Skip to content
Published September 20, 2019

“Building rockets is hard.”

These are the first words in part one of the Columbia Accident Investigation Board’s report. They called the Space Shuttle, “One of the most complex machines ever devised.” It had over 2.5 million parts, 230 miles of wire, weighed 4.5 million pounds at launch, and accelerated to 17,500 miles an hour in just over eight minutes.

And, in 2003, the Space Shuttle Columbia carried seven people out into space; but failed to bring them home safely.

When the space shuttle Columbia launched on its first flight on the 12th of April 1981 – the 20th anniversary of Yuri Gagarin’s trailblazing flight – it carried with it the promise of routine space flight. It was the first launch of the Space Shuttle program, whose goal, according to President Richard Nixon in 1972, was “to help transform the space frontier of the 1970’s into familiar territory, easily accessible for human endeavor in the 1980’s and ’90’s.”

The Challenger disaster in 1986 had raised issues with the shuttle program’s management, and resulted in the withdrawal of both commercial satellites and Department of Defense payloads; however, the program survived and was instrumental in the construction of the International Space Station.

By 2003, the shuttle program was already beyond its initial 15-year lifespan; Columbia itself was 22 years old, and had undergone many changes since its first flight, but it was still going.

Mission STS-107 was the 113th space shuttle flight, and the 28th flight for Columbia. The schedule called for sixteen days in space “dedicated to a mixed complement of competitively selected and commercially sponsored research in the space, life and physical sciences.”

These studies involved over 70 scientists and ranged from research into astronaut health, growing cell cultures in weightlessness and growing protein crystals for potential cancer therapies, to testing water recycling technology for the International Space Station and examining the physics of combustion, soot production and fire quenching processes in microgravity.

The crew of STS-107 pose for an official portrait. Source: NASA

Leading the crew was Commander Rick Husband, a 45 year old former fighter and test pilot with the US Air Force and father of two. With one previous space flight behind him, piloting STS-96 in 1999, he had already spent more than 235 hours in space. According to his wife Evelyn, “The first time Rick figured out that he wanted to be an astronaut was when he was four years old, so this was a goal that he set at a very young age.”

The pilot was William McCool, 41 years old, an American naval officer, aviator, test pilot and aeronautical engineer, avid runner and married father of three sons making his first space flight. During his flight on Columbia, he read a special poem to his wife and former high school sweetheart Lani: “I’ve witnessed the beauty of Earth from space, far, far above. What a treasure it is to behold. But I would trade this view for your embrace, my sweet love, for only you enrapture my soul.”

Payload Commander and Mission Specialist 3, Michael Anderson, was 43, a United States Air Force officer who was in charge of the scientific experiments on board. He had previously flown on STS-89, logging over 211 hours in space, and was married with two daughters. His mother Barbara said he wasn’t a public person, and when he applied to the space program he hadn’t even told them first. “He was focused, and when he decided he was going to do something, he didn’t really just talk about it. He did it.”

Mission Specialist Kalpana Chawla became the first woman of Indian descent to go into space when she first flew on Columbia in 1997, and had logged more than 376 hours in space. 40 years old, she had been married for twenty years. She was a source of great national pride to India, and had recently appeared on the cover of India Today, although she herself said she didn’t feel Indian in space; “When you look at the stars and the galaxy, you feel that you are not just from any particular piece of land, but from the solar system.”

Mission Specialist David Brown, a US Navy Captain, aviator and flight surgeon on his first mission into space at the age of 46. He was the only unmarried member of Columbia’s crew. He said that when he was a child he thought astronauts were movie stars. “And I just thought I was kind of a normal kid…  while I would’ve said, ‘Hey, this is like the coolest thing you could possibly do,’ it really wasn’t something that I ever thought that I would end up doing.”

Mission Specialist Laurel Clark was a medical doctor, US Navy Captain and flight surgeon, also on her first space mission aged 41. She was married to a fellow NASA flight surgeon and had one son. She had actually first applied to the space program when she was six months pregnant. Husband Jon said, “So she goes to the application process six months pregnant, she’s huge. I was just giving her all kinds of grief, I said there’s no way you’re ever going to get picked like that, you just don’t look the part. She stuck it out, she had a great time on her interview, and then the next cycle she applied and she got in.

Finally, Payload Specialist Ilan Ramon was a former fighter pilot and Colonel in the Israeli Airforce. He was the son of a Holocaust survivor, and the first and only Israeli astronaut. At 48, he was the oldest member of the crew, with a wife and four children. His presence was part of an international co-operation effort, and he said, “There is no better place to emphasize the unity of people in the world than flying in space. We are all the same people, we are all human beings, and I believe that most of us, almost all of us, are good people.”

At about 7:30 on the morning of January 16th, 2003, the crew were driven from their quarters at Kennedy Space Center to the Launch Complex. By 8:45, they were all aboard. The rockets fired at 10:39, and Columbia was on her way into space. Just to note, wherever times are given, they’re in Eastern Standard Time, as used by the Columbia Accident Investigation Board Report.

For 16 days, the crew were kept busy with their experiments, running educational events with schools, and maintaining the shuttle in orbit. On Flight Day 17, the 1st of February 2003, they packed everything up and got ready to come home on schedule.

The crew pose together in-flight. This image was found on an unprocessed film recovered after the disaster. Source: NASA

A video camera on the upper flight deck recorded Husband, McCool, Chawla and Clark as they prepared for re-entry into Earth’s atmosphere. Anderson, Brown and Ramon were on the mid-flight deck, so were not seen on this video. The tape shows the crew on the flight deck starting to prepare for re-entry; putting their gloves on, organising their trash, and bantering with each other. 

Back on Earth, the re-entry was broadcast on NASA’s television service, NASA Select. They hit the Entry Interface – an arbitrarily defined point at 400,000 feet – at 8:44:09am, and the commentator announced:

“Columbia’s altitude now 71 statute miles as it enters Earth’s atmosphere above the Pacific Ocean en route to the Kennedy Space Center, its speed 17,000 miles per hour.

Columbia with wings level and nose angled up at about 40 degrees to control heating as it descends into the atmosphere. It’s altitude now 68 miles. As Columbia descends into the atmosphere and approaches the continental United States it’ll perform the first in a series of four banks it performs as it approaches the Kennedy Space Center… Those designed to dissipate speed for the shuttle as it descends into the atmosphere toward landing.”

At around 8:46, on the crew’s video, they commented on the bright plasma seen outside the window.

“This is amazing, it’s really getting fairly bright out there,” said pilot McCool.

Husband replied, “Yeah, you definitely don’t want to be outside now.”

Chawla commented, “What, like we did before?”

The crew laughed, and continued to banter for a couple more minutes before focusing on their re-entry tasks.

The NASA commentator continued:

“Columbia’s course toward Florida will take it across the continental United States, crossing the California coast above the San Francisco bay area and continuing across Sacramento, California, providing a spectacular view for persons in that area of Columbia’s descent through the atmosphere… It’ll be visible as well through much of the United States’ southwest above southern Nevada and northern Arizona and central New Mexico as it continues its descent through the atmosphere, trailing a plasma trail left as it heats the atmosphere around it during its descent.”

Shortly after this, Mission Control started to see unusual readings from the orbiter’s systems. At 8:54, MMACS Officer Jeff Kling reported this to Flight Director Leroy Cain.

“KLING: FYI, I’ve just lost four separate temperature transducers on the left side of the vehicle, hydraulic return temperatures.Two of them on system one and one in each of systems 2 and 3. CAIN: Four hyd return temps? KLING: To the left outboard and left inboard elevons.”

They continued to discuss and monitor those readings. Approximately four minutes later, they received two transmissions from Rick Husband on the Columbia, but they weren’t clear.

HUSBAND: And Hous… KLING: Flight, MMACS. CAIN: Go. KLING: We just lost tire pressure on left outboard and left inboard, both tires. CAPCOM Charlie HOBAUGH: And Columbia, Houston, we see your tire pressure messages and we did not copy your last. CAIN: Copy. Is it instrumentation, MMACS? KLING: Flight, MMACS, those are also off, off-scale. 

HUSBAND: Roger, ah bu… 

This last message was cut off; according to NASA reports it was received at 8:59:32 am. It would be the last transmission received from the space shuttle Columbia.

A view of Mission Control, at around the time that contact with Columbia was lost. Source: NASA.

Mission Control continued to talk about the abnormal readings, as further readings dropped out. At this point of the flight, communication issues were not unexpected, but they soon expressed concern that they were not receiving anything at all. 

Just after nine am, the NASA commentary resumed. 

“Columbia out of communications at present with mission control as it continues its course toward Florida…Fourteen minutes to touchdown for Columbia at the Kennedy Space Center. Flight controllers are continuing to stand by to regain communications with the spacecraft…”

Columbia was expected at this point to come in range of their UHF communications system. Charlie Hobaugh, CAPCOM, started to call them.

“Columbia, Houston, comm check.”

“Columbia, Houston, UHF comm check.”

He would repeat that message another four times, without response.

Flight Director LeRoy Cain said later that he knew hope was lost when he was told at 9:04 that the long-range radars at Cape Canaveral were not tracking anything. “That was the absolute black-and-white end. If the radar is looking and there’s nothing coming over the horizon, the vehicle is not there.”

However, it was a few minutes later that he received confirmation. At approximately 9:12 am, he could be seen talking to Mission operations representative Phil Engelauf and astronaut Ellen Ochoa, who were seated behind him.

After a visibly tense conversation Cain turned back towards his desk. Although his voice was calm, his body language clearly showed his distress.

CAIN: GC, Flight. GC, Flight. Ground Control: Flight, GC. CAIN: Lock the doors.

This command lets everyone in Mission Control know that the worst has happened. The order to “lock the doors” tells them that nobody is to leave; they have to ensure that everything on their computers is backed up, and to write up their logbook notes ready for the inevitable investigation.

Outside of Mission Control, many had actually seen what happened. 

Shuttle enthusiast Chris Valentine had gone out with family members to watch the shuttle’s re-entry from an observation point 35 miles north of Flagstaff, Arizona. Their footage shows the orbiter as a bright point of light – and then clearly shows a second point of light break off from the first and fall behind. One of them called out, “Look at the chunks coming off of it! What the heck is that?” 

As others made similar films, news of the shuttle’s break-up quickly made it onto TV. 

The conversation between Engelauf, Ochoa and Cain was actually prompted by a call in to Mission Control from off-duty flight director Bryan Austin; he had seen the break-up on TV, and phoned in to tell his colleagues the bad news.

The first signs of trouble had been seen by observers on the ground at 8:53 am; a brightening of the light around the orbiter, and a noticeable streak in its trail. Four similar events followed in the next 23 seconds; eighteen over four minutes. 

The first piece of debris shed that was actually recovered was a piece of Thermal Protection System tile. It was found in a field in Texas, northwest of Lubbock. It’s estimated that this piece was shed about one minute before Husband’s last call. 

A recovered piece of Thermal Protection System tile from Columbia. Source: NASA

Across Texas, emergency dispatchers received an influx of calls caused by the shuttle’s breakup. Witnesses had been startled by sonic booms and the descent of smoking debris. One woman, driving near Lufkin, almost lost control of her car when a piece of Columbia’s debris hit her windshield.

NASA quickly activated their Mishap Investigation Team, and worked with local, state and federal agencies to retrieve whatever debris could be found. Even the smallest piece could hold the key to the investigation. Warnings were issued to the public to avoid touching any debris they might encounter – Columbia’s propellants and other substances on board could be highly hazardous – and President Bush declared East Texas a federal disaster area.

Volunteers systematically searching a field in the debris area. Source: NASA (On Flickr)

The debris field spread over 2,000 square miles; over 700,000 acres were searched by foot, 60 divers searched lakes in the debris field area, and an air search effort involved 37 helicopters and seven fixed-wing aircraft. Sadly, during this process a helicopter crash caused by a mechanical failure took the lives of two searchers, Jules “Buzz” Mier and Charles Krenek, and injured two more.

Over 84,000 pieces of debris from the Columbia shuttle were recovered; about 38% of the orbiter’s dry weight. Amongst them was a video cassette tape; the film that the crew had been recording during re-entry. The footage retrieved went up to about eleven minutes before the final breakup. 

Even before the debris had been recovered, investigators had a pretty good clue as to what could have caused the catastrophe. It was something that had happened more than two weeks before.

81.9 seconds after Columbia launched, a piece of debris detached from the External Tank and struck the orbiter’s wing. It hadn’t been noticed immediately, but it was seen the following day when footage of the launch was reviewed.

The External Tank was coated with different types of insulating foam to ensure that the liquid oxygen and hydrogen inside stayed at the right temperatures; -297° fahrenheit and -423° fahrenheit respectively, without allowing frost or ice to form on the outside. It was a piece of this foam, approximately 21-27 inches long and 12-18 inches wide, which had come off and struck the orbiter.

This was not, in itself, an especially abnormal event; foam strikes had occurred during previous launches, and all of those missions had come safely back to Earth.

However, this particular piece of foam appeared to have hit the left wing, and the first abnormal readings during re-entry were from sensors in the left wing. Surely this could not be mere coincidence?

The debris recovery operation also supported this theory. Whilst a considerable amount of debris had been recovered from the right wing, they had barely found anything from behind the wheel well of the left wing. This indicated that the left wing had been damaged earlier than the right – while the orbiter was still at such a high altitude that the pieces just didn’t make it down to the ground.

It seemed unlikely to some that a foam strike could be capable of inflicting sufficient damage to bring the shuttle down. It was, after all, extremely light material, and the piece in question was estimated to weigh about the same as a basketball.

NASA’s shuttle program manager Ron Dittemore publicly dismissed the foam strike theory at a news conference on the 5th of February.

“Right now it just does not make sense to us that a piece of debris could be the root cause of the loss of Columbia and its crew. There’s got to be another reason.”

Shuttle Program Manager Ron Dittemore, right, pictured announcing his intention to step aside. Source: NASA/Renee Bouchard

However, as more evidence was retrieved, it looked more and more likely to the investigators. They just had to prove it.

They set up a practical experiment to establish what damage a foam strike could actually cause to the materials on the orbiter’s wing, using a so-called “chicken gun”. This is a large diameter air-powered cannon used by aircraft developers to assess the risk of bird strikes; the CAIB just had to modify the chicken gun to fire foam pieces instead of dead chickens. 

Then, they needed comparable orbiter parts to fire at. This caused some issues, because there weren’t many comparable parts available, and the brittle Reinforced Carbon Carbon (or RCC) tiles at the leading edge would only be usable for one test. They conducted tests first on fibreglass panels, then focused on the RCC tiles which were most likely to be the point of impact; panels 6 to 9.

The investigators performed their first test on panel 6; this produced a crack, but it would probably not have been sufficient damage to bring down the orbiter. Some commented that this meant the foam strike theory didn’t hold. NASA specialist Scott Hubbard, in charge of the tests, said later, “That type of comment told me that it was really really important to conduct an exact test so that we would show experimentally the connection between the foam and the accident, so that the engineering staff could understand that not only in their head but in their heart.”

One of the most important points of the investigation came when the Modular Auxiliary Data System recorder was retrieved. It was extremely fortunate that this recorder was found intact, because unlike the so-called “black box” on an airplane, it was not designed to survive impacts. 

They were also fortunate in the extent of the data that it provided; the other shuttles recorded only basic data, but because Columbia had been the first orbiter in space, it had been set up to record more detailed data for development purposes. Although some of those additional sensors had since failed, and not been replaced, the investigators still had a lot more information than they might otherwise have expected.

From this data, they could see that abnormal readings had actually been recorded four and a half minutes after the entry interface point – several minutes before the first readings that had been transmitted to Mission Control. These readings were consistent with a breach in the leading edge of the left wing, and allowed them to narrow down the impact point to RCC panel 8.

Investigators set up the frame for the chicken gun test. Source: NASA

They were then able to set up a second test on panel 8; with the impact point, velocity and angle as close as possible to the actual strike on Columbia, this would be the truly revelatory test. And this time, the results were very different. It left a hole in the panel measuring 16 by 17 inches, with other cracks throughout the panel. 

Hubbard said, “When that gun fired and the hole appeared there was an audible gasp from everyone that was there… I talked to one of the engineers, a young woman that had been helping us all along, and she had tears in her eyes. and she said, “so, this is what really happened isn’t it?” I said yes, this is what really happened. This is how these people died.”

Attempting to re-enter Earth’s atmosphere with a hole like this in the wing could only have been catastrophic for Columbia.

During re-entry, temperatures at the leading edge are expected to rise to an estimated 2,500 degrees Fahrenheit. Without the protection of the RCC panel, that heat would be channeled into the wing. Major internal supports inside the wing were made from aluminium alloy, which melts at 1,200 degrees Fahrenheit. 

From the progressive loss of sensor data, investigators established that over 70% of the wires bundled in the left wing burned through in less than a minute. The damage to the wing – which was now basically melting from the inside out – also affected the orbiter’s aerodynamic profile, making it harder to control. However, initially, the flight control system compensated for the abnormal motions of the orbiter, so the crew remained unaware of the problem.

The first the crew knew of the problem was at 8:58:39 – almost five minutes after the first piece of debris was observed coming off the orbiter. They would have been alerted by the Backup Flight Software monitor to a loss of pressure in the left main landing gear tires. A few seconds later, they made their penultimate call to Mission Control; their final call just 44 seconds later. 

At this point, Husband and McCool would have been busy trying to troubleshoot the problem. Whilst they undoubtedly did everything that was within their means, sadly time was running out. The shuttle’s complete breakup followed shortly; there was between 99 and 116 seconds from the time they received that alert to the point at which their cabin depressurised.

Clark and Husband in training versions of the ACES suit. Source: NASA

The crew were wearing special pressure suits, referred to as ACES – Advanced Crew Escape Suits. Such suits had been introduced in the wake of the Challenger accident. 

However, they were not fully integrated with orbiter operations. The crew could not put their visors down during re-entry, and some operations were too difficult to complete with the gloves on. The suits could not pressurise without both of these steps complete. 

At the time Columbia entered Earth’s atmosphere, one crew member was not wearing their helmet, and three were not wearing their gloves. For those who were fully equipped, there was simply not enough time between losing control and cabin depressurisation for any of them to close their visors; just 40 seconds.

A later investigation, looking purely at aspects of crew survival, established that depressurisation of the crew module would have incapacitated the crew within seconds; the effects were so severe that they could not have regained consciousness. Even if they had been able to survive this, the rotational forces on the orbiter, combined with the failure of their seatbelt shoulder straps to lock and the fact that, unconscious or already dead, they were unable to brace themselves, inflicted further lethal trauma before the crew module itself broke up.

In total, the investigation pinpointed five events in Columbia’s breakup which had lethal potential, and established that all the crew were deceased before, or by the end of, the third.

The crew survival report concluded that “The breakup of the crew module and the crew’s subsequent exposure to hypersonic entry conditions was not survivable by any currently existing capacity.”

Remains of the crew were found, and identified by DNA. This allowed the families to hold proper funerals; Ilan Ramon’s remains were returned to Israel. Out of respect for the families, the exact details of what happened to them during their descent to Earth have never been released.

Debris from the Space Shuttle Columbia. Source: NASA/Bill Ingalls

At this point, the Columbia Accident Investigation Board could have called it a day. After all, they had the answer – the foam strike damaged the wing, and led to the orbiter’s disintegration. They chose not to, and in the introduction to Part Two of their report they wrote:

“Many accident investigations do not go far enough. They identify the technical cause of the accident, and then connect it to a variant of “operator error” – the line worker who forgot to insert the bolt, the engineer who miscalculated the stress, or the manager who made the wrong decision. But this is seldom the entire issue. When the determinations of the causal chain are limited to the technical flaw and individual failure, typically the actions taken to prevent a similar event in the future are also limited: fix the technical problem and replace or retrain the individual responsible. Putting these corrections in place leads to another mistake – the belief that the problem is solved. The Board did not want to make these errors… In our view, the NASA organizational culture had as much to do with this accident as the foam.”

When the foam strike was noticed on Flight Day Two – January the 17th – a report, accompanied by a video clip showing the debris impact, was sent to the Mission Management Team, the Mission Evaluation Room, and engineers at United Space Alliance and Boeing. 

Following NASA guidelines, a Debris Assessment Team was formed; they had their first formal meeting on the 21st of January, and were instructed to assess the potential for damage. 

The highest-ranking NASA engineer on the team agreed to take a request to the Johnson Space Center Engineering Management Directorate, asking for imaging of the wing upon which they could base their analysis. That request was ultimately denied. Instead they had to use a mathematical modelling tool which hadn’t been designed specifically for that kind of impact. 

Meanwhile, an email was sent from Mission Control to the shuttle crew, dated January 23rd.

“There is one item that I would like to make you aware of… 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.

During ascent at approximately 80 seconds, photo analysis shows that some debris… came loose and subsequently impacted the orbiter left wing… The impact appears to be totally on the lower surface and no particles are seen to traverse over the upper surface of the wing. Experts have reviewed the high speed photography and there is no concern for RCC or tile damage. We have seen this same phenomenon on several other flights and there is absolutely no concern for entry.”

On Sunday January 26, Division Shuttle Chief Engineer Rodney Rocha sent an email with the results of their analysis.

“SUMMARY: Though this case predicted some higher temperatures at the outer layer of the honeycomb aluminum face sheet and subsequent debonding of the sheet, there is no predicted burn-through of the door, no breeching of the thermal and gas seals, nor is there door structural deformation or thermal warpage to open the seal to hot plasma intrusion. Though degradation of the TPS and door structure is likely (if the impact occurred here), there is no safety of flight (entry, descent, landing) issue.”

But the analytical analysis – conducted with very limited data – had not alleviated concerns for everyone at NASA.

Images used by engineers to try to assess the damage during flight. Source: NASA

Junior engineers at NASA made further requests for high resolution satellite images to be taken of the wing, so that they could actually see whether there was any damage instead of making what amounted to an educated guess. These requests were all turned down.

Admiral Hal Gehman, head of the investigation, said, “The question of a picture was brought up to management and the management actively suppressed requests for pictures. They actually went on the aggressive to stamp out any thought of pictures.”

In one instance, a NASA employee requested assistance from United States Space Command. The request was quickly cancelled because it hadn’t gone through correct channels. An email from Roger Simpson, liaison to Space Command, read in part;

“Let me assure you that, as of yesterday afternoon, the shuttle was in excellent shape, mission objectives were being performed and that there were no major system problems identified. The request that you received was based on a piece of debris, most likely ice or insulation from the E.T., that came off shortly after launch and hit the underside of the vehicle. Even though this is not a common occurrence it is something that has happened before and is not considered a major problem.”

Another email released shows an exchange between Robert H. Daugherty, an engineer at the Langley Research Center, and Carlisle C. Campbell, an expert on the shuttle’s mechanical systems. Daugherty wrote:

“Seems to me that the benefit of an E.V.A. to go look at carnage has more pros than cons. Can’t imagine that an astronaut (even on a crappy tether arrangement) would cause MORE damage than he is going out to look for!”

Thermal analysis engineer John Kowal, at the Johnson Space Center, was concerned that the analysis by the debris impact team didn’t take all possibilities into account. 

“If the gouge were to occur in a location where it passes over the thermal barrier on the perimeter of the door, the statement that there is ”no breeching of the thermal and gas seals” would not be valid. I think this point should be clarified; otherwise, the note sent out this morning gives a false sense of security.

Jeffery Kling, a mission controller who would be the MMACS officer during Columbia’s re-entry, discussed the potential damage in Kowal’s theory in another email exchange with fellow mission controller R Kevin Mcluney and NASA contractor William Anderson.

Kling wrote:

“If there was hot plasma sneaking into the wheel wells, we would see increases in our landing gear temperatures and likely our tire pressures. If we actually saw our instrumentation in the wheel wells disappear [during] entry then I suspect that the gear will not deploy anyway because the wires that control the pyros and all the hydraulic valves would burn up, too. Ultimately our recommendation in that case is going to be to set up for a bailout (assuming the wing doesn’t burn off before we can get the crew out). The rest of the cases are great big what-ifs.”

They discussed landing and bailout options that would be available to the crew with damaged landing gear – with Anderson writing at one point, “If the wing is off, or has a big hole in it, you’re not going to make the runway, and the gear question is moot.”

However, he finished by writing, “Anyway, if there were evidence on this flight that we were missing tiles/RCC, I might be worried.”

With all requests to actually get that evidence denied, Columbia’s re-entry was scheduled as if there were nothing wrong.

In total, according to the CAIB report, there were eight missed opportunities to identify the problem, and three distinct requests for images which were denied.

NASA Administrator Sean O’Keeffe spoke to CNN following the report.

“This was a case where we missed it. You know, just flat missed it. We have to then compensate for that human characteristic and be sure that we exercise an overabundance of diligence and never let something be explained away simply because we’ve seen it before. It’s got to have a reason, it’s got to have an explanation.” 

NASA Administrator Sean O’Keeffe accepting the CAIB report from Admiral Harold Gehman. Source: NASA/Bill Ingalls

The CAIB report characterised NASA as “An agency trying to do too much with too little”. Over the years, their budget had stagnated and their effective spending power reduced; uncertainty over the future of the shuttle program and the pressure to perform on schedule for the construction of the International Space Station contributed to a very difficult environment. 

Any extensive in-flight investigation could have delayed the next shuttle launch – a mission related to construction of the International Space Station.

Gehman said, “Congress had decreed that if you don’t finish the international space station on time we’re going to cut the money and finish the program. The power of schedule and cost was so oppressive that the poor person way in the back of the room who said, “Yeah but there might be a hole in the shuttle wing” didn’t have any influence and couldn’t get heard.”

The report also pointed out that NASA had fallen into various bad habits; as O’Keeffe admitted, they had come to believe that foam strike events were not flight-safety issues, simply because they had got away with it so many times before. They put the emphasis on engineers proving that it wasn’t safe, rather than proving it was safe. 

As to whether the crew could have been saved if the damage had been recognised?

Gehman said,

“It would have been a very, very risky proposition to save the crew, but if we had known something was wrong with the shuttle we would’ve done something. We wouldn’t have just sat here for ten days and done nothing about it. We would have done something.”

A wide range of recommendations followed. Obviously, the issue of foam coming off the External Tank in the first place had to be addressed, alongside making the orbiter more resistant to damage. A more thorough program of inspections, making in-flight imaging standard for all shuttle missions, and developing more suitable computer simulation programs to assess any potential damage were also included.

From a management standpoint, they recommended better training, adopting a schedule based on their available resources rather than external deadlines, establishing an independent technical engineering authority to identify, control and analyse hazards, putting NASA’s Headquarters Office of Safety and Mission Assurance into a position of direct authority over the space shuttle program’s safety, and reorganising the Space Shuttle Integration Office.

Evelyn Husband, widow of commander Rick Husband, said, “I was convinced that it was just an accident, that there had been no way to prevent it, and the report was very disturbing to me. it became alarmingly clear that there could have been things done, and I found that very difficult to deal with.”

Following the disaster, work on the International Space Station was delayed, as the shuttle program was suspended; the ISS relied on Russian spacecraft for resupply and crew changeovers.

In July 2005, the space shuttle Discovery launched on a return-to-flight mission; however, another piece of foam came off. While it didn’t strike the orbiter, this time it was taken seriously enough to ground the shuttle for another year, before Discovery made a second return-to-flight launch.

In total, the shuttle program made 22 flights after Columbia; the final flight, STS-135, was made by the Space Shuttle Atlantis in July 2011, after which the Space Shuttle program was finally retired. 

I’d like to finish this episode with these words from the Columbia Accident Investigation Board’s statement:

“The Orbiterʼs destruction, just 16 minutes before scheduled touchdown, shows that space flight is still far from routine. It involves a substantial element of risk, which must be recognized, but never accepted with resignation. The seven Columbia astronauts believed that the risk was worth the reward. The Board salutes their courage and dedicates this report to their memory.”

Memorial service to honour the Space Shuttle Columbia Astronauts. Source: NASA/Bill Ingalls

Thanks this episode go to:

I’d like to say a special thank you to Patreon supporters:

  • Skeleheron
  • Michael Fay
  • MJ
  • Andy SJ
  • LouLi
  • Mish Liddle
  • Missa
  • Scott RC
  • Shana Gitnick
  • Teddy Gingerich
  • Will Friedrichs

and to all of you for listening and reading.

Supporting the Great Disasters podcast on Patreon can give you access to exclusive content, and helps the show keep going.


Sources, References and Further Reading

The majority of the details in this episode come from the Columbia Accident Investigation Board Report.

NASA’s History office maintains a site dedicated to Columbia, with links to many other useful resources, including the Columbia Crew Survival Investigation Report (PDF). There is also another section on the NASA website which is maintained for historic purposes, although several features no longer work.

Entry timeline of Mission STS-107 – Spaceflight Now

Some quotes are taken from the Seconds From Disaster episode on the Columbia disaster.

Selected News Articles:

    Comments are closed, but trackbacks and pingbacks are open.