5 ms·

If there are two things I've learned from these discussions it's that: 1. Nuclear advocates always talk about capital costs of building nuclear plants but neve

by jmyeet 3y ago

If there are two things I've learned from these discussions it's that: 1. Nuclear advocates always talk about capital costs of building nuclear plants but never operational costs because the LCOE of nuclear fission energy is terrible and incrasing [1] 2. Nuclear advocates like to talk about deaths because that's highly favorable to nuclear rather than, say, clean up costs. Nuclear disasters get hand-waved away even though we've had multiple of them despite there only being ~700 nuclear plants built. Chernobyl is hand-waved away because USSR. But what about Fukushima? Japan is a developed and regulated country. Tsunamis happen. So far $82 billion has been spent on the Fukushima clean up with no end in sight [2]. The final cost will probably exceed $1 trillion and take decades, maybe a century or more. The site requires active cooling still, which is where all the continaminated wastewater comes from, and the technology to clean up the core does not exist with no viable plan in sight. And that's from one nuclear power incident. [1]: https://en.wikipedia.org/wiki/Cost_of_electricity_by_source https://en.wikipedia.org/wiki/Cost_of_electricity_by_source [2]: https://asia.nikkei.com/Spotlight/Fukushima-Anniversary/Fukushima-cleanup-costs-swell-with-no-end-in-sight https://asia.nikkei.com/Spotlight/Fukushima-Anniversary/Fuku...

EmmaEngineer 3y ago
The nuclear deaths, unlike all the other sources, also ignores rest of system deaths (e.g. mining, enrichment, transport, construction and decommissioning accidents). Once you include those it looks far more like gas and oil than wind and solar (which invariably include the kitchen sink on the death toll). If you wanted to put solar on a level field with nuclear you'd only include deaths from skin cancer directly attributable to installation.
- vlovich123 3y ago
  My understanding is that those numbers for solar are installation and maintenance costs. They don’t include things like constructing the panels, mining the raw materials to make panels etc. additionally, to be comparable to nuclear, solar needs insane amounts of batteries so then you start having to include cobalt mining. There’s no perfect comparison possible because it’s an interconnected system. We still use fossil fuels to build and transport solar cells for example, so what the “true” number is is very hazy and hard to quantify. The point is nuclear and solar do not compete in the same league. Nuclear competes with coal, natural gas, and hydro. Electricity sources that have 24h reliable generation capabilities. Hydro is arguably the safest but it is quite ecologically destructive and also limited as to where you can do it.
  - EmmaEngineer 3y ago
    I doubt your numbers then, as when I googled I found “There have been no deaths caused by carrying panels or laying panels on a roof. Electricians are required to connect the panels and inverters as it is.” -- https://reneweconomy.com.au/trojan-horse-industry-angry-at-queenslands-electricians-only-solar-proposal/ https://reneweconomy.com.au/trojan-horse-industry-angry-at-q... so in the 158TWh generated by solar in Australia, that gives an average deaths per TWh of 0.003, nearly an order of magnitude lower. It's odd that you bring up cobalt, as https://www.iea.org/reports/global-ev-outlook-2023/trends-in-batteries https://www.iea.org/reports/global-ev-outlook-2023/trends-in... and https://www.energy-storage.news/lfp-to-dominate-3twh-global-lithium-ion-battery-market-by-2030/ https://www.energy-storage.news/lfp-to-dominate-3twh-global-... both suggest that cobalt is no longer relevant. It seems that sodium ion is also poised to replace lithium ion. Is cobalt the most dangerous mineral to mine? Why doesn't solar+battery provide reliable generation capabilities? There are any number of off-grid solar only projects, it's simply a matter of cost, not capability.
    - vlovich123 3y ago
      People are trying to find alternatives to cobalt. Lithium ion in general too cause it's density is not as good as we need it to be for grid scale & automotive use-cases. But that's all forward looking stuff. Nothing in your links suggests cobalt is getting removed. If anything it shows that the % of worldwide cobalt mining being used for batteries is growing - in other words, batteries are dominating the need for cobalt. I don't have any specific knowledge of mining, so I don't know cobalt specifically, but my understanding is that cobalt mining is more of a problem based on where the mines are and maybe ecological damage, not so much that mining it is especially dangerous (in other words, conflict prone regions where the mining has the effect of supporting unstable regimes / inhumane conditions / slavery). That's where I'd say uncounted deaths are happening. Re deaths per tw, keep in mind that for solar 1 person would die every 50 years. So no one dying from carrying panels / laying panels is not really surprising or indicative the numbers are wrong. Also note that most of these numbers are for industrial power generation, not roof-top solar which has a higher fatality rate. But the point is that for wind, solar & nuclear, the death rate in a given year is basically 0 (same actually for hydro because the entire death toll comes from 1 accident). That's important in terms of understanding that the safety profile for this technology is very similar & way way way below oil & gas (& just the numbers attributed directly & ignoring the very large externalities). At numbers that low, it doesn't really matter so much. > Why doesn't solar+battery provide reliable generation capabilities? There are any number of off-grid solar only projects, it's simply a matter of cost, not capability. By that logic that's also true for nuclear getting cheaper. The point is not just one of cost (& these plants are fairly expensive but not prohibitively so per se). The point is that we simply can't convert raw materials into batteries fast enough. We're trying to rapidly scale battery manufacturing capacity as quickly as possible, but to date we have 3 plants worldwide right now (2 in Australia & 1 in California which sum up to 700 MW). That's just waaay to small if you're talking about getting things powered by solar reliably globally. Also remember that batteries degrade over time which means it's a 300 MW plant today but in 20 years it's probably a 100 MW plant. It's a pretty big maintenance cost that would need to get factored in. Now comparing batteries to power generation obviously isn't fair because they just need to absorb load for a set amount of time so I don't know if these are small plants individually, but a single nuclear plant can generate GWs of power very easily without any degradation in that generation capability 24/7. Yes - the plant needs to be physically maintained & that's not cheap either - there's no perfect answer. So over time could we get solar + battery to work? Probably. But this shit is going to take a very long time - like maybe by the end of the century we could be 100% renewables. I'm skeptical but it is a possibility - my skepticisim comes from the fact that there are several high temperature manufacturing applications for which renewables don't work at all whereas nuclear can 100% easily replace fossil fuels. Meaningfully investing in nuclear would also have a significant payoff in that we could probably get off of fossil fuels within ~30 years and nuclear tech itself would change massively (molten salt reactors or thorium reactors which fail safe & can't melt down, getting cheaper plants by building more of them because right now we do a single nuclear project after very many many years, direct energy capture to avoid the need of a turbine & water, etc etc).
vlovich123 3y ago
It seems telling that nuclear, the only technology that could legitimately threaten the oil and gas industry in a short amount of time, is also the only technology that got more expensive as more capacity got installed. Why do you think that might be? Disaster cleanup is a problem for sure. I think it’s still cheaper than oil and gas because those industries externalize all their costs. Look. I’m not against solar and wind. I just think nuclear is the only way to feasibly rapidly decarbonize our energy production and solar and wind proponents always ignore that solar and wind just can’t do that on their own. Yes grid scale batteries are being invested in but it’ll take forever to build enough capacity assuming we even can in the first place - a not insignificant portion of battery production overall has to go to vehicles too and there’s no sign we’re getting our shit together to use car batteries as part of grid scale storage (the electrical distribution aspects are very difficult)
- dale_glass 3y ago
  > It seems telling that nuclear, the only technology that could legitimately threaten the oil and gas industry in a short amount of time, is also the only technology that got more expensive as more capacity got installed. Why do you think that might be? It's because every time safety fails, we find a way to solve that problem, and solving it costs money. Sometimes a lot of money. So Chernobyl absolutely confirmed that containment buildings are not optional. Fukushima resulted in increased requirements for availability of backup power in case of a disaster. Solar and wind don't really have that, because if a wind turbine falls apart in a hurricane, it's not really significantly different from any other random structure doing the same thing. Being killed by a chunk of turbine blade in a hurricane is not different from being killed by a chunk of a house, and in general not very likely because they rarely have people living right next to them. So we don't demand that wind turbines can survive anything nature can throw at it.
  - vlovich123 3y ago
    And yet, we have newer generations of reactors that fail safe & don’t need containment buildings. Why would we stick to building designs that are 70 years out of date?
ben_w 3y ago
> But what about Fukushima? Japan is a developed and regulated country. Tsunamis happen. What happened at Fukushima should have been foreseeable and thus prevented. In this regard it's an entirely valid point. When it comes to how many died, not so much: not only is it a ratio of 1 dead from the incident and 2202 from the evacuation, there were also 19,759 dead from the earthquake and tsunami that damaged the reactor. Likewise with regards to the issues other than loss of life, the impact of the reactor itself ought to be compared to the impact of the earthquake and tsunami. Besides damaging the reactor, these other impacts included 5 million tons of debris and 25 million trash; 230,000 cars and trucks being damaged or destroyed; 1.5 million households losing water and at least 4.4 million households lost electricity; 10% of fishing ports being damaged; the total collapse of 121,852 buildings, half-collapse of 281,042 buildings, and partial damage to 727,391 buildings[0]; damage to 7 dams and the collapse of one (itself leading to loss of life). [0] https://web.archive.org/web/20170922194205/https://www.npa.go.jp/news/other/earthquake2011/pdf/higaijokyo_e.pdf https://web.archive.org/web/20170922194205/https://www.npa.g...

by jmyeet 3y ago

EmmaEngineer 3y ago

The nuclear deaths, unlike all the other sources, also ignores rest of system deaths (e.g. mining, enrichment, transport, construction and decommissioning accidents). Once you include those it looks far more like gas and oil than wind and solar (which invariably include the kitchen sink on the death toll). If you wanted to put solar on a level field with nuclear you'd only include deaths from skin cancer directly attributable to installation.

vlovich123 3y ago
My understanding is that those numbers for solar are installation and maintenance costs. They don’t include things like constructing the panels, mining the raw materials to make panels etc. additionally, to be comparable to nuclear, solar needs insane amounts of batteries so then you start having to include cobalt mining. There’s no perfect comparison possible because it’s an interconnected system. We still use fossil fuels to build and transport solar cells for example, so what the “true” number is is very hazy and hard to quantify. The point is nuclear and solar do not compete in the same league. Nuclear competes with coal, natural gas, and hydro. Electricity sources that have 24h reliable generation capabilities. Hydro is arguably the safest but it is quite ecologically destructive and also limited as to where you can do it.
- EmmaEngineer 3y ago
  I doubt your numbers then, as when I googled I found “There have been no deaths caused by carrying panels or laying panels on a roof. Electricians are required to connect the panels and inverters as it is.” -- https://reneweconomy.com.au/trojan-horse-industry-angry-at-queenslands-electricians-only-solar-proposal/ https://reneweconomy.com.au/trojan-horse-industry-angry-at-q... so in the 158TWh generated by solar in Australia, that gives an average deaths per TWh of 0.003, nearly an order of magnitude lower. It's odd that you bring up cobalt, as https://www.iea.org/reports/global-ev-outlook-2023/trends-in-batteries https://www.iea.org/reports/global-ev-outlook-2023/trends-in... and https://www.energy-storage.news/lfp-to-dominate-3twh-global-lithium-ion-battery-market-by-2030/ https://www.energy-storage.news/lfp-to-dominate-3twh-global-... both suggest that cobalt is no longer relevant. It seems that sodium ion is also poised to replace lithium ion. Is cobalt the most dangerous mineral to mine? Why doesn't solar+battery provide reliable generation capabilities? There are any number of off-grid solar only projects, it's simply a matter of cost, not capability.
  - vlovich123 3y ago
    People are trying to find alternatives to cobalt. Lithium ion in general too cause it's density is not as good as we need it to be for grid scale & automotive use-cases. But that's all forward looking stuff. Nothing in your links suggests cobalt is getting removed. If anything it shows that the % of worldwide cobalt mining being used for batteries is growing - in other words, batteries are dominating the need for cobalt. I don't have any specific knowledge of mining, so I don't know cobalt specifically, but my understanding is that cobalt mining is more of a problem based on where the mines are and maybe ecological damage, not so much that mining it is especially dangerous (in other words, conflict prone regions where the mining has the effect of supporting unstable regimes / inhumane conditions / slavery). That's where I'd say uncounted deaths are happening. Re deaths per tw, keep in mind that for solar 1 person would die every 50 years. So no one dying from carrying panels / laying panels is not really surprising or indicative the numbers are wrong. Also note that most of these numbers are for industrial power generation, not roof-top solar which has a higher fatality rate. But the point is that for wind, solar & nuclear, the death rate in a given year is basically 0 (same actually for hydro because the entire death toll comes from 1 accident). That's important in terms of understanding that the safety profile for this technology is very similar & way way way below oil & gas (& just the numbers attributed directly & ignoring the very large externalities). At numbers that low, it doesn't really matter so much. > Why doesn't solar+battery provide reliable generation capabilities? There are any number of off-grid solar only projects, it's simply a matter of cost, not capability. By that logic that's also true for nuclear getting cheaper. The point is not just one of cost (& these plants are fairly expensive but not prohibitively so per se). The point is that we simply can't convert raw materials into batteries fast enough. We're trying to rapidly scale battery manufacturing capacity as quickly as possible, but to date we have 3 plants worldwide right now (2 in Australia & 1 in California which sum up to 700 MW). That's just waaay to small if you're talking about getting things powered by solar reliably globally. Also remember that batteries degrade over time which means it's a 300 MW plant today but in 20 years it's probably a 100 MW plant. It's a pretty big maintenance cost that would need to get factored in. Now comparing batteries to power generation obviously isn't fair because they just need to absorb load for a set amount of time so I don't know if these are small plants individually, but a single nuclear plant can generate GWs of power very easily without any degradation in that generation capability 24/7. Yes - the plant needs to be physically maintained & that's not cheap either - there's no perfect answer. So over time could we get solar + battery to work? Probably. But this shit is going to take a very long time - like maybe by the end of the century we could be 100% renewables. I'm skeptical but it is a possibility - my skepticisim comes from the fact that there are several high temperature manufacturing applications for which renewables don't work at all whereas nuclear can 100% easily replace fossil fuels. Meaningfully investing in nuclear would also have a significant payoff in that we could probably get off of fossil fuels within ~30 years and nuclear tech itself would change massively (molten salt reactors or thorium reactors which fail safe & can't melt down, getting cheaper plants by building more of them because right now we do a single nuclear project after very many many years, direct energy capture to avoid the need of a turbine & water, etc etc).

vlovich123 3y ago

It seems telling that nuclear, the only technology that could legitimately threaten the oil and gas industry in a short amount of time, is also the only technology that got more expensive as more capacity got installed. Why do you think that might be? Disaster cleanup is a problem for sure. I think it’s still cheaper than oil and gas because those industries externalize all their costs. Look. I’m not against solar and wind. I just think nuclear is the only way to feasibly rapidly decarbonize our energy production and solar and wind proponents always ignore that solar and wind just can’t do that on their own. Yes grid scale batteries are being invested in but it’ll take forever to build enough capacity assuming we even can in the first place - a not insignificant portion of battery production overall has to go to vehicles too and there’s no sign we’re getting our shit together to use car batteries as part of grid scale storage (the electrical distribution aspects are very difficult)

dale_glass 3y ago
> It seems telling that nuclear, the only technology that could legitimately threaten the oil and gas industry in a short amount of time, is also the only technology that got more expensive as more capacity got installed. Why do you think that might be? It's because every time safety fails, we find a way to solve that problem, and solving it costs money. Sometimes a lot of money. So Chernobyl absolutely confirmed that containment buildings are not optional. Fukushima resulted in increased requirements for availability of backup power in case of a disaster. Solar and wind don't really have that, because if a wind turbine falls apart in a hurricane, it's not really significantly different from any other random structure doing the same thing. Being killed by a chunk of turbine blade in a hurricane is not different from being killed by a chunk of a house, and in general not very likely because they rarely have people living right next to them. So we don't demand that wind turbines can survive anything nature can throw at it.
- vlovich123 3y ago
  And yet, we have newer generations of reactors that fail safe & don’t need containment buildings. Why would we stick to building designs that are 70 years out of date?

ben_w 3y ago

> But what about Fukushima? Japan is a developed and regulated country. Tsunamis happen. What happened at Fukushima should have been foreseeable and thus prevented. In this regard it's an entirely valid point. When it comes to how many died, not so much: not only is it a ratio of 1 dead from the incident and 2202 from the evacuation, there were also 19,759 dead from the earthquake and tsunami that damaged the reactor. Likewise with regards to the issues other than loss of life, the impact of the reactor itself ought to be compared to the impact of the earthquake and tsunami. Besides damaging the reactor, these other impacts included 5 million tons of debris and 25 million trash; 230,000 cars and trucks being damaged or destroyed; 1.5 million households losing water and at least 4.4 million households lost electricity; 10% of fishing ports being damaged; the total collapse of 121,852 buildings, half-collapse of 281,042 buildings, and partial damage to 727,391 buildings[0]; damage to 7 dams and the collapse of one (itself leading to loss of life). [0] https://web.archive.org/web/20170922194205/https://www.npa.go.jp/news/other/earthquake2011/pdf/higaijokyo_e.pdf https://web.archive.org/web/20170922194205/https://www.npa.g...