Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
The Medical Implications of Nuclear War, Institute of Medicine. @) 1986 by the National Academy of Sciences. National Academy Press, Washington, D.C. LONG-TERM CONSEQUENCES OF AND PROSPECTS FOR RECOVERY FROM NUCLEAR WAR: TWO VIEWS View II LYNN R. ANSPAUGH, PH.D. Lawrence Livermore National Laboratory Livermore, California INTRODUCTION I have been asked to comment on the information presented in this volume and to speculate on the long-term consequences of nuclear war and the prospects for recovery. In order to do that, it might be useful to define long term. To me this means time frames of years to perhaps even hundreds of years in terms of the ultimate response and recovery of large- scale ecosystems. Such long time frames may seem excessive, but if some of the speculated effects of nuclear war are actually realized, it may indeed take centuries before native ecosystems restabilize. Also, when referring to long-term effects of the magnitude required to have a major impact on entire ecosystems, it is clear that the driving force would not be the direct effects of nuclear war. Of potentially greater significance would be the secondary effects mediated by the intermediate- term impacts on global climate. Specifically, I refer to the speculative impacts of major decreases in the heat and light fluxes reaching the Earth's surface. Such changes are commonly referred to as "nuclear winter." UNCERTAINTIES IN THE "NUCLEAR WINTER" SCENARIO Because the long-term biological and ecological consequences depend on the short- and intermediate-term perturbations on global climate, it is necessary to translate these climatic effects into impacts on individual organisms and ecological systems. This problem is enormously complex. 566
VIEW II: ANSPAUGH 567 Any serious attempt to solve this biological and ecological problem must depend on accurate input data concerning the changes in global climate. Thus it seems prudent to examine first the validity of the projected impacts on global climate and to view these projections with a healthy dose of skepticism. I hasten to add that I certainly do not know whether a "nuclear winter" would or would not occur, because we simply do not have a thorough technical understanding of this issue. A similar statement can be made about many of the other projections concerning what might occur as a consequence of nuclear war. Figure 1 represents my reaction to much of what I have read in this volume. The projected impacts of global nuclear war involve extreme extrapolations beyond observational experience. There are a few things that are known with some certainty, such as the propagation of shock waves and thermal loadings. From this it is necessary to employ projections that involve a great deal of extrapolation, not only in terms of magnitude but of the processes themselves. As an example, there is speculation that "superf~res" might occur, but there are no firm data on which to base such speculations; and it is fair to say that there is simply no basic understanding of the physics of the process such that one can predict with any certainty the essential require- ments for the production of a superf~re. Furthermore, the whole concept of nuclear "winter" depends on fire phenomenology and how this trans- lates into the injection of smoke into the atmosphere. One must rely on estimates of combustible material, of the fraction of combustible material that would actually burn, of the fraction of burned material that would become smoke, of where within the atmosphere the smoke would be injected, of the carbonaceous content of the smoke, and of He optical properties of the smoke particles. - | Known ~ Magnitude - . and Processes L Unknown FIGURE 1 The projected impacts of nuclear war involve many extrapolations beyond observational experiences.
568 LONG-TERM CONSEQUENCES AND PROSPECTS FOR RECOVERY From these estimates of the smoke injected into the atmosphere, one must estimate the coagulation and scavenging of the injected smoke par- ticles and their diffusion and spreading. Finally, the impact of the injected smoke must be estimated in terms of the physics of visible light absorption and scattering with due regard given to the changes in the optical properties of the atmosphere as they also affect the absorption of infrared radiation emanating from the Earth's surface. Even assuming that all of these effects have been calculated properly, it is then necessary to model correctly the changes in the structure of the atmosphere including the great differences that can be expected to occur over land as opposed to over the ocean. Now, unfortunately, these projections are well beyond any existing data bases, and, more importantly, they are also well beyond current under- standing of the processes that are involved. One of the central issues in the "nuclear winter" scenario is the very important question: Have we perceived and modeled correctly all of the relevant processes that might be important in this situation, which is far beyond our observational ex- periences? Of course, this concern is not one sided. It is not known whether the current imperfect understanding leads to over- or underestimation of the projected effects on the global climate. The next desired step is to translate the projected effects on global climate into effects on biological and ecological systems. There is even a lesser ability to do this because there is not a long-range, broad-based program that attempts to model ecological systems on a sufficiently large scale. There has been a global climate modeling program for many years, and this has been applied to this problem with a great deal of success and with the achievement of substantial progress. Unfortunately, there is not an analog in biology and ecology that can be applied readily to this problem. However, the biologists and ecologists do have one compelling advan- tage in that they can actually perform some of the relevant experiments. That is, they can put biological systems into conditions that simulate the predicted effects of "nuclear winter" and make actual experimental de- terminations. Up to this time, such studies have not been done, but they are to begin shortly. ~ emphasize again that there is no global ecosystem model that can be applied to this problem. The best that there are at the moment are com- munity-scale models that operate on a geographic scale about equal to the size of a stage in an auditorium. Even these models are based largely on empirical data and do not contain a thorough understanding of the relevant processes so that they may be applied reliably to predict situations that appear to be beyond observational experience. In terms of recovery, these models do not deal with some of the essential processes such as extinction
VIEW II: ANSPAUGH 569 and migration. Thus at present biologists and ecologists do not have an ecosystem-prediction model that can be coupled to the output from the global climate prediction models. The global climate models are now being used to predict the effects known as "nuclear winter"; these effects generally consist of lower tem- peratures and lower light levels on the Earth's surface, and some predic- tions include lower amounts of rainfall. Typically, effects are predicted that are most severe for about 30 days, and then a return to normality is indicated. However, the further into time that these predictions are made, the more uncertain they become. It is fair to say that none of these predictions can be considered to be realistic and, if the predicted short- term effects should occur, it is an open question as to how long the effects might last. Some authors speculate that the modified atmosphere will restabilize with a situation that would reduce the scavenging processes; others postulate that We modified atmosphere, if modeled correctly, might be very unstable. Examine some of the predictions of the effect of nuclear war on global climate and how these predictions have changed as a function of time. In Figure 2 are plotted some data on predicted changes in temperature on the Earth's surface that I have taken from Table 2 in a paper by Covey. These data on predicted temperature changes are plotted as an approximate function of the study's completion date. The initial TTAPS study2 was with a one-dimensional model, and later studies3-7 have included three- dimensional interactive models that include the seasonal effects and other variables. What can be seen is that there has been a significant change in the predictions as the models have become more sophisticated. Moreover, the overall effect is toward less extreme changes (all changes shown here are roughly for midcontinent regions). * *During the discussion period during the symposium, Dr. Sagan objected to my char- acterization of change as shown in Figure 2. He stated that the TTAPS papers included a statement that, because of ocean buffering, "Actual temperature decreases in continental interiors might [emphasis added] be roughly 30 percent smaller than predicted here . . . Wand therefore the temperature change predicted by the TRAPS sutdy was -25°C, which is essentially the same as results of later calculations. However, I note that-35°C is used without caveat in their Figure 2, which is the key element of the paper. Furthermore, if the-35°C change is too large to represent changes in midcontinental temperatures, what does it represent? (Changes over the ocean and coastal areas would be smaller.) Finally, I note that the TTAPS paper had a companion paper (Ehrlich et al.), of which Sagan was a coauthor, and it quotes a value from the TTAPS study of - 40°C in the text and-43°C in its Table 1. If -25°C were the proper number, why the use of -40°C or -43°C in subsequent calculations?
570 FONG-TERM CONSEQUENCES ID PROSPECTS FOR RECOVERY o -10 c: ° -20 -30 -40 ·Jan. · Apr. · ~ Jul. . TTAPS Thompson Covey et al. 1983 MacCracken 1983 Aleksandrov 1984 1983 and Stenchikov 1983 FIGURE 2 Predictions of changes in temperature as a function of time. Source: Data from Covey (1985, p. 565~.~ As mentioned by Harwell in this volume, 9 these changes have an impact on the problem of predicting biological and ecological changes, too, in that they present a moving target of input data. The biological and eco- logical predictions would be much easier if the changes in global climate were so severe that the conclusion would be obvious. Now that the pre- dicted temperature changes are more moderate, the problem of predicting biological and ecological effects has become more difficult. The data plotted in Figure 2 are not the last word in predictions of changes in the global climate. Where the predictions might go from here is a question of great interest. I agree completely with the comment made by Turcoi° that astounding progress has been made in the ability to model and predict changes in global climate, and we can look to further advances in this ability. There also remains, however, the question of the validity of the input data that are used to drive these models. Coveys states, "The greatest uncertainty is the amount of smoke produced by fires." Carriers
VIEW II: ANSPAUGH 571 considered the uncertainties in input data and made estimates of uncer- tainties for the following: · fuel amount, factor of 2 · fuel that burns, factor of 2 · smoke-to-fuel ratio, factor of 3 · early scavenging of smoke, factor of 3 It is not obvious how these uncertainties should be propagated, but a simple root mean square of the above is a factor of 5. A central question is what effect these uncertainties, which relate to the quantity of smoke particles injected into the atmosphere, might have on the predicted impacts on global climate. Plotted in Figure 3 are some data taken from MacCracken,~2 who has examined the sensitivity of his cal- culated changes in temperature to the amount of smoke injected. (The temperature changes plotted here are calculated for 10 days after smoke injection into the Northern Hemisphere between 20° and 70° n. lat.) Typ- ical calculations for a base scenario have assumed that about 120 to 140 teragrams (Tg) of smoke are injected. As shown in Figure 3, the calculated change in temperature is quite sensitive to this assumption in a nonlinear way and the rate of change at 120 Tg is high. According to MacCracken's calculations, essentially the entire predicted effect on temperature would disappear if the amount of smoke injected were less by a factor of 2 or 3. . o - 10 -20 . . . . 30 60 120 240 Smoke, Tg FIGURE 3 Sensitivity of calculated changes in temperature to amount of smoke injected into the atmosphere. Source: Data from MacCracken (1985, p. 26~.~2
572 LONG-TERM CONSEQUENCES ID PROSPECTS FOR RECOVERY In view of this sensitivity, it seems prudent to reexamine carefully all input data and processes that are involved in calculating the amount of smoke that would be injected into the atmosphere. A major step in these calculations is the seemingly simple tabulation of all the combustible material that might be available to burn during some postulated scenario of nuclear war. Unfortunately, even this step is not easy and is subject to major uncertainties. Bingo has recently attempted such a tabulation, and his results for this total combustible inventory are one-third or less than those of other studies. If correct, this might be sufficient to eliminate the entire postulated effect of "nuclear winter," and it is important that this issue receive more attention and analysis. There are other issues as well. These involve the nature of the fuel and the resulting properties of the smoke that would be produced. As noted previously, important properties of smoke include its carbon content, its optical scattering and absorption properties, and whether the particles are wettable and able to serve as condensation nuclei. Thus, a major conclu- sion is that more data are needed to address these issues. Unfortunately, there seems to be very little work in progress to acquire such data. For example, two papers presented in this volumes ~4 refer to a managed forest fire in Canada. This fire appeared to have presented an opportunity to do some basic measurements on the smoke-to-fuel ratio and on the optical properties of the smoke particles, but neither of the presenters indicated that any such measurements had been made. Finally, there is the main question at issue: What would be the biological and ecological effects from possible changes in the global climate as a result of a nuclear war? These predictions have been changing, too. Ehrlich et al.8, with the assumption of a temperature change of about-40°C, predicted extremely dire consequences and questioned whether humans might survive as a species. Harwell9 presented the results of the just completed Scientific Committee on Problems of the Environment (SCOPE) study. His presentation indicated that the SCOPE study had encountered difficulty in trying to predict biological and ecological consequences of "nuclear winter," when the predictions from the global climate models were undergoing substantial changes. They therefore attempted to present predictions that were normalized in terms of predicted biological changes per unit of change in global climate (particularly temperature). Harwell emphasized a possible severe impact on agricultural systems that, coupled with a breakdown of social and transportation systems, might result in mass starvation. Generally, the ability to predict biological and ecological effects in response to climatic changes of any nature is poor. Even assuming that the predicted changes in global climate might be correct, there is not a
VIEW II: ANSPAUGH 573 good basis to predict changes in plant and animal populations. There are some good models of the physiological response of agricultural crops to lower temperatures, but, to my knowledge, there are no data on the response of crops to the simultaneous effects of lower temperature and lower levels of light flux. Furthermore, some predictions of changes in global climates indicate that the diurnal cycle might be destroyed. These effects, coupled with the predicted subsequent occurrence of greatly in- creased ultraviolet radiation, leave us with essentially no observational experience on which to draw to make predictions. Of greater interest over the long term are possible effects on native ecosystems. Native ecosystems are not like agricultural crops, which are annual and typically consist of tropical and subtropical plant species. There are no physiological models for native plants and there are even fewer relevant data that can be applied to this problem. The processes and interactions themselves are poorly understood. The simulation models that are available and that were used in the SCOPE study are limited and of questionable validity for the projected input conditions. CURRENT CAPABILITY TO PREDICT BIOLOGICAL AND ECOLOGICAL CONSEQUENCES The Lawrence Livermore National Laboratory (LLNL) and Stanford University recently cohosted a workshops with the goal of examining the current state of knowledge that can be applied to predicting the biological and ecological effects of a postulated "nuclear winter." A major result of that workshop was that the ability to make quantitative predictions is very poor, but there is a large body of expert opinion that can be used to identify and classify ecosystems with regard to their sensitivity, resilience, etc. Figure 4 is one such example from the LLNL-Stanford Workshop, in which major ecosystems were classified with respect to their suscep- tibility to the stresses following a nuclear war. Harwell9 showed similar attempts to classify ecosystems; a typical result is that the tropical rain forest is identified as the most sensitive ecosystem. This classification scheme, coupled with projected changes in global climate, allows deri- vation of some impression for the amount of damage to ecosystems that might result. The next step is to examine the characteristics of the ecosystem types in terms of the features that might be important in the recovery of that system from damage. An attempt to do that is shown in Figure 5, which is also taken from the LLNL-Stanford Workshops Again, this is not a quantitative procedure, and it is only possible to classify and rank prop- erties of resilience. This is because there is not a good understanding of
574 LONG-TERM CONSEQUENCES AND PROSPECTS FOR RECOVERY System Cold Low light Fire Toxic gas Radiation Tropical rain forest High High Moderate High Moderate Temperate forest Moderate Moderate Moderate Moderate Moderate Boreal forestLow Low Low Moderate High GrasslandModerate Moderate Moderate Moderate Low SavannaPligh High Low Low Low DesertLow (?) High Low Low Moderate ChaparralModerate Moderate Low Low Moderate TundraLow Low Moderate Low Moderate FIGURE 4 Susceptibility of the blames to the stresses following a nuclear war. several processes that would be needed in order to do this in a quantitative way. These processes include succession, local extinctions, and migration of species. It is interesting to note that Figure 5 indicates that the tropical rain forest has the poorest prospect for recovery. Thus, this ecosystem has been identified as both highly sensitive to the effects of global climatic changes and as having a poor prospect for recovery from any induced damage. Recovery Potential Environmental growth Seed Resprouting Breeding Symbiotic Ecosystem potential bank ability system interactions Tropical rain forest High Low Low Low Low Temperate forest Moderate Moderate Moderate Moderate Moderate Boreal forest Low Moderate Moderate High High Grassland Moderate High Moderate High Moderate Savanna Moderate Moderate High High Low Desert Low Moderate High Moderate Moderate Chaparral Moderate Moderate High Moderate Moderate Tundra Low Moderate Low Moderate High FIGURE 5 Components of resilience (recovery) of different blames following death of the dominant species. High means that rapid recovery is likely.
VIEW II: ANSPAUGH 575 WHAT IS NEEDED TO PREDICT BIOLOGICAL AND ECOLOGICAL EFFECTS? It would be desirable to have a suite of models, such as those shown in Figure 6, for the prediction of biological and ecological effects. One of the concepts illustrated in Figure 6 is the desirability of having models capable of dealing with a range of input data and having output specified as probability distributions. Reasonable models of global climate and individual organism response exist at present, but adequate ecosystem response and regional assessment models do not exist. Also, there are not now adequate data for input to such models, even if the models did exist. Measurements of biological response have not been conducted under environmental conditions similar to those proposed for a "nuclear winter." These experiments to determine the response of individual organisms to the cold and dark are rather simple, and even more complex experiments are possible by building large exposure cham- bers over native communities. One conclusion I offer is that we need additional research in order to understand the biological and ecological effects. The SCOPE study re- ported in this volume by Harwell9 was a tremendous undertaking, but it was largely a volunteer, short-term effort that did not produce any new data. Rather, the objective was to synthesize existing data and to rely on - n MT, smoke, etc. Global cilmate model - - - ~ 8 :_ Mortality, growth, etc. ·k - - - ~ D [f-~ ~ Temperature, light, etc. Ecosystem response model Individual organism response model . Data - - - D n Extinctions, population changes, etc. Regional- D . Tabulation of assessment _ o _ national and model _ ~ ~_ global effects Time of recovery, l species composition, etc. _ FIGURE 6 Nuclear winter assessment strategy.
576 LONG-TERM CONSEQUENCES ID PROSPECTS FOR RECOVERY Studies to be done at the University of Wisconsin, Madison (J. Palta, B. McCown, T. Tibbitts) Initial studies with wheat, potato, soybean, loblolly pine Short-term Minimum temperature of survival for different physiological periods in plant development Long-term Simulate ''nuclear winter'' conditions predicted by MacCracken, et al. for continental N. America Measure effects on plant productivity and mortality FIGURE 7 Biological studies in controlled environments. expert judgment and existing simulation models to make predictions. This effort represents an excellent beginning, but it still leaves important gaps in the needed data base and in the ability to simulate the required processes. As mentioned by other speakers, there is currently not a major research program with the goal of predicting the biological and ecological effects of nuclear war. However, a modest program has been established at LLNL. One aspect of this program is experimental; its focus is outlined in Figure 7. This study will be conducted by the University of Wisconsin, Madison, under contract to LLNL. The goal is to acquire basic data on plant re- sponses to conditions that mimic "nuclear winter." At LLNL an attempt is also being made to develop new models that are capable of simulating large-scale ecosystems and which could deal with the "nuclear winter" conditions as input data. This is an activity that has a reasonable probability of success over several years and that could have a large payoff in application to other ecological problems of a global nature. Examples are the increasing level of carbon dioxide in the atmosphere and large-scale deposition of acid. SPECULATIONS ON THE LONG-TERM CONSEQUENCES AND PROSPECTS FOR RECOVERY Finally, I return to my assignment of speculating on the long-term consequences and the prospects for recovery. My opinion is that it simply is not known what the long-term biological and ecological effects would be, because there is not a firm grasp of what the effects of nuclear war would be on the global climate. Even if these effects were known, we would still fall far short of being able to translate accurately these climatic effects into biological and ecological effects.
VIEW Il.: ANSPAUGH 577 However, it can be speculated that agricultural crops might be quite vulnerable and entire crops could be lost. Over the short term, this alone could produce a very serious problem for the surviving human population. Over the long term, agricultural crops could be reestablished, if the re- quired energy and infrastructure are available. I doubt that any major ecosystem would be completely destroyed. How- ever, I have no way of knowing that, and any impact would be very sensitive to the actual changes in global heat and light fluxes. If substantial damage should occur to some ecosystems, they would undoubtedly re- stabilize over the long term to a structure that might or might not resemble the original. OTHER REACTIONS TO THE MATERIAL PRESENTED One interesting and surprising result presented in this volume by Rot- blat~6 was his calculation of the dose that would produce 50 percent mortality (LDso) for the Japanese exposed to radiation from the atomic bomb at Hiroshima. His calculated result was lower by about a factor of 2 than that which is commonly accepted. As Rotblat pointed out, his result is preliminary and awaits the conclusion of the major reassessment of the atomic bomb dosimetry that is now under way. If his result should stand, it may have a substantial impact on any assessment of casualties due to radiation exposure. I was also surprised by the attention given to the paper presented by Greer and Rifkin,~7 wherein they pointed out that radiation and acquired immunodeficiency syndrome (AIDS) have somewhat similar effects on the immune system. While I follow the argument, this comparison fails on several significant fronts. The most important is that the effects of moderate doses of radiation on the immune system are neither critical nor life threatening and they are reversible, whereas the effects of AIDS on the immune system are the major effect of the disease and they are not reversible. Thus, I find the comparison to be superficial at best and a later characterization of nuclear war as a "global case of AIDS" i~ to be sub- stantially misleading. No one doubts that a large-scale nuclear war would be one of the worse, if not the worst, environmental disasters known to mankind; no exaggerations are needed to make that point. Finally, I have had a strong impression of an unusual willingness of participants in the symposium to accept the offered predictive results and speculations at face value. It seems unusual for this to occur when these involve extrapolations that are so far beyond our observational experience. Perhaps a reason for this could be that the process has given way to emphasis on the desired and hoped for result: a final reason why global
578 LONG-TERM CONSEQUENCES AND PROSPECTS FOR RECOVERY nuclear war must be avoided. While I have great sympathy for this view- point, I believe that we as scientists will ultimately serve society best by examining every aspect of this issue more critically. ACKNOWLEDGMENT This work was performed under the auspices of the U.S. Department of Energy by the Lawrence Livermore National Laboratory under contract number W-7405-ENG-48. NOTES ~Covey, C. 1985. Climatic effects of nuclear war. BioScience 35:563-569. 2Turco, R. P., O. B. Toon, T. P. Ackerman, J. B. Pollack, and C. Sagan. 1983. Nuclear winter: Global consequences of multiple nuclear explosions. Science 222:1283-1292. 3MacCracken, M. C. 1983. Nuclear War: Preliminary Estimates of the Climatic Effects of a Nuclear Exchange. Paper presented at the Third International Conference on Nuclear War, Erice, Italy, August 19-23, 1983. (Quoted in note 1.) 4Thompson, S. L. 1983. A Comparison of Baroclinic Eddy Heat Transport to a Simplified General Circulation Model. NCAR Cooperative Thesis No. 71, National Center for At- mospheric Research, Boulder, Colo. (Quoted in note 1.) iThompson, S. L., and C. Covey. 1983. Influence of Physical Processes in General Circulation Model Simulations of Massive Atmospheric Soot Injections. American Geo- physical Union, Fall Meeting, San Francisco, December 1983. Washington, D.C.: Amer- ican Geophysical Union. (Quoted in note 1.) 6Aleksandrov, V. V., and G. L. Gtenchikov. 1983. On the Modeling of the Climatic Consequences of Nuclear War. Moscow: The Computing Center of the USSR Academy of Sciences. (Quoted in note 1.) 7Covey, C., S. H. Schneider, and S. L. Thompson. 1984. Global atmospheric effects of massive smoke injections from a nuclear war: Results from general circulation model simulations. Nature 308:21-25. (Quoted in note 1.) ~Ehrlich, P. R., J. Harte, M. A. Harwell, P. H. Raven, C. Sagan, G. M. Woodwell, J. Berry, E. S. Ayensu, A. H. Ehrlich, T. Eisner, S. J. Gould, H. D. Grover, R. Herrera, R. M. May, E. Mayr, C. P. McKay, H. A. Mooney, N. Myers, D. Pimentel, and J. M. Teal. 1983. Long-term biological consequences of nuclear war. Science 222:1293-1300. 9Harwell, M. A., and C. C. Harwell. 1986. Nuclear famine: the indirect effects of nuclear war. This volume. i°Turco, R. P. 1986. Recent assessments of the environmental consequences of nuclear war. This volume. iiCarrier, G. F. 1986. Nuclear winter: the state of the science. This volume. i2MacCracken, M. C. 1985. Global atmospheric effects of nuclear war. Pp. 10-35 in Energy and Technology Review. UCRL-52000-85-5. Livermore, Calif.: Lawrence Liv- ermore National Laboratory. ~3Bing, G. 1985. Estimates of total combustible material in NATO and Warsaw Pact Countries. UCRL-93192. Livermore, Calif.: Lawrence Livermore National Laboratory. ~4Brode, H. L., and R. D. Small. 1986. A review of the physics of large urban fires. This volume.
VIEW II: ANSPAUGH 579 i5Kercher, J. R., and H. A. Mooney, eds. In press. Research agenda for ecological effects of nuclear winter. UCRL-53588. Livermore, Calif.: Lawrence Livermore National Labo- ratory. Rotblat, J. 1986. Acute radiation mortality in a nuclear war. This volume. Career, D. S., and L. S. Rifkin. 1986. The immunological impact of nuclear warfare. This volume. Sagan, C. 1986. Long-term consequences of and prospects for recovery from nuclear war: view I. This volume.