The sky is so big here that the towering storm clouds seem to shadow distant continents. The coppery bluffs and blue hills in the distance most certainly hide magic amid the desert scrub and creosote bush, and so it is no real surprise when the white, white discs of an obvious outer space visitation appear over the rise.

There have, indeed, been several reports of spaceship landings in this area, which boasts a thriving group of UFO buffs, but no Andromeda expedition probably looked more awesome than these gleaming monsters, spread out and listening to God only knows what, somewhere above the desert floor.

They are, of course, radio telescopes, a Very Large Array of them laid out just so, monitoring the stars while birds roost in their grids and cattle rub against their struts below.

The Very Large Array, for that is what it is called, is the most advanced outpost of radio astronomy, the science that studies the radio waves bombarding us from every direction of outer space. When completed in 1981 at a cost of the sponsoring National Science Foundation of $78 million, the VLA will involve 27 moveable radio telescopes plus a spare.

Each will be 85 feet across and hooked with the others by a computer into the equivalent of one mammoth ear 26 miles wide. Such a radio dish, if it existed, would just fit inside the 1-495 loop of the Capitol Beltway, which is about 65 miles around.

The VLA, main facility of the National Radio Astronomy Observatory, headquartered in Green Bank, W. Va., will be under the direction of a consortium known as Associated Universities Inc. About 100 scientists will work in Socorro full time.

The peculiar virtues of New Mexico have brought science here in force. It is dry, high, mostly flat and sunny 300 days of the year, with vast expanses of empty desert suitable for being bombed with negligible damage and a minimum of public notice.

The very names of the research centers evoke the thunder of the bombs and missiles tested: Los Alamos, Alamogordo, White Sands. "We're different," said John H. Lancaster, manager of the VLA construction project. "Radio astronomy is a good, clean industry - all we do is listen."

The latitude is southerly enough to provide a good look at most of the sky; the site remote enough to discourage future development that would mean interference from TV sets and automobile spark plugs in great numbers. With 60 percent of the atmosphere lying below Socorro's 7,000-foot altitude, atmospheric distortion is minimized. Fully 34 sites were examined before this one, on the Plains of San Augustin, was chosen. Biggest and Best

THE VLA IS, without argument, the biggest and best of the world's 80-plus radio astronomy installations. Scientists, already lined up several months in advance, hope to run observations and experiments to learn more about the structure of the universe, the life cycle of stars and the ultimate big questions: Where did all of it come from? How much of it is there? Where is everything going, and what are the rules of the journey?

Such issues, the concern of religion as well as science, have long been the province of the visual astronomers with their monster mirrors and sensitive cameras. The universe, however, speaks to us in every part of the electromagnetic spectrum, not just in visible light waves. It broadcasts on cloudy days when Mount Palomar's 200-inch optical telescope is blind, and it speaks from places where no light shines.

Aware of these unheard messages back in 1890, Thomas Edison proposed ringing a field with telephone wire in order to listen to the sun. It was never done (scientists now say the arrangement would not have been sensitive enough to work anyway), but the idea was the father of the white discs that now stretch to the horizon in central New Mexico.

The antennae, like giant erector sets, are assembled in a 10-story, three-sided hangar at the rate of one every seven weeks, their progress watched over by families of owls roosting in the rafters. "The owls are mascots. They eat the mice that carry the fleas that still carry bubonic plague around here," Lancaster said.

When finished, each 211-ton telescope is furnished with electronic gear so sensitive that some of it, the part that amplifies signals from space, is supercooled to - 427 degrees Fahrenheit in order to reduce interference from the machinery's own parts. Amplification is necessary; it has been estimated that all the radio waves ever collected from outer space by radio astronomy total less energy than the amount used by a cigarette ash falling 18 inches into a tray.

Once armed to view the heavens, the telescopes are trundled down ordinary railroad tracks to one of the 24 precisely measured observation sites on each of three arms of a giant Y laid out on the desert floor.

The sites are unevenly distributed along the arms of the Y, two of which stretch 13 miles from the control hub while the third, pointing not quite due north, is 11 miles long. A special 36-wheel monster transporter can move any of the antennae to any of the 72 observation sites, making possible a variety of arrangements for narrow focusing or broader sweeps of the sky. Computer Dependency

THE ENTIRE PROJECT depends to such a high degree on computers that some of the people working here occasionally feel irrelevant. The choice of the site, the shape of the grid, its skew from due north and the observation station, layout were picked by computer, all with the object of getting the most out of the rotation of the earth in helping to scan the universe.

The computers send out instructions for one one-thousandth of a second, or millisecond and receive information from the antennae for the next 50 milliseconds. All is taped (signs on the machines read "RIP" for Recording In Progress) and atomic clocks mark the tapes for precise coordination.

"We just watch the monitors to ensure the credibility of the information that comes in," said operator Kerry Hildrup, an astrophysics graduate of the University of Virginia. He admitted he found the job boring at times. That plus the isolation and the desert climate have caused a fairly high turnover among personnel, Lancaster said: 19 departures this year out of 120 staff members.

The millions of calculations each minute must allow for such irregularities as the variation in size among the sunlit and the shaded support pillars of the antennae. The computers also figure in the earth tides, or the rising and falling of the earth's surface with the pull of the moon's gravity, and the difference in that tide from one end of the VLA to the other.

The earth is slightly unsteady in its orbit, and that makes a difference."It takes 26,000 years for the earth to make one wobble around the pole, but we can see the effect here in a couple of hours," said Campbell Wade, assistant director for VLA operations of the sponsoring National Radio Astronomy Observatory.

The omnipotent computers have a degree of personality. There is "Boss," which coordinates and oversees, "Monty," which monitors the steady pulses, and "Cora," the collector, and "Corbin" the coorelator. The Problem of Size

IN THE WAY that two eyes see things from slightly different angles and synthesize the difference into depth and detail, or the way that two ears make it possible to determine the direction of sound, so pairs of telescope can be hooked together to achieve the detail of one unit the size of the distance between them.

This technique, called aperture synthesis, revolutionized radio astronomy when British scientist Sir Martin Ryle came up with it in the 1950s, he later won the Nobel Prize for his work. The problem with radio astronomy during its entire short life has been that radio waves, longer than light waves, make a much fuzzier map of their source when graphed on paper than starlight does on firm. To equal the clarity of photograhs from Mount Palomar, a single radio telescope would have to be 17 miles in diameter.

With its 27 antennae, the VLA simulates a telescope bigger than that. It will be able to make 351 different pairings to examine a radio source from as many perspectives as possible.

"If everything works properly," said Peter Napier, head of VLA's electronics division, "we should get the kind of resolution that could read the writing on a dime at the distance of 1 mile."

The VLA's nearest competitor is a 12-antenna hookup at Westerbork in the Netherlands, but those units are mounted in a straight line and so take longer for the earth's rotation to "close the circle" and provide a full picture. The VLA can complete an observation in about eight hours.

With only 12 telescopes hooked up so far, the VLA is already producing images that for the first time are equal to photographs in clarity. It is also producing images, in the initial form of grids of millions of numbers, of radio sources that optical telescopes cannot see at all.

These are made visible by IMPS, interactive map processing systems, which translate each number into a color and produce a gaudy psychedelic light show among other displays. Computer programmer Jim Torson operates the IMPS to show astronomers just where in the radio source the transmission was strongest. "I like to think of it as helping a human to understand it, to see it," he said.

The VLA already has produced discoveries that are puzzles to the astronomers. One, located by astronomer Robert Hjellming, is a "head-tail" galaxy called NGC 1265, which is 250,000 light-years across and invisible to any optical telescope. The head appears to be leaving a wake behind, like a speedboat, of electrically charged particles off to either side.

The startling thing is that one side of this wake has a major dent in it. What monstrous force or obstacle out there could divert the tail in such a manner?

Another mystery concerns an apparent quasar, or quasi-stellar radio source, located closer to our own Milky Way galaxy than anyone had previously thought possible. Solutions to the source of quasars' powerful energy may be brought nearer with further VLA work.

Other puzzles have been more down to earth. One antenna stopped functioning suddenly about 18 months ago because of a steer.

The steer, his back itchy, simply rubbed himself a bit too enthusiastically against the tubular wave guide that carries the signal from the antenna on the next-to-last leg of its journey from the stars to the computer. The tube, the diameter of a 50-cent piece at that point, snapped off. The steer was unhurt. Fences have since been installed around all the antennae.

"I wish all our phenomena were that simple," said Lancaster.