Not to derail the thread with stats or anything, but macpacheco might be interested in a table compiled by Brian Wang: Deaths-per-TWh By Energy Source:
Also see James Conca's How Deadly is Your Kilowatt?
Energy Source Death Rate (deaths per TWh)
Coal (elect, heat,cook –world avg) 100 (26% of world energy, 50% of electricity)
Coal electricity – world avg 60 (26% of world energy, 50% of electricity)
Coal (elect,heat,cook)– China 170
Coal electricity- China 90
Coal – USA 15
Oil 36 (36% of world energy)
Natural Gas 4 (21% of world energy)
Solar (rooftop) 0.44 (0.2% of world energy for all solar)
Wind 0.15 (1.6% of world energy)
Hydro 0.10 (europe death rate, 2.2% of world energy)
Hydro - world including Banqiao) 1.4 (about 2500 TWh/yr and 171,000 Banqiao dead)
Nuclear 0.04 (5.9% of world energy)
As an aside, the 0.04 nuclear deaths /TWh assumes LNT (linear, no threshold) holds and that about 4,000 -- 9,000 people will eventually die over 50 years from increased cancers -- 80 - 180/yr. However, this LNT extrapolation goes right to zero, and includes people who received or will receive lifetime radiation exposure from the Chernobyl accident less than that of lifetime background radiation in many parts of the world (Dr. Zbigniew Jaworowski 2006, reproduced in Chernobyl,The Fear of the Unknown. Flame alert: Atomic Insights is not a universally popular blog. Nonetheless, in 2012 the United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR 2012) submitted the report that, among other things, states that uncertainties at low doses are such that UNSCEAR “does not recommend multiplying low doses by large numbers of individuals to estimate numbers of radiation-induced health effects within a population exposed to incremental doses at levels equivalent to or below natural background levels.” (UNDOC/V1255385). In other words, the 4,000 - 9,000 Chernobyl death guestimate was arrived via invalid application of LNT. (From Radiation Is Not A Big Deal - UNSCEAR (a title I personally think could benefit from some of the qualification Conca later provides in the body of the article).
From RERF's Views on Residual Radiation (Radiation Effects Research Foundation, 8 December 2012):
Another oft-overlooked resource (that also does no support popular perception) is Prof Cohen's chapter How Dangerous is Radiation?.
"In terms of effects from radiation exposure immediately after the accident, results of measurements conducted so far by the government of Fukushima Prefecture on tens of thousands of people have shown that the committed dose is less than 1 mSv in more than 99.9% of such people, and the maximum dose observed in this group is as low as the global average background radiation level (2.4 mSv a year)."
Don't get me wrong: in large doses radiation can be deadly and safeguards must be maintained. Some 9,000 people die from skin cancers in the United States alone, each year. (Yet PV remains unfathomably popular </irony>.)
Those who insist that nuclear power be demonstrably 100% safe hold that technology to a far higher standard than any realistically possible alternative. Though cancer is particularly unpleasant, so are the respiratory, hormonal, and immune failures associated with coal, gas, and biofuels. After a certain point a death is a death. How many billions will unpleasantly die as consequence of global warming? How might those be averted?