The Forensics Behind the Columbia Tragedy

Feb. 4, 2003 -- A helmet, a charred tile, a fleeting image of Columbia above California, seconds of jumbled data. These are some of the myriad clues officials must now scrutinize as they face the daunting task of determining what brought down the space shuttle.

"We want to get every last shred of evidence … and put that into the pieces of the jigsaw puzzle that we will start to assemble," NASA Associate Administrator William Readdy said Monday.

As officials collect the "pieces," members of the commission that investigated the 1986 Challenger explosion estimate that a few key areas of evidence are likely to emerge:

Flight data: Data from Columbia's sensors appears to be critical in forming beginning theories about the cause of the breakup.

Debris trail: Determining where debris fell and how it fell will help in tracing Columbia's demise.

Debris analysis: By closely analyzing stress fractures and chemical content of key debris pieces, scientists can gain clues about what triggered the breakup.

Video and images: Video proved critical in understanding what happened during the explosion of the Challenger shuttle and early analysis of an incident after liftoff of Columbia suggests it may be valuable in this investigation.

Better Than Any Black Box

"The shuttle records much more data than anything you'd find on a regular aircraft," said Merritt Birky, a retired fire and explosion specialist with the National Transportation Safety Board who worked on the Challenger investigation. "So I think the data will be very significant in determining the sequence of events that led to the failure."

Data from Columbia has already provided NASA investigators with significant clues, revealing a dramatic spike in temperature and an increase in drag at the left side of the shuttle just before the orbiter disintegrated. The data originated from hundreds of sensors on Columbia that relay their information via satellites or directly to computers on the ground.

Readdy said Monday he hopes to retrieve 32 more seconds of data delivered to ground control just before communication was completely cut from Columbia. This data had been previously filtered because of its flawed nature.

"We may be able to resynchronize the data and get something from it," he said.

Debris: Where and How It Fell

Once flight data points to a particular cause, the laborious task of analyzing thousands of debris shards from the shuttle could then help confirm any emerging theories. The most telling information may be where debris fell, says John Macidull, a staff member of the presidential commission that investigated the Challenger explosion.

"It very important to establish what came off first," said Macidull. "This means not only do they have to look in Texas and Louisiana, but also all the way back to California, where there were some accounts of seeing the shuttle start to break up."

Finding pieces of the shuttle's left wing in California or at least further west than Texas, for example, might suggest this was one of the first parts to fall off the shuttle. Such evidence could confirm eyewitness accounts and images from journalists and an astronomer in California that have all suggested Columbia began breaking up over California upon its re-entry to the Earth's atmosphere early Saturday. All seven astronauts perished.

About 300 people from 30 agencies, including the FBI, Federal Emergency Management Agency, NTSB and the Texas Department of Public Safety have spanned out to help in collection. It's now estimated that debris fell over a 28,000-square-mile area, but NASA's Mike Kostelnik said today that the debris field continues to grow larger than estimated.

All debris is being recorded, mapped by global positioning systems and taken to Barksdale Air Force Base in Louisiana, where it will be catalogued.

Mapping Debris, Piece by Piece

Collecting debris in this tragedy will be a much bigger task than what Macidull faced after the Challenger explosion. The Challenger exploded 73 seconds after takeoff and its relative proximity to Earth concentrated its debris trail, although much of the debris fell over the ocean, which sometimes complicated efforts to collect the pieces. Investigators were able to reconstruct 70 percent to 80 percent of Challenger's orbiter for analysis.

Although Columbia's debris trail is vast, Macidull says most of the pieces themselves are likely to be easily identified.

"The critical pieces, as far as evidence is concerned, would not have burned very much," he said. "There would have been some burning during re-entry, but because there isn't much oxygen out there to sustain burning, it would have been limited."

Tracing what parts fell and when is not only a matter of mapping where debris shards fell, but also calculating how each piece fell. Columbia disintegrated some 38 miles above Earth, so factors like the aerodynamic shape of debris and wind and thermal jet streams all had plenty of time to influence the path of falling debris.

"You do a regression-type analysis," explained Macidull. "You get the weather people and the Gulf Stream people and do analyses of each part's shape and you can map out how they fell."

Stressed Metal

Beyond mapping, each debris item can also hold clues within its texture.

Raynald Gauvin specializes in metal analysis using electron microscopy at McGill University in Montreal. He says much can be gleaned from a stressed metal's surface.

"If you have material with a crack, you can understand how the crack started and propagated," he said. "Then you can trace the crack and see if it was brought on by a preexisting problem or by the sudden impact of the breakup."

A metal with pre-existing weakness, for example, is likely to show striations around the area of impact, whereas a metal that was in fine condition before breakup wouldn't show these ripple-like lines and the entire piece is likely to be more deformed from the impact. Using electron microscopes, scientists can further analyze if a metal's chemical composition was formulated incorrectly, which could lead to problems.

Analysis can also pick up any traces of fuel or other explosive material that could have triggered an explosion.

Never One Thing

Macidull says it's important to remember that accidents are rarely traced to a single cause, so it's important to search for clues in several different areas.

In the investigation of the Challenger explosion, for example, Macidull and others relied on video footage at launch to make an early conclusion that a problem in Challenger's solid rocket booster led to the explosion. After they pulled up debris of the solid rocket booster from the ocean they were able to confirm the booster had a leak and found that a weakly welded area in Challenger's hydrogen tank had also contributed to the disaster.

Video has also played a key role in the Columbia investigation. After watching replays of Columbia's launch, NASA officials noticed a chunk of hardened foam from the external fuel tank peeled away and struck the shuttle's left wing — the same side where data had recorded a spike in temperature during re-entry.

Dittemore said they didn't notice the incident, which has become an intense focus of the investigation, until one day after launch as they were doing routine reviews of liftoff. Further analysis of the video may help them determine how much damage the impact may have caused, although it's unlikely to be the only cause.

"It's never just one thing that happens," Macidull said. "It's almost always a series of things that, due to bad luck, come together and cause the accident."