With Discovery parked safely on the tarmac in California, the fate of the nation's manned space program now rests on the research teams assigned by NASA Administrator Michael D. Griffin to figure out why at least four large pieces of insulating foam broke away from the shuttle's enormous external fuel tank.

The teams, composed of engineers from NASA and Lockheed Martin's Michoud assembly facility in Louisiana, where the tanks are manufactured, were scheduled to give a preliminary report yesterday to Griffin and space station manager William H. Gerstenmaier. NASA has promised a fully "transparent" investigation, and initial findings may be made public as early as tomorrow.

"This is the first step back in our return-to-flight sequence," Griffin said, adding that the loss of foam insulation was "the only thing that went wrong with this mission."

An errant piece of foam caused the demise of Columbia in February 2003, and much of the subsequent 21/2-year flight suspension was aimed at resolving that problem. With the shuttle Atlantis poised for a September liftoff but grounded until NASA is satisfied that foam shedding does not jeopardize crew safety, a speedy resolution is deemed essential if the program is to have any hope of regaining its momentum.

Some NASA sources, speaking on the condition of anonymity because they are not authorized to discuss the latest findings, have said in recent days that there is good evidence the biggest foam failure is related to a repair made on Discovery's external tank during its production -- a promising determination that might allow NASA to conclude that the Atlantis tank is not similarly at risk.

That chunk, estimated to weigh 0.9 pounds, peeled off from the "protuberance air load," or PAL, ramp, which protects exterior hardware from aerodynamic stresses during liftoff.

June Malone, a spokeswoman for NASA's Marshall Space Flight Center in Huntsville, Ala., which manages the external tank, said she could not speak about investigators' findings but confirmed that workers at the Michoud plant at one point performed an abrading procedure "in the general vicinity" of the eventual failure. The aim was to smooth out a divot in the tank's foam, which is extremely heat-resistant but easily gouged.

Such fixes are considered routine and are followed by inspections to ensure that foam adhesion has not been compromised.

At least five teams are now looking at that issue and others, with each focusing on a different part of the external tank. Complicating the analysis, Discovery's tank was coated with a patchwork of two different formulations of foam: an older formulation, which is also found on the Atlantis tank and on a spare tank now at Kennedy Space Center, and a newer formulation, which covers the six tanks in Alabama that have their PAL ramps attached.

Unless explanations and ready fixes for Discovery's problems can be identified, NASA and Congress will face the difficult decision of just how much time and effort should be expended on a significant redesign so late in the aging rocket plane's career.

There is a widespread desire in the space program to speed retirement of the 25-year-old shuttle, which by any measure is deep into its autumnal years. The shuttle's primary remaining task is to bring enough components to the international space station to justify a declaration that the orbiting laboratory is "functional," and to maintain a U.S. presence in space until a new breed of space vehicle is ready for testing around 2010.

The PAL ramp underwent extensive testing in the 21/2 years after the Columbia disaster, including wind-tunnel tests late last year at the U.S. Air Force's facility in Tullahoma, Tenn.

In a report presented at a July meeting of the American Institute of Aeronautics and Astronautics in Tucson, a copy of which was obtained by The Washington Post, engineers from NASA's Aeroelasticity Branch concluded on the basis of wind-tunnel and other studies that the PAL ramps appeared unnecessary. But, in part, because one set of conditions could not be fully modeled, the team encouraged NASA to gather data from an actual flight to confirm their findings.

With those results in hand, mission controllers granted permission to install five sensors on Discovery's PAL ramp, to get real-time measures of the stresses there, according to an internal NASA memo written more than a week after Discovery's launch and examined by The Post.

The memo said preliminary data from the sensors supported the team's contention that the ramp could be eliminated without harm but added that more in-flight testing is needed before taking such a step.

The memo noted that the sensors had been installed on the liquid oxygen section of the tank -- the highest part of the tank and, therefore, the section where breakaway debris would have the highest chance of hitting the orbiter's vulnerable underside -- but that additional useful data could be obtained if sensors were also attached to the liquid hydrogen ramp on future shuttle flights.

Malone said Atlantis is equipped with sensors on the oxygen part of the tank, as Discovery was. But she said she did not know whether additional sensors have been attached to the hydrogen component.

Officials offered no specifics yesterday on how they might deal with the other parts of the tank that shed during Discovery's launch, but they expressed optimism that the problem is solvable.

"Now that we have Discovery back on terra firma, we'll go work those other issues," said William W. Parsons, the shuttle project manager. "In the next week or so we'll get preliminary data and findings about where we're headed."

Gugliotta reported from the Kennedy Space Center in Florida.