Censored information from The Effects of Nuclear Weapons

'On the one hand, a strong point can be made for the fact that civil defense, appropriate civil defense, would save a great many lives; on the other hand, one can also make a strong case that any civil defense, appropriate or otherwise, may increase the danger of a war because people will feel that they don't have to avoid it because they have civil defense.'

- Dr Frank Fremont-Smith, director of the New York Academy of Sciences Interdisciplinary Communications Program, Proceedings of the Second Interdisciplinary Conference on Selected Effects of a General War, DASIAC Special Report 95, July 1969, vol. 2, DASA-2019-2, AD0696959, page 281.

On this blog you will find links to online versions of the 1957 and 1977 editions of The Effects of Nuclear Weapons, by Glasstone and Dolan, U.S. Department of Defense. But you won't find any links to the longest version of the book, the 1962/4 edition, which will be quoted and discussed at length below. Here is the Foreword, written and signed by the U.S. Secretary of Defense Robert S. McNamara and the U.S. Atomic Energy Commission Chairman Glenn T. Seaborg (both appointed by Kennedy), to the 1962/4 edition:

'This book is a revision of The Effects of Nuclear Weapons which was issued in 1957. It was prepared by the Defense Atomic Support Agency of the Department of Defense in coordination with other cognizant government agencies ... Although the complex nature of nuclear weapons effects does not always allow exact evaluation, the conclusions reached herein represent the combined judgement of a number of the most competent scientists working on the problem.

'There is a need for widespread public understanding of the best information available on the effects of nuclear weapons. The purpose of this book is to present as accurately as possible, within the limits of national security, a comprehensive summary of this information.'

The major part which was deleted in the 1977 edition was the final chapter (Chapter 12), Principles of Protection, which on page 631 (1962/4 edition) states:

'At distances between 0.3 and 0.4 mile from ground zero in Hiroshima the average survival rate, for at least 20 days after the nuclear explosion, was less than 20 percent. Yet in two reinforced concrete office buildings, at these distances, almost 90 percent of the nearly 800 occupants survived more than 20 days, although some died later of radiation injury.

'Furthermore, of approximately 3,000 school students who were in the open and unshielded within a mile of ground zero at Hiroshima, about 90 percent were dead or missing after the explosion. But of nearly 5,000 students in the same zone who were shielded in one way or another, only 26 percent were fatalities. ... survival in Hiroshima was possible in buildings at such distances that the overpressure in the open was 15 to 20 pounds per square inch. ... it is evident ... that the area over which protection could be effective in saving lives is roughly eight to ten times as great as that in which the chances of survival are small.'

Page 645 (1962/4 edition) provides definite ideas:

'The major part of the thermal radiation travels in straight lines, so any opaque object interposed between the fireball and the exposed skin will give some protection. This is true even if the object is subsequently destroyed by the blast, since the main thermal radiation pulse is over before the arrival of the blast wave.

'At the first indication of a nuclear explosion, by a sudden increase in the general illumination, a person inside a building should immediately fall prone, and, if possible, crawl behind or beneath a table or desk or to a planned vantage point. Even if this action is not taken soon enough to reduce the thermal radiation exposure greatly, it will minimise the displacement effect of the blast wave and provide a partial shield against splintered glass and other flying debris.

'An individual caught in the open should fall prone to the ground in the same way, while making an effort to shade exposed parts of the body. Getting behind a tree, building, fence, ditch, bank, or any structure which prevents a direct line of sight between the person and the fireball, if possible, will give a major degree of protection. If no substantial object is at hand, the clothed parts of the body should be used to shield parts which are exposed. There will still be some hazard from scattered thermal radiation, especially from high-yield weapons at long ranges, but the decrease in the direct radiation will be substantial.'

A person on the ground whose clothes ignite (which is only a risk under extremely high thermal exposure to dark coloured clothing) can immediately extinguish the clothes by simply rolling over to starve the flames of oxygen.

Page 653 (1962/4 edition) correctly explains:

'Some, although perhaps not all, of the fallout in the Marshall Islands, after the test explosion of March 1, 1954, could be seen as a white powder or dust. This was due, partly at least, to the light color of the calcium oxide or carbonate of which the particles were mainly composed. It is probable that whenever there is sufficient fallout to constitute a hazard, the dust will be visible.'

Page 658 (1962/4 edition) shows how to deal with fallout on food and water:

'If emergency food supplies do become contaminated, or if it is necessary to resort to contaminated sources after emergency supplies are exhausted, many types of food can be treated to remove the radioactive material. Fresh fruit and vegetables can be washed or peeled to remove the outer skin or leaves. Food products of the absorbent typecannot be decontaminated in this manner and should be disposed of by burial. Boiling or cooking of the food has no effect in removing the fallout material. Milk, from cows which survive in a heavily contaminated area, may not be safe to drink because of the radioiodine content and this condition may persist for weeks or months.

'Domestic water supplies from underground sources will usually remain free from radioactive contamination. Water supplies from surface sources may become contaminated if watersheds and open reservoirs are in areas of heavy fallout. However, most of the radioactive fallout material would be removed by regular water treatment which includes coagulation, sedimentation, and filtration. If a surface water supply is not treated in this manner, but merely chlorinated, it may be unfit for consumption for several days after an attack. As a result of dilution and natural decay the contamination will decrease with time.

'If the regular water supply is not usually subjected to any treatment other than chlorination, and an alternative source is not available, consideration should be given in advance planning to the provision of ion-exchange columns or beds for emergency decontamination use. Home water softeners might serve the same purpose on a small scale. The water contained in a residential hot-water heater would serve as an emergency supply, provided it can be removed without admitting contaminated water. Water may also be distilled to make it safe for drinking purposes. It should be emphasized that mere boiling of water contaminated with fallout is of absolutely no value in removal of the radioactivity.'

Decontamination of streets, buildings and farm land is discussed on page 659 (1962/4 edition):

'Because of its particulate nature, fallout will tend to collect on horizontal surfaces, e.g., roofs, streets, tops of vehicles, and the ground. In the preliminary decontamination, therefore, the main effort should be directed toward cleaning such surfaces. The simplest way of achieving this is by water washing, if an adequate supply of water is available. The addition of a commercial wetting agent (detergent) will make the washing more efficient. The radioactive material is thus transferred to storm sewers where it is less of a hazard [underground drains are well shielded from people]. ... if facilities are to be provided across open country which is contaminated over large areas, bulldozing the top few inches of contaminated soil to the sides will be satisfactory only if a wide strip is cleared. Thus, if the strip is 250 feet in width, the radiation dose rate in the middle will be reduced to one-tenth of the value before clearing. A similar result may be achieved by scraping off the top layer of soil and burying it under fresh soil. Something like a foot of earth cover would be required to decrease the dose rate by a factor of ten.'

The final page of Chapter 12, page 661 (1962/4 edition), has a Conclusion which states:

'Much of the discussion presented in earlier sections of this chapter have been based, for simplicity, on the effects of a single nuclear weapon. It must not be overlooked that in a nuclear attack some areas may be subjected to several bursts. The basic principles of protection would remain unchanged, but protective action against all the effects of a nuclear weapon - blast, thermal radiation, initial nuclear radiation, and fallout - would become even more important. There is a good possibility that many people would survive a nuclear attack and this possibility would be greatly enhanced by utilizing the principles of protection in preattack preparations and planning in taking evasive action at the time of an attack, and in determining what should be done in the recovery phase of an attack.'

There is also some important censored information in the earlier June 1957 edition of The Effects of Nuclear Weapons, which is available online (see pages 514-517 of the 1957 edition): on page 514-5 three photos of successful Japanese 'blast-shielding walls' at Nagasaki which stopped damage from the nuclear explosion are shown. These walls are about 10 feet tall, and are wider at their base than at their top to avoid overturning by blast. They are made out of either precast reinforced concrete or earth-filled wooden panels. Both types fully protected vital Japanese machinery, such as electrical transformers, at 0.85 mile from ground zero in Nagasaki. Pages 516-7 of the 1957 edition shows photos of unprotected earth-moving equipment (bulldozer and road grader) badly damaged by 30 psi peak overpressure in a Nevada test: the road grader has lost tyres and the bulldozer is overturned with track damage. Another picture is shown of identical equipment completely protected after exactly the same 30 psi peak overpressure nuclear blast, because the bulldozer and road grader in this example are in an open trench (at right angles to the blast motion) which has a depth equal to the height of the equipment. Although some blast wave energy diffracted straight into the trench, the main mechanism for blast damage is wind drag, which is caused by directional dynamic pressure. Unlike the overpressure, the wind pressure does not diffract unless it is forced to do so by being blocked. Hence the blast wind blew straight over the top of the open trench, without causing any displacement or damage to the equipment. All of these photos and information were removed from all future editions of The Effects of Nuclear Weapons.

Some earlier posts related to nuclear effects and civil defence:

http://glasstone.blogspot.com/2006/08/nuclear-weapons-1st-edition-1956-by.html

http://glasstone.blogspot.com/2006/04/white-house-issues-new-civil-defence.html

The best suggestion about how to shield fallout gamma radiation in the home in an emergency remains the old but valid idea of staying in an inner room, as far from outside walls and the roof as possible, and with as much massive furniture (or any other massive objects) intervening between the fallout contamination outside and the people taking refuge, as possible.

Electromagnetic Pulse (EMP)

It is interesting to quote the scientific section from the April 1962 edition of The Effects of Nuclear Weapons introducing the electromagnetic pulse, pages 502-506 of Chapter X, Radio and Radar Effects (for the 1977 edition chapter on electromagnetic pulse, which is a very different treatment and deals with high altitude burst EMP as well as that from air and surface bursts, click here). Notice that this April 1962 section is the first mention of EMP in the Effects of Nuclear Weapons (it is not mentioned in the 1950 or 1957 editions), and that at the time of publication (April 1962) the main EMP mechanism had not even been discovered (it was discovered after analysis of the results of the Starfish Prime nuclear test on 9 July 1962):

'The Electromagnetic Pulse

'Origin of the Electromagnetic Pulse

'The electromagnetic pulse or "radioflash" which is produced at the time of a nuclear detonation is of considerable interest. It is fairly well known that even small detonations of ordinary chemical explosives can produce electromagnetic signals, so it is not surprising that substantial pulses of this type accompany nuclear explosions.

'There appears to be at least two different mechanisms whereby an electromagnetic pulse may be produced by a nuclear explosion. The first is associated with the creation by radiations from the burst of some kind of asymmetry in the electric charge distribution in the region surrounding the detonation [the 'electric dipole' Nevada test EMP mechanism]; the second is the result of the rapid expansion of the essentially perfectly-conducting plasma of weapon residues in the earth's magnetic field [the magneto-hydrodynamic late-time EMP mechanism discovered after the 1958 Teak test]. The first mechanism, often called the 'Compton-electron model' for reasons which will be seen below, is believed to be the principal means for generation of electromagnetic pulses by detonations on or slightly avove the earth's surface and by those near the 'top' of the sensible atmosphere [this belief was totally wrong, since it completely ignored the magnetic-dipole mechanism, ie, the deflection of Compton electrons by the magnetic field which was discovered after the 9 July 1962 Starfish Prime test]. The other, called the 'field displacement' model [now called magneto-hydrodynamic EMP, or MHD-EMP], might be responsible for electromagnetic signals from underground bursts where the expansion is restrained in a more or less spherically symmetrical manner by the surrounding material, or from those at such great altitudes that the only interaction of the explosion is with the geometric field [wrong, because downward travelling gamma rays will still cause the Compton effect and the mechanism for EMP then will be by the deflection of of the Compton electrons by the earth's magnetic field].

'In the Compton-electron model the photons of the initial gamma radiation leave the exploding weapon with high energies, very soon collide with electrons in the atoms and molecules of the surrounding air, and transfer to them most of their energy. These Compton electrons move rapidly away, on the average, from the center of the burst. Provided some kind of asymmetry exists, this motion is apparently one of the main sources of the electromagnetic pulse. If the explosion were perfectly symmetrical, in a uniform atmosphere, the effects would be equal in all directions; the opposite components would then compensate each other exactly and there would be no electromagnetic signal. However, there are invariably a number of unrelated factors associated with a nuclear explosion which insures the presence of an asymmetry and, hence, of an electromagnetic pulse.

'The most obvious asymmetric situation is that arising from a surface or near-surface (within 350 feet or so) burst, where the presence of the earth itself confines expansion of the weapon residues and radiation emission to the upward hemisphere. At the other extreme, where the explosion takes place high in the atmosphere, there will be very little interaction by upward-moving gamma rays because of the low air density, whereas those going downward will produce Compton electrons within a moderate distance. In both these cases, though their detailed behavior is probably different and their directions are opposite, the effective Compton-electron pulse is essentially vertical. Moreover, no matter where the burst occurs, there is inevitably some asymmetry in the emission and interaction of the photons. For example, the gamma-ray flux from an exploding weapon is itself never fully symmetric because of the presence of auxiliary apparatus, external structure, or the carrying vehicle. It should be noted that, while the 'natural' asymmetries tend to be vertical, the other type may be oriented in any direction.

'The Compton electrons created by the initial gamma radiation thus move away symmetrically, at high velocity, from the exploding weapon. Since the remaining symmetrical components still compensate each other's effects, this motion appears from a distance to be a practically instantaneously accelerated pulse of current in one direction; it is, in other words, something like an 'electric dipole' radiator of classical electrodynamics. The current pulse in the air radiates electromagnetic energy just as it would if it were flowing in a wire transmitting antenna, and this radiation constitutes the first part of the characteristic signal of the explosion.

'When the Compton electrons move away from the explosion they leave behind much slower moving positive ions, which are the other component of the ion pairs. This relative displacement of positive and negative charges produces a radial electric field. In addition, in its passage through the air each Compton electron itself produces a large number of [secondary] electron-ion pairs, perhaps 30,000, mostly toward the end of its path of 10 to 15 feet [in sea level density air]. Under the influence of the radial electric field, the large number of electrons now present will be driven back toward the burst point. This initiates a second pulse of current, but it is rapidly terminated by recombination of electrons with ions and by attachment of the electrons to neutral atoms and molecules in the air, even before the electric field is neutralized. The negative ions produced in the attachment process, and a corresponding number ofpositive ions, remain free a while longer because the ions, being heavier and less mobile than electrons, collide less frequently. This large volume of ionized gas (or 'plasma') undergoes oscillations at characteristic frequencies similar to those observed in experimental plasmas in the laboratory. The oscillations damp out in a short time, as the negative particles (ions and electrons) combine with positive ions, but while they last they produce electromagnetic waves in the radiofrequency range.

'Characteristics of the Compton-Electron Signal

'The effective rise-time of the main part of the initial signal pulse (produced by the Compton electrons) from surface or near-surface bursts is of the order of 10^{-8} second, so that oscillation frequencies as high as 100 megacycles (10^8 cycles) per second [100 MHz] may be expected. However, only a very small part of the total electromagnetic energy radiated is carried at such high frequencies. In addition, the higher frequencies are attenuated much more rapidly than the lower ones in normal propagation through the atmosphere. The frequencies of the plasma oscillations, which continue for several milliseconds and radiate considerably more energy, are much longer. These frequencies are attenuated hardly at all in normal propagation. At the lower end of the spectrum are the extremely low frequencies (in the very low kilocycle region) which might be detected very close to any such excited radiating dipole; they would exist principally in the 'induction' and 'quasi-static' fields and not be radiation at all.

'The slectromagnetic signal, as detected at a range of a hundred miles or so, thus consists of a continuous specutrum with most of its energy distributed about a median frequency (10 to 15 kilocycles per second) which is related inversely to the yield. At much longer distances, of many hundreds or thousands of miles, the form and spectrum of the pulse are determined largely by the characteristics of the medium of propagation, i.e., the 'duct' between the surface of the earth and the D- or E-region of the ionosphere.

'A somewhat similar explosively-excited vertical dipole radiator which is frequently encountered in nature is lightning, and the electromagnetic signal (or static) associated with lightning also has a peak in the region of 10 kilocycles. This must not be taken, however, to mean that there is a detailed similarity in the modes of generation of the electromagnetic signals from lightning discharges and from nuclear explosions. The transmission path largely obliterates the characteristics of the original signal in both cases.

'The Field-Displacement Mechanism [Magneto-Hydrodynamic EMP, or MHD-EMP]

'The second possibility which has been mentioned for the generation of radiofrequency signals by a nuclear explosion is considered to be of particular significance for extremely-high-altitude bursts. Immediately after the detonation has occurred, the hot weapon debris is essentially a highly ionized vapor (or plasma) which is expanding rapidly. A property possessed by all plasmas is a tendency to exclude a magnetic field, such as that of the earth, from its interior. The expanding plasma of weapon residues thus causes a violent distortion of the earth's magnetic field. As a result of the interaction between the geomagnetic field and the charged particles in the expanding plasma and in the very tenuous, largely ionized, surrounding gases, this disturbance propagates away from the burst region as a 'hydromagnetic wave'.

'The hydrodynamic wave retains its identity and characteristics in propagating over very long distances at high altitudes, but at lower levels, where it interacts with the denser atmosphere, it is detected as an ordinary electromagnetic wave or magnetic disturbance. The field-displacement mechanism is believed to be especially important at very high altitudes where the air density is low and the expansion of the debris is not impeded by the atmospheric pressure. It is probable that the same mechanism may operate to produce an electromagnetic signal from an underground burst. The expansion of the debris is here limited to a few yards and the signal is therefore small, but it may be detectable at short ranges.'

More information on EMP at nuclear tests: http://glasstone.blogspot.com/2006/03/emp-radiation-from-nuclear-space.html

Review of Dr Austin M. Brues and Dr Arthur C. Upton (Chairmen), Proceedings of the Second Interdisciplinary Conference on Selected Effects of a General War, DASIAC Special Report 95, July 1969, vol. 2, DASA-2019-2, AD0696959. The abstract of this 423 pages long report states:

'This report is a result of a second conference on the selected effects of a general war held at Princeton, New Jersey, 4-7 October 1967. Specific topics included in this particular conference were the effects of the 1954 hydrogen bomb tests in the Pacific Ocean which resulted in the fallout contamination of Marshall Island natives and of the Japanese fishermen on the Fukuryu Maru (Lucky Dragon); the ecological effects of nuclear tests in the Pacific regions; and the effects of the aircraft accident in Spain, in which nuclear weapons broke up, but did not explode ('Spanish Incident'). The conference was sponsored by the Defense Atomic Support Agency under the auspices of the New York Academy of Sciences. This volume is the second of a 3-volume series on this subject. The other two volumes have similar titles and are numbered DASA 2019-1 and DASA 2019-3.'

Page 38 notes that the 'fallout' controversy stemmed to the 1 March 1954 Castle-Bravo nuclear weapons test when 23 Japanese crew on the No. 5 Fukuryu Maru ('Lucky Dragon No. 5') tuna trawler before dawn when they observed what appeared to be a 'sunrise' in the west-southwest followed by an explosion sound 7-8 minutes later. Because of the known speed of sound, these facts prove that the Japanese tuna trawler was 150 km east-northeast of ground zero, directly on the fallout hotline and near the north of Rongelap Atoll which received the most fallout. Fallout began to arrive on the ship 3-4 hours later when they were drawing in the nets. As a result, the fish and the crew were contaminated on exposed skin.

Beta burns take two weeks to appear and at that time, on 14 March, the trawler returned to Yaizu Harbour, Japan, and the fallout effects from the test began to appear.

On page 40, Dr Merril Eisenbud states that there was no evidence of a wind change that contaminated Rongelap:

'I was then Director of the [U.S. Atomic Energy Commission] Health and Safety Laboratory and was in direct communication with one of our teams stationed in the Marshall Islands. The only wind information I have ever seen came in an official dispatch, at H - 6 hours, which arrived in New York just a few hours before shot time. From my recollection I would say that it would not have required a wind shift to dump the fallout on Rongelap. Unfortunately, the situation has never been documented in a manner that would make it available to many of us who were interested in the exact meteorlogical circumstances.'

On pages 43-44, Dr Eisenbud adds: 'the instrument on Rongerik, which was an automatic instrument [a graph recorder radiation meter] went off scale [500 mR/hour] at H + 7 hours. This was an instrument that was not part of the Task Force. It was being operated by what was basically a CINCPAC-supported civilian organization based with the Task Force but not operating as part of it. When the instrument went off scale, the operating procedure called for the aerial confirmation of this and there was not enough interest in the Task Force to authorize sending a plane over the island to see if, in fact, the instrument was working properly. As I recall it, this was delayed about 36 hours. No information beyond the initial dispatches came into the States for about two days. In other words, there was just a complete breakdown as far as information was concerned, in taking the steps that were necessary in order to evaluate the situation, and to take the necessary palliative measures. ... right up to the last minute, with the fallout lying on the ground, the people just didn't go up to investigate.'

A sea plane was finally sent to check Rongelap two days after the detonation, the delay being due to the problems of fallout at Bikini Atoll (the firing party was trapped by the fallout and had to be evacuated) which took priority over the other inhabitants of the Marshall Islands including thr U.S. weather personnel who were contaminated on Rongerik Atoll to the east of Rongelap Atoll (they avoided beta radiation burns by washing fallout off skin and changing into long sleeved shirts). It was 50 hours after burst when 16 older people were evacuated by sea plane from Rongelap when it was surveyed, and 51 hours when a ship took the remaining 48 people to safety.

Regarding the contamination to the Japanses tuna trawler, Dr Eisenbud personally had to fly to Japan and measure it, as he states on page 48:

'I saw that ship March 22, 22 days later, and by that time it was still reading generally about 110 mR per hour ... we knew what the decay-characteristics were, and if we extrapolated from that ... to H plus four hours, the integrated dose was something better than 100 R. ... By this time the ship had been hosed ...'

Dr Theodore B. Taylor commented on page 51 about the detection of fallout from the 81 kt Dog shot on a 300 ft tall tower during Operation Greenhouse at Eniwetok Atoll in 1951 (a windshift unexpectedly blew fallout over the island where the personnel were living during the tests):

'Apropos of the Dog shot, fallout was clearly audible. There were little beads of steel from the tower that condensed, and one heard this constant tinkle, tinkle of steel from the tower hitting the aluminum roofs and then rolling down the gutters and piling up in little piles on the ground.'

On page 69, the issue is addressed as to why the irradiated Japanese fishermen did not at any time radio back to their home harbour a report of seeing the explosion and being contaminated by fallout (they were in contact with their home port twice daily by radio). The reason is given, with reference to Dr Ralph E. Lapp's book about the affair, The Voyage of the Lucky Dragon, that the crew had spent two months in jail in Indonesia for poaching and feared that the Americans would arrest them for allegedly spying on the nuclear test. In passing, it is interesting that Dr Ralph E. Lapp (who died in 2004) in 1954 first alerted the world's media to the radiation dangers in his many articles for the Bulletin of Atomic Scientists about the seriousness of the Bravo test fallout and its implications for civil defence, but in 2002 he wrote a letter to the Washington Post (Thursday, November 21, 2002; Page A40), Radiation Risk Realities, in which he complained about too much fear of radiation:

Washington Post, Thursday, November 21, 2002, page A40:

'Radiation Risk Realities

'The Nov. 11 front-page story on "dirty bomb" risks, "Hunting a Deadly Soviet Legacy," needed to put the threat in perspective. The release of radioactive cesium into the atmosphere from the Chernobyl plant in 1986 was 1,000 times as great as the release in the "dirty bomb" scenario.

'In assessing radiation risk, it is essential to understand the basic facts about data accumulated during half a century of medical studies. Among a half-million Hiroshima survivors, for example, fewer than 1 percent of the observed cancer deaths were the result of the A-bomb radiation.

'How many Americans know that?

'RALPH E. LAPP
'Alexandria'

Dr Lapp's issue with the media was that it ignored the facts and exaggerated or suppressed evidence simply to stay on the fashionable bandwaggon of popular hysteria which could motivate people to buy the paper or watch the media TV reports. Facts without hysteria don't sell the media efficiently to the ignorant, unlike exaggeration. The media feels no duty to inform people of facts, just to kill off competitors by exaggerating facts to create a fictional news 'story' which scares people and creates prejudices, bias and bigotry.

Continuing the review of Proceedings of the Second Interdisciplinary Conference on Selected Effects of a General War, DASIAC Special Report 95, July 1969, vol. 2, DASA-2019-2, AD0696959, page 76 has a discussion by Dr Lin Root of the death of the radio operator, Kuboyama Aikichi, of the Japanese fishing trawler about six months after exposure (he died on 23 September 1954).

He died of a hepatitis, a liver infection, due to an infected blood transplant. Dr Root explained: 'Japanese doctors give very small blood transfusions, and Kuboyama needed a great many.'

The blood transfusions killed him with an infection when his white blood cell count has been suppressed, and were of no practical use whatsoever in combating the radiation malaise, for which blood cells created continuously by regenerated bone marrow are required. If anything useful had been done, it would have been to have given a bone marrow transplant, not infected blood.

The bad beta burns to skin on the Japanese fishermen was due to the long exposure as they hauled in nets full of tuna for several hours. The fast decay of the fallout meant that most of the contact beta exposure occurred during that time. With the Marshallese Islanders on Rongelap and Ailinginae Atolls, the beta burns occurred to moist skin with sweat glands including the neck, armpits, elbows, and the top surfaces of feet but not the bottom surfaces of feet (which were in direct contact with the contaminated ground, but were not beta-burned, because of the extra thick deal skin layer on the soles of the feet, which stopped most of the beta particles). In addition, the Marshallese women used coconut oil as a hair dressing, which was sticky and retained fallout, causing exposure of the hair roots and epilation, beginning at about the same time as the beta burns (after a latent period of 14 days or so from time of exposure).

On page 85, Dr Theodore Taylor comments on radiation hysteria by involking the example of President Kennedy's ignorance of natural background radiation and the effects of dosage:

'I think the mystique is right here at home, typified by a comment that President Kennedy made to Jerry Wiesner when they were sitting together in the White House and it was raining out. Kennedy asked Wiesner whether there was fallout in the rain that was falling on the White House lawn, and Wiesner said, "Yes, there still is." This was an intense emotional experience for the President, to see rain with fallout on the outside; nothing connected with anything in any way quantitative at all. As far as he was concerned, that rain that was falling outside was bad.'

(For a very refreshing review of the controversy over safe levels and tolerance doses versus the popular mainstream 'all radiation kills, end of story' hysteria, see Dr Daniel J. Strom's amusing 1996 report The Linear, No-Threshold Dose-Response Model: Both Sides of the Story. Pacific Northwest National Laboratory, Richland, Washington.)

An amusing argument, over whether the unknown risks of natural chemical in cranberries are similar to low level radiation risks, then occurred. Dr Stafford L. Warren responded:

'Not everybody buys cranberries and couldn't care less, but everybody is subjected more or less to the fallout.'

Dr Charles L. Dunham (of the Division of Medical Sciences, National Research Council, U.S. National Academy of Sciences) responded to Dr Warren:

'So is Vitamin A. It's toxic, too.'

On page 138 there is an interesting discussion of why the tree-climbing coconut crabs on Rongelap Atoll concentrated strontium-90, which being similar to calcium was highly diluted by the calcium in the coral (calcium carbonate) soil on Pacific Atolls like Rongelap, Bikini, Eniwetok, Utirik, etc. The tree-climbing coconut crab builds up a concentration of strontium-90 with calcium for forming a new shell when it outgrows the old one. Dr Lauren R. Donaldson explained:

'One distinct difference between the coconut crab and the usual crustacean is that as soon as the crab finishes the molting process and the new shell is formed, the crab eats the old shell and thus these minerals are returned to its body. ... So it preserves the minerals and they go on perpetuating this process year after year. This is a particular situation peculiar to the coconut crab. It's not typical of crustaceans in general.'

Dr Donaldson on page 191 presented a diagram of the distribution of beta and gamma emitting radionuclides at Rongelap Atoll in 1966 (which had been contaminated mainly by Bravo in 1954 and very slightly by Zuni in 1956), showing that Mn-54, Fe-55, Co-57, Co-60, Zn-65, Sr-90, Zr-95, Ru-106, Sb-125, Cs-137, Ce-144, and Eu-155 were present in the soil, Sr-90, Ru-106, Ce-144 and Eu-155 were in the lagoon bottom sediment, land plants took up Mn-54, Zn-65, Sr-90, and Cs-137, land crabs and rats took up Sr-90 and Cs-137, and humans took up Zn-65, Sr-90 and Cs-137. The Zn-65 in humans came from eating fish and birds, while the Sr-90 and Cs-137 came from eating land crabs and vegetation, particularly coconuts.

Lagoon algae contained Co-60, Ru-106, Ce-144, and Eu-155, while plankton contained Mn-54, Co-57, Co-60, Zn-65, Zr-95, Ru-106 and Ce-144. Lagoon fish eating the algae and plankton concentrated Mn-54, Co-60 and Zn-65, while marine invertebrates contained Mn-54, Co-57, Co-60, Zn-65, Sr-90, Ce-144 and Eu-155. Both lagoon fish and invertebrates were food for the birds, which were found to concentrate Mn-54, Co-60 and Zn-65.

Dr Lauren R. Donaldson showed a film, Return to Bikini, about the 1966 annual survey of Bikini radioactivity and environmental effects. The film discussion is on pages 230-1, and is quite remarkable.

Dr Charles L. Dunham: 'Lauren, this isn't the way I heard the story. There was a movie I saw a few years ago that was announced to the public by Ian Fleming with a 4-page spread in the London Times which showed little fish that had become disorientated, losing their way, trying to climb trees, which showed sea turtles who tried to find where to lay their eggs. They laid great quantities of eggs which were sterile and then couldn't find their way back to the sea. It showed piles and piles of tern eggs, which were also sterile, and very few terns. Now, which is the true story, sir?'

Dr Lauren R. Donaldson: 'Dr Dunham, during the period from 1946 to 1964 we were at Bikini and Eniwetok for several months most years. We made a total of 23 separate expeditions. No matter how hard we looked we could not find a mudskipper "trying to climb trees." In fact, there are no records of mudskippers at either atoll nor are there any mangrove swamps, the preferred habitat for mudskippers.'

Dr Charles L. Dunham: 'This was supposed to be an authentic movie of the aftermath of the atomic bomb in Bikini. Maybe you selected different parts of the atoll.'

Dr Lauren R. Donaldson: 'I think one would have to do more than select a different part of the atoll, in this particular case. I think even John Wolfe with his great accomplishments in environmental control couldn't build a mangrove swamp out in Bikini without an outflow of fresh water. This sort of completely falsified popular release is nothing but disgusting.'

Dr Theodore B. Taylor: 'Who made that particular movie, do you remember?'

Dr Charles L. Dunham: 'It was an Italian movie. It had a lot of other stuff in it. There were beautiful pictures, though. I must admit there were beautiful pictures of wildlife. As Lauren says, undoubtedly the ones of these mudskippers, as they call them, were taken in the mangrove swamps somewhere and there were lovely pictures of giant sea turtles laying eggs. Again they're apparently authentic pictures.'

Dr Frank Fremont-Smith: 'Maybe it was the photographer that was disorientated; thought he was in Bikini but wasn't.'

Dr Charles L. Dunham: 'That could be quite possible.'

On page 280, Dr Theodore B. Taylor discusses Dr Stonier's 1963 book, Nuclear Disaster:

'I think it's a fair statement to say that the book is essentially an anti-civil defense book; that the purpose of it is to decrease confidence in civil defense measures. The reason I'm saying that so emphatically is that there was a panel formed by the American Nuclear Society about two years ago to discuss civil defense. Eugene Wigner and I were on the side of civil defense and Stonier and someone in the Harvard Law School, whose name I've forgotten, were opposed to it. We had a very informative and worthwhile debate. He said that what he really has in mind in his writing now is to display the futility of civil defense. I think that's important because I think he would be the first to agree that he feels very strongly about this and gets emotionally involved in illustrating his point, namely, that the disaster, no matter what we do, will be so complete that we should not do anything which will indicate that people could get away with a nuclear war. I think that's his thesis. ... I think his thesis is that if we fail to prevent nuclear war, all is lost.'

On page 298, Dr Theodore B. Taylor stated his own pro-civil defense position:

'I must just say that as far as I'm concerned I have had some doubts about whether we should have had a civil defense program in the past. I have no doubt whatsoever now, for this reason, that I've seen ways in which the deterrent forces can fail to hold things off, so that no matter what our national leaders do, criminal organizations, what have you, groups of people over which we have no control whatsoever, can threaten other groups of people.'

Gamma radiation from surface burst fallout and air burst rainout

Above: rainout radiation dangers from air bursts during rainstorms can in certain cases exceed fallout dangers from surface bursts (for a good summary of rainout predictions models, see: A. Schiff, Problems with Predicting Fallout Radiation Hazard in Tactical Battlefield Situations, Lawrence Livermore National Laboratory report UCRL-51440, ADA385024, 1973). A report by W. K. Crandall, et al., An Investigation of Scavenging of Radioactivity from Nuclear Debris Clouds, Lawrence Livermore Laboratory, UCRL-51328 (1973) found that for 15 miles per hour wind, if the whole of the mushroom cloud from a 1 kt air burst was scavenged by rainfall, the peak gamma dose rate at 10 km downwind would be about 25,000 R/hr and would result in an infinite-time accumulated dose of 25,000 R. They found that at 100 km downwind, the peak dose rate from rainout would be 35-100 R/hr and the infinite dose would be 400-1,200 R. For 10 kt, assuming 100% cloud scavenging by rain, they found that the peak dose rate at 10 km downwind would be 70,000 R/hr with an infinite dose of 50,000 R; for 100 km downwind they found 1,500-2,000 R/hr peak dose rate and 10,000-15,000 R dose. For 100 kt, again assuming that 100% of the cloud is scavenged by rain, they found that the peak dose rate at 10 km downwind is 120,000 R/hr with an infinite dose of 80,000 R; at 100 km the peak dose rate is 4,500-5,500 R/hr and the infinite time dose is 30,000-35,000 R. These are far higher dose rates and doses than can occur from land surface bursts, but they are exaggerations for three reasons: (1) less than 100% of the mushroom cloud will be scavenged by rainout because not every fallout particle will be hit by a raindrop, (2) the mushroom cloud will often (especially for higher yields, such as 10-100 kt) rise above the height of the rainclouds, so that only a fraction of the activity can possibly by hit by rain droplets and carried to the ground unless there is a thunderstorm cloud with rain forming at very high altitudes, and (3) the rainout droplets arriving on the surface with fallout particles embedded in the raindrops will generally carry a large fraction of the fallout activity straight down storm sewers, where the gamma radiation is well shielded underground (by two feet of earth or so) from people on the ground. In other words, there is natural decontamination during rainout, unless the drainage is poor and puddles are left.









For previous posts on the topic of fallout from nuclear tests, see here, here, here, here, here, here, etc. The story is that extensive fallout pattern surveyed from the very first nuclear test, Trinity (a low, contaminating, tower burst on 16 July 1945) was covered up in the 1950 edition of Glasstone's official U.S. Department of Defense Effects of Atomic Weapons, which only gave the upwind fallout pattern (excluding the downwind fallout area!).

That same book did however given extensive fallout pattern data for the underwater test Baker, which was deleted from all future editions (1957, 1962/4, and 1977). I want to examine how land surface burst fallout was dealt with by Glasstone and Dolan's official manual from 1957-77.

Fallout prediction in the U.S. Department of Defense's 'The Effects of Nuclear Weapons'

The September 1950 edition, called The Effects of Atomic Weapons, includes a table of Trinity test dose rates from ground zero (it was 8000 R/hr at 1 hour at ground zero) outwards in the upwind (not downwind) direction, without explaining the larger area contaminated downwind than upwind! (The book does have an appendix about fallout plotting using meteorology, and another appendix about calculating dose rates from deposited activity, but doesn’t predict downwind dose rates.) It also has a sequence of detailed fallout and base surge predictions for underwater bursts, based on Baker test data.

The 1957 and 1962/4 editions each have different detailed prediction methods for 20 kt and 1 Mt surface bursts, based on nuclear test data from the Nevada 1951 surface burst Sugar (1.2 kt), with some clever scaling procedures to interpolate with the Eniwetok 1952 surface burst Mike (10.4 Mt) and the Bikini 1954 surface burst Bravo (14.8 Mt). The 1962/4 editions use additional data from 1954 Castle series and 1956 Redwing series tests, for upwind fallout.

The 1957 Effects of Nuclear Weapons fallout prediction for land surface bursts is based on the then-secret and therefore not cited report by D. C. Borg, et al., Radioactive Fallout Hazards from Surface Bursts of Very High Yield Nuclear Weapons, U.S. Armed Forces Special Weapons Project, report AFSWP-507, Secret – Restricted Data, May 1954, which states on page 8:

"Since the Castle Bravo shot [on a coral reef near Namu Island in Bikini Atoll] may be characterized as a hybrid between a land-surface and a water-surface shot, probably most like the former, its ground zero radiation data [which showed relatively low dose rates compared to the Elugelab Island burst Mike shot data] may not be very representative of a true land-surface detonation. For this reason, the Ivy Mike shot has been used as the primary source of data for scaling of radiation effects in the ground zero [upwind] region.

"The downwind fall-out contours constructed for Castle Bravo were based essentially upon survey data taken on the islands involved in the fall-out region."

The 1957 Effects of Nuclear Weapons data for fallout from 20 kt and 1 Mt surface bursts is therefore based on a scaling-based interpolation between 1.2 kt Sugar data and 10.4 Mt Mike data for upwind fallout, and a scaling-based interpolation between 1.2 kt Sugar and 14.8 Mt Bravo data for downwind fallout (no downwind fallout data for Mike was measured).

The scaling system is to scale the linear dimensions of the fallout pattern by the cube-root of the total weapon energy yield, and to scale the dose rate values of contour by both the cube-root of total yield and directly with the fission fraction of the explosion. This derives from the report: Scaling of Contamination Patterns, Surface and Underground Detonations by C. F. Ksanda et al, USNRDL-TR-1, September 15, 1953 (Secret – Restricted Data). The physical basis of this scaling is that the fallout pattern derives from the mushroom cloud, the radioactive parts of which (smaller than the total visible cloud, which is just water vapour at the edges), scale in vertical and horizontal extent by the cube-root of weapon yield. The fallout pattern is then like a "shadow" of the mushroom cloud, so the linear dimensions scale accordingly as the cube-root of yield, while the dose rates for each contour scale according to the depth of fallout which arrives. This fallout depth of deposit is proportional to the cloud’s vertical extent, scaling as the cube root of yield. The confirmation that this scaling system is rigorous is that the total amount of deposited activity within the fallout pattern should be directly proportional to weapon’s fission yield. The scaling confirms this, since the scaled area activity is the scales area, ie the product of two linear scaling factors or (W^{1/3})^2 = W^{2/3}, multiplied by the dose rate scaling factor, which for constant fission fraction is simply W^{1/3}. This product is [W^{2/3}].[W^{1/3}] = W, proving the physical consistency of this whole scaling system.

The 1962/4 editions of The Effects of Nuclear Weapons are different from the 1957 data as regards the prediction of upwind fallout dose rates, which are based on an internal and therefore non cited report by Frank Cluff of the U.S. Weather Bureau, The Upwind Extent of Fallout from a Large Nuclear Detonation, U.S. Department of Commerce, Weather Bureau, Washington, 20 August 1959. This suggests that the fallout dose rate at any point upwind is like rainfall deposition, being dependent on the duration that the mushroom cloud remains overhead (before it is blown downwind and fallout ends), which is about 30 minutes for Bikini and Einwetok bursts in the megaton range (1-15 Mt).

Page 2 of that report states that the data used is: "estimated from Pacific test data [which] gives the up-wind extent for several dose rate contours as a function of total yield based on a 7 knot [8.06 statute miles/hour or 13.0 km/hour] mean tropospheric wind speed and an 80 percent fission yield."

Clearly, therefore, the test data is for low average wind speeds and high average fission yield.
Figure 9.85 on page 455 of the 1962/4 edition of The Effects of Nuclear Weapons shows Cluff’s illustration of the upwind fallout, but it has corrected from Cluff’s case of 7 knots to zero wind by adding 4.03 statute miles to all of the distances of upwind contamination, and has also had the dose rate contours re-scaled to 50% fission instead of Cluff’s case of 80% fission.

If we examine the actual figures, ignoring terrain roughness (which in the Nevada desert reduced the fallout gamma dose rate at 1 m altitude by about 25%) and self-shielding of the instrument by its own battery bulk and by the person holding it (which reduces the observed reading by about another 25%), taking a 1 Mt fission surface burst in a 15 miles/hour wind with 15 degrees of directional shear with altitude from the surface to the mushroom cloud, the predicted dose rate patterns are in fair agreement for a 1 megaton surface burst.

Glasstone 1957 says (empirically) that the 1 hour reference 3000 R/hr contour extends 22 statute miles downwind; the 1962/4 revision states (empirically) that the distance is 23 statute miles, while the 1977 edition based on Dolan's computer simulations using DELFIC and SEER II fallout models gives a distance of 21 miles (ignoring terrain shielding and instrument response). DELFIC in particular is almost entirely non-empirical; the only 'normalization' or 'manual fiddle' introduced empirically into the code was the proportion of the bomb energy which is needed for cloud rise simulations to be accurate (it was found that setting this proportion at 45% gives the observed cloud rise data). The maximum width of this 1 hour 3000 R/hr contour is stated to be 3.1, 6.0, and 2.9 statute miles in the 1957, 1962/4, and 1977 editions, respectively.

In the case of the 1 hour reference 1,000 R/hr contour, the downwind distances are 40, 42, and 40 statute miles, respectively, using the 1957, 1962/4 and 1977 editions, while the corresponding maximum contour widths are 6.8, 10, and 6.9 statute miles, respectively.

At lower dose rates, disagreements are larger. However, the lethal areas where survival demands evacuation or high protection factors by buildings, are most important. Upwind fallout is a particular problem to predict. Most of the empirical data in DASA-1251 for upwind fallout is wrong because the fallout maps of Eniwetok and Bikini have been mis-scaled, as explained in an earlier post. Correcting this error, and you immediately see that the 1952 Mike surface burst produced way higher upwind dose rates than subsequent tests like 1954 Bravo and 1956 Tewa and Zuni, allowing for yield and fission proportion correction factors.

The Mike test produced more severe upwind fallout because of the way the massive steel bomb container modified the fallout distribution process and resultant particle size distribution. This would not occur with stockpiled warheads.

Glasstone 1957 states that a 1 hour reference dose rate of 10 R/hr would occur 9.35 statute miles upwind from a 1 Mt surface burst under a 15 miles/hour wind. This was reduced to 1.7 statute miles in the 1962/4 edition, but was increased to 5.8 statute miles in the 1977 edition. The upwind radius of the 100 R/hr contour at 1 hour was 3.46, -1.3 (ie, 1.3 miles downwind, not upwind!), and 3.5 statute miles according to the 1957, 1962/4, and 1977 editions. (The reason why the 100 R/hr contour is seen to commence downwind of ground zero in the 1962/4 edition stems from the assumption in that edition, on page 454, that the upwind fallout has an average arrival time of 24 minutes, so that under a 15 miles/hour wind it travels 15 * 24/60 = 6 miles downwind while falling. This model is likely wrong because it employs data from the inaccurately scaled fallout patterns in reports like WT-915, and also because it ignores the toroidal downdraft from the periphery of the rising, expanding, vortex fireball, which is entirely different in dynamics from a steady deposit of fallout from the full cloud altitude under ambient winds.)

It must be emphasised that except at ground zero and upwind, fallout dose rates never reach the the 1 hour reference level, which is an extrapolation dose rate based on the total deposit.

For example, the 1000 R/hr of gamma radiation in air 1 hour exposure rate quoted for 40 miles downwind (from 1 megaton fission surface burst on an ideal surface, neglecting instrument response, under a 15 miles per hour wind with 15 degrees shear), has an average arrival time of about (40 miles)/(15 miles/hour) = 2.67 hours, by which time the dose rate will have decayed by a factor of 2.67^1.2 = 3.24.

Hence, the maximum dose rate actually encountered at 40 miles downwind will be no more than 1000/3.24 = 308 R/hr. Terrain roughness on average ground (away from buildings which provide greater shielding) will reduce this to 70% of the ideal smooth surface value, while the shielding by the body of the person taking the measurement would each reduce this again to 75% or so, thus a person holding a radiac (radioactivity detection, identification and computation) survey meter will measure about 308 * 0.7 * 0.75 = 162 R/hr peak exposure rate. Because the instrument response is roughly the same as the bone marrow dose (the bone marrow being shielded by surrounding tissues which absorb around 30% of the gamma exposure outside the body), the instrument reading in a fallout field can be used to assess bone marrow exposure. The infinite time bone marrow dose there, allowing for a 20% reduction in dose due to the much quicker decay rate after 200 days than the t^-1.2 'standard decay law', will be 4 * 2.67 * 162 = 1727 R. This is 11.5 times the 150 R limit you would want to avoid a severe risk of acute sickness leading to mortality. Hence, you would need a protection factor of 11.5 if you were 40 miles downwind from a 1 megaton fission surface burst. Most British brick houses have that protection or higher, as long as you keep away from the outer walls/windows and away from the roof.

A lot of claims were made during the 1980s that fallout protection factors in Britain were exaggerated by the U.K. Home Office method of assessment, which uses an averaged wall thickness to include window effects, instead of treating the windows separately to the brickwork. However, this ignores firstly that the calculations assumed 1 MeV gamma ray energy, when it is known that thermonuclear weapons emit very reduced mean energy gamma rays due to fractionation and the massive contributions from very soft gamma ray emitters like neptunium-239, uranium-240, uranium-237, etc. (due to non-fission neutron captures in the thick uranium-238 'pusher' around the lithium deuteride secondary in the weapons), and secondly, the official civil defence advice was that after the bomb (but before fallout arrives) people should block up window spaces with whatever they can, such as furniture like bookcases.

If people could not do that, the fallout or rainout is visible wherever it is life-threatening, and people have time to evacuate, in the cross-wind direction. The highest dose rate fallout follows low altitude winds that you can get an idea of at ground level from the motion of low level clouds (or the stem of the mushroom, if it is not obscured by local cloud cover). The fallout from the upper parts of the mushroom cannot be assessed so easily unless the mushroom cloud can be seen from the ground, or satellite derived upper wind data is available. However, that fallout generally produces less intense dose rates, albeit over a wider area.

RAINOUT

Rainout is one topic that was substantially expanded in the 1978 and 1981 page changes to the 1972 originally 'secret - restricted data' manual Capabilities of Nuclear Weapons, edited by Philip J. Dolan, Defense Nuclear Agency, DNA-EM-1. Dolan actually did some of the research on rainout himself while at Stanford Research Institute in the 1970s. (He had much earlier done extensive research on the gamma ray spectra of different fission type fallout mixtures, including the effects of fractionation, the decay rate of the fractionated fallout, neutron induced radioactivity due to non-fission capture of neutrons, etc.)

The action of rainout is to speed up the fallout rate. Either small fallout particles get swept out of the air by much larger raindrops (this process is strictly termed 'washout'), or the raincloud and fallout cloud combine, allowing small fallout particles to be captured - as a result of their diffusion in all directions as a result of Brownian motion - by small water droplets, which naturally grow (by collisions, condensation, and turbulent attachment) into large raindrops that can then fall out of the raincloud very rapidly (this two-stage mechanism is correctly called 'rainout').

Most rainfall occurs from rain clouds at altitudes of 2.5-5 km, i.e., from the altitude range which corresponds to the mushroom top altitude for a 2 kt nuclear explosion at low altitude or surface level. Some rain comes from much higher altitudes due to thunderstorms, but this quite rare.

Rainout poses a fallout danger from low yield air bursts which would not be present in dry weather. This can affect troops and cause collateral damage by exposing civilians to radiation.

Where good drainage exists in well designed cities, the rainout danger is less severe than fallout because most of the radioactivity goes straight down the drains, where the radiation it emits is well shielded from people. Eventually most of the radioactivity ends up in sediments, and a small fraction ends is dissolved and enters rivers and the ocean, where it is diluted to insignificance compared to the natural background radiation from potassium-40.

Quantitative calculations of the rainout doses in have been published in two Lawrence Livermore Laboratory reports:

(1) J. B. Knox, T. V. Crawford and W. K. Crandall, Potential Exposures from Low-Yield Free-Air Bursts, Lawrence Livermore Laboratory, report UCRL-51164 (1971). This report calculates that if the raincloud top is at an altitude of 7.8 km, then the percentage of the nuclear mushroom cloud that is subject to rainout is 100% for 1 kt, 80% for 10 kt, and 0% for 100 kt.

It also estimates the infinite time whole body gamma dose in rads due to rainout (assuming no sheltering, and no run off into drains) at various distances downwind. Assuming that 100% of the mushroom cloud is subject to rainout from 1 kt, 10 kt, and 100 kt free air bursts (which would require thunderstorms for the 10 kt and 100 kt examples, because normal rainclouds are not high enough to completely envelop the mushroom cloud for 10 kt or more), a 40 km/hour wind and 15 degrees diffusive shear (as for Dolan, 1972 and Glasstone and Dolan, 1977), the dose at 10 km, 100 km, and 2,000 km downwind from a 1 kt burst would be 25,000 rads (rainout mean deposition time of 15 minutes after burst), 900 rads (rainout mean deposition arrival time of 3 hours), and 0.05 rad.

For 10 kt, the corresponding figures for the same distances are 50,000 rads (rainout mean deposition time of 11 minutes after burst), 11,000 rads (rainout mean deposition time of 1.9 hours), and 2 rads. For 100 kt, the figures are 80,000 rads (rainout mean deposition time of 10 minutes after burst), 30,000 rads (rainout mean deposition time of 1.6 hours), and 30 rads.

These 100% rainout figures only apply to the rainout occurring locally, within a region with the radius of the mushroom cloud. Rainout obviously can, at most, deposit 100% of the airborne radioactivity. If this happens around 10 km downwind, you get a dose there of 30,000 rads, but you then get a dose of 1,000 rads at 100 km, because the atmosphere will then be clear due to the rainout around 10 km downwind. So heavy (100%) rainout ends the fallout pattern within one mushroom cloud radius. Complete rainout always results, therefore, in relatively small, localized 'hotspots' and not a general increase in doses over massive areas within the sort of fallout pattern you get from a surface burst. Although a warm front can result in gentle rain over a wide area, once all radioactivity is deposited the air is cleared, so no further rainout of that radioactivity can occur.

(2) W. K. Crandall, et al., An Investigation of Scavenging of Radioactivity from Nuclear Debris Clouds: Research in Progress, Lawrence Livermore Laboratory, report UCRL-51328 (1973). This shows that the accuracy of the predictions above is greatest near the explosion where the doses are most serious. Accuracy degenerates at large distances where doses are small and effects are less serious. As the yield increases in reality, the percentage of the cloud which is intercepted by rain falls, so above 100 kt there is no serious rainout threat unless a thunderstorm exists at the time of detonation, downwind from the burst.

However, very close to the burst, rainout radioactivity doses from 1-10 kt air bursts can be nearly six times greater - over a small area equal to the mushroom cloud size - than dry fallout at similar distances from a land surface burst. For example, at 10 km downwind from 1 kt air burst complete rainout can produce an infinite time outdoor dose (ignoring drainage and shielding, as in the above estimates) of 25,000 rads, as already discussed. This is nearly six times higher than the equivalent dry fallout dose of 4,400 rads for a 1 kt land surface burst at the same 10 km distance etc. (using Glasstone and Dolan, 1977).

In the vast majority of cases, however, air bursts would be likely to occur in dry weather and deposit no local fallout, causing no imminent threat to life outside of the area of initial effects. The fallout would take many weeks, months and even years to gradually be deposited at very low dose rates (due to decay and dilution due to mixing and diffusion while being blown right around the world every month or so).

Even if there was heavy rainout, most of the rainwater would carry the tiny radioactive particles (mixed with the rain) straight down drains, and 1 metre of earth shields reduces the intensity of 1 MeV gamma rays by a factor of about 300. In fact, the average energy (and thus penetrating power) of fallout is considerably less than 1 MeV, so protection is quite high when fallout does down the drain.

There is an interesting bibliography of fallout computation by Richard Rowland of Kaman Sciences Corporation, California, Fallout Computer Codes: A Bibliographyc Perspective, U.S. Defense Nuclear Agency report DASIAC-SR-93-022, July 1994, database reference AD-A281 905. Six major fallout computer codes, including DELFIC (discussed in a previous post on this blog) are discussed briefly.

Rowland points out that DELFIC uses a lognormal fallout particle size distribution based on ground collected deposits of fallout from nuclear tests Teapot-Ess (1.2 kt, Nevada, 1955; shallow underground test, 20 metres depth of burst) and Small Boy (1.65 kt, Nevada, 1962; near surface burst at 3 metres above the ground). The DELFIC median particle radius is 123 microns (0.123 mm).

Other fallout prediction models use grossly different values, and so make differing predictions of the fallout distribution: WSEG-10 uses a median size of 60 microns, while DNAF-1 uses a median diameter of 229 microns. All the models use a fallout particle density of 2.6 grams per cubic centimetre. Another source of disagreement is the fireball thermal fractionation effect on the early fallout dose rate (the relative condensation rates of different fission product decay chains at different times after burst while fallout is forming). DELFIC rigorously analyses this by calculating contributions from each decay chain and summing them to achieve total dose rate.

As a result, DELFIC results in 67% of the activity being refractory (i.e., condensing before the silicate carrier material which forms the majority of the fallout mass) and this portion of activity is diffused throughout the still molten carrier material of the fallout particles in direct proportion to particle volume. It is soluble and has constant specific activity (activity per unit mass) regardless of particle size. The other 33% is from volatile fission product precursors (with low boiling points so they condense only after the silicate fallout carrier has already solidified), and this can only condense on the outside surface of fallout particles. As a result, the latter is more soluble and scales as the surface area of the particle not as the volume (i.e., specific activity, activity/mass, for volatile nuclides scales inversely with fallout particle radius).

Other fallout models deal badly with this because they ignore the fractionation with particle size and with distance from the detonation. WSEG-10 assumes K = 5180 (R/hr)/(fission kt per square kilometre) at 1 hour after burst, compared to a figure of K = 7770 for SEER3 and 6973 for DNAF-1. All these constant figures are misleading, because fractionation changes this ratio (the composition of the debris depends on particle size) with particle size (and thus with distance) from ground zero. This argument started in the late 1950s. Glasstone's 1957 Effects of Nuclear Weapons stated that the ratio is K = 3,200, the 1962 edition gave a ratio of K = 10,000 and noted that terrain irregularities reduced this by 30%, radiac meter shielding by the operator's body reduced it by another 25%, and only 60% of the fallout occurs locally anyway. Glasstone and Dolan 1977 gave K = 7,500 with the same provisos as the 1962 edition. R. R. Rapp of the RAND Corporation challenged this in his report An Error in the Prediction of Fallout Radiation (RAND report RM-5164-PR, December 1966) when he analysed the data from the 1951 Nevada 1.2 kt surface burst Sugar, which is quite smaller than the fallout prediction for a 1 kt burst in Glasstone and Dolan 1977. For Sugar, K(observed) = 2,330 (R/hr)/(fission kt per square kilometre) at 1 hour after burst, based on the integration of the measured local fallout pattern dose rates over area. Allowing for terrain shielding of deposited radiation by the surface irregularities in a desert (a transmission factor of 0.7) and shielding of the instrument by the person holding it (a transmission factor of 0.75), the for Sugar K(theoretical) = 4,440.

In the 1 kt surface burst Sugar fallout pattern, the 300 R/hr handheld-instrument-measured contour at 1 hour extends only about 1.5 km downwind, and the 100 R/hr contour extends 3.8 km downwind at 1 hour, for a mean vector wind velocity of 46 km/hour. By contrast, Glasstone and Dolan 1977, page 430, predicts a 1 hour dose rate of 3,000 R/hr at 1.5 km downwind for a 24 km/hr wind and 1.8 km downwind for a 46 km/hr wind as in the case of Sugar. Since the Glasstone and Dolan fallout predictions are for a theoretically smooth surface and responsive instrument, their dose rates need to be multiplied by 0.7 for desert terrain shielding and by 0.75 for instrument response when being carried by a person, giving 1,575 R/hr at 1.8 km downwind.

This is still over five times the measured dose rate in the nuclear test. The error is fairly similar for other dose rates. This raises serious concerns that civil defence plans against nuclear terrorism are being based on grossly exaggerated threats, and therefore will probably not be optimal in the event of a disaster.

Interestingly, Rowland also contrasts different assumptions about the decay rate of fractionated fallout, remarking that all the models apart from DELFIC and the EM-1 handbook (1992 edition) use t^{-1.2} for model the dose rate fall with time. DELFIC and EM-1, Rowland states, use a rate equivalent to about t^{-1.26}, this approximation being accurate to within plus or minus 10 percent for times after burst of 15 minutes to 1,000 hours. Obviously this is ignoring the large variations caused by neutron capture in U-238 such as Np-239 and U-237 (which results from the reaction whereby a neutron capture in U-238 is followed by two neutrons being emitted).

There is a complete experimental analysis of the effects of fractionation and neutron induced activity on fallout decay curves in Triffet and LaRivere's weapon test report WT-1317 (1961): highly fractionated (depleted) fallout near ground zero of a land surface burst gamma dose rate decays approximately as t^{-1.2}, but at a few hundred kilometres downwind where there is no significant depletion (and for unfractionated fallout from water surface bursts) the decay rate is approximately t^{-1.26}.

For more information on the dose rate to bomb fraction ratio controversy see the useful report Response to DCPA Questions on Fallout, DCPA Research Report No. 20, 1973 here. For more on the terrain shielding factor in deserts see J. M. Ferguson, Ground Roughness Effects for Fallout-Contaminated Terrain: Comparison of Measurements and Calculations, U.S. Naval Radiological Defense Laboratory, report NRDL-TR-645 (May 1963). Ferguson finds that the terrain roughness in a desert reduces the gamma dose rate by a factor of 0.6-0.7.

To see how such relatively large corrections occur, remember that on a smooth flat infinite surface half the dose rate at 1 m height comes from fallout more than 15 metres away. Then work out the angle that direct radiation from fallout at the mean distance of 15 metres is travelling to you: it is coming to you almost horizontally (only about 12% of the dose rate is from air scattered gamma rays, 88% is direct). Because it is coming almost horizontally from such a large distance, any small irregularities in the ground surface will shield a significant fraction of the radiation. Half the radiation comes from the area of a circle of 15 metres around you, an area of 700 square metres. The bulk of the radiation is not coming vertically upwards from nearby fallout such as that under your shoes, but from a very wide areas horizontally around you.

Another aspect is that if you get in an open trench below the ground surface, you will have a protection factor of about 10 from fallout radiation (i.e., you will be receiving just part of the air scattered exposure coming vertically downwards). The activity in the bottom of the trench is insignificant, just as the activity under your shoes is insignificant as a source of radiation in a fallout field. The significant source is the large surrounding area. Even in a house with no roof or windows and completely contaminated floor area, there will be radiation shielding of the large area contribution surrounding the house. Unlike gas protection, fallout shielding does not require you to keep individual fallout particles outside a house. It is not the ingress of fallout contamination which is the significant issue, but the gamma rays travelling at light speed from a very large area of surrounding ground. Anti-civil defence propaganda claims that houses with broken windows will have no fallout shielding because some fallout particles which are moving horizontally may enter the building, contaminating it. This completely misses the point about radiation shielding.

About 90% of the dose is from radiation coming horizontally from fallout deposited over many hundreds to thousands of square metres surround you. The contribution from fallout deposited under your shoes, or even over the floor of a house, is insignificant in comparison. Misunderstanding this vital point will reduce the efficiency of fallout protection.

A good analysis of this using real fallout experimentally is by C. M. Huddleston, et al., Ground Roughness Effects on the Energy and Angular Distribution of Gamma Radiation from Fallout, U.S. Atomic Energy Commission report CEX-62.81, December 1963. They found that over a cracked but fairly flat dry lake bed the fallout dose rate and angular distribution (using a collimated detector) at 3 ft (0.915 m) height is identical to that at 20 ft height over an ideal smooth surface. Hence the dose rate is 0.67 of that over a smooth surface. For 3 ft above a more rough desert terrain or a plowed field, they found that the dose rate and angular distribution was similar to the ideal theory for 40 ft height, so that the reduction factor was 0.54. Another study using Cs-137 sources for a plowed field gave a reduction factor of 0.45 (C. E. Clifford, Effects of Ground Roughness on the Gamma Dose from Cs-137 Contamination, Defense Research Chemical Laboratories, Report No. 401, Ottawa, March 1963).

***

Review of Jack C. Greene & Daniel J. Strom (Editors, Health Physics Society), Would the Insects Inherit the Earth and Other Subjects of Concern to Those Who Worry About Nuclear War, Pergamon Press, London, 1988, 78pp

This book by Greene and Strom is based on material developed for the Health Physics Society Summer School, 'Health Physics Aspects of Nuclear Attack' held at Southeastern Louisiana University, Hammond, LA, from May 28-June 1, 1984. It is a useful supplement to The Effects of Nuclear Weapons and contains five pages by Philip J. Dolan scientifically discussing the risks of nuclear terrorism (pages 17-21, discussed below).

Jack C. Greene, M.Eng., worked on the Manhattan District at Oak Ridge in World War II, became a member of the U.S. Atomic Energy Commission's Radiation Instrument Branch from 1947-51, and then joined civil defence. From 1962-73 he headed the Post-Attack Research Division of the U.S. Defense Civil Preparedness Agency (DCPA), and in in 1973 he became DCPA's Deputy Assistant Director for Research (see examples of his reports on Fallout Contamination of Food and Water, NATO Civil Defence meeting May 1966 here and Response to DCPA Questions on Fallout, DCPA Research Report No. 20, 1973 here). Dr Daniel J. Strom in 1988 was Assistant Professor of Health Physics, Department of Radiation Health, at the University of Pittsburgh. He received his Ph.D. for dose assessment research in 1984.

Wikipedia comments:

'The strange title of this second book refers to the discovery that cockroaches will withstand 67,500 rem (American variety) or 90,000-105,000 rem (German variety), compared to a human lethal exposure of only about 800 rem [4]. One theory which resulted from these observations on insects was that cockroaches, along with some simple plants and bacteria, would be likely to be the only lifeforms to survive a severe nuclear war. This theory was refuted by experience of the very rapid recovery on isolated islands exposed to close-in heavy fallout and other effects from massive hydrogen bombs at the Bikini Atoll and Eniwetok Atoll, as well as from smaller nuclear weapons in the Nevada Test Site and Australia (Montebello Islands, Maralinga and Emu Field). Full ecological recovery surveys were documented before and after each test series. (For a brief online introduction into some these studies - with specific reference to the ecological effects of the 1.69 megatons Operation Castle Nectar shot, detonated in 1954 on a barge above the crater of the 10.4 megatons Ivy Mike thermonuclear test in Eniwetok Atoll - see [5] and [6].)'

The book has a Foreword in favour of civil defence by Dr Lauriston S. Taylor, Sc.D., of the U.S. National Council on Radiation Protection and Measurements (NCRP). The Biographical Sketches at the end of the book state: 'Dr Taylor has been a leader in the field of radiation protection for more than one-half century, beginning with his appointment as Chairman of the National Committee on Radiation Protection in 1929, and continuing with his position as President of the (renamed) National Council on Radiation and Protection Measurements. He retired in 1977. He is currently [1988] Honorary President of the NCRP. ...'

Dr Taylor's Foreword states: 'I am becoming increasingly convinced that the basic problem centers about the overall ignorance of the general public coupled with an overwhelming deluge of mis-information about some of the simplest facts ... disseminated through our collective media - newspapers, radio, TV, and household magazines.

'Unfortunately, most of the public's information comes through those media which are necessarily highly competitive businesses and must make a profit to survive. By their own admissions, unexciting news is no news, and the often dull and technical discussions associated with Civil Defense make unattractive copy, so they tend to develop and emphasize, often out of context, any unusual, exciting or dramatic tidbits that come to hand ... it is truly amazing how many quickly mount the ladder into full-fledged 'facts'. Even when the right questions are asked, wrong answers are the ones likely to be given, whether through avarice or malice or ignorance. It takes real effort to dig out the real facts.'

On page 40, Dr Howard Maccabee, Ph.D., M.D., responds to the most widespread prejudices:

'I am a radiation oncologist, which means my specialty is cancer. Over the last few years, I have treated over eight hundred cancer patients. In a person's life, or in the family in which it occurs, the experience of cancer, especially if it is a cancer that is not cured, is just as stressful on an individual basis as the event of a nuclear disaster. To my knowledge, of the eight hundred cancer patients that I have treated and that have experienced this stress and grief, there has been only one suicide. The great majority of people, no matter what they have to go through to survive, will fight with every ounce of strength and with every breath to go on living even if they can maintain only a semblance of a quality of life.'

On page 48, Dr Warren K. Sinclair, President of the U.S. National Council on Radiation Protection and Measurements, points out: 'For perspective, according to the 1983 World Almanac, in 1980 there were 1,986,000 deaths in the United States, of which 414,320 were caused by malignant neoplasms (cancers), or about 21 percent.'

A great deal of technical information about alpha radiation hazards in nuclear war is summarised on pages 4-9 by Dr Edward T. Bramlitt, Ph.D., who was the Health Physicist at the Defense Nuclear Agency who was responsible for planning the decontamination of Eniwetok Atoll from 1977-80 after 43 major American nuclear tests there, most of which were surface bursts on barges or islands.

Dr Bramlitt begins by pointing out that although U-235 and U-238 emit alpha particles, they are insignificant compared to plutonium because the shorter half-life of the latter gives it a higher decay rate and hazard. The specific activity (decays per second per kg) of U-235 and U-238 are respectively 30,000 and 190,000 times lower than that of Pu-239, which has a half life of only 24,000 years.

Next, the same basic process which creates Pu-239, ie, neutron capture in U-238 in a reactor and/or in a nuclear explosion, followed by beta decay of U-239 into Np-239 and then into Pu-239, also creates some still heavier isotopes of plutonium such as Pu-240, -241, and -242, due to additional neutron captures. Dr Bramlitt discloses that about 65% of the alpha radioactivity in weapons grade plutonium is from Pu-239, 20% is from Pu-240, and about 15% is initially from Pu-241 (although during storage this rapidly decays, with a 13 years half-life, into Am-241).

Pu-242 is created in non-weapons grade reactor fuel due to its longer irradiation time. The essential point about weapons grade plutonium is precisely the fact that the fuel is removed from the reactor and reprocessed before there has been enough time for much multiple neutron capture, generally after 100 days in the reactor. This maximises the fraction of Pu-239 in the plutonium. Dr Bramlitt points out that neutron capture in U-238 during the nuclear explosion is responsible for the heavy elements in fallout:

'Environmental samples from Rongelap Atoll, which was contaminated primarily by fallout from one thermonuclear weapon test at Bikini Atoll in 1954 [Bravo], were analyzed in 1976 and found to have Am-241 making up about 30 percent of the total alpha activity. ... Pu-241 analyses indicate that Am-241 eventually will comprise about 50 percent of the alpha activity in the Marshall Islands. Following one large nuclear test at Eniwetok Atoll in 1952 [Mike], the amount of Am-241 was reported to be sufficient to eventually account for approximately 80 percent of total alpha activity. ... Pu-238 is a relatively short-lived alpha emitter (88-year half-life) which can be produced by several paths, including (n, 2n) reactions on Pu-239 [ie, the capture of one neutron, followed by the emission of two neutrons]. Pu-238 at Eniwetak Atoll typically accounts for one to ten percent of total alpha activity, although samples associated with some tests showed Pu-238 to be in the range from 30 to 50 percent. ... Bikini and Enewetak samples analyzed in this manner show Pu-240 to account for 50 to 60 percent of combined Pu-239 + Pu-240 activity. ... if 16 nCi are uniformly distributed in the lung and maintained at that level, a dose equivalent rate of 15 rem/yr will result. The maximum permissible concentration in air which leads to this dose rate is 40 picocuries per cubic metre. ... Excess deaths from lung cancers have been observed in studies of animals and people. An excess of neoplasms [abnormal tissue growths, not necessarily malignant] has been found in hamsters at 15 rads of alpha radiation and in a group of miners at cumulative dose to the bronchi of four to nine rads [doses in rads or centigray for alpha radiation are not the same as the dose equivalent in rem or centisieverts because a given dose - energy deposition per unit mass - of alpha radiation is far more ionizing and about more damaging than an equal dose of gamma radiation, so the alpha dose needs to be multiplied by a 'quality factor' of 20 to give the dose equivalent for the purpose of predicting biological effects relative to gamma doses]. A significant excess of lung cancers has been observed in a group of Hiroshima survivors who received 9.8 rads to the lung from gamma and neutron irradiation ... the latent period from radiation exposure to death from lung cancer in people is generally 10 or more years.'

We all use Am-241 in our daily lives, since it is the source used in ionization smoke detectors. The ampha particles emitted by bomb fallout have energy of 5.1-5.5 MeV, resulting in ranges in air of 3.6-4.1 cm. In skin, they have a range of about 35 microns (micrometres), which is less than the average thickness of dead skin on humans, so there is no external hazard from alpha radiation to unbroken skin, and the only risk occurs if contamination enters the body in air, water, food, or through a cut or other damage to skin.

A lung dose rate of 1 millirad per year would result from inhaling air containing 1 femtocurie per cubic metre of alpha emitters like plutonium, or from living in an area where the soil surface is contaminated with 0.2 microcuries per square metre, assuming resuspension of the deposited dust into the air by the wind, and a soil contamination of 13 picocuries per gram in the top centimetre.

Nuclear test data exists for the alpha inhalation risk (page 8). In New York, the Pu-239 concentration in air peaked at 0.45 femtocurie/cubic metre in 1959 (from global fallout after extensive atmospheric nuclear tests by Russia and America in 1958) and peaked at 1.68 in 1963 (from global fallout during even more intense atmospheric nuclear testing by Russia and America in 1962). Because of the cessation of tests at the end of 1962, the air concentration of Pu-239 after 1963 fell rapidly - it was 0.91 in 1964, 0.33 in 1965, and 0.13 in 1966.

The ground deposit of Pu-239 in New York City was also measured: it was 0.2 nanocurie/square metre in 1959 and 0.6 in 1963, rising to 1.5 in 1966. Plutonium soon gets washed out of the atmosphere and enters the deep soil or gets flushed through the drains and enters river or lake deep sediment where it ceases to give rise to biological doses.

Dr Charles J. Bridgman, Professor of Nuclear Engineering at Wright-Patterson Air Force Base, and Dr Arthur T. Hopkins, of the U.S. Air Force, on page 10 disclose that for all nuclear test data for surface bursts, an average of 70% of the radioactivity was distributed throughout the volume of melted fallout particles, and 30% was just on the surfaces of non-melted fallout particles. On page 11 they disclose that the median radius of the lognormal particle size versus activity distribution for low yield surface burst nuclear tests over Nevada desert sand is 50 microns (micrometres) compared to 200 microns for high yield surface bursts on coral rock at Eniwetok and Bikini. The finer the particle size distribution of the natural soil, the smaller the fallout and longer the fallout takes to descend, giving time for much more of the radioactivity to decay in transit during the fallout process before the particles reach the ground. Hence, a nuclear explosion in the air produces no local fallout at all because all the debris condenses into micron-sized particles which take months to years to be deposited, around the world. A surface burst on coarse sand would produce fallout like that from Nevada desert tests, while one on soft rock like coral or limestone would produce larger fallout particles which fallout quickly. A surface burst on hard rock would produce still large fallout particles and faster fallout, some of which would be gravel and rocks, while a surface burst on clay would produce very fine fallout particles more like the global fallout from an air burst. At Hiroshima, the radioactive mushroom had been blown downwind by the time the firestorm with its black rain began 30 minutes to 2 hours later, so there was no significant rainout that contributed to the initial nuclear radiation doses. A firestorm cannot interact with the mushroom cloud because it takes too long to get going.

Dr Conrad V. Chester, of Oak Ridge National Laboratory, on pages 12-13 describes the effects of a nuclear bomb explosion surface burst on a nuclear reactor. A 1-GW nuclear reactor produces 3 kg of plutonium daily. Cooling towers are vulnerable to peak overpressures as low as 1-2 psi, electric power transmission towers and cables can be vulnerable to 3-4 psi if the cables are at right angles to the blast, and auxiliary diesel generators and control rooms may be vulnerable to 25 psi.

However, the pressure vessel containing the nuclear material is very strong and needs an overpressure impulse of about 200 psi-seconds to fracture it, which means a 1 Mt weapon landing at 30-60 metres from the reactor core. If that happens, the core inventory of radioactive material will be added to the nuclear explosion fireball and will add to the fallout. The initial effect is insignificant, since in this case the 1 GW reactor core contents will only increase the 1 Mt bomb gamma dose rate at 1 hour after burst by 1%, but at great times after detonation the reactor debris contribution stands out far more, since it decays more slowly (the short lived nuclides produced in a reactor operating for months are decaying while they are created, but the long lived nuclides accumulate, so the average decay rate of the debris activity surviving from fission in a reactor is slower than the fission product mixture produced in a brief burst in a nuclear bomb).

If a nuclear reactor is hit by 25 psi peak overpressure, the control systems and cooling systems (which rely on auxiliary generators for power if the reactor is shut down) could cause the reactor core to overheat, melt down and may then gradually leak some contamination (this is not the explosive situation which occurred to Chernobyl in 1986). Even when the reactor is shut down, the decaying debris in the core produces a lot of heat which needs to be extracted, unless the design includes an efficient heat sink with natural convection possible.

On pages 17-21, Philip J. Dolan, M.Sc., discusses nuclear terrorism: 'I think it probably is inevitable that a nuclear device will be used by terrorists at some time in the future, either as a serious threat or with an actual explosion ... there is some probability that the acquisition may come about by theft of a weapon rather than by fabrication. ... No doubt, there is enough information available in the open literature to enable a group to build a nuclear warhead ... that would produce somewhere between a few tens of tons and a few kilotons of yield, which would be adequate for their purposes.'

Dolan points out that U-235 is extremely expensive to separate from natural U-238 by gaseous diffusion or centrifuges, while U-233 and Pu-239 must be produced in a nuclear reactor by neutron irradiation of Th-232 and U-238, respectively. The plants required for this would 'probably be beyond the reach of terrorist groups' so theft of nuclear components or of an actual weapon is the major concern:

'Fortunately, the plutonium that is made in power reactors is trapped in the highly radioactive residues of the fuel from which it is made. Plutonium represents only about one-half of one percent of the spent fuel from a light-water power reactor. ... More than a ton of this spent fuel must be processed to obtain enough plutonium for one weapon. ...

'All things considered, it appears that theft of already reprocessed plutonium is the most likely route for the terrorists to obtain the fuel. ... Weapon plutonium typically contains six to eight percent Pu-240 and only trace amounts of Pu-241 and Pu-242. When the reactor is run [for longer periods than when producing weapons grade plutonium] to optimize fuel usage for power production, the heavier isotopes, together with some Pu-238 that is also produced, account for 30 to 35 percent of the plutonium in the spent fuel. Pu-240 and Pu-242 fission spontaneously, producing a continuous neutron background. Pu-241 and [its] daughter products are gamma emitters. ... The neutron background also presents a pre-initiation problem that can significantly complicate the design and production of weapons. ...

'A prime threat would seem to come from some of the smaller countries with relatively undeveloped technological and industrial capabilities. Terrorist activities may appear attractive to the leaders of some of these countries who desire to exert influence beyond their own borders but who lack the military or industrial power to do so. In some cases governments have openly supported terrorist activities, while in other cases strong suspicions of such support exist in spite of denials by the governments concerned. Undoubtedly, some such governments harbor desires for nuclear weapons.'

Dr Robert Ehrlich, Chairman of the Department of Physics, George Mason University, discusses 'nuclear winter' propaganda by 'TAPPS' (R.P. Turco, O.B. Toon, T.P. Ackerman, J.B. Pollack and C. Sagan, Nuclear Winter: Global Consequences of Multiple Nuclear Explosions, in Science, v222, 1983, pp. 1283-92; the acronym TTAPS from the author surnames is deliberately meant to involk the name of the military 'lights out' bugle call, taps) on pages 21-23:

'The postulated drop in temperature following a nuclear war arises due to the blockage of sunlight caused by smoke and dust thrown up into the atmosphere. The smoke arises from fires caused by the nuclear detonations (primarily urban fires), while the dust arises due to the presence of numerous ground burst weapons. ... The rate of rainout of soot and dust depends on the turbulence in the atmosphere. In the TTAPS calculation a kind of temperature inversion results in which there is very little chance for smoke and dust to be removed by rainout. In the actual atmosphere, it seems likely that the patchiness of smoke would cause large thermal gradients and much turbulence leading to a rapid rainout. This effect could lead to a totally negligible temperature decline.

'The large temperature decline in the TTAPS calculation results primarily from smoke rather than dust, since the black soot particles tend to block light more than dust particles. ... In order for 100 megatons to give rise to a large temperature drop, one must assume that 1,000 weapons of 0.1 megaton yield are each delivered against different urban areas causing them all to burn. The TTAPS authors (Turco et al., 1983) recognize this to be a highly artificial scenario ...'

Regarding one 'disastrous' and rather suspect prediction of a 20% reduction in ozone layer thickness by a massive USSR-American nuclear war, the book points out elsewhere that this is merely equal to the natural variations in ozone:

'For purposes of comparison, the effective thickness of the ozone layer during summer is about 20 percent less over Miami than over Seattle.'

Jack C. Greene shows on pages 27-28 how the statistical distribution of fallout over America from a nuclear war would produce a spectrum of dose rate and doses. Assuming that 4,447 megatons of ground surface bursts occurred on the 50 states of the U.S. and had a fission proportion of 50%, that 1 kt of fission products deposited per square mile produces 1,000 R/hr of gamma exposure at the 1 hour reference time over average land with shielding from buildings, people, terrain and ground roughness irregularities, etc., that 80% of the fallout is deposited locally (ie, on the US), and that the land area of the 50 states is 3.6 million square miles, Greene calculates a mean 1-hour dose rate of 500 R/hr. Since the average fallout arrival time for Americans would be 2.25 hours, the mean outside dose over the first four days would be 1,100 R, which is lethal. This is the basic reason why civil defence was needed against fallout in the Cold War.

Using the conversion factor between dose rate and the mean specific activity of Nevada test fallout (5 x 10^14 fissions/gram) on page 20 of Greene's earlier 1973 report, the mean fallout mass deposit over America would be 5 grams per square foot or 54 grams per square metre, which is easily visible. (Similar to pouring 1 kilogram of sugar over a room 4.3 metres wide by 4.3 metres long.) The particles would be felt outside like sand grains bouncing off the hands, arms, and face, and could be heard bouncing off hard surfaces like car roofs, bonnets, and windows.

The statistical variation would not, of course, give everyone 1,100 R. Greene in the 1988 book shows that 10% of the U.S. land area would receive 1-hour dose rates below 110 R/hr and 4-day doses below 170 R, a total of 30% would receive 1-hour dose rates below 220 R/hr and 4-day doses below 330 R, a total of 65% would receive 1-hour dose rates below 520 R/hr and 4-day doses below 830 R, 85% would receive 1-hour dose rates below 990 R/hr and 4-day doses below 1,700 R, and the most contaminated 1% would receive 1-hour dose rates of 1,650-3,300 and 4-day doses of 3,300-10,000 R. (Page 25 of Greene's earlier 1973 report indicates that the maximum dose rates measured at 1 hour after nuclear tests, near ground zero (in the crater and on the lip of the crater) , ranged from 3,000-40,000 R/hr but are not of concern for personnel safety since other lethal effects like blast, heat and initial nuclear radiation would predominate so near ground zero.)

Elsewhere the book shows that the 1 MeV mean gamma ray energy assumed for fallout shielding calculations underestimates the protection factor because that energy is always much higher (more penetrating) than the real gamma ray energy of fallout after nuclear tests, and: 'a group of people sitting back-to-back, and fairly close to each other, would provide a significant amount of "mutual shielding". This would increase the building protection factors by multiplicative factors varying between 1.5 and 3 ... the presence of nearby adjacent structures would provide significant mutual shielding between buildings ... in all cases exposures within structures would be considerably below those predicted for buildings in isolation. Some studies have shown that additional multiplicative protection factors of 2 to 3 are quite reasonable to assume.'

On page 33, Dr Sumner Griffin shows that 75% of all American cattle can be sheltered indoors in barns with food reserves of 85 days on average; the mean protection factor of a barn against gamma radiation is 1.8 if ignoring the mutual shielding of animals by each other and ignoring ground roughness, and the corrected protection factor is about 2.5.

Finally, Dr Kenneth Skrable, Professor of Radiological Sciences at the University of Lowell, explains the mechanism of the neutron bomb on pages 51-2. He explains that the fusion of the hydrogen heavy isotopes deuterium and tritium produces helium-4 plus a neutron plus an energy release of 17.6 MeV, of which 80% (14.1 MeV) is released as the kinetic energy of the neutron and 20% (3.52 MeV) is released as the kinetic energy of the helium atom, leading to the massive neutron radiation and weak fireball of the neutron bomb: 'minimal blast and thermal radiation.'

By contrast, nuclear fission only releases a few percent of the total energy as the kinetic energy of neutrons, so the blast and thermal effects from a predominantly fission weapon exceed those from neutron radiation. Dr Skrable concludes on page 52:

'By replacing fission tactical weapons with neutron weapons, the tactical objective of the weapon will be maintained without the collateral damage and death to civilians outside the area of usefulness. Therefore, the neutron bomb, in fact, may be considered to be a bomb that benefits people, civilian people. ...

'The neutron bomb is a miniature hydrogen bomb. It is to be used as a tactical weapon against invading enemy forces. The civilian population, if ever tactical nuclear weapons had to be employed, should favor the neutron bomb over currently stockpiled fission based tactical weapons.'

The U.S. National Council on Radiation Protection and Measurements (NCRP) Report No. 42, Radiological Factors Affecting Decision-Making in a Nuclear Attack, shows that protracted exposures to radiation - such as fallout - are less dangerous than brief exposure to the same dose. For example, they show that the lethal gamma dose for 50% of humans is 450 R for a week-long exposure, but is about 600 R for a one-month long exposure. Increasing the duration, over which a given amount of radiation is spread, increases the time available for the body's repair mechanisms (both cellular repair and DNA break repair by protein P53) to reduce damage.