Clues about the timing of a giant earthquake striking Southern California -- caused by a jammed-up section of the San Andreas fault suddenly slipping loose -- may be uncovered by a new NASA radar project, scientists said today.

Scientists at the Jet Propulsion Laboratory in Pasadena are using an extremely-precise radar, strapped to the bottom of a jet flying 45,000 feet over California, to measure exact surface elevations along the southern-most section of the fault, from Riverside southeast throught the Coachella and

Imperial valleys.

For the first time, scientists will get a very-precise picture of the earth's surface near the fault. As earthquake-inducing stress builds up, they expect to be able to detect changes in elevations by overflying the area and taking new measurements, said lead scientist Andrea Donnellan.

"We'll be mowing the lawn, so to speak, mapping the San Andreas and adjacent faults, segment by segment, and then periodically repeating the same radar observations,'' she said. "By comparing these repeat-pass radar observations, we hope to measure any crustal deformations that may occur

between observations, allowing us to `see' the amount of strain.''

A special NASA jet will begin the overflights soon, and will eventually map some 600 miles of San Andreas fault and its related splinters, from north of San Francisco to Yuma, Ariz.

Earlier stress studies have identified the northern part of the Imperial Valley, 90 miles southeast of Riverside, as overripe for a big, big quake.

The US Geological Survey last year estimated that a Salton Sea megaquake would rupture the San Andreas Fault as far as Palmdale in Los Angeles County, and would kill up to 1,800 people, injure 50,000, destroy housing for 250,000 people and cause $213 billion in damages.

Scientists have long known that Riverside, Los Angeles, and San Diego are creeping north at 1.4 inches per year past San Bernardino, Lancaster and the rest of North America. Although this slippage occurs regularly along some parts of the San Andreas, it has been locked up in southern-most California for more than 320 years.

Other studies have already indicated that strands of the fault in the Salton Sea area are extremely overstressed. A recent chain of small earthquakes east of Borrego Springs and north of El Centro was feared to have been "the straw that breaks the camel's back'' and overstresses the main fault, the JPL

said in a news statement.

Of particular interest to Donnellan is the splintering network of parallel faults that spread from the Wrightwood area to the southeast, towards San Bernardino. One branch, the main San Andreas Fault, runs north of Beaumont and through San Gorgonio Pass towards Palm Springs, where it forks into

additional branches along both sides of the Salton Sea and Coachella Valley.

The other major branch is known as the San Jacinto Fault, and it runs from San Bernardino, under the interchange of the 10 and 215 freeways, and then through Moreno Valley, Hemet and the Anza area. Coincidentally, a tiny magnitude 3.3 quake struck along the San Jacinto Fault Sunday morning.

Because both faults have not moved in a major way for 320 years, damaging earth movement from those jammed-up faults could inflict severe damage in Riverside, Los Angeles, San Diego and possibly as far east as Phoenix, some scientists have said.

The current data being gathered by the airborne radar will be used a as a baseline for scientists, and will be put to use testing an earthquake prediction computer simulation called QuakeSim.