4 ms·

I'm in the anti-nuclear camp. Mainly because I think the inertia around current consumption patterns is too strong to overcome, so adding a lot of nuclear to th

I'm in the anti-nuclear camp. Mainly because I think the inertia around current consumption patterns is too strong to overcome, so adding a lot of nuclear to the mix would make things a lot more dangerous without making things much better. This is how I think about the danger of nuclear power. Imagine a really safe nuclear power plant. How often will it fail? Let's say, once every 100,000 years. Assuming this probability is uniformly distributed over the 100,000 year time period, it has a 1/100,000 chance of failing in any given year. A failure seems pretty unlikely. But, there are something like 450 operating nuclear power plants in the world. The chance of a single power plant failing in a year is then 1 - (99,999/100,000)^450 which is around 0.5%. That's starting to look a lot more likely now. After 30 years, there is a 78% chance of a nuclear power plant failing somewhere on Earth. Now what's even more interesting is if you look back on historic nuclear power accidents at INES level 6 or higher. 1957 - Kyshtym disaster, 1986 - Chernobyl, 2011 - Fukushima. And of course there have been other close calls (e.g. Three Mile Island). The cadence of these accidents seems to match the data from the thought experiment above.

chrismartin 4y ago
Current consumption patterns didn't exist a century ago. Some alternatives to petroleum-powered consumption are already cheaper and more convenient than their petroleum-powered equivalents. Today landscaping tools and heat pumps, tomorrow cars and trucks, later ships and construction equipment, and maybe someday airliners. That combined with regulatory carrots and sticks will change consumption patterns. A century ago, people in cities burned a LOT of coal for heating. Air pollution in urban areas was pretty awful. We've changed consumption patterns before and we can do it again. Imagine that we decarbonize most of our economy and mostly stop accelerating climate change, with a combination of renewables and nuclear. But somewhere in the world, a nuclear reactor melts down every 5 years. Sometimes it's like TMI and sometimes it's like Chernobyl. Is that better overall than destabilizing Earth's climate?
- mwint 4y ago
  Yes, without a doubt. Other than messing up a bunch of land, Chernobyl wasn’t even that terrible outside of a certain radius.
- atwood22 4y ago
  > Imagine that we decarbonize most of our economy and mostly stop accelerating climate change, with a combination of renewables and nuclear. But somewhere in the world, a nuclear reactor will melts down every 5 years. Sometimes it's like TMI and sometimes it's like Chernobyl. Probably it would be better to de-carbonize the economy. But the global economy is literally a tragedy of the commons. There's really no hope of massive collective action, even has humanity faces greater calamity. My point is that building a bunch of nuclear power will make it more likely that a Chernobyl happens every 5 years, but it won't have a large impact on climate change.
  - chrismartin 4y ago
    As long as we need to fill the gaps around supply of renewables, we will need to build either nuclear or fossil fuel plants. Imagine a new 1 GW nuclear plant compared to a new 1 GW natural gas plant. Which is better for the world on the margin? Also, the trend I perceive is that as societies develop, they gain both desire and ability to make the commons less tragic. China is the world's largest carbon emitter yet somehow their strategy to decarbonize feels more credible than the USA's (to the extent that we even have one at all).
    - atwood22 4y ago
      It doesn’t seem wise to vastly increase risk just for marginal capacity when we are already vastly exceeding our carbon budget.
    - Schroedingersat 4y ago
      > As long as we need to fill the gaps around supply of renewables, There are several assumptions hidden here. One is that it's a choice between building new renewable and fossil fuel plants vs. new nuclear. The real choice is between moving some fossil fuels to renewable immediately, or leaving fossil fuels online while a nuclear plant gets delayed 5 times. The second is the assumption that all capacity to consume energy much be fulfilled and must remain constant. We can turn down the aluminium smelter. We can smelt steel with electricity. We can use renewables to drive chemical reactions. We can make fertilizer with electricity. Noone dies if we stop doing these things sometimes. The only downside is it costs a little more. But this is the downside of nuclear anyway, so where's the problem? We can also time shift energy with things like thermochemical storage. A huge amount of emissions come from burning natural gas for heat. Thermochemical storage is a fraction of the cost of batteries and is compact enough to store a year's worth of energy.
- throwaway2037 4y ago
  Mister Martin: Another great post! Thank you to raise the idea of change. When I talk with people from developing countries, they frequently despair about the state of their environment, including air, soil, and water pollution. I tell them that Japan, UK, Germany, and the United States was a dirty mess in the 1960s and 1970s. Then, all of them awoke to environmentalism. Today, these are some of the cleanest countries in the world. In particular, Japan and German are still major industrial power houses (steel, chemical, manuf, etc.). Their air, soil, and water pollution declined dramatically in the last 50 years. Yes, sadly, some of this pollution has been exported to poor(er), developing countries. However, a lot of industry remains onshore and people and companies are fully committed to a clean environment. Really, I cannot stress enough how large are chemical companies in Japan. When you fly into Haneda (Tokyo) Airport, it seems like a 50km continuous line of chemical plants on the waterfront in Kanagawa-ken! The environment regulations are incredibly strict. Places that were an environmental disaster 50 years ago are now a "factory neighborhood", but no longer a polluted mess. Recovery and change is possible! For other readers, I assume "TMI" means "Three Mile Island" nuclear accident. You can read more about it on Wikipedia.
throwawaylinux 4y ago
> I'm in the anti-nuclear camp. Mainly because I think the inertia around current consumption patterns is too strong to overcome, so adding a lot of nuclear to the mix would make things a lot more dangerous without making things much better. Adding lots of nuclear would vastly reduce greenhouse gas emissions, which are the single biggest threat to humanity and life on earth. > This is how I think about the danger of nuclear power. Imagine a really safe nuclear power plant. How often will it fail? Let's say, once every 100,000 years. Assuming this probability is uniformly distributed over the 100,000 year time period, it has a 1/100,000 chance of failing in any given year. A failure seems pretty unlikely. But, there are something like 450 operating nuclear power plants in the world. The chance of a single power plant failing in a year is then 1 - (99,999/100,000)^450 which is around 0.5%. That's starting to look a lot more likely now. After 30 years, there is a 78% chance of a nuclear power plant failing somewhere on Earth. > Now what's even more interesting is if you look back on historic nuclear power accidents at INES level 6 or higher. 1957 - Kyshtym disaster, 1986 - Chernobyl, 2011 - Fukushima. And of course there have been other close calls (e.g. Three Mile Island). The cadence of these accidents seems to match the data from the thought experiment above. Yes nuclear power plants fail at some rate. Contrary to the pro-fossil-fuel / green propaganda, even quite serious failures are just not very serious in the scheme of things. A single coal mine collapse will easily kill many more people than all the direct deaths attributed to nuclear accidents combined. As for indirect, coal mines (EDIT: that should be coal power stations) will pump out chemicals and radiation that kill vast numbers of people _when they are operating normally without any failure_. Other fossil fuel plants less bad than coal, but still very damaging. EDIT: references - https://www.hsph.harvard.edu/c-change/news/fossil-fuel-air-pollution-responsible-for-1-in-5-deaths-worldwide/ https://www.hsph.harvard.edu/c-change/news/fossil-fuel-air-p..., https://cen.acs.org/articles/91/web/2013/04/Nuclear-Power-Prevents-Deaths-Causes.html https://cen.acs.org/articles/91/web/2013/04/Nuclear-Power-Pr...
- Schroedingersat 4y ago
  > Adding lots of nuclear would vastly reduce greenhouse gas emissions, which are the single biggest threat to humanity and life on earth. Does it though, or does it just distract from real solutions and prop up coal and gas for another decade? Let's say there's a trillion dollars to spend today. What is the net reduction in emissions between now and 2040 achieved by spending it on nuclear (hint: if the project takes as long as most in the west do it's negative)? What is the net reduction achieved in the same timespan by spending it on renewables? Even if there weren't a bunch of other problems (and there are) it's like claiming you need to stop and build a new engine so your car can go faster on the last lap of a race.

chrismartin 4y ago

Current consumption patterns didn't exist a century ago. Some alternatives to petroleum-powered consumption are already cheaper and more convenient than their petroleum-powered equivalents. Today landscaping tools and heat pumps, tomorrow cars and trucks, later ships and construction equipment, and maybe someday airliners. That combined with regulatory carrots and sticks will change consumption patterns. A century ago, people in cities burned a LOT of coal for heating. Air pollution in urban areas was pretty awful. We've changed consumption patterns before and we can do it again. Imagine that we decarbonize most of our economy and mostly stop accelerating climate change, with a combination of renewables and nuclear. But somewhere in the world, a nuclear reactor melts down every 5 years. Sometimes it's like TMI and sometimes it's like Chernobyl. Is that better overall than destabilizing Earth's climate?

mwint 4y ago
Yes, without a doubt. Other than messing up a bunch of land, Chernobyl wasn’t even that terrible outside of a certain radius.
atwood22 4y ago
> Imagine that we decarbonize most of our economy and mostly stop accelerating climate change, with a combination of renewables and nuclear. But somewhere in the world, a nuclear reactor will melts down every 5 years. Sometimes it's like TMI and sometimes it's like Chernobyl. Probably it would be better to de-carbonize the economy. But the global economy is literally a tragedy of the commons. There's really no hope of massive collective action, even has humanity faces greater calamity. My point is that building a bunch of nuclear power will make it more likely that a Chernobyl happens every 5 years, but it won't have a large impact on climate change.
- chrismartin 4y ago
  As long as we need to fill the gaps around supply of renewables, we will need to build either nuclear or fossil fuel plants. Imagine a new 1 GW nuclear plant compared to a new 1 GW natural gas plant. Which is better for the world on the margin? Also, the trend I perceive is that as societies develop, they gain both desire and ability to make the commons less tragic. China is the world's largest carbon emitter yet somehow their strategy to decarbonize feels more credible than the USA's (to the extent that we even have one at all).
  - atwood22 4y ago
    It doesn’t seem wise to vastly increase risk just for marginal capacity when we are already vastly exceeding our carbon budget.
  - Schroedingersat 4y ago
    > As long as we need to fill the gaps around supply of renewables, There are several assumptions hidden here. One is that it's a choice between building new renewable and fossil fuel plants vs. new nuclear. The real choice is between moving some fossil fuels to renewable immediately, or leaving fossil fuels online while a nuclear plant gets delayed 5 times. The second is the assumption that all capacity to consume energy much be fulfilled and must remain constant. We can turn down the aluminium smelter. We can smelt steel with electricity. We can use renewables to drive chemical reactions. We can make fertilizer with electricity. Noone dies if we stop doing these things sometimes. The only downside is it costs a little more. But this is the downside of nuclear anyway, so where's the problem? We can also time shift energy with things like thermochemical storage. A huge amount of emissions come from burning natural gas for heat. Thermochemical storage is a fraction of the cost of batteries and is compact enough to store a year's worth of energy.
throwaway2037 4y ago
Mister Martin: Another great post! Thank you to raise the idea of change. When I talk with people from developing countries, they frequently despair about the state of their environment, including air, soil, and water pollution. I tell them that Japan, UK, Germany, and the United States was a dirty mess in the 1960s and 1970s. Then, all of them awoke to environmentalism. Today, these are some of the cleanest countries in the world. In particular, Japan and German are still major industrial power houses (steel, chemical, manuf, etc.). Their air, soil, and water pollution declined dramatically in the last 50 years. Yes, sadly, some of this pollution has been exported to poor(er), developing countries. However, a lot of industry remains onshore and people and companies are fully committed to a clean environment. Really, I cannot stress enough how large are chemical companies in Japan. When you fly into Haneda (Tokyo) Airport, it seems like a 50km continuous line of chemical plants on the waterfront in Kanagawa-ken! The environment regulations are incredibly strict. Places that were an environmental disaster 50 years ago are now a "factory neighborhood", but no longer a polluted mess. Recovery and change is possible! For other readers, I assume "TMI" means "Three Mile Island" nuclear accident. You can read more about it on Wikipedia.

throwawaylinux 4y ago

> I'm in the anti-nuclear camp. Mainly because I think the inertia around current consumption patterns is too strong to overcome, so adding a lot of nuclear to the mix would make things a lot more dangerous without making things much better. Adding lots of nuclear would vastly reduce greenhouse gas emissions, which are the single biggest threat to humanity and life on earth. > This is how I think about the danger of nuclear power. Imagine a really safe nuclear power plant. How often will it fail? Let's say, once every 100,000 years. Assuming this probability is uniformly distributed over the 100,000 year time period, it has a 1/100,000 chance of failing in any given year. A failure seems pretty unlikely. But, there are something like 450 operating nuclear power plants in the world. The chance of a single power plant failing in a year is then 1 - (99,999/100,000)^450 which is around 0.5%. That's starting to look a lot more likely now. After 30 years, there is a 78% chance of a nuclear power plant failing somewhere on Earth. > Now what's even more interesting is if you look back on historic nuclear power accidents at INES level 6 or higher. 1957 - Kyshtym disaster, 1986 - Chernobyl, 2011 - Fukushima. And of course there have been other close calls (e.g. Three Mile Island). The cadence of these accidents seems to match the data from the thought experiment above. Yes nuclear power plants fail at some rate. Contrary to the pro-fossil-fuel / green propaganda, even quite serious failures are just not very serious in the scheme of things. A single coal mine collapse will easily kill many more people than all the direct deaths attributed to nuclear accidents combined. As for indirect, coal mines (EDIT: that should be coal power stations) will pump out chemicals and radiation that kill vast numbers of people _when they are operating normally without any failure_. Other fossil fuel plants less bad than coal, but still very damaging. EDIT: references - https://www.hsph.harvard.edu/c-change/news/fossil-fuel-air-pollution-responsible-for-1-in-5-deaths-worldwide/ https://www.hsph.harvard.edu/c-change/news/fossil-fuel-air-p..., https://cen.acs.org/articles/91/web/2013/04/Nuclear-Power-Prevents-Deaths-Causes.html https://cen.acs.org/articles/91/web/2013/04/Nuclear-Power-Pr...

Schroedingersat 4y ago
> Adding lots of nuclear would vastly reduce greenhouse gas emissions, which are the single biggest threat to humanity and life on earth. Does it though, or does it just distract from real solutions and prop up coal and gas for another decade? Let's say there's a trillion dollars to spend today. What is the net reduction in emissions between now and 2040 achieved by spending it on nuclear (hint: if the project takes as long as most in the west do it's negative)? What is the net reduction achieved in the same timespan by spending it on renewables? Even if there weren't a bunch of other problems (and there are) it's like claiming you need to stop and build a new engine so your car can go faster on the last lap of a race.