While high-level waste has received the lion's share of the public's attention, it is not the only waste problem linked with nuclear power. In fact, analyses indicate that it is not even the most important one. Several other nuclear waste issues must be addressed and we will consider them here.

In terms of numbers of deaths, the most important waste problems arise from the radioactive gas radon. We begin with its story.


Uranium is a naturally occurring element, present in small quantities in all rock and soil, as well as in materials derived from them such as brick, plaster, and cement. It is best known as a fuel for nuclear reactors, but quite aside from the properties that make it useful for that purpose, it is naturally radioactive. This means that it decays into other elements, shooting out high-speed particles of radiation in the process. The residual atoms left following its decay are also radioactive, as is the residue from decay of the latter, and so on, until the chain is terminated after 14 successive radioactive decays. One step in this chain of decays is radon, which is of very special interest since it is a gas. An atom of radon, behaving as a gas, has a tendency to move away from the location where it was formed and percolate up out of the ground into the air. Since uranium is in the ground everywhere, atoms of radon and the radioactive elements into which it decays, known as "radon daughters," float around in the air everywhere and are thus constantly inhaled by people. Since it is a gas, radon itself is very rapidly exhaled. But the elements into which it decays are not gases; hence, they tend to stick to the surfaces of our respiratory passages, exposing the latter to radiation, and thereby inducing lung cancer.

Scientists know a great deal about this problem because of several situations in which miners were exposed to very high levels of radon in poorly ventilated mines.1 In uranium mines, especially, there is often an unusually high concentration of uranium in the surrounding rock. The radon evolving from it percolates out into the mine where it remains for some time, since its escape routes are limited. The best study of this problem involved a group of 4,000 uranium miners who worked in the Colorado plateau region between 1945 and 1969. Among them, 256 men have died from lung cancer, whereas only 59 lung cancers would have been expected from an average group of American men. When this situation was recognized in the late 1960s, drastic improvements were introduced in the ventilation of these mines, reducing radon levels about 20-fold, and making present radon exposure one of the less important of the many significant risks associated with mining.

From the studies of lung cancer among miners, we can estimate the risks in various other radon exposure situations. The most important of these is in the normal environmental exposure we all receive, especially in our homes, which are often poorly ventilated. Radon percolates up from the ground, usually entering through cracks in the floor or along pipe entries. In some areas it enters homes with water drawn from underground wells. As a result of its being trapped inside homes, radon levels indoors are normally several times higher than outdoor levels.

Using the data on miners and applying the linear hypothesis leads to an estimate that this environmental radon exposure is now causing about 14,000 fatal lung cancers each year in the United States,1 several times more fatalities than are caused by all other natural radiation sources combined.

Some people, trying to conserve fuel, carefully seal windows and doors to reduce air leakage from houses. This traps radon inside for longer time periods and therefore increases the radon level in the house. If everyone sealed their houses in accordance with government recommendations, many thousands of additional fatalities each year would result. Recalling from Chapter 8 that most scientists estimate the radiation consequences of a full nuclear power program as less than 10 fatalities per year (even the opponents of nuclear power estimate only a few hundred fatalities per year), we see that conservation is a far more dangerous energy strategy than nuclear power from the standpoint of radiation exposure!

These environmental radon problems have no direct connection with nuclear power, but there are connections in other contexts.2 The most widely publicized of these is the problem of uranium mill tailings. When uranium ore is mined, it is taken to a nearby mill and put through chemical processes to remove and purify the uranium. The residues, called the tailings, are dissolved or suspended in water that is pumped into ponds. When these ponds eventually dry out, they leave what ostensibly are piles of sand. However, this sand contains the radioactive products from the decay of uranium; only the uranium itself has been removed. The most important of these decay products is thorium-230, which has an average lifetime of 110,000 years and decays into radium, which later decays into radon. In fact, most of the radon generated over the next 100,000 years will come from uranium that has already decayed into thorium-230, so removing the uranium has little effect on radon emissions over this 100,000 year period. But a very important effect of mining the uranium is that the source of this radon was removed from underground where the ore was originally situated, to the surface where the mill tailings are located. This allows far more of it to percolate into the air to cause health problems.

A quantitative calculation of this effect indicates that the mill tailings produced in providing a 1-year fuel supply for one large nuclear power plant will cause 0.003 lung cancer deaths per year with its radon emissions. This may not seem large, but it will continue for about 100,000 years, bringing the eventual fatality toll to something like 220! Fortunately, there are measures that can be, and are being, taken to alleviate this problem. If the tailing piles are covered with several feet of soil, nearly all of the radon will decay away during the time it takes to percolate up through the cover — the average lifetime of a radon atom is only 5.5 days. The law now requires that radon emissions from the tailings be reduced 20-fold by covering them with soil (or with other materials). This lowers the eventual fatality toll from 220 to 11. Even this is hundreds of times higher than the 0.018 fatality toll from the high-level waste. Anyone who worries about the effects of radioactive waste from the nuclear industry should therefore worry much more about the uranium mill tailings than about the high-level waste. However, for some reason the latter has received the great majority of the publicity and hence the most public concern. As a result, the government research program on controlling mill tailings is only a small fraction of the size of the program on high-level waste. The dollars are spent in response to public concern, rather than where the real danger lies.

By far the most important aspect of radon from uranium mining is yet to be discussed.2 When uranium is mined out of the ground to make nuclear fuel, it is no longer there as a source of radon emission. This is a point which has not been recognized until recently because the radon that percolates out of the ground originates largely within 1 meter of the surface; anything coming from much farther down will decay away before reaching the surface. Since the great majority of uranium mined comes from depths well below 1 meter, the radon emanating from it was always viewed as harmless. The fallacy of this reasoning is that it ignores erosion. As the ground erodes away at a rate of 1 meter every 22,000 years, any uranium in it will eventually approach the surface, spending its 22,000 years in the top meter, where it will presumably do great damage. The magnitude of this damage is calculated in the Chapter 12 Appendix, where it is shown that mining uranium to fuel one large nuclear power plant for one year will eventually save 420 lives! This completely overshadows all other health impacts of the nuclear industry, making it one of the greatest lifesaving enterprises of all time if one adopts a very long-term viewpoint.

Before we can count these lives as permanently saved, we must specify what is to be done with the uranium, for only a tiny fraction of it is burned in today's nuclear reactors. There are two dispositions that would be completely satisfactory from the lifesaving viewpoint. The preferable one would be to burn it in breeder reactors and thereby derive energy from it. An easy alternative, however, is to dump it into the ocean. Uranium remains in the ocean only for about a million years before settling permanently into the bottom sediments. All uranium in the ground is destined eventually to be carried by rivers into the ocean and spend its million years therein. From a long-range viewpoint it makes little difference to the health of humans or of sea animals if it spends that time in the ocean now or a million years in the future. But by preventing it from having its 22,000 year interlude within 1 meter of the ground surface, we are saving numerous human lives from being lost due to radon.

If one adopts the position that only effects over the next 500 years are relevant, there is still an important effect from uranium mining because about half of all uranium is surface mined. Approximately 1% of this comes from within 1 meter of the ground surface where it is now serving as a source of radon exposure. It is shown in the Chapter 12 Appendix that eliminating this source of exposure by mining will save 0.07 lives over the next 500 years, but in the meantime, radon escaping from the covered tailings will cause 0.07 deaths. Thus, the net health effect of the radon from uranium mining is essentially zero over the next 500 years.

We still have one more radon problem to discuss, namely, the radon released in coal burning. Coal contains small quantities of uranium; when the coal is burned, by one route or another, this uranium ends up somewhere in the ground. Again, the problem is complicated by the fact that, if the coal had not been mined, erosion would eventually have brought the coal with its uranium to the surface anyhow. The final result, derived in the Chapter 12 Appendix, is that the extra radon emissions caused by the burning of coal in one large power plant for one year will eventually cause 30 fatal lung cancers. This toll, like so many others we have encountered here, is thousands of times larger than the 0.018 deaths caused by the high-level waste produced in generating the same amount of electricity from nuclear fuel.

As discussed previously, solar electricity burns 3% as much coal as would be needed to produce the same amount of electricity by direct coal burning. Solar electricity should therefore be charged with (3% of 30 =) one death per year from radon. This again, is far larger than the 0.018 deaths per year from high-level radioactive waste.


Nuclear power plants, and more importantly, fuel-reprocessing plants, routinely emit small quantities of radioactive material into the air and also into nearby rivers, lakes, or oceans. Opponents of nuclear power made a rather big issue of this in the early l970s, but there has been relatively little publicity about it in recent years. The quantities of radioactivity released are limited by several Nuclear Regulatory Commission requirements, including requiring that equipment be installed to reduce these emissions if doing so will result in saving one life from radiation exposure for every $4 million spent.3 Since new and improved control technologies are steadily being developed, these emissions have been reduced considerably since the early 1970s, and that trend is continuing. There are also NRC requirements on maximum exposure to any individual member of the public, limiting his or her exposure to 5 mrem per year, less than 2% of that from natural radiation; 0.5% is typical for those living very close to a nuclear plant. This gives them a risk equivalent to that of driving 3 extra miles per year, or of crossing a street one extra time every 4 months.

The principal emissions of importance from the health standpoint are radioactive isotopes of krypton and xenon (Kr-Xe), gases which are very difficult to remove from the air; tritium (T), a radioactive form of hydrogen that becomes an inseparable part of water, used in such large quantities that it would be impractical to store it until the tritium decays away; and carbon-14 (C-14), which becomes an inseparable part of the omnipresent gas carbon dioxide. Of these, only C-l4 lasts long enough to irradiate future generations.

According to a recent study by the United Nations Scientific Committee on Effects of Atomic Radiation,4 the releases associated with 1 year of operation of one power plant can be expected to cause 0.06 deaths from Kr-Xe, 0.016 from T, and 0.23 from C-l4 over the next 500 years and 1.6 deaths over tens of thousands of years. This gives a total of 0.25 over the next 500 years and 1.6 deaths eventually. Nearly all of these effects are a consequence of fuel reprocessing, and they are spread uniformly over the world's population. Thus only a few percent of the deaths would be in the United States. There are improved technologies for reducing these emissions drastically, and regulations for implementing some of them have already been promulgated, but there is no rush to complete them since we are not doing reprocessing. If and when we do, it is reasonable to expect emissions to be reduced about 5-fold. In our summary we will assume that to be the case.

Note again that these effects are many times larger than the 0.0001 deaths during the first 500 years and the 0.018 eventual deaths from high-level waste. They are also larger than the 0.02 deaths estimated by the Reactor Safety Study as the average toll from reactor accidents. Here again, we see that public concern is driven by media attention and bears no relationship to actual dangers.


In order to minimize the releases of radioactivity into the environment as discussed in the last section, there is a great deal of equipment in nuclear plants for removing radioactive material from air and water by trapping it in various types of filters. These filters, including the material they have collected, are the principal component of what is called low-level radioactive waste, the disposal of which we will be discussing here. Other components are things contaminated by contact with radioactive material, like rags, mops, gloves, lab equipment, instruments, pipes, valves, and various items that were made radioactive by being in or very near the reactor, where they were exposed to neutrons. Not all of the low-level radioactive waste is from the nuclear industry. Some is from hospitals, research laboratories, industrial users of radioactive materials, and the like. These make up about 25% of the total.

In general, the concentration of radioactivity in low-level waste is a million times lower than in high-level waste — that is the reason for the name — but the quantities in cubic feet of the former are thousands of times larger. It is therefore neither necessary nor practical to provide the low-level waste with the same security as the high level. The low-level waste is buried in shallow trenches about 20 feet deep in commercially operated burial grounds licensed by the federal or state governments.

Since its potential for doing harm is relatively slight, until recently this low-level waste was handled somewhat haphazardly. There was little standardization in packaging, in handling, or in stacking packages in trenches, and little care in covering them with dirt. As a result, trenches sometimes filled with rain water percolating down through the soil, which then dissolved small amounts of the radioactivity. The caretakers regularly pumped this water out of the trenches and filtered the radioactivity out of it, allowing no radioactivity to escape.

The situation was radically changed by two very innocuous but highly publicized incidents during the l970s. In an eastern Kentucky burial ground, a place called Maxey Flats, tiny amounts of radioactivity were found off site. The quantities were so small that no one could have received as much as 0.1 mrem of radiation (1 chance in 40 million of getting a cancer). The publicity was enormous at the time. The January 18, 1976, issue of the Washington Star carried a story headlined "Nuclear Waste Won't Stay Buried," which began with "Radioactive wastes are contaminating the nation's air, land, and water." When the head of the U.S. Energy Research and Development Agency (predecessor of Department of Energy) testified for his Agency's annual budget, the first 25 minutes of questions were about Maxey Flats, and a similar pattern was followed with the chairman of the Nuclear Regulatory Commission.

After another congressional committee was briefed about the problem, its chairman stated publicly that it was "the problem of the century." I pointed out to him that due to the high uranium content of the granite in the congressional office building, his staff was being exposed to more excess radiation every day than anyone had received in toto from the Kentucky incident. He made no further alarmist statements, but as a result of all the attendant publicity the burial ground was closed.

The other incident occurred in a western New York State burial ground where there was a requirement that permission be granted from a state agency to pump water out of the trenches. In one instance, this permission was somehow delayed, in spite of urgent warnings from the site operators. As a result, some slightly contaminated water overflowed, with completely negligible health consequences — the largest doses were 0.0003 mrem — leading to widespread adverse publicity and closing of the burial ground.

A television "documentary" on the Kentucky burial ground caused a lot of problems. It showed a local woman saying that the color of the water in the creek had changed, but it did not point out that any scientist would agree that this could not possibly be due to the radioactivity. Moreover, the producer was told that the coloring was caused by bulldozing operations nearby. The same program showed a local farmer complaining that his cattle were sick — "Hair been turning gray, grittin' their teeth, and they're a-dying, going up and down in milk." There was no mention of the facts that a veterinarian had later diagnosed the problem as a copper and phosphorus deficiency, the cattle had been treated for this deficiency and had recovered, and that the TV crew had been informed about this long before the program was aired.

A large portion of a newspaper feature story was devoted to the story of a woman telling about how she was dying of cancer due to radiation caused by the leakage of radioactive material. A 0.1-mrem radiation dose has 1 chance in 40 million of inducing a cancer, whereas one person in five dies from cancers due to other causes, so there is no more than 1 chance in 8 million that her cancer was connected with the leakage. Actually, it is surely less than that, because her cancer was not of a type normally induced by radiation. Moreover, 0.1 mrem was the maximum possible dose to any person; the dose received by this particular woman was probably very much lower.

As a result of this publicity and the public's extreme sensitivity to even the slightest radiation exposure from the nuclear industry, plus some honest desire on the part of government officials to improve security, a new set of regulations was formulated.5 It requires that (1) the trench bottoms be well above any accumulating groundwater (i.e., the "water table") so as to exclude the possibility of the trenches filling with water; (2) surface covers be installed to minimize water passing through the trenches; (3) the waste be packed in such a way that the package maintains its size and shape even under heavy external pressures, when wet, or when subject to other potential adverse chemical and biological conditions; (4) packaging material be more substantial than cardboard or fiberboard; (5) there be essentially no liquid in the waste (excess water must be evaporated off); (6) empty spaces between waste packages be carefully filled; and so on.

Since the movement of water through the trenches can only be downward toward the water table, any radioactivity that escapes from the packages can only move downward until it reaches the water table, after which it can flow horizontally with the groundwater flow, which normally discharges into a river. This movement, as was explained in our discussion of high-level waste, must take many hundreds or thousands of years, because the water moves very slowly, and the radioactive materials are constantly being filtered out by the rock and soil.

In order to estimate the hazards from a low-level waste burial ground,2 we must consider two possible routes low-level waste could take to get from the ground into the human stomach: (1) it could be picked up by plant roots and thereby get into food, and (2) it could be carried into a river and thereby get into water supplies. To evaluate the food pathway, we assume the unfavorable situation in which all of the waste escapes from its packaging and somehow becomes randomly distributed through the soil between the surface and the top of the water table. For natural materials, we know how much of each chemical element resides in this soil and how much resides in food, so we can calculate the probability per year for transfer from the soil into food. This probability then can be applied to the radioactive material of the same element, and of other elements chemically similar to it. When this probability is multiplied by the number of cancers that would result if all the low-level waste were ingested by people (calculated using the methods described in the Chapter 11 Appendix for the quantities of low-level waste generated by the nuclear industry), the product gives the number of deaths per year expected via the food pathway.

It is estimated that it would take something like 800 years for any of the radioactivity to reach a river. In order to estimate conservatively the hazards from the drinking water pathway, we therefore assume that all radioactivity remaining after 800 years reaches a river, and as in the case of high-level waste explained in Chapter 11, one part in 10,000 of it enters a human stomach.

Adding the effects of the food and drinking water pathways, we obtain the total number of deaths from the low-level waste generated by one large nuclear power plant in 1 year to be 0.0001 over the first 500 years and 0.0005 eventually.2

Siting of Low Level Waste Burial Grounds

While we have shown that the hazards from low-level waste burial are very minimal, the public perception is quite the opposite. We have already noted how the burial grounds in Maxey Flats, Kentucky, and western New York State were shut down by adverse publicity. A third commercial burial ground near Sheffield, Illinois, was permanently closed when the space allocated in the original license became filled and an application for additional space was rejected. I was involved in the hearings leading to that decision. The setting had a strong "backwoods" flavor, and emotion ran rampant. No one seemed interested in my risk analyses and other quantitative information. The county supervisors decided not to increase the allocation.

That left three remaining burial grounds for commercial waste — at Barnwell, South Carolina, Beatty, Nevada, and Richland, Washington. Barnwell now accepts more than half of the total volume of waste, which contains more than three-fourths of the radioactivity. Richland accepts most of the remainder. With all the bad publicity, these sites became political liabilities to the states' governors, leading them to make threats that they might stop accepting waste from out of state. Politicians loudly proclaimed that it wasn't fair for their state to accept everyone else's waste. All three sites escalated their charges to customers constantly, nearly 10-fold over the past decade.

In order to resolve these problems, Congress passed legislation in 1980, amended in 1986, requiring each state to take responsibility for the low-level radioactive waste generated within its borders. The law contained a series of deadlines. By July 1, 1986, each state had to pass legislation specifying either that it would build a facility or that it had made arrangements for its waste to be sent to another state's facility; by January 1, 1988, each state was required to have legislation specifying the site selection procedure. Failure to meet any deadline meant that waste from that state must pay a hefty surcharge when sent for burial, and after January 1, 1993, the three states which now have burial grounds may refuse to accept waste from other states.

In response to this law, states formed compacts. For example, the Appalachian Compact consists of Pennsylvania, West Virginia, Delaware, and Maryland, with a disposal site in Pennsylvania. Present arrangements are to have sites in at least 13 states, with 6 of these scheduled to be operational by January 1, 1993. After that, states not affiliated with these 6 must make arrangements with 1 of them, or store their waste until their facility is available.

The law seems to be working reasonably well. There are pressures to extend deadlines, but so far the government has refused to do so. There is a great deal of activity as contractors go through the process of site selection. Local opposition is often vocal, and there is lots of negative publicity, but progress is being made.


Although not found in nature, elements heavier than uranium, called transuranics (TRU), are produced in reactors and therefore become part of the waste, both high level and low level. These elements, like plutonium, americium, and neptunium, have special properties that make them more dangerous than most other radioactive materials if they get into the body. Since they are often relatively easily separated from other low-level wastes, it is government policy to do this where practical, and to dispose of these TRU wastes by deep burial somewhat similar to that planned for high-level waste. The first repository for this purpose, designated for waste from military programs, is now under construction in New Mexico. The amount of radioactivity in the TRU waste is many times less than in the high-level waste. Furthermore, elevated temperatures, which many consider to be the principal threat to waste security, are not a factor for TRU waste. Therefore, while there has been little study of commercial TRU waste disposal (there will be no commercial TRU waste until reprocessing is instituted), it seems clear that it represents less of a health hazard than high-level waste.


Table 1 lists the health effects from electricity generation that have been discussed in this and the previous chapter in terms of the number of deaths caused by 1 year's operation of a large power plant during the first 500 years, and eventually over multimillion-year time periods.2 The minus sign for radon emissions from nuclear power indicate that lives are saved rather than lost.

This table is the bottom line on the waste issue. It shows that, in quantitative terms, radioactive waste from nuclear power is very much less of a hazard than the chemical wastes, or even the radioactive wastes, from coal burning or solar energy. Almost every technology-based industry uses energy derived from coal and produces chemical wastes, and in nearly all cases, these are more harmful than the nuclear waste. This is true even ignoring the lives saved by mining uranium out of the ground; if the latter is included, nuclear waste considerations give a tremendous net saving of lives. By any standard of quantitative risk evaluation, the hazards from nuclear waste are not anything to worry about.

The problem is that no one seems to pay attention to quantitative risk analyses. There have been several books written (always by nontechnical authors) about the hazards of nuclear waste, without a mention of what the hazards are in quantitative terms. In one case the author interviewed me, at which time I went to great lengths in trying to explain this point, but he didn't seem to understand. When I explained what would happen if all the radioactive waste were ingested by people, he was busily taking notes, but when I tried to explain how small the probability is for an atom of buried waste to find its way into a human stomach, the note-taking stopped, and he showed impatience and eagerness to get on to other subjects. My impression was that he was writing a book to tell people about how horrible nuclear waste is, and his only interest was in gathering material to support that thesis.



Deaths Caused
SourceFirst 500 years  Eventually
  High-level waste0.00010.018
  Radon emissions0.00-420
  Routine emissions (Kr,Xe,T,14C)0.050.3
  Low-level waste0.00010.0004
  Air pollution7575
  Radon emissions0.1130
  Chemical carcinogens0.570
Photovoltaics for solar energy
  Coal for materials1.55
  Cadmium sulfide0.880


The real waste problem has been waste of taxpayers' money spent to protect us from the imagined dangers of nuclear waste. One example is the handling of high-level waste from production of materials for nuclear bombs at the Savannah River plant in South Carolina.6 This military high-level waste is considerably less radioactive than the commercial waste discussed in Chapter 11, so the DOE developed and carefully studied several alternative plans. One relatively cheap ($500 million) one was to pump the waste dissolved in water deep underground, whereas a much more expensive ($2.7 billion) one was to handle it like civilian high-level waste, converting it to glass and placing it in an engineered deep underground repository. The DOE study estimated that the first plan would eventually lead to eight fatal cancers spread over tens of thousands of years into the future (assuming that there will be no progress in fighting cancer). The U.S. DOE therefore decided to spend the extra $2.2 billion to save the eight future lives, a cost of $270 million per life saved. If this $2.2 billion were spent on cancer screening or highway safety, it could save 10,000 American lives in our generation. As one example, we could put air bags in all new cars manufactured this year and thereby save over a thousand lives per year for the next several years. Or we could install smoke alarms in every American home, saving 2,000 lives each year for some time to come.

But even these comparisons do not fully illustrate the absurdity of our unbalanced spending, because people living here many thousands of years in the future have no closer relationship to us than people now living in underdeveloped countries where millions of lives could be saved with this $2.2 billion. Some people have argued that it doesn't really help to save lives in countries suffering from chronic starvation because doing so increases their population and thus aggravates the problem. But that argument does not apply to areas where catastrophic 1-year famines have struck. For example,7 during the l970s, there were two such famines in Bangladesh that killed 427,000 and 330,000, a localized famine in India that killed 829,000, and one each in the African Sahel and Ethiopia that were responsible for 100,000 and 200,000 deaths, respectively. If we place such a low value on the lives of these peoples, how can we place such a high value on the lives of whoever happens to be living in this area many thousands of years in the future?


The most flagrant waste of taxpayer dollars in the name of nuclear waste management is going on at West Valley, New York, about 30 miles south of Buffalo.8 Since the West Valley problem has been widely publicized, it is worth describing in some detail. This was the site of the first commercial fuel-reprocessing plant, completed in 1966 and operated until 1972, when it was shut down for enlargement to increase its capacity. During the following few years, government safety requirements were substantially escalated, making the project uneconomical: the original cost of the plant was $32 million, and the initial estimated cost of the enlargement was $15 million, but it would have cost $600 million to meet the new requirements for protection against earthquakes. (All areas have some susceptibility to earthquakes, but it is minimal in the West Valley area.) It was therefore decided to abandon the operation, raising the question of what to do with the high-level waste stored in an underground tank.

The potential hazard was that the radioactive material might somehow leak out, get into the groundwater, and be carried with it into a nearby creek which runs into Lake Erie. Lake Erie drains into Lake Ontario and eventually into the St. Lawrence river, and the three of these are used as water supplies for millions of people. How dangerous would this be?

If all of the radioactive waste stored at West Valley were dissolved in Lake Erie now, and if it passed unhindered through the filters of water supply systems with no precautions being taken, we could expect 40,000 eventual fatalities to result. However, the radioactivity decreases with time by about a factor of 10 per century for the first few hundred years, so that if it were dumped into Lake Erie 400 years from now, only six fatalities would result; and if the dumping occurred more than 1,000 years in the future, there would probably not be a single fatality.

How likely would it be for wastes to get into Lake Erie in the near future? Let us suppose that all the containment features designed into the system failed, releasing all of the radioactive material into the soil. The nearest creek is several hundred feet away, and water soaking through the soil would take 10 to 100 years to traverse this distance. But the radioactive material would travel much more slowly; it would be effectively filtered out as the water passed through the soil and would consequently take 100 times longer — a total of at least a thousand years to reach the creek. We see that this alone gives a very high probability that the material will not get into the creek or lakes until its radioactivity is essentially gone.

But how likely is a release into the soil? The initial protection against this is the tank in which the waste is contained. It is basically one tank inside another, so that if the inner tank leaks, the radioactive material will still be contained by the outer tank and a warning about the situation will be given. In addition, there are three further barriers keeping it from getting into the soil. First, the tanks are in a concrete vault which should contain the liquid. Second, the concrete vaults are surrounded by gravel, and there are pipes installed to pump water out of this gravel. If the radioactivity managed to get into this region, it could still be pumped out through these pipes; there would be plenty of time — many weeks, at least — to do this. Third, the entire cavity is in a highly impermeable clay that would take a very long time for the liquid to penetrate before reaching the ordinary soil. There is still one last barrier worthy of mention; the water flow in the creek is sufficiently small that during the 10 or more years it would take the groundwater to reach it, a system could be set up for removing the radioactivity from the creek water.

Some perspective on the danger of leakage into the ground may be gained from considering a Russian program in which more than twice the radioactive content of the West Valley storage tank was pumped down a well into the ground. This was done as an experiment to study movement of the radioactivity through the ground with a view to using this method for large-scale high-level waste disposal. At last report the results were consistent with expectations and the plans were proceeding.

Up to this point we have been assuming that the radioactive materials are in solution in the waste storage tank, but actually 95% of them are in a solid sludge which is lying on the bottom of these tanks. This sludge would be much less likely to get out through a leak, to penetrate the concrete vault, and to be transported through the ground with groundwater; even if it were dumped directly into Lake Erie, most of it would settle to the bottom, and even if it got into city water supplies, it would very probably be removed by the filtration system. The consequences of release into Lake Erie that we have given earlier are therefore probably 10 times too high.

In summary, if there should be leakage from the tank, it would very probably be contained by the concrete vault. If it were not, it could be pumped out with the water which permeates the surrounding gravel. If this should fail, it would be contained for many years by the thick clay enclosing the entire cavity. When it did eventually get through to the surrounding soil, the movement of the radioactive materials would be sufficiently slow that they would decay to innocuous levels before reaching the creek. It would not be difficult to remove the radioactive material from the creek itself if this were necessary; if, as seems virtually certain, the material was delayed from reaching Lake Erie for at least 400 years, less than one fatality would be expected. If all else failed, any excess radioactivity in Lakes Erie and Ontario would be detected by routine monitoring operations, allowing precautions to be taken to protect public health.

But what if there were a violent earthquake? A structural analysis indicates that even the most violent earthquake believed possible in that area would not rupture the waste storage tanks. (Such an earthquake is expected only once in 16,000 years.) One might consider sabotage of the tank with explosives; but the tank is covered with an 8-foot thickness of clay which would be extremely hazardous to dig through unless elaborate protective measures were taken. If the tank were successfully ruptured, all of the other protective barriers would remain intact, so in all likelihood no harm would result. Saboteurs have many more inviting targets available if their aim is to take human lives. As an example, they could easily kill thousands of people by introducing a poison gas into the ventilation system of a large building.

A very large bomb dropped from an airplane could reach the waste and vaporize it: if this happened, several hundred fatalities would be expected, but far more people would be killed if this bomb were dropped on a city. These same considerations apply to a possible strike by a large meteorite or the development of a volcano through the area. These latter events are, of course, extremely improbable.

Up to this point we have ignored the effects of the radioactive materials permeating the soil in the event of a leak in the tank followed somehow by bypass of the concrete vault, the gravel pump-out system, and the thick clay lining. While in the soil, the radioactive materials could be picked up by plants and get into human food. How much of a hazard would this be? If all the radioactivity in the West Valley waste storage tank were now to become randomly distributed through the soil from the surface down to its present depth, if its behavior in soil is like that in average U.S. soil with the same percentage of land area used for farming, and if no protective action were taken, we would expect 30 fatalities. If the situation were postponed for 100 years, 3 fatalities would result, and if it were postponed for more than 180 years, we would not expect any. Our assumption here that the material becomes randomly distributed through the soil up to the surface is probably a very pessimistic one. Also, in the very unlikely event in which there could be a problem, it would easily be averted by checking food grown in the area for radiation and removing from the market any with excessive radioactivity.

In 1978, the DOE set about deciding what to do about this waste tank.9 The simplest solution would be to pour cement mix into the tank to convert its contents into a large block of cement. This would eliminate any danger of leakage. The principal danger would then be that groundwater could somehow penetrate successively through the clay barrier, the concrete vault, and the stainless steel tank wall to dissolve away some of this cement. Each of these steps would require a very long time period. For example, although the sides of swimming pools and dams are cement, we note that they aren't noticeably leached away in many years even by the soaking in water to which they are exposed; moreover, groundwater contact is more like a dampness than a soaking. If the material did become dissolved in groundwater, all the barriers to getting into Lake Erie outlined above would still be in place and would have to be surmounted before any harm could be done. Even this remote danger could be removed by maintaining surveillance — periodically checking for water in the concrete vault and pumping it out if any should accumulate. The cost of converting to cement would be about $20 million, and a $15 million trust fund could easily provide all the surveillance one might desire for as long as anyone would want to maintain it.

If this were done, what would the expected health consequences be? I have tried to do risk analyses by assigning probabilities, and I find it difficult to obtain a credible estimate higher than 0.01 eventual deaths. It would be very easy to support numbers hundreds or thousands of times smaller.

However, this management option is not being taken. Instead the DOE has decided to remove the waste from the tank, convert it to glass, and bury it deep underground in accordance with plans for future commercial high-level waste. This program will cost about $1 billion. Spending $1 billion to avert 0.01 deaths corresponds to $100 billion per life saved! This is going on at a time when the same government is turning down projects that would save a life for every $100,000 spent! That is our real waste problem.

One last item deserves mention here — the radiation exposure to workers in executing the plans described above. It turns out that exposure is greater in the billion-dollar plan that was adopted than in the plan for conversion to cement, by an amount that would cause 0.02 deaths (i.e., a 2% chance of a single death) among the workers. Since this is more than 0.01 deaths to the public from the conversion to cement, the billion-dollar plan is actually more dangerous.

I have met the government officials who chose the billion-dollar plan, and have discussed these questions with them. They are intelligent people trying to do their jobs well. But they don't view saving lives as the relevant question. In their view, their jobs are to respond to public concern and political pressures. A few irrational zealots in the Buffalo area stirred up the public there with the cry "We want that dangerous waste out of our area." Why should any local people oppose them? Their congressional representatives took that message to Washington — what would they have to gain by doing otherwise? The DOE officials responded to that pressure by asking for the billion-dollar program. It wasn't hurting them; in fact, having a new billion-dollar program to administer is a feather in their caps. Congress was told that a billion dollars was needed to discharge the government's responsibility in protecting the public from this dangerous waste — how could it fail to respond?

That is how a few people with little knowledge or understanding of the problem induced the United States Government to pour a billion dollars "down a rathole." I watched every step of the process as it went off as smooth as glass. And the perpetrators of this mess have become local heroes to boot.


No discussion of nuclear waste problems could be complete without including leaking waste storage tanks. There was a great deal of publicity about them and their hazards, especially in the early 1980s.

When high-level waste is first isolated in a chemical-reprocessing plant, it is stored in underground tanks for a few years before being converted into glass. One of these tanks might handle the waste accumulating from 50 power plants over several years, a very substantial quantity of radioactivity. This raises the question of the dangers from possible leaks in the tanks.

This question arose very early in the history of nuclear energy. During World War II, the Hanford Laboratory was established in the desert of central Washington State to produce bomb material in reactors and separate it in chemical-reprocessing plants, leaving the waste in underground storage tanks. These tanks were made of ordinary steel, which was known to be readily corroded by the waste solution. It was therefore assumed that the tanks would eventually leak, releasing the radioactive waste into the soil. The thinking at that time was that this would be an acceptable situation — if the radioactive material was eventually to be buried underground anyway, why not let some of it get there through leaks in storage tanks? However, as public concern about radiation escalated over the years, this procedure became less acceptable, and in the 1960s the technology was changed. The new tanks were thereafter constructed of stainless steel, which is much less easily corroded, and facilities were included to keep close track of any corrosion that might occur. But more important, the new tanks were constructed with double walls. Thus, if the inner wall developed a leak, the liquid would fill the space between the walls, thereby signaling the existence of the leak. In the meantime, the liquid would still be contained by the outer tank wall, which would leave plenty of time to pump the contents into a spare tank.

These new tanks, which have been used in several locations for many years, including the West Valley tank described above, have had no problems. But some of the old single-wall tanks at Hanford have developed leaks, as was to be expected. On such occasions, the practice has been to pump the contents into a spare tank, but on one occasion in particular, a leak went undetected for 7 weeks, resulting in discharge of a substantial quantity of radioactivity into the soil. Although there has been extensive adverse publicity from the incident — and there was irresponsible negligence involved — there have been no health consequences. Moreover, it is most difficult to imagine how there can ever be any, due to the following considerations.

All of the significant radioactive materials are now absorbed in the soil within a few feet of the tank, still 150 feet above the water table. As rain-water occasionally percolates down through the soil, some of this material may be expected eventually to reach the water table after several hundred years. By this time, all of the radioactive materials except Plutonium will have decayed away. The groundwater in the water table takes about 20 years to reach the river, but plutonium in groundwater is constantly filtered out by the rock and hence travels 10,000 times more slowly than the water. It would therefore take 200,000 years to reach the river, by which time it would have decayed away.

However, even if all of the plutonium involved in the leak (about 50 grams) were dumped into the river now, there is less than a 1% chance that even a single human health effect would result. While the analysis given here contains no controversial elements and has never, to the best of my knowledge, been scientifically questioned, I have frequently heard word-of-mouth claims of detrimental health effects, both present and future, from the leaking tanks at Hanford. Rarely are such effects not at least hinted at in popular books and magazine articles about radioactive waste.

Some readers may be surprised by the above statement that 50 grams (1.8 ounces) of plutonium dumped in a river is so harmless. Part of the reason for this is that only one atom in 10,000, or 0.005 grams, can be expected eventually to enter human stomachs (see Chapter 11 Appendix). Another part of the reason is that when plutonium does enter a stomach, 99.99% of it is excreted within a few days, so it has little opportunity to irradiate the vital body organs. The stories one hears about the high toxicity of plutonium are all based on inhaling it into the lungs, rather than ingesting it with food or water.


Spent fuel must be shipped from reactors to reprocessing plants, and the high-level waste derived from it must be shipped to a repository for burial; therein lie substantial potential hazards. After all, inside plants the radioactivity is remote from the public, but in transport close proximity is unavoidable. Moreover, accidents are inherently frequent in transportation — half of all accidental deaths occur during the few percent of the time we spend traveling.

We will confine our attention to the shipping of spent fuel, since that is where there has been the most experience; other high-level waste shipments will be handled analogously. One general safety measure is to delay shipping as long as possible to allow short-lived radioactivity to decay away. For spent fuel, the minimal delay has been 6 months. But the most important safety device is the cask in which the spent fuel is shipped. It typically costs a few million dollars, and one can well imagine that a great deal of protection can be bought for that kind of money.

These casks have been crashed into solid walls at 80 miles per hour, and hit by railroad locomotives traveling at similar speeds, without any release of their contents.10 These and similar tests have been followed by engulfment in gasoline fires for 30 minutes and submersion in water for 8 hours, still without damage to the contents. In actual practice, these casks have been used to carry spent fuel all over the country for more than 40 years. Railroad cars and trucks carrying them have been involved in all sorts of accidents, as might be expected. Drivers have been killed; casks have been hurled to the ground; but no radioactivity has ever been released, and no member of the public has been exposed to radiation as a consequence of such accidents.

If we try to dream up situations that could lead to serious public health consequences, we are limited by the fact that nearly all of the radioactivity is solid material, unable to leak out like a liquid or a gas. There is no simple mechanism for spreading it over a large area even if it did get out of the cask. In almost any conceivable situation, significant radiation exposure would he limited to people who linger for several minutes in the immediate vicinity of the accident; hence the number of people exposed would be relatively small.

When all relevant factors are included in an analysis, studies indicate that with a very full nuclear power program there would eventually be one death every few thousand years in the United States resulting from radioactivity releases in spent fuel transport accidents.11 Of course, there would be many times that number of deaths from the normal consequences of these accidents.

Some people have posed the problem of terrorists blowing up a spent fuel cask while it is being transported through a city.12 To study this, Sandia National Laboratory carried out tests with high-explosive devices and lots of instruments to help determine just what was happening. Their conclusion was that even if several hundred pounds of high explosives were used on a cask traveling through downtown Manhattan at noon on a week day, the total expected number of eventual deaths would be 0.2; that is, there is only a 20% chance that there would be a single death from radiation-induced cancer.

A more important effect is the slight radiation exposure to passers-by as trucks carrying radioactive waste travel down highways. Some gamma rays emitted by the radioactive material can penetrate the walls of the cask to reach surrounding areas. (Due to weight limitations, it is not practical to have as thick a shield wall on a truck as is used around reactors.) Even with a full nuclear power program, however, exposures to individuals would be a tiny fraction of 1 mrem. The effects would add up to one death every few centuries in the United States.11

Some perspective on these results can be obtained by comparing them with impacts of transport connected with other energy technologies. Gasoline truck accidents kill about 100 Americans each year and injure 8 times that number. It has been estimated that coal-carrying trains kill about 1,000 members of the public each year. Clearly, the hazards in shipment of radioactive waste are among the least of our energy-related transportation problems.

In spite of the very long and perfect record of spent fuel transport and its extremely small health effects, a great deal of public fear has been generated. Many municipalities, ranging from New York City down to tiny hamlets, have consequently passed laws prohibiting spent fuel transport through their boundaries. It does not take many such restrictions to cause extreme difficulty in laying out shipping routes. A New York City ordinance, for example, prevents any rail or truck shipments from Long Island to other parts of the United States.


In the late 1970s, there was a great deal of publicity about a disastrous accident involving high-level nuclear waste in the Soviet Union. The incident reportedly took place near the Kyshtym nuclear weapons complex in the southern Ural Mountains during the winter of 1957-1958. First reports13 were from Z. A. Medvedev, a Soviet scientist now living in England, who pieced together information from rumors circulating among Russian scientists and from radiation contamination studies reported in the Russian literature. A second Soviet scientist, now living in Israel, reported having driven through the area in 1960 and observing that the area was not inhabited; his driver told him that there had been a large explosion there. Medvedev theorized that this situation had resulted from a nuclear explosion of radioactive waste, and according to rumors he had heard, there had been many casualties and serious radioactive contamination over an area of several thousand square miles.

Following these reports, extensive studies were carried out at Oak Ridge National Laboratory,14 and later at Los Alamos National Laboratory,15 using information culled from a thorough search of the Russian literature and U.S. Government (CIA) sources. It was concluded that the incident resulted from some very careless handling of radioactive waste in the Russian haste to build a nuclear bomb. It was deduced that there had been a chemical explosion due to use of a reprocessing technology that is very efficient but was rejected for use in the United States because of the danger of such an explosion. Both groups concluded that it could not have been caused by a nuclear explosion of buried radioactive waste as had been conjectured by Medvedev, because the ratio of the quantities of various radioactive materials was grossly different from that produced by a nuclear explosion. Moreover, a large nuclear explosion requires that very special materials be brought together very rapidly, in a tiny fraction of a second, and that very little water be present, whereas waste is normally dissolved in water, and the movement of materials is very slow. The U.S. researchers became quite convinced that the area contaminated was much smaller than Medvedev had estimated — rumor mills often exaggerate such figures. None of the accidents that seemed even remotely possible to the American researchers could have administered radiation doses approaching a lethal range. It is therefore highly dubious whether there were any casualties from radiation. Moreover, there is nothing in the Russian scientific literature on medical effects of radiation to suggest that any had occurred in this incident.

The most definite conclusion of both U.S. studies was that no such accident could possibly occur with American waste-handling procedures — answering that question was the prime reasons for these studies. None of the ingredients that could lead to either a nuclear or a chemical explosion is present in the U.S. technology.

As a result of their new Glasnost policy, the Soviet Union released a report16 on the incident in 1989. It confirmed that on September 29, 1957, there had been a chemical explosion of the type conjectured by the U.S. studies in a high-level waste storage tank. The area treated as contaminated was about 400 square miles. About 600 people were evacuated in the first few weeks, and the number eventually grew to 10,000. There were no casualties from the explosion, and medical and epidemiological studies over the past 30 years indicate no excess mortality rates or excess disease incidence among those exposed, and no genetic effects in their progeny. Nearly all of the land has now been restored for use, mostly as farms. The report emphasizes that experience and data from the Kyshtym incident have been very useful in dealing with consequences of the Chernobyl accident

Opponents of nuclear power in the United States used the Kyshtym incident as an indication of the hazards of radioactive waste. However, it is important to emphasize that the technology that led to the chemical explosion in Kyshtym has never been used in the United States because that danger was recognized. Such an accident, therefore, most definitely cannot happen here.


This chapter and the previous one have presented detailed scientific analyses of the radioactive waste problem. The principal results of these analyses are contained in Table 1, with additional information in the section on waste transport. They show that the hazards of nuclear waste are thousands of times lower than those of wastes from other technologies. However, the American public has been badly misinformed about these dangers and has become deeply concerned about them. As a result, billions of dollars are being wasted, but fortunately, even with all this unnecessary spending, the cost of radioactive waste management increases the price of nuclear electricity by only about one percent.


We have stated that by far the most important radiation health effect of nuclear power is the lives saved by mining uranium out of the ground, thereby reducing the exposure of future generations to radon. Let us trace through the process by which this is calculated.2

The radon to which we are now exposed comes from the uranium and its decay products in the top 1 meter of U.S. soil, since anything percolating up from deeper regions will decay before reaching the surface. From the quantity of uranium in soil (2.7 parts per million) and the land area of the United States (contiguous 48 states), it is straightforward to calculate that there are 66 million tons of uranium in the top meter of U.S. soil. This is now causing something like 10,000 deaths per year from radon, and will continue to do so for about 22,000 years, the time before it erodes away. This is a total of (10,000 x 22,000 =) 220 million deaths caused by 66 million tons of uranium, or 3.3 deaths per ton. As erosion continues, all uranium in the ground will eventually have its 22,000 years in the top meter of U.S. soil and will hence cause 3.3 deaths per ton.

In obtaining fuel for one nuclear power plant to operate for 1 year, 180 tons of uranium is mined out of the ground. This action may therefore be expected to avert (180 x 3.3 =) 600 deaths. However, to be consistent we must also consider erosion of the mill tailings covers. When this is taken into account, the net long-term effect of mining and milling is to save 420 lives.

On the short-term, 500-year perspective, we can ignore erosion, so uranium deep underground has no effect. However, about half of our uranium ore is surface mined, and about 1% of this (0.005 x 180 =) 0.9 tons is taken from the top 1 meter. This saves (0.9 x 3.6 =) 3.3 lives over the next 22,000 years, or (500/22,-000 x 3.3 =) 0.07 lives over the next 500 years. We must subtract from this the lives lost by radon emission through the covered mill tailings, 0.003/year without the cover, x 1/20 to account for attenuation by the cover, x 500 years = 0.07. Thus the net effect over the next 500 years of mining and milling uranium to fuel one power plant for 1 year is to save (0.07 - 0.07 =) 0.00 lives. That is the result used in Table 1 in this Chapter.

We next consider the effects of radon from coal burning. Coal contains an average of 1.0 parts per million uranium — some commercial coals contain up to 40 parts per million. When the coal is burned, by one way or another this uranium is released and settles into the top layers of the ground, where it will eventually cause 3.3 deaths per ton with its radon emissions, as shown above. The 3.3 million tons of coal burned each year by a large power plant releases 3.3 tons of uranium, which is then expected to cause 3.3 deaths per ton, or a total of 11 deaths. These will be distributed over about 100,000 years, with some tendency for more of them to occur earlier; thus about 1% of them, or 0.11 deaths, may be expected in the next 500 years. That is the figure in Table 1 in this chapter.

On a multimillion-year time perspective, the uranium in the coal would have eventually reached the surface by erosion even if the coal had not been mined; these 11 deaths would therefore have occurred eventually anyhow, and hence can be discounted. However, the carbon, the main constituent of coal, is burned away, whereas if the coal had been left in the ground and eventually reached the surface by erosion, this carbon would not have emitted radon since it contains no uranium (the uranium in the coal has already been accounted for). Since that carbon is missing, its time near the surface will be taken by other rock which does contain uranium, 2.7 parts per million on an average, or (3.3 x 2.7 =) 9 tons in the 3.3 million tons of rock which replaces the carbon. Multiplying this by 3.3 deaths per ton of uranium gives 30 deaths, the result we have used in Table 1, in this chapter.

[next chapter]