Hacker News Reader

Comment by jbullock35 | Hacker News Reader

The linked tweet doesn't provide a citation or context for the "helped prevent ~2M deaths" claim. But there is work in this area which suggests that "~2M lives saved" in the last 50 years underestimates the benefits of nuclear power by a lot.

One commenter here writes

> What is the actual claim here? Nuclear power didn't prevent any deaths, after all everyone will eventually die for one reason or another. So it must be something like two million people would have died earlier without nuclear power, but by how much?

That's a better way to think about the problem: not "lives saved," but life-years saved. Calculations of life-years saved are necessarily sketchy, but there is at least one promising preliminary effort [1]. See especially page 22.

[1] https://storage.googleapis.com/production-sitebuilder-v1-0-1...

View on HN

Nuclear power should be exactly comparable to other types of power.

Deaths during construction. Carbon produced in construction.

Deaths caused by mining and transporting fuel. Carbon produced by that.

Deaths caused by consuming fuels. Carbon for that as well.

Deaths caused by accidents during the lifetime of the power source.

Deaths caused to wildlife and pollution.

Deaths caused by decommissioning and deconstruction. Carbon for that too.

Then lifetime for nuclear power station, vs oil, coal, gas, wind turbine, solar.

Finally you can do an unbiased comparison of each.

Nuclear power has saved many lives by not requiring dangerous coal mining, particulate emissions, and climate change effects.

That's the real trolley problem: they have to be powered by SOMETHING.

The nature of the carbon production is important. Carbon from wired electricy is different from embodied carbon from battery electric vehicles, and both are different from CO2 equivalent emissions from the chemistry of the industrial process itself.

After all, end game is you use a carbon neutral technology to power itself, emissions from energy usage go to zero. Similarly, solving carbon emissions for batteries cleans up energy storage.

Also, deaths due to climate change dwarf all the other sources combined (except, maybe, for deaths from coal emissions), so direct deaths are a second order effect that can be ignored in practice.

OurWorldInData has a lot of data for these types of comparisons: https://ourworldindata.org/safest-sources-of-energy

but feels says nuclear bad...

A retrospective analysis of the benefits of nuclear power avoids accounting for one of the major consequences: the nuclear waste, which remains harmful/deadly for thousands of years.

And so is CO2...

Lead and mercury spewed out by coal plants ends up in the ocean and is also deadly for very long. I'd rather have poisonous metals in storage than floating around and accumulating in the food chain.

>which remains harmful/deadly for thousands of years.

Which it has in common with things like acid mine drainage and waste from a bunch of other industrial processes.

Of course, solving the former "just" requires employing a small army of personnel to make sure that the effluent is getting cleared from those mines properly, maintaining tailings ponds (and fixing them when they leak), and (since this is the perennial hypothetical for nuclear waste disposal) understanding that a bunch of people from the hypothetical neo-Stone Age civilizations are going to die once that goes away because they're necessarily going to be incapable of chemical analysis of water downstream.

Perhaps it's unfortunate for most of humanity that storing waste in on-site casks is an effective enough solution; if the waste was more dangerous, we'd have dealt with it already. Instead, we can just leave it sitting there and tilt at windmills.

I'd rather have an exploded reactor and evacuated surrounding area every 50 years than I would an uninhabitable planet. But maybe that's just my obsolete 1970s/80s optimism showing through; modern humanity is much happier with the historically-compatible "use unreliable sources of power and disconnect the poor quarter at night" strategy.

Sadly coal miners are exposed to more radioactive material.

Nuclear waste is either or. Either we survive as a species and know where it’s buried in that 10,000 year container deep underground where everyone knows. Or we fail as a species and lose the technology to dig that deep ever again.

And it doesn't stop after the coal is mined, either. Coal ash and tailings are quite toxic.

Now that there are cleaner alternatives we don't need new nuclear anymore to save lives.

The only cleaner alternative with fully predictable and controllable power delivery is hydro.

Even hydroelectric power isn't entirely predictable:

https://www.cnbc.com/2021/08/06/california-shuts-down-major-...

And good luck if a dam fails:

* https://en.wikipedia.org/wiki/1975_Banqiao_Dam_failure

* https://en.wikipedia.org/wiki/Dam_failure

Just like each nuclear plant is a "ticking nuclear bomb", each dam is a ticking dam bomb :) I forgot where I heard this from. Anyway, a dam failure can cause hundreds of thousands of deaths within minutes.

And hydro is a full capacity already pretty much everywhere in the EU at least

In Czechia, we don't have enough wind or sun for generating much power, and the available space for hydro isn't ideal either.

If we don't need anything except the cleaner alternatives, we should start by banning and issuing laws against burning fossil fuel for connecting to the energy grid. This naturally include imports, as buying energy generated by burning fossil fuel outside the nations borders is just as bad as if the burning occurred inside.

At that point people can choose between all the non-fossil fuel alternatives and pick what attributes they favor.

No you're right china gets a free pass to strip mine for batteries and solar cells with slave/prison labour... then we all sing that solar in the west is great...

I'm just waiting for the day all of the humans required to keep a one of the older, less safe designs running all become unavailable and the people required for disaster management when it melts down are also unavailable and the world experiences a completely unmitigated meltdown. I really don't think this is out of the realm of possibility within the next hundred years. The assumption here is a societal collapse of sorts where say 90% of the population go away. IIRC there were something like 600,000 people who responded to the cleanup of Chernobyl, and something like 4% of the radioactive material of a single reactor was all that leaked into the environment. I sort of see nuclear power plants (at least certain designs) as a ticking time bomb that'll potentially leave large parts of the planet uninhabitable (or maybe just really high rates of thyroid cancer in the remaining population) for the 10% that remain in a post collapse world.

Nuclear does have a great record on paper. It's the non-linear behavior that scares people. And when troops roll up and start shelling nuclear power plants and people go "oh... oh I didn't consider how it might function under these conditions" that that non-linearity rears its head again and people get pretty cold on nuclear when faced with situations like that.

> I sort of see nuclear power plants (at least certain designs) as a ticking time bomb

But that's just progress, right? Even now we still do stupid things with regular chemicals despite knowing not to.

Every chemical process evolves and gets safer. 1st gen nuclear power stations were/are potentially able to meltdown. Newer designs can't. But we can never be safe from human stupidity and wars.

For example - https://en.wikipedia.org/wiki/List_of_industrial_disasters

And even recently: https://en.wikipedia.org/wiki/2020_Beirut_explosion

Somehow people will ignore those disasters as they're not nuclear, but Chernobyl+Fukashima everyone remembers. It just feels like "meltdown and waste" whilst genuine concerns, aren't really the conversation-ending arguments people pretend them to be.

(edit - I live within 15 miles of a nuclear power station, and I'm completely comfortable with that).

I think you miss my point. My frame of reference is society has collapsed and the _people_ that keep nuclear safe are no longer around and this has the potential to cause a Beyond Design Basis Accident the likes of which could render large portions of the planet uninhabitable.

I'm generally fine with nuclear as long as society can maintain itself and the data back up what the proponents of nuclear claim. I'm just suspecting that every reactor that could and would meltdown if humans were no longer around to attend to it being left to meltdown completely unmitigated would be a disaster of epic proportions.

Agreed.

But I think the accident severity distribution for a nuclear power plan is dramatically more heavy tailed than that for other severity distribution for other power sources. That is, the impact of an individual accident for a nuclear plant can reasonably be orders of magnitude beyond the impact for a non nuclear plant.

I don’t think that’s usually taken into account and I do think that swings the cost benefit analysis into a net negative relative to other options.

Life-years saved is a step in the right direction, but still the wrong metric to focus on in my opinion.

Healthy life-years is a much better metric to optimize.

If we get people living to 200, but 120 of those years are in a miserable and extremely debilitated state, I consider that a failure compared to average life span of 100, where only 5 years are in an extremely debilitated state. i.e. getting 95 good healthy years + 5 bad ones >>> 80 healthy years and 120 bad ones.

I suppose you can make an argument that extending your miserable years can help a person survive long enough to eventually take advantage of body and mind rejuvenation tech, so there is some point where absolute maximization of life-years might make sense.

This is QALYS. It's how the UK allocates health care spending.

The UK allocates healthcare in a pretty humane, efficient and, ultimately, (whisper it!) communist manner.

That doesn't sound very communist to me. Moreover, while the British certainly prefer their healthcare system to the American system (and I probably agree), I've never heard them describe their system as efficient (on the contrary, they seem to talk about it as a money pit).

The amazing thing is that despite being a money pit, it still manages to deliver better overall value than the US "system".

It's not really amazing, it's just that not everything is in reference to the US system. When Brits comment about their system being a money pit, they aren't always thinking about the US system as their frame of reference. This seems to be a stumbling block for us Americans, or at least for Americans who are passionate about healthcare reform.

Experts regularly rate it as one of the most efficient health systems, so whoever you're talking to is ill informed. Probably listened to too much propaganda from people who want to adopt the widely hated and less efficient American system instead.

Hm, as a German, I shudder at the thought of potentially being dependant on a system that regularly breaks down during standard influenza waves in the winter [1] and that has fixed-amount budgets for individual sickness [2], so your tooth better splits during the first few months...

Is the NHS better than the US system? Maybe. But it is by far inferior to the French, the Spanish or the German healthcare systems.

[1] https://www.theguardian.com/society/2017/sep/24/winter-flu-c... [2] https://www.thetimes.co.uk/article/dentists-run-out-of-cash-...

You're comparing apples to oranges by using a different metric than the parent is using.

I was talking specifically about efficiency:

> Despite regular criticism, the National Health Service (NHS) is effective when comparing the government spend versus the outcomes of human development and lifespan. While critics point to Switzerland achieving the lowest death rates in Europe, or Germany having a consistently quality health care system, they don’t compare these to the level of expenditure.

> In fact, the UK has a higher life expectancy than Germany, with lower health care funding than in other large economies. The Swiss spend $3,079 more on health care per person than the UK, while Germany spends $1,993 more per person than the UK. Critics can’t ignore these massive differences in expenditure when debating the outcomes that the NHS provides. If the UK matched Swiss expenditure per person, an additional $206 billion would be spent on health care annually.

The correlation between the lifespan of a whole population and the dollar amounts spent by the government (which BTW probably shouldn't be compared across countries with different costs of living), is presumably pretty confounded.

I'm not trying to build this argument from scratch, I'm just telling people who get their info from other sources that the commonly accepted opinion of the experts in this field is that the NHS is an efficient health service.

If you don't accept that then feel free to write a letter to their academic journals and correct their work. I’m sure they’ll be really interested in your uninformed knee-jerk reactions to things that your gut doesn't like the sound of.

Point is: as a patient, I don't care much about "bang for the bucks", I care about whether or not I get the medical attention I need when I need it.

> In fact, the UK has a higher life expectancy than Germany

Life expectancy seems to be very similar, tbh [1]. The metric "money per patient vs. like expectancy" also does not include differences in cost of living (which is higher in Germany and Switzerland) and demographic differences (on average, the UK is younger than Germany and Switzerland [2])

[1] https://www.wolframalpha.com/input?i=life+expectancy+uk+vs+g...

[2] https://www.wolframalpha.com/input?i=median+age+uk+vs+german...

> Experts regularly rate it as one of the most efficient health systems, so whoever you're talking to is ill informed.

Maybe.

> Probably listened to too much propaganda from people who want to adopt the widely hated and less efficient American system instead.

Brits don't have much propaganda for the American system, and I doubt the British psyche frames every aspect of British life in reference to American life. That said, your zeal for the UK system and passionate distain for the American system are duly noted.

>Brits don't have much propaganda for the American system, and I doubt the British psyche frames every aspect of British life in reference to American life.

The rest of the world actually pays more attention to america than america does to it.

A lot of brits have family and friends in america. Theyre at least dimly aware of the insane medical bills and bankruptcies.

> communist manner

I think the word you are looking for is "collectivist".

The idea that people are simultaneously intelligent enough to understand the somewhat nuanced argument that nuclear power is safe, but also so stupid that "2 million lives saved" makes more sense to them than just a number of QALYs is, I think, an abject failure of science journalism. If you explain everything like everyone is 5, you reap what you sow.

For those unfamiliar with the unit, QALY is quality-adjusted life year. [1]

[1] https://en.wikipedia.org/wiki/Quality-adjusted_life_year

This is a strange shift in responsibility. Wouldn’t it be better to say: “Coal shortened the life of so and so many people”? After all If there were neither coal power (or other air polluting power) nor nuclear those lives wouldn’t have been shortened either.

Yes. This is an attempt to boost nukes, but however much better it makes nukes look, it makes renewables several times moreso.

If it is retrospective, it doesn't really say anything about renewables because for most of the last 50 years they haven't been [meaningfully] feasible. The debate going forward is still entirely valid because nuclear has upsides that contrast with renewable. It should be part of the discussion.

It is still a rather strange shift in perspective. We don’t count the people saved by taking public transit, we count the people killed by cars. Speculating what was and wasn’t feasible for the past 50 years is even stranger.

Usually these comparisons have an implicit all else being equal because you can get into some real funky speculations with alternative history without it. For example, if we had neither coal nor nuclear how different would the energy market be today? Would there have been more innovations in renewables and the current technology would have been reached in the 80s? Or would we have implemented our infrastructure to maximize energy efficiency? How many lives would that have saved?

You get the point. This practice gets unnecessarily complicated really fast. I agree with parent that the only reason to say that “nuclear saves lives“ over “coal destroys lives” is that you have an agenda to promote nuclear over alternatives without mentioning those alternatives.

City planners do make estimates on how much traffic is reduced by introducing more and new public transit. They also make estimates on reduced traffic accidents when building news roads and redirect traffic away from more dangerous paths and those close to high risk areas like schools.

We can also compare similar sized locations and compare outcomes. How many lives dies to traffic accidents in cities prioritizing bike lanes and mass transit compared to same sized cities that prioritize car travel?

We can also count how many people that die to air pollution related illness in countries with a lot of coal power and compare that to cities with a lot of nuclear power. We can also look at countries like Island or Norway, through countries that can exclusively depend on hydro/geothermal power has a fairly low sample size. With coal we can get a fairly massive sample size just looking at the scope of a city.

Indeed. The logic goes: Vehicular traffic kills so-and-so many people. We can significantly lower that number by reducing vehicular miles traveled. Options include better bike infrastructure, better public transit, mixed zoning, lower speed limits, etc.

If a city engineer would invert the logic and start with “Bicycle lanes have saved so-and-so many lives”. That would be rather odd wouldn’t it?

If I search for "Bicycle lanes lives saved" (and alternative version of the search string) I find plenty of articles and even studies with those words in the title. Here is one:

https://www.forbes.com/sites/jeffmcmahon/2022/03/09/bicycle-...

In the United States, just under 1 percent of commuters travel by bicycle, Patz said. Increasing that slice to 6 percent—the level in Madison, WI—would avoid 20,000 premature deaths per year, he said

> That would be rather odd wouldn’t it?

Maybe I read too much news but No, I don't find it odd. Building safer roads, reducing traffic, improving traffic safety, does save lives. Reducing air pollution saves lives too.

It's not about "boosting nukes" it's about countering nuclear energy FUD (mostly about how dangerous nuclear energy is). Renewables would probably be comparable, but I'm very skeptical about your "several times better" claim. Nuclear energy and renewable energy are both very low carbon sources with very few non-pollution-derived deaths (rooftop solar exceeds nuclear energy for deaths, even including nuclear anomalies like Chernobyl IIRC). So there's no way that the carbon gap between fossil fuels and nuclear is significantly larger (much less 3x larger) than the gap between fossil fuels and renewables.

What stops nuke construction today is that it costs a hell of a lot more, and you don't get any kWh out for years, if ever. Most usually, lately, you pour in $billions for years and get nothing out but a lot of scrap concrete.

I don't know that it costs "a hell of a lot more" for a unit of reliable energy, but yeah, you can deploy renewable energy in smaller units more quickly (it might take you a decade to build out a GW of nuclear capacity or a GW of solar capacity, but at least in the case of solar you don't need the full GW to complete before you can start selling power). However, to be clear, this is the economic argument, not the health argument that the OP was debating.

Nuclear power is clean at the plant but the fuel has a very dirty production pipeline. Further, plants are estimated to have a carbon impact around the same or worse than natural gas because their supply chain is so expensive. Instead of the carbon coming from burning the fuel, it comes from the enormous amount of work that goes into keeping a nuclear power plant running.

HNers love to shout about the environmental impact of semiconductor production for solar panels and mining for batteries and even permanent magnets used in electric motors. But apparently reactor fuel comes out of the ground and is refined based on ...wishful thinking?

It's all moot. Nuclear power is not being deployed by utilities because it is one of the most expensive forms of energy generation. With almost a century of enormous subsidization and research, it has only increased in cost. Wind and solar have plummeted in cost. Wind is currently the cheapest, with solar hot on its tail. Solar continues to fall at around 10% a year.

Last year in the US alone, and only accounting for grid-scale projects, wind and solar have replaced nuclear at a 6:1 ratio. When you account for commercial and residential solar and wind, it's an even higher ratio.

Solar cells cost $76 per watt in the seventies and now it costs thirty cents per watt. That's a 250 fold reduction.

Wind has gone from 55 cents per kwhr in the eighties, to 5 cents today.

Nuclear? More expensive than it was in the eighties.

With HVDC transmission technology seeing wider and wider use, and well as battery technology plunging in cost: everyone in the industry and government sector considers "what energy source do we go forward with" as solved.

It's not just cheaper: it doesn't have any of the numerous concerns nuclear does. Solar panels can be recycled (or simply resold for applications where the 10-20% reduced capacity doesn't matter, ie where there's plenty of roof/land available relative to energy needs), and the biggest problem with wind turbines has been figuring out how to recycle the turbine blades when they're EOL, something that is now solved. GE is turning blades into concrete, and Vestas has figured out how to recycle blades into like-new fiberglass.

Nobody sees nuclear as a viable path forward except the nuclear industry.

It's unclear that battery scaling will happen in time to reach sharply negative carbon emissions, or that solar and wind will meet future energy demands. Our kids need to put all the carbon back in the ground, which means we need to put the first 10-20% back ourselves. That would double global energy consumption for decades, assuming we started today.

Your supply chain carbon arguments have a fundamental flaw: Transportation and the grid need to decarbonize regardless of which technology the power plant uses.

Solar panel production is being bootstrapped with coal based power plants. That doesn't imply coal plants will be needed moving forward, or that industrial (bursty, scheduled consumption) electicity will be scarce.

The fact that US nuclear plants are riduculously economically inefficient doesn't imply plants in the rest of the world are similarly bad, or that the technology is infeasible. The US environmental movement intentionally sabotaged the industry through the second half of the 20th century. Also, efficient, safe nuclear plants have been operated for decades in other countries.

Anyway, we're clearly headed to a situation where we have effectively infinite clean electricity during the day, thanks to solar.

The real questions are whether energy storage devices are cheaper than nuclear, and how economical it is to time shift energy-intensive industrial processes (basically building 2-3x more factory, but idling it at night).

I'm hoping there is some bursty carbon capture technology on the horizon, where buying more equipment for a higher burst capture rate is cheap, so the electricity produced for the quarter of the day where there's a massive surplus isn't wasted.

We need to do that anyway, and it would greatly reduce nuclear, hydro and energy storage demand.

Cost for storage is falling even faster than solar or wind ever did.

All that is unclear is what will end up cheapest, in any particular area. In many places, the poorer round-trip efficiency of chemical synthesis (hydrogen, ammonia, hydrocarbons) is offset by the massive usefulness of synthesizer output after the tanks are full. That is, when the tanks have been topped off, the excess output generates revenue.

Solar generating capacity has become so cheap that needing 2x, 3x, 4x more, for whatever reason, is no blocker.

Even if batteries and solar get another 10x cheaper, that's not enough to make it through winter in some populated areas.

A whole home backup battery is $10K's, and likely stores under 24 hours of normal usage. Buying 90 of those per house (for three months of power backup) would cost millions per house today.

A bright overcast day in California can easily drop solar output by 50-75%. Up north, it's much worse, and those days can come in multi-month bursts.

Much of the US south will be uninhabitable according to current climate projections, which increases demand up north, creating distribution problems that haven't been solved. Also 40+ state megastorms are becoming common in the US. What happens if the whole continent is cold and cloudy at the same time?

I guess you're propsing synthetic natural gas, or similar?

Batteries are the most expensive storage medium. No utility will use those to keep more than a few minutes or hours of backup. Pumped hydro is cheap and reliable, but each site has a strict upper limit on how much it can store.

Synthetic anhydrous ammonia, and hydrogen, are the favored utility-scale media. They can be stored, but more importantly they can also be shipped in from the tropics, and most importantly can be sold when the tanks are full.

Ammonia is not great for shipping to due low gravimetric energy density. At the same time, ammonia cycle efficiency in a fuel cell is low, and fuel cells are expensive.

Hydrogen is a shipping nightmare, as you well know.

So at this point we are comparing the cost of a full NH3/H2 backup energy generation, plus the cost of a full renewable energy grid with at least 50% overproduction and hours of battery storage, plus the cost of storing 100 million tons of NH3/NH2, as well as the cost of NH3/NH2 production, and funding of generally idle overproduction in countries with a better climate (!!)

Liquified hydrogen carried at atmospheric pressure just needs insulated tanks. We already have a lot of experience shipping liquified natural gas. It can probably be carried in the same tanks.

You can burn ammonia where you burn natural gas today. It is probably not suitable as fuel for aircraft, but otherwise it is entirely good enough. Ships burning bunker oil can be converted burn ammonia just by changing out tankage and plumbing.

But ammonia fuel cells will only get better, never worse.

Building out a H2/NH3 system will be massively expensive, but a tiny fraction of the cost of that much nuke capacity. And, the H2 and NH3 are valuable industrial feedstock, so none of the production capacity ever sits idle.

Doesn't storage and transport of hydrogen introduce a lot of difficulties that don't exist with other liquified gases, such as penetrating into steel and making it brittle?

Hydrogen embrittlement is a problem at high pressure and temperature. Liquifying sidesteps that.

Scaling battery storage does seem easier than scaling nuclear. They are mass producible as small modules and can utilise less skilled labour in factories. Grid scale developments have negligible civil engineering requirments, ridiculously quick to build and can work on small or large sites. They are easy for investors to understand and insure. And the underlying technology can piggy back on the cell phone industry and will be required for electric cars.

Also, by the time we get to nuclear scale out it may well be competing with hydrogen also.

I don't think people understand the scale of storage required to make decarbonization through renewables feasible. The USA consumes 500GWh of electricity every hour. And electricity is only about a third of total energy consumption. 8 GWh is one minute of electricity storage, 20 seconds of total energy storage. We'll need and estimated 3 weeks of storage to decarbonize through renewables: https://pv-magazine-usa.com/2018/03/01/12-hours-energy-stora...

By comparison, the US just needs to build four nuclear plants for each one that exists currently. To achieve a 100% nuclear electricity grid. A bit less than that for decarbonization since we have some hydroelectric power.

Just four new nuclear power plants for every existing plant is still a lot of plants.

There are 93 active reactors in the USA today. That means we need to build between 350 to 400 more reactors. There are currently only 2 new reactors under construction and 11 more planned (6 of these are NuScale modes with only 77 MW Capacity). The last reactor to open was at the Watts Bar plant in Tennessee in 2016, and before that 1996 in the same plant. Both of these reactors started construction in the 1970s.

If we “just” need 350-400 more, it is gonna take us ages to get there.

Building nuclear plants at the same rate as we did in the 1970s would get us there in 3 decades. You write as though this is some Herculean task, but if the US didn't reduce the pace of nuclear construction in the 1980s we would have had a decarbonized grid by 2000. It's not just hypothetically possible, this pace of nuclear construction has historical precedent.

By comparison, nobody has any realistic plans to decarbonize fully through renewables. The plan is to burn gas for ~40% of our electricity when intermittent sources aren't producing, and cross our fingers while we hope for a miraculous breakthrough in energy storage. Plans to decarbonize through renewables assumes that hydrogen storage, thermal batteries, or something else will provide effectively free energy storage.

It is not fair to extrapolate the building speed of the 1970s to today. Especially not with nuclear reactors. We don’t build things like in the 70s anymore. There are safety and environmental standards which weren’t being observed back then. And that’s a good thing. It is not a coincidence that things slowed down in the 80s. The 3 mile island partial meltdown happened in 1979 and it left a mark. If we would still be building and running our reactor like we did back then, but had 5 times more reactors to maintain, we can assume that the probability of a full meltdown and a major nuclear accident would be far higher. Chernobyl serves as a reminder of what can happen if safety standards are not adequate.

The labor and logistics situation is also completely different from today then it was in the 70s. Today’s megaprojects tend to go way over budget and suffer significant delays. Land acquisition is more complicated (and again that is a good thing as many building project back then tended to displace a lot of minority residents) and a much smaller proportion of our labor pool are construction engineers. This leads to a longer and more expensive building process. As an example the only two reactors currently under construction have gone at least $10 billion over budget and are 6-7 years behind schedule. And these are reactors in the existing Plant Vogtle in Georgie. I imagine a whole new power plant would suffer even worse logistics problems.

> Plans to decarbonize through renewables assumes that hydrogen storage, thermal batteries, or something else will provide effectively free energy storage.

This is only partially true. Plans also include a significantly more efficient energy usage. People are pressuring governments to significantly increase investments in green infrastructure. This includes more robust energy grid, electrifying railways and ferry terminals, increased building standards for efficient power usage and retrofitting old buildings to similar standards. We don’t need to replace carbon power 1:1 if we patch up the inefficiencies of our current energy use. There are also hopes for better battery technology and carbon capture. A carbon capturing gas burning power plant is not an unrealistic advancement, nor is solid state and molten salt batteries (all of which exist today at varying levels of economic feasibility; none less feasible then new new nuclear). Meaning that we can bridge that ~40% without emitting more greenhouse gasses and without new nuclear.

Three mile island was due to a pressure valve failure, not a design flaw in the pressure vessel, secondary storage, or steam generators - the components of a reactor that form most of the cost of construction. And the reactor meltdown was contained. The safety standards held. I'm not sure how people try and spin this into evidence that the reactors were unsafe.

You're correct that the 1970s construction isn't the same as modern construction though. During the 1970s, series of the same reactor designs were built. As opposed to most modern construction, which are typically the only reactors of a given kind built in a given country. Serial production is much cheaper than one-off production.

Electrifying railways and other transportation is going to increase electricity usage, not decrease it. Only about a third of total energy consumption is electric, and most of the rest needs to be electrified as a prerequisite to decarbonization. So assuming that electricity consumption is going to decrease is a very dubious prospect.

I didn’t bring up Three Mile Island to argue that enhanced safety standards back then would have prevented the partial meltdown. I brought it up to demonstrate that it caused a shift in policy which resulted in slow-down in reactor build-up. Three Mile Island happened despite the safety standards at the time. We don’t know the probability distribution of a nuclear meltdown but we can safely assume that it is a function of relevant safety standards and a total number of reactors. USA decided to mitigate the risk by both limiting the growth of new reactors and increasing the standard. This was a logical choice. If USA had done neither other accidents would be happening now with greater probability then it did in the 1970s.

The point I'm making is that the increased cost of nuclear wasn't due to greater safety standards. It was due to lower rates of construction, losing the economics of scale of serial production.

I’m backtracking the debate here so feel free to ignore this.

I just watched the latest Technology Connections video about heat pumps[1] where he talks about the total energy savings in replacing gas stoves with heat pumps to provide residential heating, with total energy savings up to 5 times. This goes to show my previous point about how “[w]e don’t need to replace carbon power 1:1 if we patch up the inefficiencies of our current energy use.”

So to answer your concern that:

> Electrifying railways and other transportation is going to increase electricity usage, not decrease it.

All else being equal, yes. However we are not going to replace the energy consumption 1:1. Electrical railways have the potential to be far more efficient in the energy use. Especially if you consider the potential increase in traction allowing better track usage which will pull super inefficient cars and lorries from the road and onto the tracks.

Also a minor correction:

> So assuming that electricity consumption is going to decrease is a very dubious prospect.

No, I don’t assume that. However I am hopeful that we can easily decrease our total energy consumption as we electrify. Meaning that if we are still pumping gas to put into our newly electrified infrastructure and consumer systems, there is a potential that we need to pump less of it for the same benefits because of how much more efficient these new electrified systems are over the old fossil fuel powered ones.

1: https://www.youtube.com/watch?v=MFEHFsO-XSI

Even if heat pumps are more efficient, replacing gas heating with heat pumps (like what much of Germany is going to have to do) is still an increase in electrical load. Similarly with railway. Electric rail engines may be more efficient, but the existing diesel engines are totally unrepresented in current electrical load.

Even if total energy consumption goes down, huge amounts of fossil fuel usage in transportation and industry needs to be electrifed. The fact that they may be more efficient doesn't change the fact that they are totally unrepresented in current electrical energy usage. I stand by my point: decarbonization means our electricity usage is going to grow significantly as existing fuel-based energy is converted to electrical energy input.

But hitting those big numbers is the whole point of scalability. It may be easier to build a few hundred giga factories than a few hundred nuclear plants. What you get is less useful, but politically and technically simpler. And work in other industries make storage more viable. Those HVDC links that America should build anyway make storage more viable. Reducing costs in renewables make storage more viable. Even reducing costs in nuclear power could make it more viable.

The Tesla giga factory produces 35GWh of batteries per year. It would take ~14 years for a giga factory to produce just one hour's worth of storage. We need days at least, potentially weeks depending on the mix of renewables. Again, the scale just isn't there for batteries.

This is why renewable evangelists assume something like hydrogen storage, or compressed air will deliver nearly-free storage. But that's just hope, who knows if these approaches will actually be competitive.

By the time those 14 years are up we would also have had to throw out the batteries multiple times from degradation.

And 35 factories lets you hit the 12hr mark within five years and deliver an 85% decarbonised grid. That may not be better but it could be more achievable.

> but the fuel has a very dirty production pipeline

Not really. You need very little uranium to operate a plant, and a lot of it comes from decommissioned bombs. When we were mining it in the US, it caused far fewer problems than coal or natural gas.

>Nobody sees nuclear as a viable path forward except the nuclear industry.

This is provably false on many fronts. There is a growing recognition that nuclear is the only low carbon source that can produce substantial carbon-free power in all weather conditions and all latitudes.

"There were over 50 additional nuclear reactors under construction in 2020, and hundreds more are planned primarily in Asia."

https://www.weforum.org/agenda/2022/01/energy-nuclear-power-...

https://ourworldindata.org/safest-sources-of-energy

https://www.weforum.org/agenda/2020/07/nuclear-power-energy-...

https://www.mercurynews.com/2022/03/23/tech-billionaires-ral...

Nuclear may or may not become the dominant energy source, but to say "Nobody sees nuclear as a viable path forward" is simply incorrect.

The 2M number is referenced here[0]

[0] https://blogs.scientificamerican.com/the-curious-wavefunctio...

The study is from NASA: https://pubs.acs.org/doi/abs/10.1021/es3051197?source=cen&

Yeah, if people can throw around "deaths due to air pollution" it seems pretty legitimate to throw around "lives saved due to nuclear".

It might seem legitimate but I disagree that it is. It needlessly promotes one alternative over unmentioned others, giving people the sense that the only mentioned alternative is the only sensible one.

Take a version of the trolley problem for example. A trolley is heading towards group of people tied to the tracks. You have the option to pull a lever and divert the trolley to another pair of tracks, but it will hit an engineer working on those tracks. You state that pulling the lever saves all these lives, but unmentioned is the alternative to derail the trolley to a runaway zone which saves everybody.

In this version of the trolley problem you can count the deaths due to inaction. Is it right however to shift the rhetoric and talk about how many lives saved due to one particular action without mentioning there was an alternative?

It's an understandable impulse to assume that every article is framed in reference to the thing you're passionate about, but considering renewables weren't generally feasible (i.e., hydro and geothermal are older but not generally available for obvious reasons) until relatively recently (and even still, its feasibility for reliable base load generation is dubious), it's a certainty that this article is positioning nuclear relative to fossil fuels, not renewable energy. Even if renewables were available the whole time, it's perfectly reasonable to talk about the lives saved by nuclear even if renewables would save a few more (and to be very explicit, the renewables-only crowd reliably exaggerates the lives saved--some renewable energy sources kill more people than nuclear, even when accounting for disasters like Chernobyl).

By the same logic, Hydropower has saved 3 Million lives in the last 50 years. Efficiency and insulation probably saved even more.

Sure. Did you think this was a counterpoint to my comment? Or what's the relevance?

You said:

> to be very explicit, the renewables-only crowd reliably exaggerates the lives saved--some renewable energy sources kill more people than nuclear, even when accounting for disasters like Chernobyl.

I don't see how that's meant to be a rebuttal. I don't know the figures, but I wouldn't be surprised if hydro (replacing carbon) has saved a lot of lives. My quote was referring to people claiming that per unit energy, renewables are dramatically safer. Hydro is great where available, and renewables are important. There just isn't a compelling reason to rule out nuclear energy at this point.

One thing that always bugged me about the trolley problem is that it omits the question: “Why are all these people tied to the tracks? Who is responsible for tying them? Why am I responsible for pulling the lever, but not the criminal for tying the people? Or even the police for failing to stop the criminal?”

This is the same sense that I get when I see people promoting nuclear as live saving technology. It is not on nuclear plants to save people from pollution, it is on the coal industry to stop polluting.

If people were angels, there would be no need for government. Unfortunately, the fossil-fuel industry is going to pollute unless they're prevented by regulation. The industry selects for people who are willing to put profits over moral responsibility, so society can't rely on them not to pollute any more than we can rely on murderers not to murder.

Reply on news.ycombinator.com