Effects of Nuclear Weapons on the Atmosphere
F. Sherwood Rowland, 1985
F. Sherwood Rowland is a professor of chemistry at the University of California, Irvine and recipient of the 1995 Nobel Prize in Chemistry. He has received worldwide recognition for alerting nations to the danger posed by chlorofluorocarbons when released into the atmosphere through aerosol propellants, refrigerants, and solvents. Honors conferred upon Dr. Rowland include election to membership in the National Academy of Sciences, fellowship in the American Academy of Arts and Sciences and the American Geophysical Union, and the Tyler Prize for Environmental Achievement, the world prize in ecology and energy.
Whiteley: Your consideration of nuclear winter begins with several aspects of our current situation. The first is that the atmosphere has no national boundaries. What do you mean?
Rowland: Well, basically the atmosphere mixes extremely rapidly. Something which is released in Los Angeles will reach New York in a few days, it will reach Western Europe in a week or two, and completely around the world in a matter of a few weeks. The mixing to the southern hemisphere is a little bit slower, but still takes place within a year or so, so that anything that you put into the atmosphere that has a lifetime of a year or more, and many gasses that are released to the atmosphere do have lifetimes of that length, will be mixed everywhere, so that one problem isn't restricted just to an individual country. If you release it in the United States it's a problem for the world; if you release it in Africa, it's a problem for the world. That makes it a general problem for everyone.
Whiteley: Let's take three current problems that you've singled out. First is acid rain.
Rowland: Well acid rain comes about from the release of sulfur and nitrogen oxides, partly from - mostly from burning fossil fuel, either by burning it in power plants which releases a lot of sulfur, or in automobiles which releases a lot of nitrogen oxides. These then move, and they become a regional problem. If it's released in the midwestern part of the United States, it will show up in Quebec, and it becomes a U.S./Canada problem. If it's released in West Germany or in England, it shows up in Scandinavia, and again it's an international problem of countries that are separated by a thousand miles or so. That's a regional problem. There are two others that have been considered extensively in the atmospheric community in the last ten years or so. One of these is the "greenhouse effect" which comes also from burning fossil fuels which releases carbon dioxide. And here it becomes a world problem because carbon dioxide spreads over the entire world, and the amount of carbon dioxide has been moving up steadily over the last twenty-five years.
The third problem is the problem of stratospheric ozone, and this comes about - it becomes a problem because there are gasses being released by man which get up into the stratosphere and can affect the ozone. But before they get up into the stratosphere, they mix thoroughly all around the world.
Whiteley: The point is that the atmosphere already is fragile, already has some significant problems that have no national boundaries, so the beginning considerations of nuclear winter must take into account the fact that we are already causing problems in the environment.
Rowland: We are already quite concerned about the possible changes in the atmosphere. We know the changes are going on; it's the possible consequences of these changes for a release of gasses such as carbon dioxide and chlorofluorocarbons and sulfur dioxide into the atmosphere, and the effects that they have - biological, climatic - and the effects that they have on countries well away from those in which the release took place.
Whiteley: The phenomena of nuclear winter is one that has come to the attention of the community four decades into the nuclear age. Prior to the consideration of nuclear winter, what were the major atmospheric concerns of nuclear weapons?
Rowland: I would say - make it broader than just atmospheric concerns - the release of nuclear energy in a bomb produces a tremendous amount of blast and heat and radiation and then fallout. And over the time period since 1945 when the first bombs were set off, there has been immense concern about all of these effects. The initial bombs which were set off over Japan in 1945 were at the level of about 20 kilotons. Now one hears the bomb strength measured in megatons, which is a thousand - a megaton is a thousand kilotons - and the talk is about full exchange between two countries such as the United States and the Soviet Union. You're talking 5,000 to 10,000 megatons, each megaton being 50 times the effect of Hiroshima or Nagasaki. Enormous amounts of blast energy released over cities, or over military targets, and radioactive fallout coming downwind from this, and a typical calculation of what the effects would be from such a nuclear exchange are of the order of 750 million people killed immediately, and another several hundred million people who had been seriously maimed and affected by one or the other of these effects. Now, the nuclear winter is a further consideration that is added onto the top of this, and this has to do with longer term atmospheric effects, and longer term meaning that they might last weeks or several months.
Whiteley: Let's take the components of that nuclear winter consideration one at a time, and ask you to elaborate on their potential effects. The first would be smoke in the troposphere.
Rowland: Well, the question of nuclear winter is - the whole phrase "nuclear winter" has to do with the interference with the arrival of sunlight at the surface of the earth, and this interference can come about by putting into the atmosphere something which will absorb sunlight before it reaches the surface. The main considerations here have to do with dust, which would be kicked up by the nuclear explosions, and with smoke, which would come from the fires that would be set off by such nuclear explosions. The fires - there's a question of soot for instance coming from a burning city is one of the important considerations in nuclear winter. If you put up black particles into the atmosphere, you have the possibility that they will reflect sunlight upward and away from the earth, preventing it from reaching the surface of the earth, and ending up cooling the earth because it fails to receive that sunlight, and it is this possibility that has led to the terminology "nuclear winter."
Whiteley: Given the general effects that are identified, one is smoke in the troposphere. Why will that linger?
Rowland: Well, the question of whether - how long it will linger is very much a wide open question. The difficulties that one has in trying to estimate how long such smoke will last - clearly if you burn - if you see forest fires now, some of the smoke that goes up from the forest fire lasts long enough to go across the country. It lasts a few days, and then eventually dissipates and is taken out. The kind of question that comes up in nuclear winter is that such fires would be - there would be many such fires, literally thousands of them burning simultaneously, and the question of how rapidly that would dissipate is one that isn't so easily answered. You can't just say that if one forest fire is typical then a thousand will behave the same, because when you have all of them doing it at the same time, putting tremendous amounts of soot in the air, you may change the characteristics which remove smoke from the air.
Whiteley: Second was dust in the stratosphere.
Rowland: Well, there's no question that anything you put up into the stratosphere takes much longer time to come down. Particles that are put up in the stratosphere typically take several years to be removed, so anything that gets up that far gets mixed around the world, and gradually comes down over a time scale that's not weeks, but years.
Whiteley: Why does not gravity have a greater effect on particles like dust?
Rowland: Well, gravity has an effect on things in proportion basically to their size. Now obviously gravity works on everything, but the question of whether dust falls out is an intimate question of how big the particles are, because if you have them small enough, then they get blown about and the mixing by the winds overcomes the pull of gravity on very small particles; larger particles fall out very quickly. If you look after a volcano, for instance, the large particles come out 50, 100, 200, 300 miles downwind from the volcano, but the smaller particles go up higher and last much longer.
Whiteley: A third effect was radioactive fallout.
Rowland: Yes. Any of the nuclear exchanges is going to involve the release of a substantial amount of radioactivity, and this radioactivity will be distributed on particles, and much of it will come out in the immediate 150 or 100 miles downwind of wherever the bomb was let off. And some of it will be carried up much higher. The radioactive fallout will be very intense immediately after a nuclear explosion, and then gradually dies out. And some of these things have short half-lives of - some of the fission products that come from nuclear explosions have half-lives of days, weeks, months; some of them last 25 years, some of them 100 years. So it will be a contamination which is very intense early and then will gradually die out, but which will last for many tens of years.
Whiteley: A fourth effect is the destruction of the ozone layer. What are the effects of that?
Rowland: Well, the ozone is present in the upper atmosphere, in the stratosphere. It's produced by the action of sunlight on molecular oxygen, molecular oxygen being one of the two major components of the air, the other being nitrogen. Nitrogen is about 4/5 of the atmosphere, oxygen is about 1/5. And sunlight acting on oxygen produces another form of oxygen which has three atoms instead of two, and this is the one that is called ozone. And ozone can be removed from the atmosphere by interactions with some of the other chemicals that are released, and in particular it can be attacked by nitrogen oxides, and setting off a nuclear weapon will cause some of the nitrogen and oxygen in the atmosphere to combine to form nitrogen oxides. These nitrogen oxides would then have an effect on ozone, and would remove some of the ozone that's present in the atmosphere.
Whiteley: What are the effects on people of depleted ozone?
Rowland: The question of what the overall effects would be depends on how much depletion, and how long. The ozone serves to shield the whole surface of the earth from the arrival of ultraviolet radiation from the sun. The effects of ultraviolet radiation on humans, particularly the damaging ultraviolet radiation, is to cause skin cancer. On other biological species it may be more severe because they happen to be smaller, and the radiation penetrates more into the depths into the biological species, than is just taken out on our skin. But that's - one of the possible effects is to certainly increase ultraviolet radiation with whatever effects that will have on each of the individual biological species that's at the surface of the earth.
Whiteley: I'd like to ask you to step back from the immediate effects in this identification of some of the long term effects known as "nuclear winter," and to approach the problem as a scientific problem, as the National Academy of Sciences recently did. As a scientist begins to examine a problem like nuclear winter, where do you begin?
Rowland: Well I think the place that Crutzen and Birks who really brought this subject up in 1980 and 1981, the place they began was starting to ask questions about forest fires and burning cities, that up to that time there had been a lot of consideration given to radioactive fallout to the blast and the heat, but not much consideration to effects in the atmosphere that might last for weeks or months. One such effect that was known was the possible effect on ozone. And so Crutzen and Birks started out to calculate what the effects on ozone would be, but then they began to ask the question well, what about the fires, wouldn't they put things into the atmosphere as well? And now the kind of question that one has as a scientist approaching that problem is if you put it in a blunt way, what kind of soot would be produced if you set off nuclear explosions over Los Angeles, and there aren't any answers to that. You can try to make approximations to it, you can try to estimate what comes out when you burn forest fires, but that's not necessarily the same as burning Los Angeles or Paris, or some other cities - and so you're left immediately with some big uncertainties. And so you try to do the best you can of estimating how much burning there would be, how much carbon would be put in the air, and that was the beginning of the calculations on nuclear winter.
Whiteley: So that began with a set of assumptions. For example, in a National Academy of Sciences report, that the military action would be divided into air bursts which would bring up some kinds of particles and some kinds of effects, and ground bursts, which would have a different effect. How do those affect your modeling?
Rowland: Well, in the particular case of Crutzen and Birks, they were part of a larger study that was published eventually by the Swedish magazine, AMBIO to propose a scenario of a possible nuclear war, and then asked what would be the consequences of this, and in that they had very carefully specified there would be so many bombs delivered at such and such locations, total bomb tonnage, so much in the northern hemisphere, so much in the southern hemisphere. Everything was completely fixed, the time of day that it happened and so on. Then they were asked to calculate for that. But in general, if they had asked it a little bit differently, then you would have had a different set of assumptions, and the question of whether you have 5000 megatons distributed between two warring powers, or whether you have something which they kind of describe often as a surgical strike of 100 megatons, or something like this. This is an assumption as to what the nature of a particular nuclear war might be, and then you try to go from that. But clearly there has to be some kind of decision as to what you think is going to be the starting point for it before you can start doing calculations.
Whiteley: Given those set of assumptions, a second set of assumptions were made about the types of particles, the types of chemicals. How does a scientist begin to think about those?
Rowland: Well, the starting point for that, what they were doing was to look at the existing knowledge of fires. There are measurements on some of the forest fires, there are measurements of a variety of kinds of chemicals that are given off when some of the materials of modern commerce are subjected to fire and to decomposition from heat, and then you try to go from that to estimating what it would be like if you had this on a very large scale, if there were very many fires and tried to multiply up from the measurements that are actually known. But this process of multiplying up from small fires to very large fires, and from very large fires, to very large fires going simultaneously in hundreds of places, has its own errors in it that make any conclusion that one draws quite uncertain.
Whiteley: You've indicated, however, that there have been some natural phenomena over the years that have been studied, and can provide some insights into the worldwide effects of even limited explosions of a non-nuclear nature.
Rowland: Well, it's certainly true that when you have a large volcanic explosion, the volcano El Chichon which went off very recently, immediately put particles into the atmosphere which spread worldwide and were picked up in the stratosphere everywhere. The time scale for this mixing was of the order of a few months to a year or two. The explosion of the volcano Tambora in 1815 put particles into the atmosphere that stayed for several years and caused severe climatic disturbance in New England and in Europe in the summer of 1816. So we know from these experiences that it is possible for volcanic dust to be put into the atmosphere, and last for a period of time of several years with appreciable effects. The Summer of 1816 was very cold throughout most of the northern hemisphere. This was a volcanic explosion in tropic regions, but which spread around the world. The volcanic explosion of Krakatoa in 1883 also produced effects in the atmosphere that could be seen for several years.
Whiteley: In the years ahead when there are two- or three-dimensional models that allow a much more complex set of interactions in making predictions, what will it be possible to do in terms of studying nuclear winter?
Rowland: Some of the uncertainties can be narrowed and reduced. Others, I think, are not really going to be approached. It's going to be very difficult to know what the particle size will be from a burning city if you haven't made measurements on a burning city. And we don't have any measurements, we hope we don't get any measurements for this. So that will be an uncertainty that simply remains. That you'll have to have some sort of approximation, some sort of estimate of what you think it will be like, and in that case there will surely also be disagreements among various scientists who say I think that's probably a little bit overestimated one way or the other. And these uncertainties will multiply, and so in the end your conclusions are going to have big uncertainty bars that maybe something very severe might happen, maybe something less severe would happen. And you're not going to be able to choose between these regardless of how good your model is, because some of the input data is going to be still not measurements, but just calculations.
Whiteley: Having identified the significant components of nuclear winter and the kinds of considerations that have led scientists to assert that nuclear winter is a real phenomena, that there is much that can be known about it based on prior knowledge as well as modeling, the problem still exists of the nature of scientific uncertainty and how, in a world with a fragile atmosphere, your fellow citizens should think about the nature of scientific investigation on a problem of major significance. What models are available to make even more precise the current levels of analysis?
Rowland: Well, the way that the atmosphere is modeled now is often described in terms of one-dimension, two-dimension, three-dimensional models. In a one-dimensional model, all that one considers is the change in altitude, in height. And you take approximate characteristics that you say are sort of typical of the world, and try to estimate what will happen at different altitudes above that particular location. In a two-dimensional model what you do is to see how this would happen as a function of the latitude; that is, now instead of asking the question for sort of an average place for the world, you ask the question well, what would it be like over someplace in the tropics, and what would it be like in the polar regions, and what would it be like in the north temperate zone, and would it be the same in the southern hemisphere, and so on. And then in a full three-dimensional model, you put in the change across longitude as well, which would mean that you ask whether there would be differences on the west coast and the east coast of the United States, and whether there would be differences between Western Europe and Canada and so on. The ultimate goal for almost all modelers is to have a good working three-dimensional model with full chemistry, and this is enormously complex, and it is well beyond anything which is currently available. So that in terms of atmosphere prediction, what one does is to decide which model is best for the circumstances that you're considering, and for many of the cases of global effects, the first approximation is done with the one-dimensional model. You ask what would it be like at various altitudes, and take the present atmosphere and try to simulate the present atmosphere, and then you put into that some change - the change could be putting in carbon in the form of soot, or the change could be putting in sulfur dioxide from burning fossil fuels, or it could be nitrogen oxides from burning, from automobiles, and then you ask what will that do to the model.
Whiteley: Given the limits of scientific uncertainty about the problems of nuclear winter, the National Academy of Science nevertheless came to some general kinds of conclusions. What for you are the central known facts about nuclear winter?
Rowland: I think that there will certainly be, if there were such a large scale nuclear exchange, there would be large amounts of soot put into the atmosphere; there would be large amounts of dust put into the atmosphere. Some of this would last for weeks, maybe for months. In many locations, especially inland locations, the sunlight would be cut off for substantial periods of time, and so the temperatures would be much less at the surface of the earth. How much less is quite uncertain. It could be 20 or 30 degrees Fahrenheit less, and if that occurred in the summer, then it would be a very cold summer. It would have very substantial effects on any biology that was taking place such as growing food crops. If it happened in the winter the temperature drop might be different than in the summer, but it would just make that much worse all of the effects that you have from the radiation and the blast. If you were suddenly to kill 750 million people in the north temperate region, and have another 350 million that were severely injured, disrupted most of your civilization, then you clearly are going to have major, major problems even if you didn't add to that severe climatic disturbance. It seems to me that the consequence of an all out nuclear exchange may not be the end of life - of human life on earth, but it would be the end of present civilization.
Whiteley: Professor Rowland, thank you for sharing with us today your insights from the scientific community about the phenomena of nuclear winter.