Leaders of NASA's Mars Surveyor 2001 mission may be making a number of design changes to minimize the chances that whatever failure doomed the Mars Polar Lander doesn't recur with the 2001 lander. The latter is the primary target of a top-to-bottom examination because it is essentially a copy of the Polar Lander.

Changes will likely include adding a more robust and capable communications system, and perhaps a so-called black box -- a device similar to an aircraft flight data recorder that helps crash investigators learn what went wrong after airline disasters. Instead of a recording on magnetic tape and sending out a beacon ping, though, it would store information in computer memory and send out a UHF radio signal.

"It would record the last ten minutes of data and, if its power vanished, it would turn on and broadcast," said George Pace, the project manager for the Mars Surveyor 2001 mission at NASA's Jet Propulsion Laboratory. "Eventually an orbiter going overhead it would pick this up and within a few minutes it would be able to read out the whole memory."

Had such an instrument been part of the Mars Polar Lander, it is likely that NASA scientists and engineers would already know what went wrong with the spacecraft when it reached Mars on December 3, Pace said.

Another of the Polar Lander's shortcomings that engineers may now add to the 2001 lander is a system that will allow mission controllers to track the spacecraft's position during atmospheric entry, descent and landing, Pace said.

Engineers last heard from the Polar Lander as it approached the martian atmosphere. Its next scheduled transmission was to be after it reached the surface, but controllers never received a signal.

It is possible that the craft landed on the surface safely, but was tilted in an awkward direction that made its antennas useless, said Jeff Plaut, deputy project scientist for the 2001 mission. A third likely addition to the lander will be a more versatile direct-to-Earth communications system so that the orientation of the lander would not have a big impact on the spacecraft's ability to send and receive signals, Plaut said.

Still, there is no way to know what actually happened to the Mars Polar Lander, a fact that has raised concern among many of the scientists building instruments for the 2001 mission. Some worry that a long investigation or major design changes could delay the scheduled April 2001 launch. If the lander cannot launch in April of 2001, the next available launch window is during the summer of 2003.

"In the space business when a spacecraft fails and you don't know why, you ground the vehicle," said Steve Squyres, who leads the rover team for the 2001 mission. "Whether or not it's going to be ready for launch in 2001 or not isn't clear." Officials have been discussing the possibility of relaxing the launch schedule if necessary, Squyres said.

The communications components can be added relatively easily without threatening an on-schedule launch, although they could muscle out some of the science instruments that are manifested for the mission, according to Plaut.

"There are pretty tight constraints on mass. There would most likely have to be some trade made if these components are added," he said.

Any more-comprehensive changes could complicate the task of making the target launch date.

Whether the craft is ready depends on what the investigation of the Polar Lander failure reveals, Pace said. "As far as any other design changes, in January we're going to be onto a process where we're looking at all the elements of entry, descent and landing and see if we can find a smoking gun."

So far nothing clearly points to why the Polar Lander was lost, he said, but additional tests of the flight hardware and the deployment sequences the spacecraft initiates as it approaches and enters the martian atmosphere might reveal some clues.

One of the only clues NASA now has about the fate of the Polar Lander is the fact that the two Deep Space 2 probes have also not been found. These probes were supposed to be deployed from the lander as it separated from its cruise stage, falling to the ground independently of the lander.

The fact that controllers never heard their signals could mean that something went wrong during the separation phase, and the lander capsule and probes never successfully separated from the spacecraft's cruise stage, Pace said.

Another possible saboteur is the landing site itself, even though the site selection team originally was confident that the site was a beauty. "Maybe there's something that we really don't understand about the terrain and where we were landing at the south pole," Pace suggested. Soft or puffy soil, unseen trenches or steep slopes could have spelled the death of the lander and both probes.

The equatorial region where the 2001 mission is headed is much better understood, Pace said, but rocks could still pose a threat. For that reason, the lander team may decide to strengthen the spacecraft's legs for the 2001 mission.

Still, work is proceeding on schedule for a launch in April 2001. NASA has not yet made any final decision on changes for the spacecraft, adding that he expects decisions to come by the end of January -- still plenty of time to implement some design changes. The mission schedule has nearly two months of padding built in, when engineers and builders can fit in additional work. If necessary, more time could be gained by working crews on weekends and overnight shifts to make the launch deadline, Pace said.

The $282 million 2001 mission includes a Mars orbiter, which is to be launched a month before the lander. Although its instruments are flying aboard a platform very similar to the Mars Climate Orbiter, which was lost when it reached Mars September 23, Pace said he isn't worried that a hardware flaw could affect the 2001 orbiter.

An investigation into the Climate Orbiter failure blamed human error and the navigation and control processes of the mission. While these processes will be redesigned for the next mission, the spacecraft design is sound, Pace said.