As investigators increasingly focus on a breach at or just behind the leading edge of the shuttle Columbia's left wing as the root cause of the Feb. 1 disaster, engineers poring over telemetry from the doomed ship are zeroing in on exactly where the breach must have occurred - and how it must have propagated - to explain the orbiter's response to the resulting aerodynamic forces that ultimately ripped the ship apart.

Within 600 seconds of the shuttle's plunge back into the discernible atmosphere 400,000 feet above the Pacific Ocean, a deadly plume of super heated gas managed to work itself into the ship's left wing, burning through a thick bundle of wires carrying data from various wing sensors and then into the left main landing gear wheel well. Once inside the wheel well, the plume apparently burned its way through the heat-softened corners of the left landing gear door, spewing out in jets at right angles to the flow of hot air rushing across the belly of the ship.

During the second day of a hearing by the Columbia Accident Investigation Board, NASA specialists outlined a wide-ranging, multi-center effort using wind tunnels, computational fluid dynamics and other sophisticated tools to develop scenarios that best explain the telemetry from the stricken spacecraft.

While the work is far from complete, they are zeroing in on a breach at or very near the leading edge of the left wing. And they are beginning to rule out scenarios that received attention earlier in the investigation, including a breach at or near the left main landing gear door and the earlier-than-normal onset of turbulence and high heating that might have resulted from any inherent "roughness" in the surface of the left wing or from unusually high atmospheric density.

Instead, investigators are assuming a small breach near the leading edge of the left wing allowed hot gas to begin entering the structure 488 seconds after entry interface, the moment Columbia began encountering the discernible atmosphere. The breach probably was there before the shuttle dropped out of orbit, but for the purposes of developing credible scenarios, investigators are assuming it opened up 488 seconds past "EI" because that's when the first signs of a temperature increase inside the wing showed up in telemetry from the shuttle.

"We're looking at about the first 600 seconds of entry, what happened from entry interface to the point where we believe there's a breach in the wheel well and the temperatures start rising," Stephen Labbe, chief of NASA's applied aeroscience and computational fluid dynamics branch, told the CAIB. "So if we can get that solved, we feel we'll have made a significant contribution to the investigation."

Entry interface occurred at 8:44:09 a.m. on Feb. 1. At 8:52:05 a.m., telemetry shows Columbia's autopilot starting to adjust the shuttle's wing flaps, or elevons, to counteract subtle but unusual aerodynamic forces acting on the left wing. Twelve seconds later, a landing gear brake line temperature sensor registered a "one bit flip" change, indicating an increase of about 1.5 degrees. The sensor was located in the left main landing gear wheel well.

At 8:52:59 a.m., data from a temperature sensor at the back of the wing suddenly went "off-scale low," indicating a complete failure. Other sensors at the back of the wing soon followed suit. The wiring for those sensors was routed forward, along the left side of the landing gear wheel well and then across the well's forward face just a foot or so from the area of the left wing's leading edge where engineers believe the breach occurred.

At 8:54:10 a.m., temperatures began rising dramatically across the wheel well and 10 seconds after that, telemetry indicates an unusual trend acting to pull the shuttle's nose to the left somehow reversed, as if the left wing suddenly gained more lift.

Labbe painted an intriguing picture of how the data might fit together. The brake line temperature increase, he said, "suggests a breach, a first initial breach into the wing."

"There must have been ingestion of hot gas in order to create that change in the wheel well," he said. "The second one is a burn through of the wire bundle that holds all of those instruments so that whatever was being ingested had to be able to burn through that wire bundle. When we get to the wheel well breach, we see a significant rate of change. Instead of just drifting up, now we see a large increase in the rate of change and that corresponds to a change in the aerodynamic trend where (the shuttle's rolling moment) was drifting negative and now it's starting to go back positive."

In the shuttle coordinate system, yaw and roll are considered positive when the shuttle is moving to the right and negative when moving to the left. During normal re-entries, the shuttle's flight control system keeps both centered around zero. But in Columbia's case, both roll and yaw went negative very early on, indicating aerodynamic forces were acting to pull the ship's nose to the left while at the same time trying to roll the ship slightly in the same direction.

Around 8:53:46, the first known piece of debris fell away from Columbia as documented in amateur video shot during the shuttle's descent across California. At 8:54:07 a.m., the fifth such "debris shedding" event was noted. The unusual change in the shuttle's rolling moment - from negative to positive - occurred at 8:54:20 a.m., just 10 seconds after the onset of a rapid climb in wheel well temperature.

At 8:54:22 a.m., a sensor mounted on the sidewall of Columbia's fuselage, well above the left wing, recorded an unusual temperature rise. Eleven seconds after that, a bright flash was seen by ground observers, followed seconds later by what appears to be a relatively large piece of debris falling away, the sixth such event and one of the two most significant.

"Somewhere between debris 5 and 6 is when we see this event where the rolling moment was drifting negative, the change in rolling moment, and it changes direction and starts this positive trend," Labbe said. "And we think this is a very key point for us in trying to understand what happened. Something changed about the configuration, some damage, and since we know we were shedding debris, something significant happened there that changed the trend on rolling moment."

Board chairman Harold Gehman then asked what sort of change in the aerodynamics of the spacecraft could cause it to suddenly start rolling back to the right.

"You've asked the $64,000 question there, I believe, and that's what our work is going to be," Labbe replied. "What it suggests early on is that I was losing lift on the left wing. And then something changed to start creating lift on the left wing, or pushing up on the left wing. ... The damage is so significant, it's creating locally a very high pressure that's on the lower surface of the wing and starting to push up on the wing."

An opening on the underside of the wing could provide that sort of lift, but engineers believe the left main landing gear door remained in place until vehicle breakup or just before. But jets of hot air spewing out around the corners of the door probably could provide sufficient pressure to reverse the rolling moment.

"I think if you have a jet, if it's coming out with a strong enough rate that you create a jet or create enough flow out of there, it will set up a shock in front of that, which will create a high pressure, which would be on the lower surface, which would push up on the wing and probably create more lift," Labbe said.

It's also possible disturbed air flow caused by a worsening breach near the leading edge could have had the same result. John Bertin, a professor of aeronautics at the U.S. Air Force Academy, told the board the loss of two or three reinforced carbon carbon panels from the leading edge could create a shock zone of higher pressure.

"That could have caused the pressure to be higher and giving you an asymmetric force," he said.

NASA's integrated aerodynamics, aerothermodynamics and thermal analysis review is attempting to figure out, using wind tunnel tests, complex computational fluid dynamics and other tools just what sort of breach must have occurred to explain what happened to Columbia.

"So what kind of hole or damage can be created in from entry interface to 488 seconds that could produce that initial change in the instrumentation?" Labbe asked, referring to the initial brake line temperature rise. "Then we go on to the next step. Step B is we burn through that wire (bundle) in another 42 seconds. So if we pick a location and we have a burn through, can it then also burn through the wire 42 seconds later?

"Then we have the breach into the wheel well at 600 seconds where we see the rate of change and of course, that has to be consistent with the initial breach and the burning through the wire. So you can see how we're trying to piece all these together. And then finally, we see this change in the fuselage wall temperatures and whatever is producing that, is the damage consistent?"

Contact with Columbia was lost a few seconds after 9 a.m. when the vehicle, yawing sharply to the left, broke up high above Texas. Because of the shuttle's high altitude - more than 200,000 feet - the dynamic pressure acting on the shuttle was roughly equivalent to what one would feel at sea level in a 150 mph wind. One maneuvering thruster, or just a degree or so of aileron trim can keep the ship properly oriented.

Toward the end of Columbia's flight, however, the shuttle was experiencing some 160,000 foot pounds of force acting to the left.

"That requires all four (right-firing yaw) jets, three or four degrees of aileron, sideslip, everything the vehicle had to try to counteract that moment, it was using," he said.

"Hold your arm outside of a car, you can feel (the wind) trying to pull your arm back," he explained. "That's a moment of what you're feeling about your shoulder and you're talking maybe, you know, 10 pounds, 20 foot pounds of moment. Not very much at all. And we're talking about over a hundred thousand food pounds of moment."

Even with all the damage that must have occurred because of the initial, quickly worsening breach, "the flight control system still was commanding the vehicle to do exactly what it guidance was telling it to do," Labbe said. The unusual change in sign indicating a positive roll due to increase lift on the left wing was "almost like the damage has returned the vehicle back to its original flight characteristics," Labbe said. "But then, of course, we see a rapid increase and then essentially going off the cliff there at the end."

Damage patterns in recovered debris and ongoing analysis of telemetry indicates the initial breach probably occurred at or near RCC panels 6 and 9 on the left leading edge. Or just behind them. So-called carrier panels, covered with heat-shield tiles, are bolted in place just behind the RCC panels to provide a flush surface between the carbon-carbon composite material and the tiles permanently bonded to the wing's underside.

Some investigators believe it is likely a damaged carrier panel or a broken/missing tile on a carrier panel just behind the RCC panels in question could have provided the entry point for the plume of hot air that ultimately destroyed the left wing. The day after launch, ground radars detected an object separating from Columbia that was roughly the same size as a carrier panel.

Whether a carrier panel came off before or during entry, "it would be very easy to have damage occur and such that the RCC panels themselves would come off," said board member James Hallock. "They're only held on by, I believe, two bolts. Get this thing out, get some heat in there in the right place on those bolts and they could come loose and that's what could be happening, all this stuff is starting to open up into a larger area."

Engineers are conducting wind tunnel tests and carrying out complex computational analyses to determine how such a breach might worsen.

"We're going to be looking at multiple panels in this thing and other panels missing and that's really where our future work is focused, is to first do a survey of the wing leading edge and then start looking at other damage scenarios that try to produce that," Labbe said. "And then eventually, get our higher fidelity CFD (computational fluid dynamics) analysis to get to the actual flight conditions."

One issue that remains to be resolved is whether the initial breach was present before re-entry began, which most observers believe, or whether it opened up during the descent, which Labbe's scenario suggests. In recent weeks, much has been made of NASA's decision not to request high-resolution spy satellite photography of Columbia in the days after launch to look for signs of damage.

Whether the breach was the result of some weakened component failing during descent or whether a visible defect was present prior to entry is significant.

"It bears on a lot of things because if the fault just manifests itself right here (during entry), even though aerodynamic pressures are practically nothing but maybe enough to remove something or cause something that was weakened (to fail), then all this stuff about on-orbit photography becomes irrelevant," Gehman said. "It's important to know whether or not the orbiter had a pre-existing condition, which then didn't manifest itself heat wise until you got enough heat."

Regardless of when it actually opened up, the presumed breach between RCC panels 6 and 9 was in a particularly bad area. As it turns out, RCC panel 9 is precisely where two hypersonic shock fronts interact during shuttle descents.

"For the shuttle, without damage, the sweep angle (of the wing) is such that the interaction effects are relatively benign," Bertin said. "So that while there's a shock-shock interaction, the highly swept leading edge prevents you from having strong interactions."

But if multiple RCC panels were missing, providing what amounts to an unswept wing leading edge, "you'd have strong interactions and very large heating going on," Bertin said.

A blurry photograph of Columbia taken by off-duty personnel at Kirtland Air Force Base in New Mexico shows the shuttle within a minute or so of its destruction. The photo has been enhanced now and it clearly shows signs of significant damage to the left leading edge area. Bertin believes the imagery is actually showing the high-density shock zone that formed just in front of the damaged area.

"So if you were missing, maybe not one panel but maybe two panels and maybe it's downstream from the initial problem that you had and stuff like that, then you've got a little like two teeth missing from the leading edge and you've got a little notch in there," he said. "Now the flow can go in that notch and create a shock pattern that in my mind kind of looks like what the Kirtland photograph might be telling you."

He said the initial breach might have started out small, but it would have quickly worsened.

"I'm assuming RCCs possibly were lost in time and a very early one, maybe one would be missing, maybe more," Bertin said. "But then because the under structure is exposed, that some additional damage occurred and other ones would have come off in some fashion. With just one missing, you could get the damage that maybe was observed eventually. But from seeing the Kirtland (photo), I'd think you'd have a pretty good piece missing."

Once the plume from the breach got into the wing, it quickly made its way into the left landing gear wheel well. The exact mechanism is not yet known but investigators have recovered debris from the landing gear door attachment system that indicates some areas in the well were subjected to temperatures high enough to melt titanium, nearly 3,000 degrees. Interestingly, sensors in other areas of the well did not record anything out of the ordinary.

"When this plume entered, it can bounce around," Hallock said. "This is a very rarified atmosphere still and so we could be having 3,000 degrees at one end, which is melting titanium, and at the other end we'll have a sensor that doesn't even know much is happening. You've got to have enough molecules in there, atoms in there, to be able to convey the temperature itself.

"Temperature is a quantity that depends on having something for it to register on," he explained. "You've seen in these reports where they say astronomers have found a place where the temperature is 10 million degrees and this is a place in space itself. Why? There's two or three molecules moving around with such energy that yes indeed, if you had a thing that those things could hit against, they would register a temperature that high.

"So don't think of it as being like heating up the air in this room. It's very different because there are so few molecules there that you can get a great gradation right across the entire area."

Gehman said the board was particularly interested in the sort of chemical reactions that can occur when aluminum is subjected to extreme temperatures.

"What we're leading to, of course, is a scenario in which a new path into the wheel well could be developed in seconds," he said. "We're trying to establish just how rapidly other kinds of processes can rearrange the inside of the wheel well and then match up with these temperatures. Right now, we're still in the exploratory stage of that."

In other developments, Gehman said the board is close to releasing "interim recommendations" to NASA. He would not discuss what those recommendations might require, but after the news conference he told reporters one of them will focus on how NASA and the operators of military imaging systems like spy satellites should work together in the future.

"The business of NASA and the various intelligence gathering agencies of United States getting their act together so at they don't talk past each other - we're talking about on-orbit photography and things like that - NASA doesn't need to wait, they can go ahead and get that sorted out right now," Gehman said. "I'm not going to comment on exactly what could be done, but one of the earliest recommendations we think is coming up for board consideration is guidance to NASA to get their act together."

He said NASA personnel were not at all up to date on the capabilities of modern military imaging assets and that no interface existed to bridge that gap. Agency managers decided early on not to request spy satellite imagery of Columbia's left wing because they did not believe the systems had enough resolution to see presumably small signs of damage on the leading edge or in the heat-shield tiles.

While Gehman did not discuss the capabilities of modern imaging systems, Ted Molczan, a respected satellite observer, has calculated that Columbia could have been imaged on multiple occasions by classified optical imaging satellites. Resolution would have varied from six to 16 centimeters, or 2.3 to about six inches.

"We think both sides of the house bear equal blame," Gehman said. "Improvements in on-orbit photography that have come along, nobody from that part of the house has come down here (to NASA), knocked on their door and said, you know, things have changed since we made this agreement. And NASA doesn't have the right people cleared to receive the data. ... They made decisions without having the right information. The point is, this is a system that broke."