5 ms·

Nuclear power has the lowest deaths per unit of power of all commercial power generation methods [1]. The externalities of nuclear power are highly visible and

Nuclear power has the lowest deaths per unit of power of all commercial power generation methods [1]. The externalities of nuclear power are highly visible and concentrated in a small physical and temporal range. This makes them obvious and scary. The externalities of other power generation methods are diffused through time and space, to the point that they appear mundane if they are noticed at all. Coal power kills 1000x as many people per unit as nuclear. Natural gas kills 40x as many. Solar kills 4x as many. Humans have shown themselves to consistently be incredibly bad at comparing chronic and acute risks at scale. 1: https://www.engineering.com/DesignSoftware/DesignSoftwareArticles/ArticleID/21308/Whats-the-Death-Toll-of-Nuclear-vs-Other-Energy-Sources.aspx https://www.engineering.com/DesignSoftware/DesignSoftwareArt...

pfdietz 4y ago
> Solar kills 4x as many. I dispute this number. Are you considering roof mounted residential solar? Most solar in a solar-powered world will be on the ground.
- opo 4y ago
  This does likely include roof mounted residential solar. >Most solar in a solar-powered world will be on the ground. It is probably more accurate to say that most solar in a solar-powered world should be on the ground. The reality is that roof-top solar is the most expensive form of power in the world and yet it continues to be heavily subsidized. These subsidies are in the form of direct tax breaks and usually higher electricity prices paid by those who can't install it. A dollar wasted on roof-top solar would be MUCH better spent on solar panels installed on the ground by the utility.
- kwere 4y ago
  solar panels arent born in nature and dont degrade biologically. also consider output by scale. most of heavy production occur in xinjiang where there isnt a safety culture, installations are anothoer issue. btw the data on solar is understimated for sure due to chinese data quality https://www.statista.com/statistics/494425/death-rate-worldwide-by-energy-source/ https://www.statista.com/statistics/494425/death-rate-worldw...
jmyeet 4y ago
> Nuclear power has the lowest deaths per unit of power of all commercial power generation methods This is a poor metric to use. The official death count from Chernobyl was 31. It may be as high as 50. There are probably more deaths attributable through long-term effects, etc but you start getting into subjective modeling to figure out a number for that. It's almost 40 years later and the absolute exclusion zone is still 1,000 square miles. Treating this as only 31 (or 50) deaths grossly under-represents the magnitude of the disaster.
- pfdietz 4y ago
  Chernobyl is estimated to cause 27,000 fatal cancers in the larger population. You can't dismiss these just because they can't be directly demonstrated. Technology regulation is not criminal law; the technology doesn't have to be shown guilty beyond reasonable doubt.
  - jmyeet 4y ago
    You realize that supports my argument, right? The comment I was replying to was arguing nuclear power caused the least deaths and I was explaining how that's a bad metric because it doesn't capture the impact and damage of the absolute exclusion zone. So if the actual death count is 27,000 instead of 31 or 50 that's actually much worse, which further undermines that commenter's argument.
    - pfdietz 4y ago
      I should have written "one can't" rather than "you can't". Sorry that I implied something personal.
- wskinner 4y ago
  This fairly thorough analysis comes up with a higher estimate of around 300-500 [1]. Of course, deaths per unit of energy is not the only meaningful metric, and it ignores the environmental impact that you mention. But have you considered the environmental impact of using fossil fuels for energy? 1: https://ourworldindata.org/what-was-the-death-toll-from-chernobyl-and-fukushima https://ourworldindata.org/what-was-the-death-toll-from-cher...

pfdietz 4y ago

> Solar kills 4x as many. I dispute this number. Are you considering roof mounted residential solar? Most solar in a solar-powered world will be on the ground.

opo 4y ago
This does likely include roof mounted residential solar. >Most solar in a solar-powered world will be on the ground. It is probably more accurate to say that most solar in a solar-powered world should be on the ground. The reality is that roof-top solar is the most expensive form of power in the world and yet it continues to be heavily subsidized. These subsidies are in the form of direct tax breaks and usually higher electricity prices paid by those who can't install it. A dollar wasted on roof-top solar would be MUCH better spent on solar panels installed on the ground by the utility.
kwere 4y ago
solar panels arent born in nature and dont degrade biologically. also consider output by scale. most of heavy production occur in xinjiang where there isnt a safety culture, installations are anothoer issue. btw the data on solar is understimated for sure due to chinese data quality https://www.statista.com/statistics/494425/death-rate-worldwide-by-energy-source/ https://www.statista.com/statistics/494425/death-rate-worldw...

jmyeet 4y ago

> Nuclear power has the lowest deaths per unit of power of all commercial power generation methods This is a poor metric to use. The official death count from Chernobyl was 31. It may be as high as 50. There are probably more deaths attributable through long-term effects, etc but you start getting into subjective modeling to figure out a number for that. It's almost 40 years later and the absolute exclusion zone is still 1,000 square miles. Treating this as only 31 (or 50) deaths grossly under-represents the magnitude of the disaster.

pfdietz 4y ago
Chernobyl is estimated to cause 27,000 fatal cancers in the larger population. You can't dismiss these just because they can't be directly demonstrated. Technology regulation is not criminal law; the technology doesn't have to be shown guilty beyond reasonable doubt.
- jmyeet 4y ago
  You realize that supports my argument, right? The comment I was replying to was arguing nuclear power caused the least deaths and I was explaining how that's a bad metric because it doesn't capture the impact and damage of the absolute exclusion zone. So if the actual death count is 27,000 instead of 31 or 50 that's actually much worse, which further undermines that commenter's argument.
  - pfdietz 4y ago
    I should have written "one can't" rather than "you can't". Sorry that I implied something personal.
wskinner 4y ago
This fairly thorough analysis comes up with a higher estimate of around 300-500 [1]. Of course, deaths per unit of energy is not the only meaningful metric, and it ignores the environmental impact that you mention. But have you considered the environmental impact of using fossil fuels for energy? 1: https://ourworldindata.org/what-was-the-death-toll-from-chernobyl-and-fukushima https://ourworldindata.org/what-was-the-death-toll-from-cher...