Nuclear Winter: U.S. Government Thinking During the 1980s

The work of physicist Harold Brode at the Pacific Sierra Corporation influenced nuclear winter research, Members of the TTAPS group had worked with Brode and the group cited his work in the original December 1983 Science article hypothesizing about the phenomenon. Brode was less interested in the global impact of fire effects than in the problem of developing predictive models that could fully account for the local impact of nuclear weapons effects. He and others at PSC saw such techniques as essential because of “the enormous carnage and confusion accompanying a nuclear attack, and the unprecedented potential for casualties and damage from the subsequent fires.” Looking back at the destruction of Hiroshima and Nagasaki would support that perspective, he argued, because “the damage due to fire was much greater and more complete than that due to blast from the nuclear bombs or from equal tonnages of high-explosive bombs.”

Brode and his colleagues further explained that the “vast amounts of gas, smoke, hot air, and ashes generated by the fire may cause high casualties” and “the very size of the burning region precludes escape for most of those caught within the area.” The existence of high yield thermonuclear weapons made predictive techniques all the more important since the point was “not only that much damage is not counted, but also that much larger attacks than necessary may be planned.”

This study is difficult for the mathematically-challenged (such as the present writer), but the text clearly conveys the importance of Brode’s effort. In this study Brode and his colleagues sought to “delineate the damage expected from a superfire, and … [to] aid both those planning or assessing nuclear weapon attacks and those planning defense against such attacks.” They supported the development of computer programs that could make it possible to make “general predictions of the extent of fires in unspecified urban areas,” predictions that would be “as reliable as those for blast damage.”

By the late 1980s Brode had developed predictive models for nuclear fire-storms that could be incorporated into target planning and provide for more reliable understanding of the destructiveness of nuclear weapons. While discussion with officials at the Pentagon and Omaha would go far, in the end military officials chose not to adopt Brode’s models, partly for budgetary reasons, but mainly because they were not necessary to achieve the requisite levels of damage – which was after all their assigned objective– without considering fire effects or having to think deeply about how much devastation the weapons could actually cause.[10]

Michael MacCracken, a physicist and climate expert at Lawrence Livermore Laboratory, played a central role in the lab’s research on nuclear winter. In this conference paper, representing his initial take on the recent research by TTAPS and other analysts, MacCracken wrote that the climatic effect of nuclear war that was of “most concern … is the very large, nearly instantaneous injection of soot … which is a strong absorber of solar radiation.” The resulting “hundred to thousand fold increase in atmospheric soot concentration would prevent sunlight from reaching the surface over those latitudes wherever substantial amounts of the smoke spread.” While the scenarios that could produce soot concentrations were uncertain, they indicated the “potential for substantial impacts on the hemispheric, and perhaps global, radiation balance and climate.” Thus, McCracken estimated that nuclear war could produce “substantial cooling … that would last weeks to months over most continental regions of the Northern Hemisphere.” This finding, significant cooling if not necessarily a deep freeze, would inform his later work on the subject.

In early 1983, while the TTAPS authors was preparing their Science article and presenting their findings, the Department of Defense tasked the National Research Council (the operating arm of the National Academy of the Sciences) to study the “possible atmospheric effects of nuclear war." The Council tasked two committees, a Committee on the Atmospheric Effects of Nuclear Explosions and a Commission on Physical Sciences, Mathematics, and Resources, to prepare a report. By late 1983, they had prepared a draft that they shared with external reviewers for feedback. Comments by Livermore’s Michael MacCracken have been released and illustrate an aspect of the peer-review process involved in the production of such studies. The text that MacCracken was commenting on is unavailable, but his serious and encouraging observations showed a collegial approach.

Having no doubt that a major nuclear war could have “potentially very severe derivative consequences,” McCracken nevertheless found that the report did not “describe consistently the uncertainties and underlying assumptions regarding our present knowledge” of those effects. He was especially concerned about emissions from fires, optical properties of smoke, and climatic effects. For example, with respect to the latter, he argued that “we have so far only the very crudest indications of what will occur.”

The Defense Nuclear Agency, which was responsible for surveying weapons effects, was initially reluctant to support a comprehensive research effort.[11] But seeing “serious potential for severe consequences” from a large-scale nuclear exchange, and also “very large uncertainties” about the impact, DNA initiated a study to establish more certain knowledge. Besides seeking to develop a “reasonable predictive capability” to determine the effects, DNA also wanted to establish a “scientific basis” for gauging the impact on the climate and the environment. Interestingly, one of the charts [PDF p. 15] includes the phrase “atmospheric trauma” to characterize the cooling impact of the smoke produced by nuclear explosions.

The problem of nuclear winter briefly emerged in a talk between Reagan and British Labor Party leader and nuclear critic Neil Kinnock. The two had a friendly discussion, with each arguing for their positions, although Kinnock credited the recent U.S “build-down” position and both agreed on the need to prevent further nuclear proliferation. Kinnock mentioned the “widely discussed” idea of nuclear winter in his comments on the danger of nuclear proliferation, and Reagan later acknowledged that a “‘nuclear winter’ was theoretically possible.”

Elements of the Department of Energy such as Lawrence Livermore Laboratory were investigating “nuclear winter” before it became a public issue. Stimulated by recent research on the possible environmental impact of nuclear war, especially by the TTAPS group and a 1982 article by Paul Crutzen and John Birks, scientists at Livermore, including Michael McCracken, had begun a full research program by late 1983. This report provides useful background on previous investigations starting with a 1945 study on the “Ignition of the Atmosphere with Nuclear Bombs," the atmospheric fallout research of the 1950s, and a 1975 National Academy of Sciences study on the global effects of nuclear war. With the problem of soot and smoke already highlighted by recent publications, Livermore initiated a major research program directed by Michael May involving six tasks, including nuclear war scenarios, fire ignition, “microphysics and chemistry” of material injected into the atmosphere, and atmospheric modeling. For example, with respect to war scenarios, LLNL would investigate the “fuel loading and fire potential of typical targets,” especially cities. This report provides details of the ongoing work on the six tasks.

As the NAS prepared its report (see related item, Document 3), Michael MacCracken provided more detailed observations on the latest draft. His comments were again detailed and prefaced with congratulations: the draft is “greatly improved and should be issued.” But he expressed some criticism, for example, that the text had underplayed “the potential interactions between the various individual influences (smoke, dust, and chemistry).”

This report provides a detailed account of the follow-up of the Livermore task groups studying aspects of the nuclear winter problem. For its work, the atmospheric modeling group was using Oregon State University’s Global Circulation Model (GCM) (which would soon be used to track climate change). In addition to the work that Livermore was conducting, the White House science adviser had asked the National Oceanic and Atmospheric Administration (NOAA) to develop a national plan, to be completed in September, for five years of research. Moreover, the Defense Science Board had established a small task force to study “Atmospheric Obscuration.” Michael May and others with the lab had also attended conferences and gave presentations in London and the Soviet Union.

While U.S. and other scientists were advancing their initial findings about nuclear winter, Soviet officials and researchers began making similar claims.[12] U.S. intelligence community analysts were not sure how seriously the Soviet leadership was taking the subject but believed that it “will remain interested” because of its “profound implications” for their strategic position. According to the analysts, Soviet nuclear winter research was derivative and used by political leaders to justify their positions on nuclear weapons. So far, Soviet science had not made important advances in nuclear winter research or even developed “realistic Nuclear Winter climate modeling,” in part because of “limited computer capabilities.” “Even the fastest Soviet scientific computer, the El’brus-1, is still less than one-tenth the effective speed of a Cray.” Crays were then the world’s fastest computers. The report was scornful of Soviet research as “fall[ing] far short of normal scientific standards” and amounted to “more replication than verification” of research findings elsewhere “because it lacks original Soviet data or models.”

So far, U.S. intelligence had not gleaned any information on secret Soviet research on nuclear winter. The U.S. had new intelligence from a recent defector, but information on the “Soviet Nuclear Winter Network” in this report has been withheld in its entirety. The analysts suspected that “officials in the weapons development structure” were following the work being done at the Soviet Academy of Sciences. Soviet nuclear planners would want to “know to what extent their strike plans and US retaliatory strikes would lead to adverse consequences in the USSR.” If the Soviets ever put any significant resources for research into the NW problem it has not been disclosed.[13]

Whatever was going on behind the scenes, the analysts argued that the Soviets found it politically beneficial to trumpet concern about nuclear winter. However, nuclear winter played out politically and diplomatically, the analysts did not foresee that it would produce “any changes to Soviet nuclear weapons policies or programs” because the “scientific evidence is not yet convincing.” Yet, if Moscow came to “accept Nuclear Winter effects as both credible and profound, it could lead to serious contradictions between these new considerations, on the one hand, and Soviet doctrine and weapons employment policies on the other.”

A puzzle in the story of Soviet nuclear winter research is the disappearance of physicist Vladimir Aleksandrov. In 1985, while in Madrid for an international conference, he vanished without a trace. The case was never solved.[14]

During a visit to Washington, Joseph Knox of LLNL had acquired a copy of the “prepared release,” the summary and conclusions, for the NRC/National Academy of Sciences study on nuclear winter. On the text, he made a number of marginal comments, which he acknowledged to the readers of his cover memo, might be read as conveying the “belief [that] the report could have been better.” But he was far more positive: “I am relieved that the NAS report is as good as it is.” He remained concerned that some of the uncertainties about smoke and soot emissions from fires “were deemphasized a bit.”

The NRC report issued a few weeks later avoided definitive language or quantitative conclusions, but the committee found that a nuclear exchange involving 6500 megatons (half of the world’s stockpile) could have “long-term climatic effects with severe implications for the biosphere.” The nuclear exchange would “insert significant amounts of smoke, fine dust, and undesirable chemical species into the atmosphere,” and the “depositions could result in dramatic perturbations of the atmosphere lasting over a period of at least a few weeks.” There was a “clear possibility that great portions of the land areas of the northern temperate zone (and a larger segment of the planet) could be severely affected,” with possible effects including “major temperature reductions (particularly for an exchange that occurs in the summer) lasting for weeks, with subnormal temperatures persisting for months.” Uncertainties notwithstanding, the report thus suggested even greater danger from nuclear war.

Analysts at the RAND Corporation were among those in the defense establishment who looked at the nuclear winter problem. Jeremiah Gertler was disturbed that some commentators had concluded that nuclear winter had no immediate implications for policy because it was just another “horror” or as the “ultimate deterrent” it did not require compensatory action or was unproven and not worth considering. After reviewing the TTAPS analysis of nuclear winter, Gertler concluded that, even if the theory could not be proven, if it became “widely accepted” or remained “unproven,” it had policy implications that needed to be considered.

Gertler’s paper focused on several problems. For example, nuclear winter posed serious challenges to the credibility of U.S. deterrence. Popular acceptance of the theory could give NATO allies “much greater reason to doubt our willingness to defend them with nuclear weapons.” That U.S. allies, the U.K. and France, had their own nuclear forces and the capacity to “initiate a nuclear winter” could also have a “detrimental impact on the future of the NATO alliance.”

Gertler suggested fixes to U.S. strategic attack policy that could reduce nuclear winter effects, for example, by avoiding highly combustible cities, using lower-yield weapons, earth penetrators, and enhanced radiation weapons (“neutron bombs”), and by making high altitude detonations that would have an electro-magnetic pulse effect. Moreover, if policymakers assumed that nuclear winter was possible, in the event of a nuclear crisis National Command Authorities needed an estimate of the “boomerang effect of each use of nuclear weapons” and “the likely effects on the United States of every given detonation before they issue orders.”

In terms of comparative conduct, if both Moscow and Washington believe that an unsurvivable nuclear winter could occur, “the constraints on both nations' behavior should be similar.” If one or the other government, however, did not take nuclear winter into account, it might be “less restrained in its nuclear-use policies than one that does, “as it might believe that a nuclear war could be fought and won.”

In contrast to Gertler’s report for RAND (Document 10), the Pentagon’s nuclear winter report sent to Congress over the signature of Defense Secretary Caspar Weinberger in March 1985, was, as Lawrence Badash put it, “cheeky” in arguing that, even if the nuclear winter thesis was correct, it required no change in U.S. nuclear policy. Acknowledging the possibility that nuclear war could induce climate change, the Weinberger report emphasized the uncertainties: “for any major nuclear war, some decrease in temperature may occur over at least the northern mid-latitudes. But what this change will be, how long it will last, what its spatial distribution will be, and, of much more importance, whether it will lead to effects of equal or more significance than the horrific destruction associated with the short-term effects of a nuclear war, and the other long-term effects such as radioactivity, currently is beyond our ability to predict, even in gross terms.”[15]

Much of the report was descriptive of research programs under way and some of it was argumentative, with some criticism focused on the TTAPS paper and its one-dimensional model, even though subsequent research had gone beyond it.

The Pentagon argued that the nuclear winter theory did not require any change in the U.S. nuclear posture: deterrence policy, including the Strategic Defense Initiative, were “fundamentally sound.” The report acknowledged that “reducing unwanted damage must be an important feature of our policy.” Thus, the U.S. had reduced the total number of its nuclear weapons, cut the total yield of the stockpile, and had been developing more accurate nuclear and non-nuclear systems. While reform of nuclear targeting procedure was underway at the Defense Department, it was actually not until the late1980s that a serious effort began to scale down the number of nuclear weapons in U.S. war plans and to reduce the level of destructiveness.[16]

For its chief Senate sponsor, the Pentagon report was a letdown. Senator William Cohen (R-ME), who had requested a detailed and thorough evaluation of the nuclear winter problem, pronounced that “In view of what I asked for, it's unlikely that they could have made an evaluation in 17 pages.” It had not responded to the law, did not discuss the possible impact of nuclear winter on military operations, and evaded the required full discussion of biological or environmental effects. White House science adviser George Keyworth found the report to be “mostly mush.”[17]

The Pentagon produced another report the following year, which key Senators also found wanting, but it is not publicly available. [18]

The reports that the various task groups prepared at Livermore Laboratory (described in Documents 5 and 7) are unavailable, but in an LLNL publication Michael MacCracken wrote this review of the research on fire, smoke, and climate. An innovation he employed was the use of three-dimensional models “to address some of the shortcomings in the early climate calculations.” The models included the Oregon State University General Circulation Model (see Document 8) and a model developed by the Lab, GRANTOUR, which could calculate the movement of smoke particles and the processes that influenced their concentration and distribution. Both the GCM and GRANTOUR models could be “coupled in a fully interactive manner.”

With the TTAPS study focusing on urban targets, Livermore was working with DNA to look at target scenarios in detail. Some targets, such as ICBM sites in prairies, would not generate much fire, while “the proximity of military targets to developed areas and forests would … alter the number and kinds of fires started.” With its modeling efforts, LLNL could forecast the distribution of smoke during the days and weeks following a nuclear war that injected 150 teragrams (150 million tons) of smoke into the troposphere. One conclusion was that “our interactive simulations predict relatively large decreases in surface temperatures for several geographical regions. Because these sharp temperature drops start from summer rather than global-average conditions, however, we do not find indications of the extended deep freeze predicted by TTAPS.”

That finding was not meant to be predictive: the recent work was in the nature of “sensitivity studies of what processes are important in determining the potential climatic perturbation of large amounts of smoke in the atmosphere.” MacCracken concluded that much more effort was necessary to “understand whether substantial cooling is likely or is a remote possibility.”

Taking an even-handed approach to the nuclear winter problem, the GAO delivered a report to Congress that assessed it as a “plausible theory with numerous uncertainties in critical areas such as war scenarios, fire research, and climate modeling.” Exemplifying the impartial approach was the assessment of the original TTAPS paper: Noting that TTAPS had been subject to various criticisms (e.g., for using annually averaged temperatures and neglecting heat transfer from oceans to land) and the uncertainties about physical and chemical effects, GAO nevertheless found that the authors were “successful in focusing attention on a potentially important phenomenon and giving added support for studying the nuclear winter theory.”

The GAO looked closely at the state of nuclear winter research and its prospects. As noted, it pointed to areas of uncertainty where further research was necessary, such as nuclear war scenarios, fire characteristics, smoke production and transport, effects from dust, toxic chemicals and gases, and the biological/agricultural effects. Improvements in computer modeling, which necessarily require high-speed machines, was an important research area but was limited by lack of data on smoke and fire. “Until scientists receive more accurate data to apply to their computer climate models and improve their software to simulate atmospheric responses more accurately, modeling results will only show a plausible range of nuclear winter effects under strictly prescribed conditions.” For example, more research was essential to “reliably model how long a smoke cloud would last if the particles were injected into the stratosphere where there is no moisture condensation to rapidly remove them.”

To develop a plan for further research, the director of the National Climate Program Office of the National Oceanic and Atmospheric Administration (NOAA) chaired an Interagency Research Program (IRP), which identified priorities and recommended funding levels for the various agencies. According to the GAO, for fiscal year 1986 the IRP provided only $2 million in new funding beyond the $3.5 million that had had already been slated. Although the NOAA-chaired committee had discussed the possibility of a $50 million, 5-year program the final report included no proposals for future funding levels.

The fairness of the GAO report drew the ire of the White House Office of Science and Technology Policy (OSTP). According to the GAO, “OSTP expressed the view that the report, in its discussion of policy issues, was giving more validity to the nuclear winter theory than was warranted and suggested the tenor of the report be changed.” The GAO retorted that “it does not agree.”

OSTP also wanted GAO to delete the discussion of funding options, citing a “nondisclosure proviso,” but GAO refused, arguing that the information was widely known. Indeed, a 1985 article in Science had disclosed that $50 million had been under consideration .but that federal officials were reluctant to “extract the new funds from ongoing programs” and that George Keyworth, the White House science adviser was loath to “press the issue hard enough.”[19]

At the request of the Department of Agriculture, a U.S. government interagency committee sponsored a review of some of the findings of a two-volume report, The Environmental Consequences of Nuclear War, published in 1985 by the Scientific Committee on Problems of the Environment (SCOPE) part of the International Council of Scientific Unions (ICSU) (now the International Council for Science).[20] Some of the TTAPS authors, such as Richard Turco, contributed to the SCOPE report as did Michael MacCracken and Paul J. Crutzen. The first volume had focused on the physical and atmospheric effects of nuclear war while volume two looked at the ecological and environmental impact. The SCOPE report did not use the term nuclear winter, but it nevertheless treated atmospheric perturbations caused by mass fires as integral to the terrible devastation along with the blast and radiation effects. Thus, the authors found that the smoke and soot produced by a major nuclear war could significantly reduce sunlight and produce major drops in temperature.

According to Volume II of the SCOPE report, the effects of a nuclear war could disrupt agricultural production to the point where those who survived the “acute phase,” where smoke drastically reduced sunlight and temperature, would “starve during the chronic phase.” While the authors carefully qualified their conclusions, pointing to the limits of the models, they noted that even small changes in temperature could have a devastating impact on agriculture in the Northern Hemisphere. It was such findings that the Department of Agriculture wanted investigated.

The Committee on Interagency Radiation Research and Policy, set up by the White House in 1984, established a special task group of agricultural and meteorological experts from universities and government to review the findings of Volume II. The project team did special simulation studies based on models of corn and soybean growth, which helped lead them to the conclusion that there “is a high chance of complete crop failure if temperatures drop by 15°C even for one month during a crop season, as far south as 36°N latitude.” The report agreed with the SCOPE finding that smaller changes in temperature could cause crop failure at higher latitudes. Further, they agreed that the SCOPE report on Volume II “does present a wide range of possible post-nuclear war catastrophes in our life support system” and “clearly establishes the probability of mass starvation for the surviving populations as a direct consequence of nuclear attack.”

Yet, while finding that the SCOPE report’s methodology was “sound” the task group also concluded that it understated the effects of nuclear war on agriculture. Pointing to areas that should have received additional emphasis, they observed that the destruction of dams would cause “loss of large areas of irrigated agricultural land” while the destruction of the U.S. food and agriculture infrastructure would cause “severe disruption of production, processing, and distribution.” In their executive summary, the authors found that “the worst case scenario in Volume II may be a gross underestimation of the likely effects of a nuclear war on human populations.”

That White House’s science advisers had authorized this review of the SCOPE report is interesting in itself, and that its findings supported nuclear winter theories was also important. Yet, for reasons unknown, this report received no notice at the time, with apparently no publicity from the White House, although it would be cited in studies by experts.