SCIENTIFIC HURDLES

By Amy Wilentz.

An all-out Soviet nuclear attack on the U.S. could entail an intercontinental blitzkrieg: thousands of missiles launched from enemy territory, letting loose tens of thousands of deadly warheads surrounded by a nebula of hurtling decoys and debris. In half an hour, this lethal ''threat cloud'' would be over the U.S., raining destruction on cities and military targets alike. Trying to stop this deluge would require enormous technological breakthroughs in at least four areas: sensors, lasers, particle beams and computer programming. Should such advances occur, SDI proponents argue, a reasonably effective Star Wars defense would reduce to virtually zero the number of Soviet intercontinental ballistic missiles (ICBMs) getting through outer space to their targets. But critics respond that virtually zero is not enough when nuclear weapons are involved. Moreover, the Soviets have other ways to deliver a bomb--from offshore submarines or cruise missiles, for example, neither of which could be intercepted by proposed SDI technology. SDI planners see their defense as a multilayered ''architecture'' that could blunt a Soviet attack during the three distinct stages of delivery: the missile's boost, or launch, phase; midcourse, essentially the intercontinental space flight after the nuclear warheads and decoys have been released; and terminal, or re-entry, when the deadly warheads drop back into earth's atmosphere heading toward their targets. The most important of these is the boost phase, during which an ICBM's multiple warheads are still onboard and can be knocked out with a single shot. Hitting a missile in boost, says Stanford Physicist Sidney Drell, ''is like tackling the quarterback before he can throw the ball.'' SDI Director Air Force Lieut. General James Abrahamson told the TIME conference it represents the ''big payoff'' of Star Wars. Boost phase provides certain other opportunities for the defender. As missiles rocket through the atmosphere, their thrusters emit a hot, bright tail of fire, making them an excellent target for heat-seeking infrared sensors. SDI researchers hope to develop small, inexpensive but highly accurate self-guided missiles known as ''smart rocks,'' which could home in on a rapidly moving missile or warhead and destroy it by force of impact. But if boost-phase kill is attractive, it is not easily achieved. Because infrared sensors cannot ''see'' around the curve of the earth, they must be in an orbit high enough to spy into Soviet territory. Some would even have to be fixed in geosynchronous orbit, 22,300 miles up. Smart rocks would also have to be launched from space in order to hit a missile during boost. One plan would fire the rockets from ''gun pods'' in low orbit so they could speed to the vicinity of a rising Soviet missile. But Ashton Carter of Harvard, an SDI skeptic, points out that such sensors and gun pods would be vulnerable: ''Hovering a couple of hundred kilometers over enemy territory is a very uncomfortable place to operate.'' In fact, the entire SDI apparatus for boost- phase sensing and shoot-down would have to be predeployed in space and would therefore be extremely susceptible to a pre-emptive enemy attack. ''It is easier to destroy the space-based components of a strategic defense system,'' says former Secretary of Defense Harold Brown, ''than it is to destroy the ballistic missiles.'' Nor would defenders have much time to identify and hit a missile during this initial stage. Today the Soviet ICBM boost phase lasts up to five minutes. ''Fast-burn booster'' technology now in development may cut that time to as little as 50 seconds--a ''short fighting window.'' Missiles that escape the boost phase and enter midcourse flight present an overwhelming problem. By that time they have released their re-entry vehicles (warheads aimed at U.S. targets), as well as thousands of decoys and reflective metal scraps known as chaff, forming a threat cloud of up to 1 million objects. The challenge for a defense during this 20-minute midcourse flight is to pick the RVs out of this debris and disable them. Infrared sensors are ineffective at this stage, so recent research has concentrated on an interactive sensor, a stream of highly accelerated uncharged atomic particles that would penetrate an object and ''see'' what is inside. When these neutral particle beams hit a massive object like a warhead, gamma rays are emitted. Decoys, which have very little mass, give off virtually no emissions. ''When you get a signal,'' said SDI's Chief Scientist Gerold Yonas, ''it's the warhead. When you don't, it's a decoy.'' At present the paraphernalia needed to produce these beams is so large that it would be impossible to put in orbit as a fighting machine. Once the real warheads have been identified, they could be targeted for kill by laser weapons, intense beams of light that could destroy a Soviet missile or warhead by burning through its skin or, in a pulsating version, by hitting it with a sledgehammer-type blow. SDI scientists have been exploring the merits of deploying several types of laser weapons in space. Chemical lasers, generated by the reaction of gases such as hydrogen and fluorine, are now considered too unwieldy for space deployment. When they are ground based, their long-wavelength beam would be too ineffective to penetrate the atmosphere and make a missile kill. Today the hottest option is the free- electron laser, generated by the action of electromagnetic fields on electrons. Although it might also be too big to lift into orbit, this laser has a shorter wavelength, which gives it the potential to penetrate the atmosphere from the ground. Whatever ground-based laser weapons are chosen, their beams would have to be bounced off high-tech mirrors that would retarget them from space. Warheads that survive the boost and midcourse onslaughts hurtle toward earth in a ''terminal'' phase, the last 125 miles and the final two minutes of their mission. Back in the atmosphere, space-related problems no longer deter the defender. An RV can be detected by standard imaging radar and shot down, preferably with smart rocks. But little time remains once the RVs are spotted, which means a defense runs the risk of being overwhelmed. In addition, the Soviets could blind radar with nuclear bursts in the sky and skew targeting by outfitting their RVs with stubby wings that would allow them to maneuver and escape the defensive rockets. Like fast-burn missiles in boost phase and decoys in midcourse, stubby wings are just one of the available conventional methods the Soviets might use to counter complicated Star Wars technologies. Whatever obstacles SDI must overcome in developing sensors and weapons are dwarfed by the difficulty of coordinating these elements into an overall defense. Says SDI Critic John Pike of the Federation of American Scientists: ''The issue is not whether some individual SDI elements work, but whether it would work at the systems level.'' For SDI to be effective, battle- management computers must coordinate sensing devices, track myriad Soviet warheads throughout their flight, aim the U.S. weapons, assess the success of a hit and then retarget. In addition, the space-based portion of the SDI system, including the lethal smart rocks and beam weapons, would be orbiting the earth. All battle- management information, including detection and tracking of Soviet ICBMs, would have to be relayed from one satellite complex to the next, as parts of the entire affair moved alternately in and out of range of the U.S.S.R.'s threatening zone. The computer software needed to direct such a defense is vastly more complicated than any yet operating: Yonas estimates that up to 50 million lines of information code would be necessary. The space shuttle's operating software consists of only about 500,000 lines, and still computer glitches have resulted in frequent launch delays. Yet because the stakes are so immeasurably high, the SDI system would have to work perfectly the first time out. Yonas concedes that ''there is no way we could go into battle without a system that has been highly tested.'' But there is no way to test the system under real battlefield conditions. Final testing will be done in simulations by the ''national test bed,'' a supremely sophisticated computer-video model of a nuclear battle incorporating SDI components. The people who program the test bed must try to anticipate every countermeasure the Soviets might conceive. Said Stanford's Drell: ''You can't say to the Russians, 'Hold your attack. I'm not quite ready.' '' SDI advocates cite promising advances in complex technologies. But few scientific experts find it possible to put faith in the ability of such a system to operate in a nuclear showdown. At the TIME conference, Drell quoted a pertinent scene from Shakespeare's Henry IV, Part I: '' 'I can call spirits from the vasty deep,' says Glendower. 'Why, so can I, or so can any man,' replies Hotspur. 'But will they come when you do call for them?' '' Given the unique mission of SDI, and the stakes involved, that question is critical.

CHART: TEXT NOT AVAILABLE BOX: How to Master Starspeak When Star Wars enthusiasts speak of an engagement in the national test bed, they are not talking about what one might think, even when rubber mirrors and hardbodies are tossed in. The SDI program, making no effort to construct a multilayered shield for the English language, has launched a lot of new lingo. A sampler: National Test Bed: A high-tech computer and video operation designed to simulate space-battle scenarios. Hardbody: A targeted missile, often hidden from infrared detection by its huge plume of heat and gas. Smart Rocks: Small kinetic-energy projectiles that can be hurled at missiles or warheads. Pop-Up: The fast launch of a missile carrying a nuclear weapon to generate an X-ray laser that can shoot from space; because the system is not already in orbit, it is not vulnerable to a pre-emptive strike. Rubber Mirror: A computerized mirror of thin glass on honeycomb panels; the panels are controlled by microchips and mechanical arms that enable them to compensate for the distorting effects of the earth's atmosphere on laser beams. Interactive Discrimination: A system that transmits neutral particle beams to distinguish warheads, which emit gamma rays, from decoys, which do not. Red Team: A group of scientists within SDI whose task it is to develop and analyze possible Soviet countermeasures to the program. Space Mines: Orbiting explosives designed to threaten satellite defenses. Precursor Bursts: Nuclear explosions in space designed to foil defensive systems by creating a background of nuclear emissions, magnetic pulses and heat that can fool sensors.

With reporting by Michael Duffy and Bruce van Voorst/Washington