Radioactive water is leaking into the sea, there’s a little plutonium in the soil, and traces of nuclear fallout have been detected in places as far apart as Kuwait and Maryland. In a few parts of Japan, you’re also not supposed to eat the broccoli or the beef.

The effects of the accident at the Fukushima Daiichi nuclear plant grow by the week, creating a lengthening catalogue of worries and proving once again that nuclear power frightens people as few other technologies do.

But when the dead and sickened are added up, how dangerous is it really?

The partial meltdown in Japan has injured 23 people and exposed as many as 21 to levels of radiation higher than is considered safe to receive in one year. Two workers are still missing but are assumed to have been killed by the earthquake or tsunami, not the nuclear accident. No people in the “plume zone” outside the plant have been contaminated to a degree that is expected to affect their health, based on radiation readings so far.

In the months after the world’s worst nuclear disaster, in Chernobyl in 1986, about 50 people died. In the next-biggest accident, at Three Mile Island in 1979, no one did.

History suggests that nuclear power rarely kills and causes little illness. That’s also the conclusion engineers reach when they model scenarios for thousands of potential accidents.

Making electricity from nuclear power turns out to be far less damaging to human health than making it from coal, oil or even clean-burning natural gas, according to numerous analyses. That’s even more true if the predicted effects of climate change are thrown in.

Compared with nuclear power, coal is responsible for five times as many worker deaths from accidents, 470 times as many deaths due to air pollution among members of the public, and more than 1,000 times as many cases of serious illness, according to a study of the health effects of electricity generation in Europe.

“The costs of fossil fuels come out quite high, while the costs for nuclear generally come out low,” said Anil Markandya, an economist at the University of Bath in England and scientific director of the Basque Centre for Climate Change in Spain, who co-authored the study published in the Lancet in 2007.

Even in the wake of the Fuku
shima Daiichi disaster, Markandya and many others who have done similar work can’t imagine a situation — a realistic one, that is — in which the health cost of nuclear power would equal that of coal.

Or even come close.

The hidden costs

About half of the electricity in the United States is made with coal-fired plants and about one-fifth with nuclear power. Many experts think there is an urgent need to determine what role nuclear power should play in feeding America’s energy-hungry future.

To inform that discussion, economists, engineers and epidemiologists have teamed up to determine the full economic, health, social and environmental consequences of generating electricity with various fuels. Most of this work has been done in Europe, where the acceptability of nuclear power, and the fraction of electricity generated with it, differs greatly among nations of the European Union.

The goal is to capture not only the costs reflected on a person’s monthly utility bill but the many hidden ones borne by individuals, communities and governments. In this way, analysts seek out the “impact pathway” of each fuel — every effect it has, direct and indirect.

For power plants (and also hydroelectric dams and wind farms), this includes the land to site them; construction, operation and decommissioning costs; and the humans who are killed or injured along the way. That means accidents and black lung disease in coal miners; radiation exposure in uranium miners and millers; and deaths and burns in oil-rig fires.

The impact pathway also includes what happens to the public — collisions with coal trains; asthma, respiratory disease and heart attacks caused by smokestack soot and gases; and emissions’ effects on agricultural production.

Health consequences are measured two ways.

Occupational deaths in mines, oil rigs or power plants are counted directly. Death and illness in the public is determined by epidemiological studies, such as ones estimating the fraction of hospital admissions for emphysema that can be attributed to air pollution. Those impacts are then given a monetary cost that is added to the price tag of a kilowatt hour of electricity. (The cost is the value of a life lost by premature death, or diminished by illness, that economists use in other analyses.)

The calculations can be very fine.

In “Full cost accounting for the life cycle of coal,” published this year by a team of 12 researchers led by Paul R. Epstein of Harvard Medical School’s Center for Health and the Global Environment, the ledger included .02 cents per kilowatt hour for mental retardation caused by mercury in coal-plant emissions.

Using similar methods, Markandya and his co-author in the Lancet study, Paul Wilkinson of the London School of Hygiene and Tropical Medicine, found that in Europe coal is responsible for .12 deaths from accidents, 25 deaths from pollution and 225 cases of serious illness per terawatt (1,000 billion kilowatt) hour of electricity generated. In comparison, nuclear causes .02 accidental deaths, .05 pollution deaths and .22 cases of illness.

This human health cost is much higher in some parts of the world than others.

It’s especially high in China, where three-quarters of the electricity is made by burning coal, mining accidents kill about 6,000 people a year, and hundreds of millions of people are affected by air pollution. In some inland cities, the economic cost to human health of making electricity from coal is as much as seven times higher than the cost of generating the electricity, according to a calculation by Stefan Hirschberg at the Paul Scherrer Institutin Switzerland, which has done energy system analysis for the European Commission.

Nuclear power’s advantage over fossil fuels is even more dramatic when carbon dioxide emissions are considered.

Many experts think greenhouse gases are a future threat to health. Some say the threat is already here, and point to 30,000 heat-related deaths in Europe in August 2003 as evidence. Coal produces 1,290 grams of CO 2 per kilowatt hour in direct (smokestack) and indirect (mining, transport) emissions, while nuclear produces 30, according to the Lancet study.

Built into the calculations are the consequences of what are called “beyond-design” nuclear accidents — events similar to what is underway in Japan. However, there aren’t enough big nuclear plant accidents to provide a statistically meaningful estimate of their frequency, effects and costs. According to a database compiled by the Paul Scherrer Institut, from 1970 to 2008 there were 1,686 accidents in the coal industry, 531 in the oil industry and 186 involving natural gas in which five or more people died. There was just one such nuclear accident — at Chernobyl 25 years ago this month.

To better estimate the potential impact of nuclear catastrophes, analysts break down plant operations into thousands of different actions and then estimate the probabilities of hypothetical accident sequences. Hirschberg and his colleagues used a Swiss nuclear plant to come up with such an estimate. They calculated that nuclear accidents in Europe can be expected to cost .007 lives per gigawatt year (1 million kilowatt years), compared with .12 lives for coal, .02 lives for oil and .06 for natural gas.

Radiation’s toll

There is also much uncertainty about how many people might be harmed by a big nuclear accident.

At Chernobyl, two people died during the accident and 28 others died of radiation illness in the first four months afterward. (Some estimates of the early deaths put the number as high as 57 ).

Since then, there have been 6,800 cases of thyroid cancer in people who were children at the time of the accident, according to a recent report by the U.N. Scientific Committee on the Effects of Atomic Radiation, with the number still rising. As of 2005, only 15 were fatal.

To date, there is no clear increase in leukemia or other cancers, or deaths from non-cancer diseases. However, various expert groups estimate that 4,000 to 33,000 premature deaths might occur as a consequence of the accident.

In general, the hazards of radiation are less than most people think.

Since 1950, Japanese and American researchers have followed 120,000 residents of Hiroshima and Nagasaki, the cities on which the United States dropped atomic bombs in 1945 to end World War II. Three-quarters of the people in the Life Span Study were exposed to the blasts; one-quarter were away at the time. The number of deaths attributable to the bombs is estimated by comparing survival in the two groups.

Through 2000, 42,304 of the people in the study had died. Of those deaths, 822 were “excess” — probably a result of the radiation.

Nuclear’s ‘dread factor’

Many critics of nuclear power say none of this truly accounts for the technology’s hazards.

“To replace carbon pollution with radioactive pollution is not a healthy solution,” said Epstein, the Harvard physician. “Even if the events are rare, what’s happening now in Japan demonstrates how profound and long-lasting these impacts can be.”

At a recent briefing by Physicians for Social Responsibility, David Richardson, an epidemiologist from the University of North Carolina, said that “the unsolved problems of long-term storage and its contribution to nuclear proliferation” are two reasons besides accidents that make nuclear power unacceptable.

Future accidents at storage sites are considered by energy analysts. But because modeling suggests they’re improbable, they don’t affect the calculations much. Mental-health effects of nuclear accidents are part of the calculations, too, but the doomsday fear of them and threat from nuclear proliferation are not.

“There is a kind of dread factor for nuclear which is very hard to quantify,” Markandya said. He added after a pause, “In the end . . . if people feel really uncomfortable with nuclear power, then they ought to go against it.”