The torch of plasma, or superheated atmospheric gas, that investigators believe brought down the shuttle Columbia might have been leaving the craft's wheel well, not entering it, having cut a channel of destruction all the way from the leading edge of the wing near the fuselage, officials involved with the inquiry said yesterday.
The new theory of the loss of the shuttle is gaining prominence on the board led by Adm. Harold W. Gehman Jr., but no single theory has emerged as a clear leader yet, one official said.
"The good news is that we're now talking about scenarios," the official said, indicating that NASA and the board are beginning to close in on the causes of the shuttle's breakup.
The picture of the shuttle's final moments is emerging as thousands of small points of data resolve into a larger image.
The data the investigative board is drawing on include the condition of debris, the readings from sensors captured in NASA's computers as the shuttle began to go out of control, and the information gleaned from computer models that researchers run to see which best match the known facts.
About 22,000 pieces of the shuttle have been found, and investigators are learning much by identifying which parts are still missing. Many of the reinforced panels at the leading edge of the left wing have been found, for example, but some have not - most compellingly, in the area close to where the wing was attached to the fuselage.
The notion that the plasma was already inside the shuttle structure when it reached the wheel well would help to explain one oddity of the collected debris so far: damage to a wheel-well door that appears to show a hole that goes from the inside out, with the damage perpendicular to the direction of travel.
Part of the challenge is to distinguish between damage from the breakup and damage caused by falling 200,000 feet through the atmosphere. Heavier shuttle pieces were scorched by the heat of entry; lighter pieces, such as tile, lost speed more quickly and heated up less.
Physics offers clues, including the melting points of metal and other materials used in the shuttle. By noting which of the recovered materials have melted and which have not, the investigators can "follow the heat," as board member Roger E. Tetrault put it last week.
Another mystery in the loss of the shuttle is why, if pieces were falling off as early in the descent as California, neither the crew nor NASA noticed the damage.
The board official speculated that this might be explained as the loss of tile and insulating blankets from the top of the shuttle, which under normal flight conditions is exposed to far less heat and pressure than the toughened underside.
A breach at the leading edge of the wing, or at the boundary where the panels meet the more delicate insulating tile, could cause a "shock front" that could send vortices of hot plasma spilling over the top of the shuttle and ripping off tiles and blankets, he said.
"If they were to come off, you might not have any real hint of that," he said.
Investigators say other evidence seems to point to this conclusion, including a fuzzy photograph taken as the shuttle passed over Albuquerque, N.M.
The photo appears to show disruption at the leading edge of the wing and some kind of flow trailing off the back of the wing. The latter could be evidence of the plasma, the official said.