"We are completely redoing the analysis from scratch," shuttle program manager Ron Dittemore said Monday. "We want to know if we made any mistakes."
Meanwhile, members of the independent board investigating the Columbia disaster arrived via helicopter Tuesday morning to get a first-hand at the look at the debris scattered over the countryside.
"It makes the accident more personal to us. This prevents it from becoming an abstract event," said retired Navy Adm. Harold W. Gehman Jr., chairman of the commission. He said the commission has 20 teams of experts in various fields working independently.
On Monday, investigators located the spacecraft's nose cone in a heavily wooded area of eastern Texas. A crew was to return to the site Tuesday to excavate the cone, found partially buried in a hole 20 feet across.
"It's basically the front of the nose cone," said Warren Zehner, an Environmental Protection Agency on-scene coordinator. "It's reasonably intact."
The nose cone is one of the largest and most recognizable parts found so far and could provide insight into how the shuttle disintegrated over Texas on Saturday, killing all seven astronauts aboard. The accident left a debris trail covering hundreds of square miles in at least two states, Texas and Louisiana.
But Dittemore now says NASA was also interested in any pieces that may have fallen from Columbia as far west as New Mexico, Arizona or California. The FBI was checking reports of possible debris in Arizona.
The reason for that interest is electronic data from Columbia that showed the shuttle began having problems earlier in the flight, as it came over the country's western coastline. That data pointed to extreme heat as a possible cause of the accident.
As the shuttle crossed the California coast, sensors in the left wheel well alerted Mission Control to a buildup of heat. Another sensor, above the wing on the left side of the fuselage, showed an even higher temperature spike — possible signs that some heat tiles had cracked or broken off.
The shuttle's flight computers struggled to compensate for heavy "drag on the left wing" before control was eventually lost, a scenario which also could have been caused by a damaged wing.
NASA is now re-examining the shuttle's launch to see how that damage may have occurred.
About 80 seconds after Columbia lifted off, a large piece of thermal insulation broke loose from the external fuel tank and struck the left side of the orbiter, reports CBS News Correspondent Bob Orr.
A videotape released Monday by NASA shows the debris clearly hitting the bottom side of the left wing and disintegrating into what appears to be a cloud of dust.
The impact by the 2½-pound, 20-inch fragment may have damaged the heat tiles, which keep the ship from burning up during re-entry into the atmosphere.
"We're making the assumption from the start that the external tank was the root cause of the problem that lost Columbia," Dittemore said. "That's a fairly drastic assumption and it's sobering."
While Columbia was still in orbit, NASA engineers analyzed launch footage frame-by-frame and were unable to determine for certain whether the shuttle was damaged by the insulation.
But they ran computer analyses for different scenarios and different assumptions about the weight of the foam, its speed, and where under the left wing it might have hit, even looking at the possibility of tiles missing over an area of about 7 inches by 30 inches, NASA said.
The half-page engineering report — issued on Day 12 of the 16-day flight — indicated "the potential for a large damage area to the tile." But the analyses showed "no burn-through and no safety-of-flight issue," the report concluded.
High-level officials at NASA said they agreed at the time with the engineers' assessment.
"The best and brightest engineers we have who helped design and build this system looked carefully at all the analysis and the information we had at this time, and made a determination this was not a safety-of-flight issue," Michael Kostelnik, a NASA space flight office deputy, said at a news conference Monday.
However, the incident during this launch was not the first time NASA officials considered the danger of tile damage.
The Columbia sustained significant tile damage in 1997, after NASA stopped using the coolant Freon in production of foam that coats the external fuel tank, a NASA engineer said at the time. The change was made because of the potential environmental damage Freon can cause.
In his December 1997 report, Greg Katnik, a mechanical systems engineer at Kennedy Space Center, raised the possibility that the new foam may have had some unknown characteristics that were not compatible with the severe conditions of takeoff.
Kalpana Chawla, one of the seven astronauts killed Saturday, was on the 1997 Columbia mission.
Dittemore said he knows of at least two other shuttle launches in which foam came off and damaged the shuttle, though nowhere near to the extent suspected in the case of Columbia's last launch. One involved Columbia, the other shuttle Atlantis.