The Effects of Nuclear War

APPROACH

The scope of this study is both broader and narrower than that of most other studies on this subject. It is broader in three respects:

1. it examines a full range of possible nuclear attacks, with attacking forces ranging in extent from a single weapon to the bulk of a superpower’s arsenal;
2. it deals explicitly with both Soviet attacks on the United States and U.S. attacks on the Soviet Union; and
3. it addresses the multiple effects of nuclear war, indirect as well as direct, long term as well as short term, and social and economic as well as physical.

Those effects that cannot be satisfactorily calculated or estimated are described qualitatively. But this report’s scope is narrower than most defense analyses because it avoids any consideration of military effects; although it hypothesizes (among other things) missile attacks against military targets, only the “collateral” damage such attacks would inflict on the civilian society are examined.

The approach used was to look at a series of attack “cases,” (table 1) and to describe the various effects and overall impact each of them might produce. By analyzing the impact of the same attack case for both a U.S. attack on the Soviet Union and a Soviet attack on the United States, the report examines the significance of the different kinds of vulnerabilities of the two countries, and offers some insights about the consequences of the differences between the two countries’ nuclear weapon arsenals. The cases were chosen primarily to investigate the effects of variations in attack size and in the kinds of targets attacked. It is believed that the analysis is “realistic,” in the sense that the hypothetical attacks are possible ones. Patterns of nuclear explosions were examined that are not very different from those that, OTA believes, the existing nuclear forces would produce if the military were ordered to make attacks of the specified size on the specified targets.

Table 1. -Summary of Cases

Case	Description
1	Attack on single city: Detroit and Leningrad; 1 weapon or 10 small weapons.
2	Attack on oil refineries, limited to 10 missiles.
3	Counterforce attack; includes attack only on ICBM silos as a variant.
4	Attack on range of military and economic targets using large fraction of existing arsenal.
For each case the first section describes a soviet attack on the United States and the following section a US attack on the Soviet Union

Case 1: In order to provide a kind of tutorial on what happens when nuclear weapons are detonated, the study describes the effects of the explosion of a single weapon. Then it examines the effects of such an explosion over a single U.S. city (Detroit) and single Soviet city (Leningrad) of comparable size. The base case was the detonation of a 1-megaton weapon (1 Mt = energy released by one million tons of TNT), since both the United States and the Soviet Union have weapons of roughly this size in their arsenals. Then, in order to look at the ways in which the specific effects and overall impact would vary if other weapons that might be available were used, the effects of a 25-Mt weapon over Detroit, the effects of a 9-Mt weapon over Leningrad, and the effects of 10 weapons of 40 kilotons (kt) each over Leningrad are described. An attempt was made to describe as well the effects of a small weapon in a large city (such as a terrorist group might set off) but was unsuccessful because the effects of such a weapon in a metropolitan setting cannot be inferred from the existing body of knowledge regarding military weapons. This is explained in the body of the report.

The casualties from such attacks could range from 220,000 dead and 420,000 injured to 2,500,000 dead and 1,100,000 injured (many of the injured would wind up as fatalities), depending on the details of the attack and the assumptions made regarding conditions. The discussion in chapter II shows how the time of day, time of year, weather conditions, size of weapon, height of burst, and preparation of the population could all make a great difference in the number of casualties resulting from such an attack. The extent of fire damage is a further uncertainty, Even if only one city is attacked, and the remaining resources of a nation are available to help, medical facilities would be inadequate to care for the injured. A further imponderable is fallout (if the attack uses a surface burst), whose effects depend on the winds.

Case 2: In order to examine the effects of a small attack on urban/industrial targets, the study examines a hypothetical attack limited to 10 SNDVs (strategic nuclear delivery vehicles, the term used in SALT to designate one missile or one bomber) on the other superpower’s oil refineries. In “planning” this attack, which is not analogous to any described in recent U.S. literature, it was hypothesized that the political leadership instructed the military to inflict maximum damage on energy production using only 10 SNDVs without regard to the extent of civilian casualties or other damage, It was assumed that the Soviets would attack such targets with SS-18 missiles (each carrying 10 multiple independently targetable reentry vehicles, or MlRVs), and that the United States would use 7 MlRVed Poseidon missiles and 3 MlRVed Minuteman III missiles.

The calculations showed that the Soviet attack would destroy 64 percent of U.S. oil refining capacity, while the U.S. attack would destroy 73 percent of Soviet refining capacity. Calculations were also made of “prompt fatalities,” including those killed by blast and fallout, assuming no special civil defense measures: they showed about 5 million U.S. deaths and about 1 million Soviet deaths. The results were different for the two countries for several reasons. Soviet oil refining capacity is more concentrated than U.S. oil refining capacity, so that a small attack can reach more of it. At the same time, Soviet refineries tend to be located away from residential areas (the available data on population location deals with where people live rather than with where they work) to a greater extent than U.S. refineries. A further difference is that a limitation on the number of delivery vehicles would lead each side to use weapons with many MlRVs, so the United States would attack most of the targets with Poseidon missiles which have small warheads, while the Soviets would use SS-18 intercontinental ballistic missiles (ICBMs) which carry much larger warheads, and large warheads cause more damage to things not directly targeted (in this case, people) than do small warheads.

One can only speculate about the consequences of such extensive destruction. There would have to be drastic changes in both the U.S. and Soviet economies to cope with the sudden disappearance of the bulk of oil refining capacity. Productivity in virtually every industrial sector would decline, and some sectors would be largely wiped out. There would have to be strict allocation of the remaining available refined petroleum products. Some Soviet factory workers might end up working in the fields to replace tractors for which fuel was unavailable. The United States might have to ban commuting by automobile, forcing suburban residents to choose between moving and long walks to a bus stop. The aftermath of the war might lead to either an increase or a decrease in the amount of petroleum products required by the military. Changes in people’s attitudes are impossible to predict. Calm determination might produce effective responses that would limit the damage; panic or a breakdown in civic spirit could compound the effects of the attack itself.

It is instructive to observe the asymmetries between the problems which the United States and the Soviets would face. Soviet agricultural production, which is barely adequate in peacetime, would probably decline sharply, and production rates would slow even in essential industries However, the Soviet system is well adapted for allocating scarce resources to high-priority areas, and for keeping everybody employed even if efficient employment is unavailable. The relative wealth and freedom of the United States brings both advantages and disadvantages: while agriculture and essential industry would probably continue, there would be a staggering organizational problem in making use of resources that now depend on petroleum — one must ask what the employees of an automobile factory or a retail establishment on a highway would do if there were virtually no gasoline for cars,

A major question relating to these results is how much they could vary with changed assumptions, The figures for fatalities were based on air bursts, which would maximize destruction of the refineries. (As an excursion, U.S. fatalities were recalculated on the assumption of surface bursts, and use of the best fallout shelters within 2 miles of where each person lives. This reduced fatalities by one third, ) There was no data available on the types of Soviet residential construction in the vicinity of oiI refineries: treating it parametrically gave casualty figures of about 1,500,00 if the construction is all houses, and about 800,000 if it is al I apartment buildings. Perfect accuracy was assumed for missiles that are in fact somewhat inaccurate — some inaccuracy might reduce the extent of damage to the refineries, but it might well increase the number of deaths.

Case 3: In order to examine the effects on civilian populations and economies of counterforce attacks, the study examined attacks on ICBM silos and attacks on silos, bomber bases, and missile submarine bases. Such attacks have received fairly extensive study in the executive branch in recent years, so OTA surveyed a number of these studies in order to determine the range of possible answers, and the variations in assumptions that produce such a range, An unclassified summary of this survey appears as appendix D of this volume. (The complete survey, classified secret, is available separately. )

A counterforce attack would produce relatively little direct blast damage to civilians and to economic assets; the main damage would come from radioactive fallout, The uncertainties in the effects of fallout are enormous, depending primarily on the weather and on the extent of fallout sheltering which the population makes use of. The calculations made by various agencies of the executive branch showed a range in “prompt fatalities” (almost entirely deaths from fallout within the first 30 days) from less than 1 to 11 percent of the U.S. population and from less than 1 to 5 percent of the Soviet population. This shows just how great a variation can be introduced by modifying assumptions regarding population distribution and shelter.

What can be concluded from this? First, if the attack involves surface bursts of many very large weapons, if weather conditions are unfavorable, and if no fallout shelters are created beyond those that presently exist, U.S. deaths could reach 20 million and Soviet deaths more than 10 million. (The difference is a result of geography; many Soviet strategic forces are so located that fallout from attacking them would drift mainly into sparsely populated areas or into China. ) Second, effective fallout sheltering (which is not necessarily the same thing as a program —this assumes people are actually sheltered and actually remain there) could save many lives under favorable conditions, but even in the best imaginable case more than a million would die in either the United States or the U.S.S.R. from a counterforce attack. Third, the “limited nature” of counterforce attacks may not be as significant as the enormous uncertainty regarding their results.

There would be considerable economic damage and disruption as a result of such attacks. Almost all areas could, in principle, be decontaminated within a few months, but the loss of so many people and the interruption of economic life would be staggering blows. An imponderable, in thinking about the process of recovery, is the extent of any lasting psychological impacts.

Case 4: In order to examine the kind of destruction that is generally thought of as the culmination of an escalator process, the study looked at the consequences of a very large attack against a range of military and economic targets. Here too calculations that the executive branch has carried out in recent years were used. These calculations tend to assume that Soviet attacks on the United States would be a first strike, and hence use most of the Soviet arsenal, while U.S. attacks on the Soviet Union would be retaliatory strikes, and hence use only those weapons that might survive a Soviet counterforce attack. However, the difference in damage to civilian populations and economies between a “first strike” and a “second strike” seems to lie within the range of uncertainty created by other factors.

The resulting deaths would be far beyond any precedent. Executive branch calculations show a range of U.S. deaths from 35 to 77 percent (i. e., from 70 million to 160 million dead), and Soviet deaths from 20 to 40 percent of the population. Here again the range reflects the difference made by varying assumptions about population distribution and sheltering, and to a lesser extent differences in assumptions about the targeting policy of the attacker. Soviet casualties are smaller than U.S. casualties because a greater proportion of the Soviet population lives in rural areas, and because U.S. weapons (which have lower average yields) produce less fallout than Soviet weapons.

Some excursions have been run to test the effect of deliberately targeting population rather than killing people as a side effect of attacking economic and military targets. They show that such a change in targeting could kill somewhere between 20 million and 30 million additional people on each side, holding other assumptions constant.

These calculations reflect only deaths during the first 30 days. Additional millions would be injured, and many would eventually die from lack of adequate medical care. In addition, millions of people might starve or freeze during the following winter, but it is not possible to estimate how many. Chapter V attempts to calculate the further millions who might eventually die of latent radiation effects.

What is clear is that from the day the survivors emerged from their fallout shelters, a kind of race for survival would begin. One side of the race would be the restoration of production: production of food, of energy, of clothing, of the means to repair damaged machinery, of goods that might be used for trade with countries that had not fought in the war, and even of military weapons and supplies. The other side of the race would be consumption of goods that had survived the attack, and the wearing-out of surviving machines. If production rises to the rate of consumption before stocks are exhausted, then viability has been achieved and economic recovery has begun. If not, then each postwar year would see a lower level of economic activity than the year before, and the future of civilization itself in the nations attacked would be in doubt. This report cannot predict whether this race for economic viability would be won. The answer would lie in the effectiveness of postwar social and economic organization as much as in the amount of actual physical damage. There is a controversy in the literature on the subject as to whether a postttack economy would be based on centralized planning (in which case how would the necessary data and planning time be obtained?), or to individual initiative and decentralized decision making (in which case who would feed the refugees, and what would serve for money and credit?).

An obviously critical question is the impact that a nuclear attack would have on the lives of those who survive it. The case descriptions discuss the possibilities of economic, political, social, and psychological disruption or collapse. However, the recital of possibilities and uncertainties may fail to convey the overall situation of the survivors, especially the survivors of a large attack that included urban-industrial targets. In an effort to provide a more concrete understanding of what a world after a nuclear war would be like, OTA commissioned a work of fiction. It appears as appendix C and presents some informed speculation about what life would be like in Charlottesville, Va., assuming that this city escaped direct damage from the attack. The kind of detail that such an imaginative account presents—detail that proved to be unavailable for a comparable Soviet city–adds a dimension to the more abstract analysis in the body of the report.

Civil Defense: Chapter III provides some basic information about civil defense measures, discusses the way in which they might mitigate the effects of nuclear attack, and discusses the uncertainties regarding their effectiveness. There is a lively controversy among experts as to the effectiveness of existing Soviet civil defense programs, and another controversy as to whether existing U.S. programs ought to be changed. The major points in dispute were identified, but no attempt was made to assess the merits of the arguments. For the purposes of this study, it was assumed that the existing civil defense programs, as described in this report, would be in effect, and that a full-scale preattack evacuation of cities (sometimes called “crisis relocation”) would not take place. This assumption was made because it appeared to be the only way to describe existing vulnerabilities while avoiding predictions about the course of events leading up to a nuclear war. While both the U.S. and the Soviet Governments profess to believe that urban evacuation prior to an attack on cities would save lives, ordering such an evacuation would be a crisis management move as well as a civil defense precaution.

Long-Term Effects: While the immediate damage from the blasts would be long term in the sense that the damage could not be quickly repaired, there would be other effects which might not manifest themselves for some years after the attack. It is well established that levels of radiation too low (or too slowly absorbed) to cause immediate death or even illness will nevertheless have adverse effects on some fraction of a population receiving them. A nuclear attack would certainly produce both somatic effects (largely cancer) and genetic effects, although there is uncertainty about the numbers of victims. OTA calculated the ranges of such effects that might be produced by each of the attack cases analyzed. Cancer deaths and those suffering some form of genetic damage would run into the millions over the 40 years following the attack. For the comprehensive attack (Case 4), it appears that cancer deaths and genetic effects in a country attacked would be small relative to the numbers of immediate deaths, but that radiation effects elsewhere in the world would appear more significant. For counterforce attacks, the effects would be significant both locally and worldwide.

A 1975 study by the National Academy of Sciences (NAS)¹ addressed the question of the possibility of serious ecological damage, and concluded that while one cannot say just how such damage would occur, it cannot be ruled out. This conclusion still stands, although the NAS report may have been more alarmist about the possibility of damage to the ozone layer than recent research would support.

Table 2 summarizes the results of the case studies.

Table 2. – Summary of Effects

Case	Description	Main causes of civilian damage	Immediate deaths	Middle-term effects	Long-term effects
1	Attack on single city: Detroit and Leningrad; 1 weapon or 10 small weapons,	Blast, fire, & loss of infrastructure, fallout is elsewhere,	200,000 - 2,000,000	Many deaths from injuries; center of city difficult to rebuild,	Relatively minor
2	Attack on oil refineries, Iimited to 10 missiles,	Blast, fire, secondary fires, fallout, extensive economic problems from loss of refined petroleum,	1,000,000 - 5,000,000	Many deaths from injuries; great economic hardship for some years; particular problems for Soviet agriculture and for U.S. socioeconomic organization	Cancer deaths in millions only if attack involves surface bursts
3	Counterforce attack: Includes attack only on ICBM silos as a variant	Some blast damage if bomber and missile submarine bases attacked	1,000,000 - 20,000,000	Economic impact of deaths, possible large psychological impact	Cancer deaths and genetic effects in millions, further millions of effects outside attacked countries
4	Attack on range of military and economic targets using large fraction of existing arsenal	Blast and fallout, subsequent economic disruption; possible lack of resources to support surviving population or economic recovery, Possible breakdown of social order. Possible incapacitating psychological trauma.	20,000,000 - 160,000,000	Enormous economic destruction and disruption, if Immediate deaths are in low range, more tens of millions may die subsequently because economy is unable to support them. Major question about whether eco- nomic viability can be restored–key variables may be those of political and eco- nomic organization. Unpredictable psychological effects,	Cancer deaths and genetic damage in the millions; relatively insignificant in attacked areas, but quite significant elsewhere in the world. Possibility of ecological damage.
For each case the first section describes a Soviet attack on the United States and the following section a U.S.attack on the Soviet Union

1. Long-Term Worldwide Effects of Multiple Nuclear-Weapons Detonations (Washington, D.C.: National Academy of Sciences, 1975).

---