or so. What it did—and the reason it was valuable—it gave us a three-dimensional view of many of the things we had only seen in one dimension, by landing on relatively flat planes, elsewhere. Three dimensions, to a geologist, is like ice cream: it is extremely important and delicious to have. And for a geologist, there's a tremendous amount of exploration yet to be done.

MICHAEL GRIFFIN: The mistake of the post-Apollo years—a mistake, frankly, which belongs squarely in the Nixon administration of now nearly 40 years ago, in fact, 40 years ago next year—was the decision that the United States would no longer go beyond low Earth orbit.

NARRATOR: Even NASA's plans for low Earth orbit were soon scaled back. President Nixon cut the planned space station but continued to support the shuttle.

HOWARD MCCURDY: So astronauts, for 20 years, just flew into space and did some experiments and then flew back. Well, that was never what the shuttle was designed to do. It was designed to be a transportation system, to go someplace else, but it eventually became, because of the lack of money, its own destination.

JOHN LOGSDON: So then, in order to be justified, for the first time, the budget office said, "It has to be cost-effective. You have to show that it's a better way, a more economical way than using expendable launch vehicles."

And that required more launches, and so you needed more customers. And the only other customer in town was the Department of Defense and the intelligence agency. So you had to go to the intelligence agencies and say, "Would you use this vehicle?"

HOWARD MCCURDY: You had to be a space truck. It had to carry military satellites; it had to carry scientific satellites; it had to carry people, of course.

JOHN LOGSDON: It has to be able to take off, fly around the Earth and land at the same place.

NARRATOR: Protected by more than 20,000 thermal tiles, Columbia, the first shuttle orbiter, also needed to survive the fierce heat of reentry and land safely, like an airplane. It could carry a crew of seven and a payload about the size of a school bus.

But combining all the shuttle's functions with cheapness and reusability proved impossible in the end. The shuttle was less like a flatbed truck and more like a Formula One race car, requiring extensive re-fitting after each flight.

ROGER LAUNIUS: The thing that they learned in the context of flying the first few missions is how hard it was to turn the shuttle around, to process it. It took months on end for all of these orbiters to be prepared to return to flight, and in some cases longer than even a few months.

NARRATOR: The defining concept behind the shuttle, the idea that reusability would cut costs, was not turning out that way.

HOWARD MCCURDY: The per-flight costs of the shuttle range from a half a billion to over a billion dollars every time you fly. You can build a sports stadium with that. So that was one of the great quiet secrets about the shuttle, that the shuttle was going to cost as much to fly as the equivalent rocketry of the 1960s and 1970s. It's very, very expensive. Part of NASA's culture is to believe that if the mission is successful no one will ever ask what it costs.

NARRATOR: During the 1980s, the annual number of missions gradually increased: four missions in '83, five in '84, and a record nine missions in 1985. Launch rate was a priority. NASA was subsidizing commercial launches to stay competitive with the European Space Agency. NASA had reduced the fee for a satellite launch to such a low price, the line of customers was out the door.

DON NELSON (NASA Engineer): Launch schedule was a prime consideration. We kept saying that safety is the number one consideration, but launch schedule was right up there with it. We were showing the congress and the American public that, "We've got a space truck here that's ready to go, let's go start doing other things in space." So we're extremely important that we meet those schedules and keep the operating cost of the vehicle down.

JOHN LOGSDON: The engineers down in...close to the vehicle, knew how risky this was, but the upper management chose to ignore the messages of problems and said, "We've committed to do this, and we're going to do it."

HOWARD MCCURDY: And if you look at all the responsibilities, the functions, that were placed on the shuttle, they all, kind of, had conflicting design requirements, as a result of trying to get too many people agreeing on a single vehicle. The accident, in effect, was planned from the moment of the conception. It might take 30 years to happen, but it was eventually going to happen.

NARRATOR: The odds would finally catch up with the shuttle on January 28, 1986. In a fireball, the shuttle Challenger exploded 73 seconds after launch, killing seven astronauts, including high school teacher Christa McAuliffe.

MISSION CONTROL (Archival Footage): Obviously a major malfunction.

NARRATOR: Her presence on the mission was regarded as a sign that the era of safe and routine space travel had finally arrived.

CHRISTA McAULIFFE (Archival Footage): And I just hope everybody tunes in on day four, now, to watch the teacher teaching from space.

NARRATOR: The loss of Christa McAuliffe and the Challenger crew was a severe blow to NASA and to the nation. A distinguished board was assembled to investigate the accident. They found that the shuttle had launched on too cold a day, compromising a thin pressure seal called an o-ring, allowing hot gases to escape and ignite the main fuel tank.

The board also learned that several engineers objected to the launch but had been ignored. NASA, it was revealed, had deep flaws in its culture that were compromising safety.

DON NELSON: The Challenger accident was a surprise in two ways to us. It was a surprise just to have the accident. The other surprise is that there were engineers that knew the problem that caused the Challenger accident, and they had been trying to bring it forward to the management, and the management wouldn't listen to them.

ROGER LAUNIUS: After the Challenger accident, in 1986, the D.O.D. decided that this was just something that they would not be able to support any longer. And they made the decision, at that time, to pull the military payloads off the space shuttle and to launch all of them on expendable launch vehicles.

NARRATOR: When the program resumed, in 1988, the shuttle had lost its military and commercial payloads. NASA's focus then returned to science and discovery. Ulysses, Galileo, Magellan, probes and telescopes examined the far reaches of the solar system and beyond. But the most spectacular mission of them all was the repair, in orbit, of the Hubble space telescope.

Story Musgrave performed many of those repairs.

STORY MUSGRAVE: Space has the opportunity to offer you, number one, exploration. It's what kind of universe we've got, what's our place in it, you know? People love Hubble images. It tells them where they came from. It tells them where they are going, brings the world together.

HOWARD MCCURDY: After the Challenger accident, NASA got six, 10, 15, 20 successful flights; the agency drifted back into the old idea that it was just an airplane. People inside NASA became so convinced that this was routine and operational that, in 1996, the agency provided that the shuttle operations would be turned over to a private contractor.

NARRATOR: The shuttle workforce had peaked at 30,000, in 1993, but, by 2002, it had been cut nearly in half.

HOWARD MCCURDY: Private industry can go in with its cost-cutting models that are used in grocery stores and parts dealerships, that allow you to squeeze increasingly large numbers of people and money out of a program and run it as close to the margin as you possibly can. But when you push the shuttle to the margin, that's very dangerous.

NARRATOR: In 1998, assembly of the long-delayed international space station began. The shuttle would finally have somewhere to go. But the space station brought another round of scheduling and budget pressures.

SCOTT HUBBARD (Columbia Accident Investigation Board): At the time of the Columbia launch, NASA's new administrator, Sean O'Keefe, who was a budget person, had come in to fix what was regarded as the big problem of NASA, which was an inability to manage cost.

SEAN O'KEEFE (NASA Administrator, 2001–2004): Indeed, the reason I became NASA administrator was predominantly because the program management on the international space station had become quite confused. The shuttle is a very expensive piece of machinery to fly. The safety regime of this is what makes it so expensive.

DON NELSON: In the shuttle design, where we made a major miscalculation is that we could design it so that the factor of safety would always be that it would fail safe. But there were a number of places on it where it would not fail safe.

NARRATOR: Failing safe would mean that a crew could survive a vehicle failure. The capsules of Mercury, Gemini and Apollo had an auxiliary rocket that could pull the crew to safety if the main rockets failed. But as the Challenger accident cruelly demonstrated, the shuttle had no such fail safe. Even if the rockets didn't fail, the orbiter was still vulnerable during liftoff.

During the shuttle's development, there was concern that ice fragments falling from the shuttle's huge external tank could damage the orbiter. The tank contains hydrogen and oxygen, kept liquid at supercool temperatures. To prevent ice—condensed from the moist Florida air—from forming on the exterior required that foam insulation be applied, a material as light and as brittle as Styrofoam. But this presented its own problems.

On the very first flight of the shuttle, in 1981, engineers noticed hundreds of fragments of foam breaking loose and hitting the underside of the orbiter.

ALLEN RICHARDSON (Boeing Engineer): We had all this concern, early on, about the foam. And we found about 50 gouges in it, on the bottom side of the vehicle.

NARRATOR: Engineers worried that severe damage to the underside of the orbiter could allow super-heated atmospheric gases to leak in during re-entry, which would be catastrophic. But over time, as the shuttle continued to land safely, the foam-shedding issue became less and less of a concern.

DOUGLAS OSHEROFF (Columbia Accident Investigation Board): After STS 1, 2, 3 and 4, it was pretty clear that there was no way of eliminating impacts to the orbiter at the level that were forbidden, so they just basically ignored that. And, you know, there were...I would say, on virtually every flight, there were small pieces of foam which would impact on the undersides of the wings, typically causing damage to the thermal tiles.

NARRATOR: More than 20 years later, in 2003, as Columbia sat on the launch pad for what would be her final mission, NASA was still studying the problem.

SEAN O'KEEFE: When you see something, however abnormal, often enough, you begin to think it's normal. The fact that it's happened several times and always explainable upon return and inspection, and you look at what the damage was or the effect of whatever, and said, "Okay, that's within a margin of acceptable damage or consequence and so therefore not safety- or flight-threatening."

NARRATOR: Just three months earlier, during the launch of the shuttle Atlantis, a chunk of foam had gouged a small crater in one of its booster rockets, missing a critical electronics box by inches. NASA ordered an investigation, but it was delayed. The launch of Columbia could not wait. Its science mission had to be completed, to make way for the construction of the international space station.

WILLIAM HARWOOD (CBS News Space Analyst): The mission wasn't a very sexy mission, as missions go. It was a very mixed bag sort of mission. It was a bunch of experiments that had been, kind of, delayed because they never were fit in somewhere else, and, finally, they all got thrown together on this flight.

NARRATOR: Eighty-one seconds after liftoff, a large piece of foam insulation broke off the main tank and hit the orbiter on its left wing. Engineers had seen the impact while reviewing film on the second day of the flight, but mission managers decided the impact did not pose a safety problem.

The astronauts successfully completed their scientific work, and the flight was otherwise uneventful.

NARRATOR: On February 1, 2003, NASA personnel, families and reporters gathered near the landing strip at the Kennedy Space Center to welcome home the Columbia astronauts.

JON CLARK: We were at the landing site. And we wait there, on the bleachers, with the other families and family members, and there's not nearly as much ceremony with landing as there is for launch.

WILLIAM HARWOOD: On that morning...it was a Saturday morning, and we were not going to broadcast the landing, live. But I was there on standby, like I always am. And I was writing my landing story.

NARRATOR: After more than a hundred successful landings, the ground control team in Houston was thoroughly familiar with re-entry procedures. But less than a minute after Columbia crossed over the California coast, the first alarm sounded in mission control.

MAINTENANCE, MECHANICAL, ARM AND CREW SYSTEMS OFFICER (Archival Footage): Flight desk.

HOUSTON CONTROL (Archival Footage): Go ahead, MMACS.

MMACS OFFICER (Archival Footage): F.Y.I., I've just lost four separate temperature transducers on the left side of the vehicle.

HOUSTON CONTROL ROOM (Archival Footage): Are you telling me you lost them all at exactly the same time?

MMACS OFFICER (Archival Footage): No, not exactly. They were within probably four or five seconds of each other.

HOUSTON CONTROL ROOM (Archival Footage): Okay. Where is that instrumentation located?

MMACS OFFICER (Archival Footage): All four of them are located on the aft part of the left wing.

NARRATOR: A few minutes later, another faulty reading.

MMACS OFFICER (Archival Footage): I've lost tire pressure on left outboard and left inboard, both tires.

HOUSTON CONTROL ROOM (Archival Footage): Columbia, Houston. We see your tire pressure messages, and we did not copy your last.

SPACE SHUTTLE (Archival Footage): Roger, uh, but...

NARRATOR: That was the last message transmitted by Columbia.

LEROY CAIN (STS-107 Flight Director): At some point, several minutes into it, the thought of the debris that we had seen from the foam and the discussions that we had on orbit came into my mind because it was the left wing. And that gave me kind of a sinking feeling, and...but again, fortunately, your training kind of kicks in, and it forces you to not get zeroed in on any one specific area until you have good reason to do that. But it did sort of cross my mind that this could be a very bad situation if, in fact, we had a breach in that wing. And that's what was causing all this, all this to happen.

HOUSTON CONTROL ROOM (Archival Footage): Columbia, Houston. U.H.F. comm check.

JON CLARK: The clock's ticking down. They'd lost comm. Now that wouldn't...if you're going to bail out you could...you'd still have some communication. So that made me think, "Well, something...maybe it's more serious than just the loss of a landing gear system."

LEROY CAIN: A few minutes into the loss of comm, I knew we were coming close to where we should have been picking up Myla, the ground station. And so I asked the flight dynamics officer, you know, "When do you expect tracking?" And when he told me, "We should have had it about a minute ago," that was really bad news.

STORY MUSGRAVE: The weather was beautiful, they had had no technical problems, so I went out there expecting it will come by.

JON CLARK: And then all the cell phones started going off, with the astronauts that escort us, the family escorts. And then, boom, all of a sudden, we're rushed to vehicles.

STORY MUSGRAVE: And it never came.

WILLIAM HARWOOD: We have...a friend of ours, he lives in Dallas, Texas, a young kid, a teenager who was out watching the landing that morning because the track of the orbiter was going to go right over Dallas. And just after I'd sent my friend a note saying, "I don't know about you, but I'm getting nervous," he sent me a note from his friend in Dallas—instant messaging once again—and this kid's note was, "OMG (oh my god). The shuttle broke up. The shuttle has broken up."

NARRATOR: While flight director Leroy Cain was struggling to understand the situation, shuttle enthusiasts, watching the sky in Texas, could see exactly what was going on.

LEROY CAIN (Archival Footage): T. C. flight, T. C. flight. Lock the doors.

HOUSTON CONTROL (Archival Footage): Copy.

LEROY CAIN: No phone calls offsite, outside of this room. Our discussions are on these loops, on the recorded loops only. No data, no phone calls, no transmissions anywhere into or out, okay?

NARRATOR: The unthinkable had happened again, but Leroy Cain knew what he had to do.

LEROY CAIN: At that point in time, of course, I knew that we had lost the vehicle. I knew that the vehicle had broken up over Texas. So you go into a mode of making sure we capture all of our notes and all of our data and all of the screens and, and the configuration. And all of that is very well laid out in a checklist that has to do with these kind of contingencies. You, when you're training, you hope that you never have to use it. And on that day, we had to use it. And that's the procedure that we were getting into when we asked the ground controllers to lock the doors.

NARRATOR: Again, seven astronauts had perished, prompting another soul-searching investigation.

SEAN O'KEEFE: Ironically, in November of 2002, we actually conducted an exercise where we activated a board. We had listed all the members of the board who would be involved in an investigation and so on. And the day this happened, at about 9:30, that plan was activated.

NARRATOR: The Columbia Accident Investigation Board was headed by retired Navy admiral Hal Gehman. Scott Hubbard was the designated NASA representative. At the time he was the head of NASA's Ames Research Center.

SCOTT HUBBARD: The first month of the board was very much devoted to the recovery of the debris, the operations in Texas, these 20,000 people that were out there searching for what they could recover, both from the astronauts remains, as well as the shuttle, itself.

SEAN O'KEEFE: The debris was spread over a 200-mile segment from West Louisiana all the way to Dallas. And I mean it covered an area 200 miles long and about 10 miles wide.

SCOTT HUBBARD: And this was a very emotional time. I remember standing there, and that got to you, you know, you really felt a sense of loss, you know, these people had died in this tragedy.

Initially, we were very skeptical that the debris could tell us anything, but over time, as we looked at what places were burned, what pieces had a sharp edge because it had been through a very hot environment, I think the debris is telling us something.

NARRATOR: Although only about 40 percent of the actual orbiter was recovered, the condition and location of the pieces that were found led investigators to the wing that had been hit by foam. The accident board also studied footage of the foam impact 81 seconds after liftoff.

STORY MUSGRAVE: ...a hard hit in a football game, when you see someone, "Bang, oh!" You know, you kind of cringe because you have the empathy to know, "If that had been me, I'd have been hurt."

NARRATOR: This was the impact that engineers had noticed while reviewing images of the launch on the second day of the flight.

SCOTT HUBBARD: They knew, on the second day of the mission, that a piece of foam hit the wing leading edge, that it probably weighed...probably was the size of a small briefcase and probably weighed two pounds.

NARRATOR: The foam came from the base of a ramp that connected the external tank to the orbiter. It was foam from the same area that had endangered the shuttle Atlantis three months earlier. But that was all the engineers knew.

RODNEY ROCHA (STS-107 Division Chief Engineer): There was also an uncertainty about where the foam had hit. Was it the front of the wing, where the carbon panels were? Was it the tile acreage on the bottom? Or was it the landing gear door?

NARRATOR: Rodney Rocha was the division chief engineer for the Columbia flight. After seeing the video, he and other engineers had wanted NASA to obtain additional images to check the shuttle for damage.

RODNEY ROCHA: We have ground-based telescopes, and—"we" meaning the military does—and satellites up there. I'm aware that we used some of that in the early shuttle flights in the '80s. Some of that was used and was classified. So that's why, in my e-mail request, I said let's ask, and I actually said, "let's beg" for outside agency assistance. And I explained why. "We're highly uncertain about this problem. There's too many possibilities here. Some of them are very bad."

NARRATOR: On the fifth day of the mission, the debris assessment team, including Rocha, made an initial estimate that the fragment of foam was several hundred times larger than average.

WILLIAM HARWOOD: This was a big foam strike. It was the biggest one anyone had ever seen. It was a large piece that clearly came off, hit the underside of the wing, and no one knew where it hit.

NARRATOR: On the eighth day of the mission, the crew was informed of the incident as part of a routine e-mail. It said, in part, "This item is not even worth mentioning, other than wanting to make sure that you are not surprised by it in a question from a reporter. There is absolutely no concern for entry."

But Rodney Rocha was very concerned. And his requests for additional images were denied.

RODNEY ROCHA: We felt were in a topsy-turvy world. It's like someone saying, "I want you to tell me how bad that car accident is that you just heard out the window." And you..."I want you to tell me if we need an ambulance or not." And you say, "Well, I'll go look out the window," and you say...and someone says, "No, you may not look out the window. You do your analysis first, and you tell me if you need to call an ambulance first. It's a lot of trouble to call an ambulance. You tell me, you do an analysis first. You tell me an answer first." How can you possibly get out of that, that kind of uncertainty? It's impossible.

NARRATOR: On day nine, the mission managers met to resolve the foam issue. Their meeting was recorded.

WOMAN (Archival Audio): Okay, good morning, and welcome to the N.R.T. Okay, go ahead now, Scott.

MAN (Archival Audio): Okay, we've received the data from the systems integration guys. The analysis is not complete, but I'm, kind of, just jumping to the conclusion of all that, but we do not see any kind of, you know, safety of flight issue here, yet, in anything that we've looked at.

WOMAN (Archival Audio): No safety of flight, no issue for this mission, nothing that you're going to do different? That may be a turnaround? That's it? Alright, any questions on that?

WILLIAM HARWOOD: So this is foam, like the stuff on the external fuel tank, and it doesn't weigh a thing. I mean, this is almost weightless. And I think that, even among very smart engineers, there is a very human perceptual quality here that, "This doesn't weigh anything; so what if it hits you?"

NARRATOR: But after the accident, the investigation team took a harder look at the foam.

SCOTT HUBBARD: I said, "Doug, you won a Nobel Prize in Physics, you know, let's just do some high school arithmetic here, okay?

DOUGLAS OSHEROFF: We knew that the time, the transit time for the foam was, I think, .16 seconds.

SCOTT HUBBARD: Here's how fast it's going, okay? And we do some simple high school physics.

DOUGLAS OSHEROFF: The distance is equal to one-half-A.T.-squared, and you calculate the acceleration, you calculate the velocity.

SCOTT HUBBARD: It would impart a force of roughly three thousand pounds, you know, roughly a ton of force.

DOUGLAS OSHEROFF: But, in fact, NASA management claimed they didn't understand this. I mean this was very trivial stuff.

WILLIAM HARWOOD: And the analogy they always used was as if a Styrofoam cooler on a car in front of you, in a pick-up truck or something, blows out on the road and hits your windshield when you're going 70 miles an hour, what happens? And the answer is nothing happens. The Styrofoam breaks up in little pieces and falls off, it doesn't break your windshield.

SCOTT HUBBARD: We had a picture of the crew in every single room at the Columbia board. And we would always remind ourselves, "We're here, in part, at least, because of these people who died and what we owe to them and their families to figure out what happened."

ROGER LAUNIUS: Scott Hubbard, who was a member of the Accident Investigation Board, ran around all over NASA—and at that time he was the center director of the Ames Research Center—he ran all over NASA saying, "The foam did it. The foam did it. The foam did it," telling anybody who would listen, "Here is how. This is why. And the foam did it." And yeah, there was resistance to that. People didn't want to believe it. And until they demonstrated it, they didn't really believe it.

WILLIAM HARWOOD: NASA has this terrible expression called "It's in-family." In other words, "We've seen this before. It never did anything before. I don't think it'll do anything now."

NARRATOR: The investigation board assembled parts from other shuttles to devise a conclusive test. Everyone agreed, the leading edge of the wing was probably where the foam had struck. The board went ahead and constructed a life-sized model of the wing, with plans to test-fire fragments of foam at it, at different speeds and angles.

JOHN SCHWARTZ (The New York Times): I had engineers at NASA, very smart people, saying, you know, "They're going to fire a piece of foam at that wing, and it's going to bounce off. Foam can't do that." And these were not dumb people. These were intelligent people. But the physics of it was counter-intuitive.

SCOTT HUBBARD: It was a very powerful emotional moment, because we had done all this work for months to get everything ready to do this test. It was very hot, down there in San Antonio. It's, you know, a hundred degrees every day, and the place is full of snakes and stuff; it's all outdoors. And so I was standing there, with all these other officials, and two astronauts came up to stand right behind me. And they counted it down to the firing of the gun, and the gun, you know, "boom!" And we looked, and all of a sudden there was this giant hole in the wing leading edge and at first I went, "Yes!" And then I went, "Oh." Because part of me said "Okay, we've demonstrated it," but then part of me said, "Oh, this is how these people actually died."

There was one engineer, who came from Johnson Space Center, and she actually had tears in her eyes, because she, you know, didn't want to believe that the foam did it. But this was so dramatic a test, it showed that there was just no doubt.

STORY MUSGRAVE: After the accident the people said, "Well, during launch, a problem happened and it was all over for the crew then." It was not all over then. And you don't look to spy satellites or anything like that; it's "Do a space walk; it's just do a walk." And there was no hearing that language from the NASA, "Just do a walk."

RODNEY ROCHA: Assuming we would've seen the hole in the wing, we would've known...we would've known it was a carbon panel. We would've known which panel. We would've known the size. And then my colleagues and the contractors would've come up with some kind of response to that.

Hundreds, if not thousands, of people would've put their brains together and started thinking pretty smartly about how to respond to this. And we would've told the crew, given them some strong suggestions about what to do. And I think they would've done it.

Now, would that have worked? I'll...I guess I'll never know.

DOUGLAS OSHEROFF: We actually tasked NASA to put together a team of engineers to imagine that it's day five of the flight and they've just learned that there's a big hole in the left wing of Columbia. So what could they do about it?

One group said, "Well, Atlantis is close enough to being ready to launch that we, basically, kind of, wind down, as far as possible, all activities on the shuttle, and without sacrificing any of the normal safety procedures for preparing Atlantis for launch," that they could get the Atlantis up. It would have to, sort of, essentially, park next to Columbia, and then they would transfer the astronauts, one at a time. And then, in fact, you could bring those guys back.

NARRATOR: The board found that, although the odds of success were low, the rescue scenario might have worked. But the Columbia crew was unaware of the danger until the very end, as confirmed by one last piece of evidence.

WILLIAM HARWOOD: During the search for wreckage, I interviewed a guy who one day was...they'd gone out, because people were calling in, you know, "There's debris here. There's debris there."

And these guys, they'd sent somebody out to check out everything. And these two guys went to this house, and it turns out whatever it was it was nothing or something. And they walk back to their truck, and they're on the side of a country road in Texas. And they're standing by the back of this pickup truck, and one of the guys looked down on the side of the road, between the asphalt, where there's a little gravel and then grass. He sees this cassette.

And it looked like a, like an audiocassette, you know? Likeit's Texas—it'd be country and western music or something. But it was a little thicker than an audiotape. And he picked it up, and he said, "You know, that's interesting."

It didn't look like an audiotape. And he put it in the bag just for the heck of it, and they turned it in. And that was the tape from the flight crew's own video camera that Laurel Clark was using during re-entry to film the crew.

COLUMBIA CREW VIDEOTAPE: Two up, looking good.

WILLIAM HARWOOD: And that is the only video you have of the crew on re-entry day. And I remember the day that that video was released, and NASA put it up on the satellite, and we all watched it. That's what got me, because you're watching the crew on the flight deck, and you knew what had happened to them, to the second. And you knew that when this tape ended—because it was burnt, you know, from falling back to Earth, the ends not on there—when the tape stops, they're all alive and healthy, and they're a few minutes away from death, from violent death.

COLUMBIA CREW VIDEOTAPE: I'm trying to see if I can get another of that window view yet.

This is amazing...is really getting...it's really bright out there.

Yep. Yeah, you definitely don't want to be outside now.

What, like you did before?

NARRATOR: As the orbiter descended back towards Earth, the heat produced by re-entry brought the shell of the vehicle up to the standard 1,600 degrees Celsius. Shuttle-watchers in California witnessed the fall of the first fragments of debris.

Superheated gas had already penetrated the left wing, through the hole made by the foam impact. It slowly began to destroy the vessel from inside out.

The shuttle's autopilot fired thrusters to compensate for the drag on the damaged wing, but it was too much. Columbia broke apart over Texas, 16 minutes before it was expected to land in Florida.

SCOTT HUBBARD: At the beginning of the Columbia board, we didn't know much about the cause. We didn't know much about what the Shuttle program was supposed to be doing. Later on, we came to understand that there was an unspoken contract between the crew and the people on the ground that the crew will do all they can to make the mission successful, and the people on the ground—the engineers and the managers—will do everything they can to keep the crew safe and to bring them home safely.

And I think we developed a sense that maybe the astronauts felt like the people on the ground hadn't done everything they could have done, and some things fell through the cracks.

NARRATOR: When members of the Columbia Accident Investigation Board released their findings, in 2003, they were unflinching in their critique of what they called NASA's "broken" safety culture. They concluded that the management team operated outside of the rules, even as it held its engineers to a stifling protocol. But the board members didn't limit their judgment to dysfunctional management.

SCOTT HUBBARD: I remember making the comment, and I wasn't alone, that if your going to risk people's lives, that ought to be for a purpose. And out of these discussions, grew what's in the last part of the report, which is to say, "Look, NASA needs to figure out why it's doing this." And the space science program is well established with the scientific community, but why are we doing the human spaceflight piece? And it's my belief—I believe this is accurate—that that report lead directly to the "Vision for Space Exploration."

NARRATOR: The "Vision for Space Exploration" was announced by President Bush in January, 2004. It directed NASA to build new spacecraft that could take astronauts beyond the confines of low Earth orbit, and it mandated the retirement of the shuttle.

JACK SCHMITT: The decision has been made, and I think it's a correct one, to go back to a heavy-lift launch foundation, because it offers us so many other things we can do. The shuttle can't go to the Moon, and they can't go to Mars. And it really does not give us the capability to build craft and put craft into space that can do those two things.

NARRATOR: To speed the transition, the new suite of space vehicles, called Constellation, will be based largely on proven technology.

MICHAEL GRIFFIN: Constellation, actually, in its architecture, utilizes a number of pieces of both Apollo technology, from several decades ago, and shuttle technology.

NARRATOR: Solid rocket booster and fuel tank design from the shuttle, engine and rocket design based on the Apollo moon rockets. But unlike Apollo or the shuttle, the new system launches crew and cargo separately.

MICHAEL GRIFFIN: Every time the shuttle lifts off with a cargo and a crew, it carries 120 tons of payload to orbit, of which 20 tons is cargo and all of the rest is support elements for the crew and the cargo itself. So, in that sense, unless your objective is to put people in orbit, most of the shuttle capability is wasted. On the other hand, if your objective is to put people in orbit, then what are you doing with this 15-foot diameter, 60-foot long payload bay in the back?

NARRATOR: In addition to their new capabilities, the spacecraft eliminates the major safety flaws the Columbia board identified with the shuttle.

SCOTT HUBBARD: You can get an immediate factor-of-10 improvement in safety just by moving the vehicle from the side to the top, like in Apollo.

NARRATOR: A capsule on top of the rocket cannot be hit by the kind of debris that doomed Columbia. It also allows for a crew evacuation system that could have saved the Challenger astronauts.

MICHAEL GRIFFIN: We are returning to a design which has a crew escape system. There will be an escape tower, an escape rocket on top of the Orion crew vehicle that, in the event of an emergency, will allow the crew a second chance.

NARRATOR: Before being retired, the shuttle will continue to be used to finish construction of the international space station. Cameras on the space station, as well as on the shuttle's robotic arm, are being used to detect damage to the orbiter. New techniques have been developed to limit the amount of foam shedding and even repair damaged tiles in space. After more than two years in dry dock, the shuttle resumed flight on July 26, 2005, with mission STS-114. Aboard the Discovery, once in orbit, Commander Eileen Collins spoke.

EILEEN COLLINS (NASA Astronaut Archival Audio): As the crew looks back at our beautiful planet and then outward towards the unknown of space, we feel the importance today—more than any time—of space exploration, to all those who are living on Earth. Our flight is the next flight of many in the human exploration of the universe.

And finally, we reflect on the last shuttle mission, the great ship Columbia and her inspiring crew: Rick, Willie, Mike, K.C., Dave, Laurel and Ilan. We miss them and we are continuing their mission. God bless them tonight and God bless their families. Good night.

NASA CONTROL ROOM (Archival Audio): Thank you, Eileen and the STS-114 crew. We miss them too.

NARRATOR: Could the astronauts have been saved? On NOVA's Space Shuttle Disaster Web site, examine how a high-risk scenario to rescue the crew might have unfolded. Find it, on PBS.org.

Major funding for NOVA is provided by the following:

One of the biggest challenges for geoscientists is trying to find the oil and gas that's hidden below the surface of the Earth. Lately, our researchers have developed a new technology called R3M. You can't see it, and you can't feel it, but the Earth has electromagnetic waves, and if you build the right kinds of listening devices, you can make sense of those waves. So using those tools, we can supply more of the energy that we need with less of an impact on the environment.

And David H. Koch. And...

Discover new knowledge: HHMI.

And by the Corporation for Public Broadcasting, and by contributions to your PBS station from viewers like you. Thank you.

To order this NOVA program, for $24.95 plus shipping and handling, call WGBH Boston video at 1-800-255-9424.

NOVA is a production of WGBH Boston.