Washington — The National Academy of Sciences, in a study commissioned by the Department of Defense, gives credence to the scientific theory that nuclear war could lead to nuclear winter.
In the words of the study released yesterday, there is a ''clear possibility'' that in the event of a nuclear war the sky would darken over much of the Northern Hemisphere, causing ''severe'' temperature drops that might last for months.
If true, the nuclear-winter scenario could result in vast changes in US military strategy, experts say. It raises a number of thorny questions: Is the launching of nuclear weapons, even against an undefended enemy, self-destructing? Is nuclear war thus ''impossible,'' as some scientists claim? Or will superpowers be tempted to launch limited first strikes, in an effort to head off a full-scale conflagration?
''Even a small possibility of (nuclear winter) must be considered very seriously, and the administration is studying the potential implications of this new facet of nuclear war, both for our strategy of deterrence and for our strategic modernization program,'' said Richard Wagner Jr., assistant secretary of defense for atomic energy, at a congressional hearing earlier this year.
The nuclear-winter theory was first proposed one year ago, at an international conference, by a group of scientists that included James Pollack of the NASA Ames Research Center and Carl Sagan of Cornell University. In essence, the theory states that a major nuclear exchange would raise so much debris and light so many fires that dust and smoke would block the sun for months on end, lowering temperatures as much as 75 degrees F. and causing severe environmental damage.
Some nuclear scientists - notably Edward Teller - dismiss the theory as pure guesswork. But the Reagan administration takes it seriously. Last February, the President's science adviser, George Keyworth II, ordered the National Oceanic and Atmospheric Administration to plan an organized campaign of study of the climatic effects of nuclear war. The Defense Nuclear Agency has begun spending about $2 million a year on research into the subject. The National Academy of Sciences report was the first fruit of this DNA money.
The report says that much study needs to be done, that any conclusions it draws about the climatic effects of nuclear war are ''emphatically of an interim character,'' and points out that there is ''already a sobering list of relatively well understood consequences'' of a nuclear weapons exchange.
But the study concludes that in the event of nuclear war sunlight-obscuring plumes would quickly cover much of Eurasia, North America, and the North Atlantic. If the attacks occurred in the summer, mean continental temperatures in the Northern Hemisphere could drop as much as 10 to 25 degrees Centigrade, the report says. There would be less variation in winter.
''There is a clear possibility that great portions of the land areas of the northern temperate zone (and, perhaps, a larger segment of the planet) could be severely affected,'' the report says.
Nuclear warheads, particularly those set to explode at ground level, would raise tremendous amounts of dust high into the air, the report says. Large amounts of undesirable chemicals, such as carbon monoxide, would be released.
But the most dire environmental effects of nuclear war would come not from dust or chemicals but from smoke, rising from massive fires in cities and forests, concludes the National Academy of Sciences study. For their baseline, NAS scientists assumed that about half the world's nuclear weapons would be detonated in any major exchange, igniting firestorms in 1,000 of the world's largest cities.
Tiny black smoke particles from such fires would sponge up the sun's rays very effectively, and could block a considerable portion of the heat and light that normally reach Earth's surface.
''Light levels could be reduced by a factor of 100 in regions'' covered by nuclear smoke, says the report.
Though target areas would feel the brunt of these climatic effects, southern areas of the world might be affected, too: ''Significant amounts of dust and smoke could drift to and across the equater as early as a few weeks after a nuclear exchange,'' the NAS study concludes.
Among the questions that still need much study, say NAS scientists: How long would these effects last? How much smoke will nuclear fires generate? How high will this smoke go, and how long will it stay in the air?