NASA badly needs a shuttle success. NASA plans hinge on a successful shuttle. With Discovery's imminent launch, the US prepares to return to space. But big questions over the direction of US space policy remain unanswered. [BY]By Robert C. Cowen, Staff writer of The Christian Science Monitor

MARS, whose fiery red disk now brightens the night sky, tantalizes American space scientists with visions of renewed planetary exploration. But they can't effectively pursue this vision or any other space enterprise until NASA gets its shuttle fleet back in regular service. The space agency has been tantalizingly methodical about this, preferring to troubleshoot possible malfunctions rather than stick to a preset schedule. As Richard Truly, associate administrator for spaceflight, has observed, ``We're going to wait until we have it right.''

Last week, the National Aeronautics and Space Administration decided it does ``have it right.'' It set Discovery's departure from launch complex 39-B at the Kennedy Space Center for Sept. 29.

Admiral Truly has warned that it may take two or three tries to get the spaceship off the pad because of weather delays or other factors.

But it now seems likely that Discovery's five-man crew will soon take their spacecraft back into its native orbital environment.

Led by mission commander Frederick Hauck, the crew for the four-day flight includes pilot Richard Covey and mission specialists John Lounge, David Hilmers, and George Nelson.

NASA has made more than a hundred major modifications to Discovery. These many improvements don't make Discovery essentially a new spacecraft. But this mission will, in a sense, be a second shakedown flight for the shuttle system.

There is more at stake, however, than just a successful test flight. Much of what the United States currently plans to do with its space program depends on the timely return of regular shuttle service.

The space agency has planned a tight shuttle launch schedule, beginning with Discovery and running through the first space station assembly mission in April 1994.

The shuttle manifest has been eased by shifting some military and commercial missions to unmanned rockets. The Mars Observer, for example, now is set to go on a Titan 3 booster in September 1992. And a 1992 Pentagon payload will move to an Air Force rocket.

NASA will update the manifest again in December. Whether it retains this busy schedule depends on the timely success of Discovery's second debut.

About six hours into the Discovery mission, the astronauts plan to tackle their most important task - deploying the 5,000-pound Tracking and Data Relay Satellite (TDRS). Once this is clear of the orbiter, mission commander Hauck and pilot Covey will move Discovery about 41 miles behind and 18 miles above the satellite. Then the Air Force-supplied Inertial Upper Stage (IUS) rocket will boost TDRS to a height where it can be injected into an orbit 22,300 miles high. This is the Clarke (geosynchronous) orbit - named for science writer Arthur C. Clarke - where a satellite remains over a given region on the ground.

The TDRS system is as essential to NASA's future orbital operations as is the present network of ground tracking and communications stations. This network currently links controllers and satellites for only 15 to 20 percent of an orbit. Much of the time, spacecraft are out of contact. With at least two TDRS satellites in place, controllers should be in touch with spacecraft, including shuttles, about 85 percent of the time.

Challenger took the first TDRS up on the sixth shuttle mission in April 1983. Despite a malfunction with the IUS booster, controllers managed to nurse that satellite into position. It now sits over the Atlantic just east of Brazil. A second TDRS was lost in the Challenger explosion Jan. 28, 1986. NASA plans to position the new TDRS-3 south of Hawaii.

Under current plans, Discovery is to take up TDRS-4 next February. It will replace the aging TDRS-1, which will become an on-orbit spare. When the TDRS system is operating successfully, NASA plans to phase out much of its present ground tracking network.

This is a crucial transition for US civilian space operations. Both manned missions and unmanned satellites will be almost totally dependent on orbiting communications satellites. Failure of either the satellite or ground-station portion of the TDRS system could mean loss of much important data from the Hubble Space Telescope and other major astronomical observa-tories NASA plans to launch in the 1990s, among other spacecraft. It's an example of the increasing interdependence of NASA programs which the Congressional Budget Office has warned puts individual missions at greater risk.

Then there's the retrieval of the Long Duration Exposure Facility (LDEF). Challenger set the 8,000-pound satellite on orbit in April 1984. Thirty feet long and 14 feet in diameter, LDEF carries 57 experiments in trays to test the effects of weightlessness and exposure in space on various substances, including seeds and electronic materials. It was to demonstrate what NASA hoped would become a useful type of orbiting research facility. But LDEF's planned recovery by shuttle was repeatedly postponed until the Challenger accident put it on indefinite hold.

Now the drag of the thin residue of atmosphere at orbital height is causing LDEF's orbit to decay. If the massive satellite were to spiral in, large chunks of it would survive frictional burn-up and hit the ground. NASA wants to bring it back. But it has put off the recovery from July to November next year, just a few months before LDEF is expected to begin tumbling to an extent that would make it impossible for astronauts to get hold of it.

Other time-sensitive missions include the launch next year of Galileo, to study Jupiter and probe its atmosphere, and of the Magellan Venus radar mapper. The shuttle Atlantis is to launch Magellan next May and Galileo five months later. While there is a leeway of a few weeks in each of these launch windows, the timing is essentially set by the configuration of the planets. So NASA has protected these launches by rescheduling other missions as Discovery's tentative launch date moved from spring to fall.

One of the most significant postponements has been the slippage of the Hubble Space Telscope launch, already delayed four years, from June 1989 to no earlier than February 1990. In compensation, project engineers will check out the telescope and develop its software to a point where the observatory could fly on 3 months' notice should a major defense or other payload cancellation open up an earlier shuttle slot.

Tomorrow: Manned vs. unmanned space programs.