Ways in which the world climate might be catastrophically disturbed have received a great deal of attention as a result of three series of papers. The first was led by a 1967 paper by Manabe and Wetherald1, which reported that atmospheric CO2 levels were rising, and forecast that a doubling of CO2 levels would cause an increase in average world temperatures of about 2.3ºC as the result of the greenhouse effect [see Box - The Greenhouse Effect]. It was followed by many papers2 which have forecast increases in average world temperatures over the next century by as much as 6ºC as the result of man-caused increases in levels of CO2. Some of these papers have warned of severe effects on ecosystems and the fate of nations. A recent EPA report3 has concluded that greenhouse warming over the next fifty years is unavoidable, and urges planning for the effects.
The second was led by a 1980 article by Luis W. Alvarez and his associates4, which reported finding a distinct clay layer at the Cretaceous-Tertiary (C-T) boundary containing the element iridium whose isotopic composition was characteristic of meteorites rather than terrestrial rocks. They suggested that this was evidence of the impact of an asteroid about 10 kilometers in diameter 65 million years ago. They hypothesized that such an impact would darken the skies with dust, which might reduce surface temperatures enough to cause the mass extinction of the dinosaurs and many other species then living, the fossil record of which seems to coincide with the C-T boundary. Further work has found such layers at many sites around the world and tended to confirm the impact hypothesis. There is still controversy concerning where the impact occurred, to what extent it caused the extinctions during this era, and whether the most important climatic and biological effects were the result of darkening, cooling, heating, flooding, or other mechanisms5.
The third series of papers began with a 1982 article by Paul J. Crutzen and John W. Birks6, who forecast that a nuclear war would loft enough smoke and dust to cause a darkening of the skies and enough hydrocarbons and oxides of nitrogen to cause dangerous levels of air pollution. It was followed by a 1983 article by Turco, Toon, Ackerman, Pollack, and Sagan (called TTAPS)7, which predicted that multiple nuclear explosions could, depending on their number, yield, and type of target, blanket the entire planet with a cloud of soot and dust for 3-12 months that would reduce sunlight by up to 99 percent and reduce average temperatures to as low as -40ºC (-40ºF) in the interiors of Northern Hemisphere continents. They called this scenario a "nuclear winter". A companion article by Paul Ehrlich and his associates8 argued that dark and cold could combine to cause the death of much unprotected life and deal a blow to food production that could cause massive starvation. These papers have led to a surge of work in many research centers around the world, most of which have thus far confirmed the general thrust of the TTAPS model, and to a strategic reassessment by nuclear planners. The National Climate Program Office of the National Oceanic and Atmospheric Administration has been appointed to coordinate the preparation of a National Plan of Research on this subject, with the participation of several agencies, including the Defense Nuclear Agency, the Office of the Secretary of Defense and the Federal Emergency Management Agency. An item for this research is to be included in the DOD budget report to Congress in March, 1985.
The TTAPS group used a one-dimensional radiative-convective model (RCM) which examined how dust and soot produced by various nuclear war scenarios would rise and settle out and affect light absorption and globally-averaged temperatures at various levels in a typical vertical column of air above a continent and above an ocean. Their model predicted that even limited nuclear wars could inject enough dust and soot into the stratosphere to adversely affect global climate.
If a completely opaque cloud blanket lasted long enough, surface temperatures on an oceanless planet would eventually descend to a fairly uniform level of about -55ºC (-67ºF) worldwide. The TTAPS group predicted that a low of -25ºC (-13ºF) could be reached in continental interiors within two weeks for their reference 5000 megaton scenario, but cautioned that transfers from the reservoir of latent heat of the sea might reduce the temperature drop by 30-70 percent, especially near coastlines. MacCracken, using a two-dimensional circulation model, has predicted that average temperatures would not drop lower than about -5ºC (23ºF) because of transfers of heat from the oceans9. Covey, Schneider and Thompson, of the National Center for Atmospheric Research (NCAR), using a three-dimensional global circulation model (GCM)10,11 have predicted temperature drops to about -15ºC (5ºF) in continental interiors of the Northern Hemisphere. Their model also indicates that the blanket might spread worldwide within a few weeks. Similar work in the Soviet Union has been done with similar results11.
The author has confirmed the main thrust of the nuclear winter scenario with a fairly simple one-dimensional radiative-convective model (RCM) on a 32-bit UNIX supermicro using the Q'Nial and C languages. A supercomputer like the Cray 1 is not needed for this, although even that machine is not sufficient for the kind of high-resolution three-dimensional modeling needed to simulate the complex processes involved and answer some of the questions raised by the study. In retrospect, it seems surprising that the nuclear winter scenario wasn't discovered ten or twenty years ago.
The author has also developed a statistical biological impact model (BIM) [See Box - Biological Impact Model]. The TTAPS and other models have tried to predict average temperatures. However, more important for predicting biological effects are the extreme temperatures that might last longer than various life forms subjected to them. Of course, temperature averages would be important for their impact on things like fuel consumption. The BIM applies a statistical approach to try to predict the probability that temperatures might exceed the tolerances for various higher lifeforms. Although crude, it indicates that for a spring event, most parts of continents from about 60ºN to 20ºN, even coastal areas, would be likely to experience at least one cold spell severe enough to kill most unsheltered life and wipe out agricultural activity, and for the tropics and Southern Hemisphere, cold enough to kill rain forests and have a devastating effect on agriculture. Most people could endure the cold. The main problem would be the loss of all of the first and most of a second year of food production, One uncertainty is whether crops in the Southern Hemisphere could be harvested after a spring event in the Northern Hemisphere. It is possible that excessive precipitation could wipe them out.
The author has observed the damage done by a week of 108ºF temperatures in 1980 and by a week of sub-freezing temperatures in the winter of 1983-84 to vegetation in his area of South Texas, especially to shrubs and citrus trees. These highs were only about 2ºC above normal extremes. The lows were not as low as single-night extremes in many previous years, but lasted longer. Plants and animals are vulnerable to temperatures that exceed usual extremes by even a few degrees for more than a few days. Most species are more susceptible to highs than to lows5, and start to die at about 42ºC (108ºF), few if any surviving beyond 50ºC (122ºF). Many species can endure drastic drops in temperature, especially warm-blooded animals and plants with deep root systems.
The closest thing we have to an experimental test of these models is the terminal Cretaceous event. Although there are significant differences between the various nuclear scenarios and impact by an asteroid or comet, there are some important similarities. Emiliani et al have suggested that 70 percent of animal genera and 30 percent of plant genera vanished during this period and that the pattern of extinctions indicates the cause was elevated temperatures. They calculated that a sea impact would loft enough H2O vapor to raise temperatures 8-10ºC. Considering how much it would take to wipe out 70 percent of the animal genera then living, and the similarities with a nuclear war, the models that predict catastrophic cooling or heating [see Box - Nuclear Summer] following a nuclear war can be said to have persuasive experimental support.
However, Hickey has pointed out that the pattern of plant extinctions does not fit the model: they occurred over several million years, and mostly in northern temperate latitudes14. Tropical species were not affected as much as might be expected from either cooling or warming. Indeed, most of the extinctions seem to have occurred in Asia east of the Ural Mountains and in North America west of the Rockies, which Emiliani suggested might have been the result of a giant tidal wave from an impact in the Bering Sea area5. Hickey does say, however, that signs of ecological instability in plant communities seem to support the impact hypothesis. Emiliani has argued15 that the fossil record shows most evolutionary successions involved not the victory of a new competitor for the same niche over an old one, but extinction of a species from other causes, perhaps disease, leaving a vacuum which was not always filled by a better-adapted species. He suggested that this might be the explanation for the disappearance of Neanderthal Man. We can speculate that the rate of extinction may have accelerated during this period, not directly from the short-term physical effects of the impact event, but by it creating conditions favorable to the development of plagues.
What would happen in the United States and other nations if there was a distant nuclear war or nuclear winter was triggered deliberately [see Box - How it Might Happen]? A preliminary analysis of USDA data indicates that stocks of food in pantries and supermarkets could feed U.S. residents for about 30 days, and stocks in warehouses another 60-90 days. After that, they would have to live on feed stocks, which might last a year with tight rationing. Such feed stocks are not well distributed, and converting them to human consumption would present processing problems. Other nations would be in much worse shape. FAO estimates world food reserves at about 33 days now.
A nuclear winter could wipe out all of one year of agricultural production, and severely impair production during the second. Much livestock might not survive, and seed stocks needed for replanting might be lost. It might take up to three years to get agricultural production to a level sufficient to feed everyone now living. By the time it could be done, there would not be nearly as many people to have to feed. The process of desertification might be accelerated and run to completion within a few years, especially if the nuclear summer scenario is valid, which could make modern civilization impossible to sustain, and reduce humanity to scattered bands of nomads. When well-fed people look upon the people suffering from famine in Africa, they could be looking at where they will be some day.
Nations should make plans and preparations for these contingencies. The models need to be developed further, and the results widely discussed. Improved international cooperation in the gathering and analysis of meteorological data must be sought. Emergency plans for gathering and analyzing global meteorological data following a nuclear episode need to be made. Countries should consider building reserves of food and other supplies, recognizing that such preparations could provoke a great deal of political resistance. Such preparations would be especially difficult for democratic countries.
It is possible that the Soviet Union has been preparing to deliberately trigger a nuclear winter [see Box - How It Might Happen (winter war scenario)] for twenty years, although their behavior in 1971 [see Box - How It Almost Happened] suggests that they were unaware of it at that time. After initial cooperation on the modeling effort, they have suddenly gone secretive on the subject - a bad sign. Widespread discussion of the winter war scenario would make it much less likely, by depriving them of the advantage of surprise. In view of possible complications like the nuclear summer, however, anyone contemplating trying the winter war scenario should think again. They might be faced with having to use further nuclear detonations to inject enough dust into the stratosphere to offset the nuclear summer effect, and in the process accumulate intolerable levels of radioactivity.
The winter war scenario might be deterred by a U.S. policy that world climate is a "vital interest" of the United States, and that we would respond to any attempt to trigger a nuclear winter with an all-out nuclear attack. The alternative would be a "food race" in which the United States and the Soviet Union tried to outdo each other stockpiling in preparation for a nuclear winter, a race that would threaten to leave the rest of the world, literally and figuratively, out in the cold.
It was Pierre Laplace, back in the 18th century, who first compared the operation of the atmosphere to a greenhouse, the glass of which admits visible light, but does not permit the escape of the infrared radiation into which the visible light has been converted after absorption by the objects within it. The result is a higher temperature inside than outside. Laplace did not know about infrared radiation, but he got the main idea right. Of course, the greenhouse also traps the air inside, which would otherwise carry away the heat through convection.
In the atmosphere, the function of the greenhouse glass is performed by several substances, mainly water vapor (H2O) and carbon dioxide (CO2), which are transparent to visible light, but absorb most of the radiation in the infrared part of the spectrum with a wavelength less than 8 and greater than 13 microns. H2O makes the most important contribution to the effect. When air containing H2O and CO2 absorbs infrared radiation, it heats up. If the air is at the surface, conduction keeps the surface warm. Infrared absorbed some distance above the surface heats the air there, but the H2O and CO2 reradiate infrared radiation, part of which is directed upward and escapes into space, and part of which is directed downward to heat up the surface. In the absence of H2O, the surface temperature would average about -13ºC. Since the average temperature is about 15ºC, greenhouse warming is 28ºC. Actually, it would be greater than that, but the presence of clouds, ice, and vegetation increases the albedo, or fraction of light reflected, of the earth from about 0.14 to about 0.33. This reduces the amount of light that reaches, and is therefore absorbed by, the surface, resulting in cooler surface temperatures. Of the warming, the CO2 contributes about 2ºC.
H2O is in dynamic equilibrium in the troposphere. As more evaporates, clouds form, which reduce heating from sunlight and increase precipitation. H2O vapor is prevented from entering the stratosphere in significant quantities by the cold trap at the tropopause, which forms the boundary between the lower troposphere and the stratosphere. The tropopause varies in altitude from about 15 km in the tropics to about 10 km at higher latitudes.
Most workers estimate that the effect of doubling the concentration of CO2 in the atmosphere would be to increase average global temperatures about 2º to 4ºC. CO2 alone would only cause about a 1.8º immediate rise, but there is a positive feedback effect, as such a rise would also increase the atmospheric H2O content. This could take up to 40 years because it would require that long to heat the upper 70 m of the oceans. The uncertainty concerns the effect of these changes on average cloud cover. Such a doubling could occur if present trends continue in the burning of fossil fuels, although various mechanisms might be expected to remove much of the CO2 introduced in this way, especially if the rate of introduction is slow. If enough of the biomass on the surface of the earth were quickly oxidized, as might occur through "rapid desertification" following a nuclear war, it could increase CO2 in the atmosphere by a factor of about two, for an increase in temperature of from 2º to 4ºC. The burning of most stored fossil fuel supplies and materials derived from petrochemicals, which comprise a significant part of cities, could bring the increase up by another O.5ºC.
Other substances could contribute to greenhouse warming. There are at least 30 compounds that might be produced by nuclear detonations in significant quantities that could have an effect. Many of them become smog when exposed to sunlight. It is estimated that the ingredients of current air pollution make a contribution to greenhouse warming comparable to that of CO2. The most important of such substances are those which absorb strongly in the infrared window from 8 to 13 microns. One of these is ozone (O3). Others include methane (CH3) and ammonia (NH3). NH3 is rapidly removed by photodissociation and combination with H2O, but many of the others are stable enough to last for several years. It might not take much to close the infrared window, and this could have a significant effect on greenhouse heating.
While the cold trap now prevents significant amounts of H2O from entering the stratosphere, if the cold trap failed or was penetrated, and enough H2O was transported into the stratosphere, it could cause significant greenhouse warming of the surface. It is estimated1 that the injection of 6.6x1013 kg of H2O could raise temperatures about 2ºC and 4x1014 kg of H2O about 8ºC. It could take several years for such H2O to precipitate out.
Much of this nuclear winter modeling has been done assuming that detonations would be confined to the northern hemisphere between 30º and 60ºN latitude. This simplifies a model, but is not realistic. The Ambio reference scenario12 is closer to what is likely to happen. This scenario assumes that major population and industrial centers in Asia, Africa, Australia, and Latin America would be targeted, and that the Southern Hemisphere would receive 173 out of the total of 5742 megatons. The TTAPS group did consider a scenario of 300 megatons in the Southern Hemisphere in combination with an exchange in the Northern Hemisphere, but most other groups have ignored that scenario.
There are reasons to believe that the Soviet strategy is to rapidly occupy Western Europe, using few if any nuclear weapons there, while their homeland is sacrificed. They would not want anything left anywhere in the world that could present a threat to them until they could consolidate their position. "Non-alignment" would afford no protection from nuclear attack. In this scenario, nuclear detonations would not be confined to the Northern Hemisphere. The author's reference "Soviet first strike" scenario assumes a Soviet attack of 5000 megatons, 300 megatons into the Southern Hemisphere, and a U.S. response of 700 megatons, the force expected to survive the Soviet first strike, mainly SLBMs and airborne and naval cruise missiles.
This Soviet plan has been made much less likely, however, by the installation in Western Europe of Pershing II and cruise missiles, which effectively denies the Soviets their strategic objective and refuge by insuring that Western Europe would be destroyed in a nuclear war. It has forced them to rethink their strategy.
Most of these models have also neglected detonations at sea. Both superpowers have a major stake in tracking and targeting each other's missile submarines. Because of the uncertainty in the position of submarines, larger warheads with a larger kill radius are likely to be used, at least 1 megaton. Detonations are likely to be several hundred meters below the surface. The "Soviet first strike" scenario assumes about 300 megatons would be used in this way. There would be little dust and no smoke, but enough H2O vapor might be injected into the stratosphere to have a long-term climatic impact.
A second scenario has been examined in some detail. In this scenario the Soviet Union stockpiles food and other supplies for several years, then, in early spring of 1992, deliberately triggers a nuclear winter by detonating a number of nuclear devices on its own territory in Siberia. This reference scenario assumes 200 megatons of airbursts, to start firestorms in forests, followed a few minutes later by about 5000 megatons of groundbursts, from devices buried in a hexagonal pattern a couple of hundred meters below the surface, which would loft a great deal of dust and inject the smoke from the fires into the stratosphere. The Soviets would claim that they had been attacked by the United States, to turn world opinion against it, and then proceed to "fill the empty seats" left by the nuclear winter. They could move into various countries with offers of food, then take over. If the first nuclear winter doesn't give them world domination, they could repeat the procedure, or threaten to do so, until the world capitulated or they were left the only survivors. This is called the "winter war" scenario.
If this scenario seems a bit brutal even for the Russians, one might consider their predicament. Stalin left the Soviet vlasti (rulers) in a bind. For at least another generation, it is going to be very difficult for them to safely reform the way China has begun to do. Only the ascendance of the Party members who came of age in the Khrushchev Era offers any hope of that. Those before and after are unlikely to change anything. In the meantime, they face a depletion of their natural resources and a decline of the population of ethnic Russians relative to the population of Soviet Asiatics. They cannot win a nuclear war. Their climate is bad, and their economy doesn't work very well. The Soviet intelligentsia are unreliable, and the narod (masses), always unruly, are getting more skeptical and impatient. The Polish Solidarity movement threatens to spread to other satellites and even into the Soviet Union proper. China has taught them that promoting "Communist" regimes in other countries is more likely to create enemies than friends, unless the country is either militarily occupied or heavily subsidized. Occupying Afghanistan and subsidizing Cuba is getting expensive, and they can hardly look forward to trying to extend these models to control of the entire world. Now China is making reforms that shake Leninist doctrine to its foundations, and is setting an example that threatens the nomenklatura (Soviet ruling system). In the meantime, more and more nations are acquiring nuclear weapons. They once looked forward to Malthusian catastrophe to create opportunities for them, but an overpopulated world is just as unmanageable for them as is it for the West, and threatens to make the world not worth dominating. A nuclear winter, considered in the context of the winter war scenario, might be perceived by them as having the specifications of an almost ideal strategic weapons system. They may think that, with the advantage of surprise, it could be used to depopulate most of the rest of the world while leaving most industrial capacity intact, with little or no risk of retaliation. They may ignore warnings from their scientists about the risks and uncertainties, and see it as the only way out of their dilemma.
An experimental test of the nuclear winter hypothesis nearly happened in 1971. At a diplomatic reception in Moscow, a Soviet diplomat approached an American diplomat and asked him "Would the United States stand by if we launch a nuclear attack on China?" The American immediately said, "No, we most certainly would not stand by!" The Russian was a little taken aback by this immediate and emphatic reply, and said "Perhaps you should check with your superiors on this." The American said, "I will, but I know what their answer will be!"
The purpose of the Russian's question was to remove the ambiguity that arose after the split between the Soviet Union and the Peoples' Republic of China as to whether the U.S. nuclear umbrella, which protected all countries not in the Soviet camp, now extended to protect China. By his reply, the American said that it did. The Soviet Union was then in an advanced stage of preparation for a nuclear attack on China's military and industrial facilities, which would also have caused the death of at least 300 million Chinese. A few weeks later, a higher-level Russian official asked the same question of a higher-level American official and got the same answer. Finally, Leonid Breshnev asked the same question of Henry Kissinger. He got the same answer, and decided not to go through with it. Shortly thereafter, the Chinese found out how the U.S. saved them from nuclear attack, and on April 6, 1971, they invited the U.S. ping-pong team to Peking. The rest is history.
This was probably the closest the world has come to a nuclear war since 1945, and it would not have involved nuclear detonations on the territory of the U.S. or its allies. The yield of the detonations would likely have exceeded 300 megatons, which would have been sufficient to cause at least a mild nuclear winter. However, we would not have been prepared at that time to study it properly.
Looking at the period following the nuclear winter, the author's RCM and BIM indicate that temperatures might increase above normal levels, to four-day highs as much as 12ºC above normal extremes. This would be the result of many small contributions to the greenhouse effect, from CO2, H2O, O3, CH3 and various aerosols injected into the troposphere and stratosphere, from CO2 from the decay of dead plant and animal life, and from reduced surface albedo from rapid desertification. Positive and negative feedback factors were considered. The model predicts that the "cold trap," which prevents H2O from entering the stratosphere, will collapse as the stratosphere is heated by the dust and soot, and that convective activity from the oceans and from patchiness in the cloud cover will allow as much as 5x1014 kg of H2O to enter the stratosphere. As the dust and soot clear, the cold trap should drop and most of the H2O vapor precipitate, but as much as 5x1013 kg of H2O could remain in the stratosphere, enough to cause a greenhouse warming of the surface of up to 8ºC. The model also predicts that 300 1-megaton deep seabursts could put as much as 5x1013 kg of H2O into the stratosphere, enough to cause about 1-2ºC of greenhouse warming. The model predicts about 3ºC of heating from increased CO2 and another 3ºC from about 30 other substances, mainly O3 and various hydrocarbons. These effects are not all additive, so it seems unlikely that warming by more than 12ºC would result, but even 6ºC would be enough to drastically affect most lifeforms, and 10ºC could bring sustained highs fatal to most land life on earth. The model indicates that the worst of this scenario, called the "nuclear summer", would last until about 3-5 years after the nuclear war, but temperatures elevated by 3-6ºC could persist for many decades. Given certain reasonable assumptions, the long-term biological effects of the nuclear summer could be worse than those of the nuclear winter. Transition from the nuclear winter to the nuclear summer would be extremely complex, and the model does not attempt to predict how that might happen, except to suggest that stratospheric H2O vapor might accelerate the removal of dust and soot as it precipitates, shortening the nuclear winter somewhat. Conversely, dust and soot might remove more of the H2O than the model predicts.
Some of the substances created by nuclear detonations are oxides of nitrogen (NOx). Injection of NOx into the stratosphere could be expected to deplete ozone (O3) there, which acts to screen out ultraviolet (UV-B) radiation that can be injurious to most life forms. The TTAPS and Ehrlich papers warned of the impact of UV-B on ecosystems. The absorption of UV-B by O3 is also the cause of heating of the upper levels of the stratosphere, which causes it to be convectively stable. Destabilization of the stratosphere, when combined with a collapse of the cold trap [see Box - Nuclear Summer], could contribute to long-term greenhouse heating.
At lower altitudes, NOx is a major contributor to photochemical smog and the production of ozone (O3). Other substances, such as hydrocarbons, could be expected to be produced in significant quantities. Crutzen6 has argued that the Ambio reference scenario could result in O3 levels of 160 ppmv (parts per million by volume), and pointed out that daily exposures to comparable levels of O3 could seriously affect plant growth and increase plant mortality. This could further contribute to the adverse biological effects of a nuclear war.
This model uses statistical methods similar to those used to predict 50- and 100-year highs and lows for things like temperatures, rainfall, and flooding. For any given point of interest on a continent, a four-day moving average is computed over historical temperature data on daytime highs and nighttime lows at that point over a twenty-year period. Next, the thus-averaged daytime highs and nighttime lows reached within twenty days before or after a given date are found, and used to compute the probability H(T,t) that daytime highs of at least TºC above the mean for date t will be sustained for at least four days during the twenty days before and twenty days after t. Similarly, the probability L(T,t) for nighttime lows is computed. The mean difference D(t) between the mean daytime highs and nighttime lows is also calculated. By choosing a given probability, say .75, we get provisional figures of merit for temperature variability associated with that point and day of the year.
Variability will be affected by changes in temperature gradients, which will be perturbed, so we look for a correlation between our figures for variability and data on temperature gradients, either in the neighborhood, or upwind from, the point of interest. For purposes of predicting lows during a nuclear winter, it is thought that the dominant temperature gradients will be those from mid-continent to mid-ocean in the vicinity of the point of interest, so such data are used in calculations intended to project the likely increase or decrease in temperature variability for the point of interest. Using this approach on some of the averages from the NCAR GCM, the BIM indicates that most interior continental sites in temperate and subtropical latitudes of the Northern Hemisphere might have a 75 percent probability of reaching -25ºC for at least 96 hours. This is likely to be fatal to most exposed life.
Copyright © 1984 Vanguard Institute
17707 Blanco, San Antonio, TX 78232
This document also available as a text file and a
zipped WordPerfect file.