Chart: The Deadliest Energy Sources in the World

Read time: 2 mins
Deaths per terawatt hour by energy source

How deadly is your energy source? The very real and lethal effects of our global energy choices become clear in this interactive data visualization, showing the death rate, as measured by the number of deaths per terawatt hour (TWh), for each of the major global energy sources, e.g., coal, natural gas, oil, nuclear, hydro, peat, and biomass. Take a closer look at the chart here:

Chart of deaths per terawatt per energy source

Click on the image to view the full size.

The left column displays the percentage of world energy production for each energy source while on the right you can see the corresponding number of attributed deaths per terawatt hour.

To give some perspective to the value of a terawatt hour: “All wind turbines installed by the end of 2009 worldwide are generating 340 TWh per [year], equivalent to the total electricity demand of Italy, the seventh largest economy of the world, and equaling 2 percent of global electricity consumption,” according to a report the World Wind Energy Association published last year (PDF). Meanwhile, wind energy’s cumulative death rate is 0.01 deaths/TWh (.xls file), based on 44 deaths since 1975, which Paul Gipe, author of Wind Energy Comes of Age, calculated in 2010.

This data visualization comes via IBM’s research site, Many Eyes, where you can click through to interact with the data. It doesn’t show here the data for solar (rooftop’s 0.44 deaths/TWh) or wind, which the original source has claimed both have a higher number of deaths per TWh than nuclear (0.04 deaths/TWh). However, this doesn’t take into account the most recent calculations for wind energy (0.01 deaths/TWh), making wind – and not nuclear – the least deadly energy source based on these metrics.

Coal, unsurprisingly, comes out the deadliest energy source, due to the clear health impacts of air particulate pollution. All the more reason the Environmental Protection Agency should be strictly regulating coal power’s many hazardous air pollutants.

Nuclear energy’s low death rate may seem somewhat surprising because of its pervasively toxic stigma (Homer Simpson working at a nuclear plant controls? – radiation! meltdown! Chernobyl!). However, the low rate here is likely due in part to the fact that nuclear supplies a larger chunk of energy worldwide in comparison to say, solar, and that deaths associated with nuclear energy are harder to link directly because the adverse and fatal effects can develop over a much longer period of time (e.g., cancer). Deaths from solar (rooftop, anyway) tend to be from falling accidents during installation and maintenance on roofs and shouldn’t be dismissed in the push for clean energy because these deaths are easily preventable.

Hat tip to GOOD for highlighting the chart.

Image: Many Eyes

Get DeSmog News and Alerts


Brian Wang did account for the long term Chernobyl deaths in his calculation, which leads to the 0.04 (or 0.037) deaths per TeraWattHour in the figure.

“Averaging about 2100 TWh from 1985-2005 or a total of 42,000 TWh. So those 50 deaths would be 0.0012 deaths/TWh. If those possible 4000 deaths occur over the next 25 years, then with 2800 TWh being assumed average for 2005 through 2030, then it would be 4000 deaths over 112,000 TWh generated over 45 years or 0.037 deaths/TWh.”

I guess one might claim there will be more deaths from cancer from nuclear accidents (GreenPeace certainly do claim there is more). And one could adjust accordingly. On the other hand, as more modern, ever safer, reactors are built, and older reactors retired, the chance of an accident having serious consequences diminishes, and the deaths per unit power would decrease. It’s unclear to me that the same could happen with wind: the world is just starting installing wind in riskier locations, such as offshore.

One interesting issue is that as renewables’ penetration into the grid increases, their capacity factors may decrease because of the best sites having been taken. This would result in Deaths/TWh delivered increasing, other things being held equal. Another issue is that, even given a constant capacity factor, a system may require that more units be installed to maintain a certain base level of production (the market would need to be oversupplied), leading to more deaths. Alternatively, as wind power’s proportion increases, the system may require that turbines deliberately reduce their output (lowering their capacity factor). This would likely increase Deaths/TWh. These last two issues are all tied up with how the “smart grid” and demand management evolve, but they will inevitably have an effect on the death rate.


the data behind this is poorly researched and somewhat questionable. I commented on this also at Barry Brook’̈́s site. E.g., if you look at the original site from which most of this is coming

you find that nuclear is rated at 0.04 deaths per TWh, but when you look at the
grey table given there – taken from an IAEA publication – you read

Occupational fatalities: 0.04
Public fatalities: 0.65

(And I suspect the 0.65 is based on the widely-circulated, but very conservative
4000 Chernobyl figure). So, your source doesn’t tell the whole story…

Then, the referenced ExternE 1995 paper that is given as the source for these numbers I haven’t been able to find… the ExternE paper that is linked to, is a different one, from 2001.

This is the digging I have done so far… now it’s your turn ;-)

Mark Z. Jacobson”s peer review paper rated nuclear as safer than coal but much more dangerous than wind, solar, hydro or geothermal or even oil or natural gas. He included the risks of nuclear proliferation in this weighing, and how much blame civilian nuclear deserves for nuclear proliferation is controversial. (Would Pakistan, India and Israel really not have sought and obtained nuclear weapons if they did not have civlian reactors, given great power sponsorship of their nuclear weapons programs?) Even that aside, nuclear still ends up more dangerous than solar, geothermal, wind and hydro and wave. But if you discount proliferation dangers, fewer fatalities than oil, and gas. With or with out proliferation, nuclear is less dangerous than coal. And ethanol is apparently worse than everything.

Jacobson, Mark Z. 2009. “Review of solutions to global warming, air pollution, and energy security.” Energy and Environmental Science 2009(2):148-173.

(Note: actually without nuclear proliferation, nukes match coal with CCS, that is coal without climate effects. But since the amount of CCS currently in use is a rounding error, we can say that actually existing coal rates much worse than nukes with or without proliferation in Jacobson”s study. Again only one study, but unlike the sources in this post, peer reviewed.

Gar Lipow, thanks… I knew about Jacobson but had forgotten, good reference.

About proliferation, yes one can argue about how much of a difference civil nuclear programmes make. But I’m not convinced by the disinterestedness of much of the argumentation I’ve seen.

E.g., if civilian nuclear tech doesn’t help proliferation, then why do folks oppose Iran getting enrichment technology, which they claim is for civilian use only? And I remember the story of how Pakistan got hold of the Dutch ultracentrifuge secrets, c/a a soft-spoken young researcher named Khan.

And several other countries got, or almost got, from Pakistan. So I would say it matters.

Yes, on proliferation Im not convinced either. But I think the argument is good enough that it is worth weighting things both ways. Incidentally am I the only whose apostrophe key does not work on this site? Works fine everywhere else.

I replied to Martins comment, listing a peer review source that put nuclear ahead in fatalities of many of the sources the chart says it is safer than. Im guessing that because I put in a link, it put the comment into moderation.

BTW I cant use apostrophes on this site, but they work everywhere else. Is that just me, or that general to this site. Ive read it for a long time, but seldom commented on a post.