Nuclear Winter and the

End of the Arms Race

By Carl Sagan and Richard Turco

Random House. 499 pp. $27.95

THE GLOBAL warming of the Cold War has undercut the relevance of a variety of recent books. One example is A Path Where No Man Thought: Nuclear Winter and the End of the Arms Race by Carl Sagan and fellow atmospheric chemist Richard Turco.

Back in 1983, Sagan alarmed the world when he said that a nuclear attack would generate enough soot and dust to make the noon-day sun no brighter than a full moon, lowering the temperature as much as 35 degrees Centigrade, freezing plants and animals. Ultimately, mankind itself might be starved into oblivion in a nuclear winter.

Fears of global nuclear war have diminished dramatically in the last few years. Concern about possible nuclear winter has been replaced by fear of the greenhouse effect -- the slow warming of the Earth because, in large part, of man's burning of fossil fuels, resulting in rising sea levels, shifting climate and growing desertification of agricultural lands. Our accelerating desire for fossil fuel-based industrial growth might lead to starvation and suffering -- a long hot petroleum summer.

For Sagan and Turco, however, nuclear winter remains too great a threat to ignore. They offer the following calculation: "The lost productivity of people killed by nuclear winter can be estimated from the gross World Product as no less than $10 trillion a year, so that if . . . it required a century to reestablish the equivalent of the current global technical civilization, the cumulative productivity loss would be well in excess of $1,000 trillion." By then assessing the "likelihood" of global nuclear war "even as low as 0.1 percent per year, it would then follow that we should be spending a trillion dollars a year to prevent nuclear war" (since 0.1 percent of $1,000 trillion is $1 trillion).

The book's primary flaw is the abundance of such pseudo-calculations. The estimate assumes that all global nuclear wars result in a nuclear winter that destroys global civilization for a century. Yet this scenario requires, according to Sagan and Turco, each side to detonate 3,000 to 6,000 nuclear warheads on the other, and even so, the effects of those explosions must be on the severe end of what scientists now think likely. Rather than assess a probability, it is perhaps best to say that this is a "highly remote" outcome for global nuclear war, as Steven Schneider of the National Center for Atmospheric Research puts it. Most scientists now believe that a post-nuclear war climate change would be a nuclear "autumn" that wiped out crops for a year, rather than the cataclysm Sagan and Turco describe.

Nevertheless, Sagan and Turco use their calculation to argue, "In summary, we can say that the risks posed to the human species by nuclear winter at the most probable levels of severity now scientifically established are unacceptable by any standards that are currently applied to environmental issues." That is, we should worry more about nuclear winter than global warming.

The attempt to overplay the effects of nuclear winter runs through the book. At one point Sagan and Turco imply that the direct effects of nuclear weapons -- the initial blast and fires -- will kill only the quarter of a nation's population that lives in cities, while "nuclear winter is a way for nuclear weapons to find and kill those who live far from cities." Yet in the Soviet Union and United States, between 50 percent to 60 percent of the people live in cities, and the radioactive fallout alone can kill those elsewhere. All this could be seen as splitting hairs were it not for the serious consequences the overselling of nuclear winter has had.

Consider the following confused New York Times editorial from earlier this year about a disagreement between Environmental Protection Agency chief William Reilly, who wanted to commit the United States toward stronger action to slow the greenhouse effect, and White House Chief of Staff John Sununu, who "is said to be skeptical of the computer simulations on which forecasts of global climate are largely based." According to the editorial, "The simulations are full of uncertainties, as their creators admit. The same computer programs once predicted that weeks of icy darkness would follow a nuclear war, an estimate now revised to a mild chill. The forecasts of greenhouse warming are not yet robust enough to form the basis of policy." The statement is doubly incorrect. The revised estimate -- nuclear autumn -- kills tens of millions of people, not exactly a mild chill. Far more important, the computer simulations that suggest there will be global warming are not the simplistic ones that initially overestimated the effects of nuclear winter. Quite the reverse, it is the more sophisticated computer programs -- three-dimensional models of the atmosphere -- that revised nuclear winter into nuclear autumn and that now raise the specter of global warming. Unfortunately, given how scientific results are often prematurely and incorrectly presented to the public the truth matters less than perception.

The authors do not acknowledge any responsibility for the public's misperceptions. Instead, they spend much of the book trying to explain how they never overestimated nuclear winter and how that theory must be the key element of any reformulation of nuclear policy. Yet, one need not believe in nuclear winter to believe in the need for drastic reductions in nuclear arsenals.

If Sagan's popularization of nuclear winter has had costs, it also has had undeniable benefits. Sagan rightly claims credit for helping to spur funding of atmospheric computer models that are now being used on the greenhouse effect. By raising awareness of the dangers of nuclear war, he has perhaps reduced the likelihood of such a war and, in so doing, made books like this one increasingly unnecessary. So, contrary to what Sagan and Turco argue, we should be less concerned about nuclear winter than global warming.

America is engaged in a conflict in the Persian Gulf to maintain easy access to petroleum, so that we can continue our profligate overuse of it as the world's single largest producer of greenhouse gases. We are in Saudi Arabia not to make the world safe for democracy, but, at least in part to make the world safe for global warming.

As we reformulate America's security policy in the new world order, it should be with a clear eye toward energy and environmental security, not to the perils of a global nuclear war, now fortunately much less likely.

Joseph Romm is assistant for international security to the president of the Rockefeller Foundation.