Skylab is coming home to earth -- all 79 tons of it sometime around July 13. As been reported steadily over the past six months, when it does come down it will spray debris along a tract 200 miles wide and up to 4,000 miles long, somewhere between the latitudes 50 degrees North and 50 degrees South -- a belt that excludes the north of Canada, most of the Soviet Union and, oddly, the British Isles. Fully 90 percent of the world's population will be threatened by Skylab's descent, and that descent isn't far away. It is time, now, to decide that kind of warning we will give to people who will be jeopardy during its final hours.

Much of Skylab's debris can be expected to burn up a reentry, and there is a reasonable chance that the surviving pieces will fall into the board reaches of the world's Ocean. Still, a substantial possibility remains that some of the pieces of Skylab will fall on populated land.

This is the first satellite reentry to drop pieces that could cause a heavy loss of life. Following basic aerodynamic data, solid forged-metal remnants weighing above 1,000 kilograms can be expected to hit at more than the speed of sound. A metal piece the size of Chicken Little's acorn would have about the effect of a construction rivet dropped from the top of the World Trade Center. Those larger chunks could also create sonic booms. Anyone who has questioned the Concorde flights can sympathize with the people caught in such a cacophony.

Furthermore, one of the large objects, travelling at such speed, might collide with an immovable object -- like a building foundation -- and generate terrific heat and an explosion. The devastation would cover a wider area than just the point of impact. The larger parts from Skylab are, in effect, capable of producing damage as great as non-nuclear weapons.

NASA has estimated, based on mathematical models and experience with descending communications satellites, that Skylab will break up into not more than 500 pieces of significant size -- that is, weighing a pound or more. This may be so. The data is incomplete. But pieces weighing under a pound can also be lethal. Competent analysts guess that there will be upwards of 5,000 such pieces of debris, and possibly as many as 30,000 total items at earth impact.

NASA has said that there is only one chance in four that Skylab will come down over land, since three-quaters of its orbital path is over the ocean. This sounds reassuring, but what about that fourth chance -- for example, a "footprint" of descending debris crossing through Chicago and Charleston, of Vienna and Tehran, or striking New Delhi or Peking? Based on the official 500-piece estimate, NASA has projected that the odds of hitting at that is not reassuring.

In the face of this inexorable oncoming event, there isn't much left to be done. All of NASA's heroic efforts in 1977 and 1978 to maintain Skylab aloft in its orbit -- until the Space Shuttle could be brought along -- are now of no avail. The shuttle's first engine blew itself to bits in April, during the first efforts to get in a 500-second "firing" test.

But it will be possible this month to give advance warnings in areas where Skylab could fall, and to give those warnings in ample time -- for example, 36 hours -- for orderly cautionary measures to be taken.

Unfortunatley, at present, NASA's official plan is to relay only timing and orbital path data, as projected with successively increasing accuracy by NORAD's Space Defense Center in Colorado Springs. These figures will be delivered within the United States to the Federal Preparedness Agency, FAA and the news media. Foreign governments and other institutions will depend on a second relay by our Department of State.

This focus carries two problems. First, the technical issue of projecting exactly where Skylab hits sould be less important than the goal of minimizing the chances of its killing someone. By releasing information in such a way as to emphasize the final impact projection -- effictively a two-hour warning -- the opportunity for everyone to get in a thorough response would be cut, in comparison with issuing full warning schedules beginning at the 36-hour mark. Also, the two-hour warning has provoked professional psychiatric objections that such an annoucement might trigger an urban panic -- certainly a result that would overshadow the statistical danger from Skylab.

The second problem is with the raw data itself. Skylab's descent will be accompanied by a slowing down, as the space station hurtles into the atmosphere. Its velocity will reduced from 17,650 mph to less than 1,500 mph. As a result, the pieces will not be able to maintain their track along the normal orbital path; they will fall from far to the west. Few, if any, of the receiving organizations being served by NASA and State have the computing programs on hand to convert the orbital path figures to actual risk patterns. Considering errors, the area at risk is 500 miles wide.

To reach people effectively, a broader warning system must be put into operation. At the 36-hour point, timings for the beginning of reentry [the start of Skylab's slowdown] can be narrowed to a 12-hour period. The satellite's position and direction can be plotted continuously for this whole period of risk. Then, mathematical functions for reentry projection can be included, together with adjustments for engineering uncertainties.

The overall result from the 36-hour mark would be risk-pattern schedules for everyone who might be hit. If that 4,00-mile long "footprint" could fall on a country, or a city, its inhabitants would be told -- to within an hour, at most -- when any lethal debris could be expected to fall. For the case where the satellite missed -- by falling earlier or later on, further down the risk pattern -- then the worst will have been for someone to have stayed indoors unnecessarily.

On the other hand, this is summer of most of the people under Skylab's orbit -- we are outdoors a lot and more than usually vulnerable. Perhaps that warning to those in the threat zone to stay indoors, if they judge the odds harsh enough, would save a life. Surely, foreign airlines warrant this special attention. There's not much we can do about people's property.

Skylab is American; it has "made in USA" all over it. We are liable worldwide for damage or injury it might cause. But that is only the start.

With all the $50 billion invested in our space program, it would be demeaning to let this predictable event cause unnecessary loss of life. It is imperative that we provide to the world at large an adequate warning, as many hours in advance as possible, of the incoming threat. We have created it, through our own mistakes, and we are morally responsible for it. CAPTION: Illustration, no caption