At first, investigators could only speculate about events that took place 40 miles above the Earth, where little behaves as it does at sea level.
With Columbia scattered across the Southwest, they had few physical clues at hand.
There was, however, no shortage of theories.
Astronomers speculated that an erupting solar plume of charged particles had damaged the shuttle's fragile electronics. Other experts wondered if years of weathering had corroded the shuttle's airframe or leached the strength from its protective sheathing.
Had the pilot made a mistake? Had a landing-gear door opened too soon, or had a tire exploded? Perhaps Columbia had collided with orbiting junk or a micrometeoroid. Ice falling from the external fuel tank could have damaged it.
A falling piece of foam insulation could have damaged it too.
Over and over, engineers replayed a blurred video of insulating foam splattering on the left wing during liftoff. Methodically, they studied all the circumstances: unusually strong wind shear, more swiveling of the rocket boosters than usual, violent sloshing of liquid oxygen within the fuel tank.
Everything had seemed well within design limits, they said.
Sean O'Keefe, the head of NASA, would later concur. He dismissed those who kept talking about foam as "foamologists."
It was O'Keefe's misfortune to arrive at NASA 13 months before the accident. A former financial watchdog at the Pentagon, he knew his way around budgets and Washington bureaucracy, but he knew little about spaceflight.
In the aftermath of the accident, the tasks he faced were all but irreconcilable.
He had to defend NASA while finding fault with it. He had to straighten out a human spaceflight program he only sketchily understood. And he had to persuade Congress to cover the costs at a time of war and soaring deficits.
The investigation would need all the credibility it could muster.
To provide it, O'Keefe appointed an Accident Investigation Board barely 90 minutes after Columbia was lost. For chairman, he chose Hal Gehman, an owl-eyed career Navy man who had led the Pentagon's investigation into the terrorist bombing of the destroyer Cole.
Though appointed by NASA, Gehman, a retired four-star admiral, had no intention of doing its bidding.
Adroitly, he marshaled congressional support to set his investigation distinctly apart from the space agency. In time, the board would amply prove its independence.
NASA was duty-bound to assist the Gehman probe. Even so, it bridled at its demands and penetrating questions, which ranged far beyond the simple mechanical causes of the shuttle failure.
NASA itself was on trial.
Moving more nimbly than NASA's bureaucracy, an enterprising band of volunteers began their own inquiries that would turn up some of the first clues.
One was Tony Beasley, who had seen unusual flares trailing from the spacecraft as it passed over Caltech's radio observatory in the Owens Valley. His colleagues and his neighbors saw them too.
As an astronomer, Beasley knew the importance of quickly sharing reliable eyewitness accounts. Within an hour, he had made a detailed report of his sighting, first through Caltech, then through radio, newspaper and television interviews.
Almost as quickly, NASA's shuttle managers rejected his information in a public briefing on the accident.
Beasley swallowed whatever anger he felt.
The next day, Paul Dimotakis, a leading authority on hypersonic flight and applied physics at Caltech, suggested to Beasley that they might be able to track down the source of those mysterious flashes on their own.
Dimotakis, a 58-year-old native of Greece, was a soft-spoken, courtly man with an immigrant's fierce pride in his expertise. Few in the world were as intimate with the mysteries of the high-speed turbulence that had ripped Columbia apart.
The astronomer and the physicist were certain that the first debris fell not over East Texas, where so much attention centered, but a thousand miles to the west.
They enlisted Brian Kern, another astronomer from Caltech, and three experts in space navigation from the Jet Propulsion Laboratory in Pasadena.
They called themselves the skunk team.
Beasley eventually counted 6,541 images and 34 video recordings of the shuttle's last reentry, covering its entire flight path across the United States.
Amateurs had taken them all.
Beasley and Dimotakis studied two shaky videos shot with camcorders.
Immediately, they noticed that in the first video, recorded by a man near Reno, the shuttle passed Venus in the sky.
They seized the clue.
By knowing Columbia's planned trajectory, the relative position of the planet and the constant rate of the video frames, they worked out the timing, speed and direction of the flight to within a fraction of a second.
The second video was from Springville, Calif., near Sequoia National Forest. In it, they spotted the star Deneb and the compass star Polaris. Other stars — Alpha Cepheus, Vega and Beta Cassiopeia — could be plotted. This allowed them to confirm the timing of the flashes and the position of the shuttle.
By enhancing the stars digitally and then working with the constellations they revealed, the two men determined that Columbia's actual flight path had been barely a mile off its predicted reentry. Navigation experts at the Jet Propulsion Laboratory quickly confirmed their work.
The skunk team could then estimate the mass and size of the debris the spacecraft was shedding.
By knowing the properties of the exotic materials used in shuttle construction — recorded in three 22-year-old notebooks archived in the Caltech library — they calculated how quickly each piece was left behind by the spacecraft.
A theory took form.
The brightest flash — an event that NASA several weeks later would label "Flash #1" — was most likely a vapor of burning aluminum droplets, spewing perhaps from the spacecraft's melting framework.
"You see a bright flash and then a haze that is left behind," Beasley said. "The haze slows to essentially zero almost instantaneously, so it is impossible that it is any sort of solid material."
Perhaps something inside the shuttle had exploded. Melting metal scattered "like dumping a bucket of tinfoil chaff out of an airplane," Beasley said.
But what could have blown off so early in reentry without destroying the entire spacecraft?
Working with little sleep, the skunk team focused on a smaller but more substantial flash in the videos.
NASA would later label it "Debris Point #6."
From its speed, Dimotakis deduced that this chunk of Columbia might be up to 4 feet long.
Anxious to find it, the team arranged to charter a small plane, with Caltech donating the money to pay for it.
By this time, it was Saturday afternoon, two weeks after the accident. They would start their own air search Sunday morning, as soon as Dimotakis could refine his calculations for the most likely landing spot.
It was midnight by the time he and Beasley had finished cross-checking their results.
They canceled the flight.
They had underestimated how far something flung at 4 miles a second from an altitude of 200,000 feet would fly before hitting the ground.
The debris, they figured, had traveled 240 miles. It could have landed anywhere in an area 30 miles by 10 miles centered on the Gunlock Reservoir in southeastern Utah.
They sent their analysis to NASA and to the independent investigating board. Civil Air Patrol pilots were soon searching the area. Nothing turned up.
Beasley later drove out to scout the terrain.
The countryside that had seemed so pristine on the map was littered with a century's worth of mining equipment, auto parts and other discards.
The crucial piece of debris was out there somewhere, Beasley believed. "Some cowboy will kick it over in 50 years."
Not far from the Johnson Space Center in Texas, David Whittle, 61, was sitting down to breakfast that first Saturday in February. The 36-year NASA veteran planned to spend the day debugging his pastor's computer network at the League City United Methodist Church.
But his bag was packed, just as it had been for years.
The phone rang. It was Mission Control.
Whittle left his pancakes on the table.
Whittle had three jobs in the shuttle program, the most important of which he had never actually performed — running the salvage efforts after a catastrophic accident. It was his job to locate, recover and catalog any wreckage.
During simulations involving hundreds of NASA workers, Whittle had watched as they ditched the shuttle in the ocean, diverted it to emergency landings in Europe, staged high-altitude bailouts and launched rescue missions to the International Space Station.
In the simulations, everyone survived.
Even after the 1986 Challenger accident, NASA officials could not bring themselves to practice for the death of a crew. Nor did they ever simulate the aftermath of a catastrophic accident over heavily populated areas.
Now, no one knew where the crew's remains might be found. No one knew how many roads were closed, buildings evacuated or bystanders injured by falling debris.
Whittle headed for Barksdale Air Force Base near Shreveport, La., to set up headquarters for the search effort. He landed with no plan in hand. Instead, he improvised.
Based on initial reports, the wreckage of Columbia was spread across one-fifth of the country. Emergency dispatchers in Texas and Louisiana were fielding calls about debris.
"Everything we found was strewn along 60 or 70 or 80 or 100 miles," Whittle said. "Interested in the left wing? It is strewn over 100 miles. Interested in the right wing? It is spread over 100 miles. The crew cabin? It is along 100 miles."
NASA investigators were especially anxious to find recorders, cameras and computers — anything with a memory.
Unlike the other shuttles, Columbia had a nervous system of 721 sensors that monitored stress, pressure, vibration, strain and temperature. All of that data spooled onto a tape recorder under a crew seat.
None of it had been transmitted to Mission Control.
"We got so many false reports of black boxes being found intact," said John Hunt, a senior avionics expert at United Space Alliance, which runs shuttle operations for NASA. "I was on an emotional roller coaster."
One morning in early March, a field technician called to report that searchers had recovered a 35-millimeter camera from the shuttle that might have pictures of debris falling from the external fuel tank during launch. Hunt could barely contain himself.
Then he saw the photographs of the find.