No one knows the actual cost of French Nuclear. France has yet to dismantle one of its end of life reactor.
Everyone know that the provisioned costs don't make sense. It was pointed by the national court of auditors in 2012 and 2014 [1] and again in a report from the national assembly in 2017 [2]. We are not even taking about puny mistakes. Even the actualisation ratio used is garbage.
This is true but people tend to forget one thing: decommissioning doesn't necessarily means dismantling. Dismantling is a tough challenge (mainly because of the civil engineering part btw, which becomes difficult to manage when you add the constraints from nuclear: “What do you mean when you say we can't use dynamite? oO”) but it's only a legal obligation not an absolute one. With government support to change the law, EDF could just unload the fuel (price is well known, they do it all the time), and pickle the primary circuit in place (the price is also known, they did it a few times already) and now you have a place where all contamination (which comes from neutronic activation) is trapped inside the structure (Co60 in the rebar of the structure mostly) and the biggest risk is asbestos… (like your nearest decommissionined building in your neighborhood, even houses).
This option leaves you with a few ugly buildings in the middle of nowhere, but the cost unbeatable (even way less than decommissioning thousands of wind turbines).
> This option leaves you with a few ugly buildings in the middle of nowhere, but the cost unbeatable (even way less than decommissioning thousands of wind turbines).
Aren't you forgetting something? Buildings slowly deteriorate, even if not used. So now you have to maintain the disused reactor, for ... how long exactly?
> Buildings slowly deteriorate, even if not used. So now you have to maintain the disused reactor, for ... how long exactly?
And here comes radioactive decays to the rescue! Half-life of Co60 is a little bit more than 5 years, for instance. The important thing to consider is that dangerous radioactive materials are also the ones with the shortest half-life (because the same number of atoms emits more radiation per second), so as long as you took out the fuel[1] you don't have anything really dangerous for too long. It could still be a problem when decommissioning (because you spill everything out in a short time and you have workers just nearby) but if you let the building just decay slowly, you don't have issues.
[1]: the fuel is a bit special, because it has long half-life but it's still dangerous for two reasons. 1) it's alpha emitters, the worst kind of radioactive substance. 2) the concentration is enormous.
Alpha radiation is easily shielded from external exposure, but for that exact reason it's also the most damaging; gamma radiation, for instance, will mostly pass right through you, while alphas will deposit all of their energy right into your cells. They're shielded by your dead skin, but if you ingest or inhale them, then there's no dead skin to protect you, and all that radiation goes straight into your cells.
In fact, according to the weighting system that converts joules of energy absorbed (Grays) to severity of radiation dose (Seiverts), a joule of alpha radiation is 20 times worse than a joule of X-rays, beta radiation, or gamma rays.
And it's ingestion or inhalation that we're worried about from environmental contamination; not that the environment itself would become so radioactive, Fallout-style, that you'd take rads just from standing around - but that radioactive dust from demolition might get into the air, or that contaminants might leak into the groundwater.
Yeah, but - context. We're talking about leaving a disused reactor around in some type of safe manner. Ideally it won't be leaking... but some radioactive material is still in it. Aren't the beta & gamma radiators more dangerous, even as far as irradiating other parts of the structure itself is considered?
Neither beta nor gamma radiation will "activate" other parts of the structure to make them radioactive themselves; only neutron radiation (which can convert stable isotopes into unstable isotopes with more neutrons in them) can do that.
Nor will they weaken the structure.
So as long as people don't enter the reactor without proper precautions, there's really no reason to worry about radiation inside it. And this problem can be solved pretty well with a fence and warning signs.
> So as long as people don't enter the reactor without proper precautions, there's really no reason to worry about radiation inside it. And this problem can be solved pretty well with a fence and warning signs.
The buildings being designed as a bunker also helps. Just add a single guardian and his dog just to be sure nobody is actively trying to break through the concrete walls and you're good to go.
> I'm also not sure where you got the idea that alpha emitters are the worst kind of radioactive substance; alpha radiation is easily shielded.
Ahah! I considered making an appendix especially for this one because I expected some people to make this mistake, so here we are. Notice that if you're French, the mistake isn't yours but it's the official physics course for French high-schoolers which is to blame.
First of all, alpha rays are helium nucleus, they are really heavy compared to electrons (beta rays) and thus, much more energetic (energy of an order of magnitude of a few MeV vs hundreds of keV), and gamma rays in the case of radioactivity are even lower (40 keV in case of iodine for instance).
If you stand in front of a radioactive source, the radiations comes from in front of you, you can really easily shield against alpha rays (because they are big!), but you can't really shield against gamma rays (because they are just photons), then gamma rays are the most dangerous in that specific context.
But most people aren't physicists or nuclear workers, and you don't usually end up being irradiated by a radio source (the incident you talked about earlier is a good counter example though). The major risk faced by a population is not direct irradiation, it's contamination: that means, you eat food or drink water which contains some radioactive element. And now you get the radioactive source right in your body (let say the thyroid, if we're talking about radioactive iodine). Here, there is no possible shielding, so the total energy is what matters. And regarding the different kinds of rays and their ability to pass through matter, gamma rays have some chance to exit your body without ionizing a single cell, beta have less chance, and alpha have zero chance to go out.
So yes, in terms of radio protection of populations, you fear alpha rays way more than others. And if you operated a plant you are actually allowed to release a little (and subject to regulations of course) amount of beta-emitters (tritium is released in low quantities quite often for instance) but you aren't allowed to release any single atom of alpha emitters.
Also, the list you quote contains fission products (from krypton to C14) as well as activation products. Under normal conditions (I mean, no critical accident like TMI, Chernobyl or Fukushima), most fission products stays in the fuel rod, and then they won't remain in the decommissioned plant.
Zn65 and Fe59 decay quickly, Tritium will be slowly released in the water nearby (yes, that's the normal procedure and it's happening during all the plant's life) and then you have Co60.
The parent talk about how the building material like rebars absorbs the isotopes, and then how long it can afterward be harmful.
Do those isotopes also get absorbed, and if so, at what rate? Since I don't know the physic I don't know if all isotopes can get absorbed at the same rate, but my intuition is that the answer is no. I would also guess that nuclear plants get exposed to different amount of each type of isotope, so the above table would need to include both in order to compare the radiation risk after X years.
Isn't cesium contamination the reason why the descendants of the Bikini Islanders still can't move back to their ancestral home? Same group as potassium so it gets sucked up into fruiting bodies (eg coconuts)
Cs137 is a real problem, it's radioactive enough and at the same time it has a long half-life for this kind of product. And as you say, it's metabolized so it can spread through the food chain…
Fortunately, it's a fission product so it's only released when you melt your reactor or when your reactor is in fact a bomb… (Mandatory reference to the Silly Asses short story from Isaac Asimov: https://en.wikipedia.org/wiki/Silly_Asses).
Because the GP took a list that mixes both fission products and activation product.
Fission products are directly created when splitting an uranium nucleus in half. They stay in the fuel rod unless it's damaged in a catastrophe. With a bomb you don't have rods, so everything is just released in the air.
That list is a bit misleading for anyone skimming; it has a bunch of numbers in years and in days. I was puzzling over why there is no natural Zinc-65 until I read it more carefully.
Half life is almost as confusing as statistics. A halflife of 5 years means 10% of the radioactivity is still there 16 years later. You'll still get 1% of the rems 33 years later, and that's if the materials don't migrate.
Alpha decays are still problematic for any material that can by aspirated or ingested. Demolition means flying dust. A leaky building means groundwater contamination. A lot of these substances are also heavy metals, so even without the alpha decay problem they are highly toxic.
Can you keep a building full of alpha sources water tight for five or six half lives?
There would be no alpha sources left in the plant, as they are confined in the fuel rods, that would be taken away from the plant. (And be managed like spent fuel, which is a totally different problem than decommissioning).
> A halflife of 5 years means 10% of the radioactivity is still there 16 years later.
But it's very likely that 10% is irrelevant.
Radioactivity is not a all or nothing thing. It exists all around you right now.
Camping out at a old abandoned nuclear power plant 20 years after it's decommissioned probably gives you less radiation exposure then you would get from a flight from California to Hong Kong.
Good question. The dismantling narrative probably comes from a mix between wishful thinking from politicians and bad habits of lying all the time from the nuclear industry… I mean, the nuclear story is even brighter if we can promise we'll clean up after us right?
"This is the €25bn deep geological storage facility for France’s high and medium-level radioactive waste... [France] produces enough toxic radioactive waste every year to fill 120 double-decker buses (about 13,000 cubic metres worth, or 2kg a year for every French person)." [0]
This figure sounds excessive, and it's not clear what is counted: no break down between high-level/low-level waste, and does not state if this is before or after chemical reprocessing, if this is before or after vitrification.
Going from first principles, France has 133 GWe of nuclear capacity, and each Gw reactor produces 27t of unprocessed waste a year. France reprocesses used fuel, of which > 95% is un-burnt uranium - this goes back into reactors.
Conservative estimate suggest that they produce <200t of waste a year. That's 3g per person!
Of that, only 1/5 is real long-lived waste.
There is no industrial process that produces less waste. These volumes are easily manageable.
Which is much less waste than solar produces. We barely even try to keep the waste from solar panels segregated from the environment because it is impossible with volumes that high. It is impossible to try to keep it to the same standards as nuclear but the damage is probably about the same.
They've got large quantities of Cadmium Telluride and Cadmium Sulfide, the Cadmium portion of which is quite toxic. Not to mention that burning coal releases a ton of radiation into the air also. This isn't whataboutism, far from it, in an ideal world there'd be no waste. But in the real world we have to reduce harm and balance options against each other. Nuclear waste can be quite effectively dropped into long-term geological storage facilities.
The reality is that no such facilities exist. At the Indian Point reactor in New York, tons of radioactive materials are held in insecure, leaking ponds that are emitting waste into the groundwater and Hudson River. That facility isn’t the worst and isn’t unique.
Ewaste is a problem for sure, but it doesn’t present the same complexity as nuclear waste products.
Almost all of the pro-nuclear arguments depend on an ideal state that never will and never did exist today. We have the reactors that we have because the military was in the midst of a massive buildup of atomic weapons during that timeframe.
As it stands, those plants are often not viable — my state directly subsidizes nuclear plants because the operational expenses and capital costs make the electricity produced more expensive than the open market will pay. And that’s after the taxpayer implicitly taking on the long term expenses associated with decommissioning and dealing with the plants for decades after retirement, at taxpayer expense.
> Ewaste is a problem for sure, but it doesn’t present the same complexity as nuclear waste products.
It isn't complex because it is unsolvable. When large amount of solar panels reach their end of life we are guaranteed to get large quantities of carcinogens in landfill that eventually leaks into the water table. Many sets of landfill are guaranteed to be poorly managed, and even those that are well managed are not scrutinised to the same extent as a nuclear related activity and will be of lower quality.
With nuclear, if we successfully traverse a complicated path then there will be no carcinogens in the water supply in the next 100 years. If we do it wrong there will be small quantities of carcinogens in the water supply.
> Almost all of the pro-nuclear arguments depend on an ideal state that never will and never did exist today.
What we are doing with nuclear today is working better than what we were doing with solar today. There is less waste and the harm it does is less than the harm of solar waste.
Anti-nuclear arguments just don't seem grapple with these questions of scale. Carcinogens sourced from e-waste are at least as bad as carcinogens from radioactive materials. As far as I can tell, large quantities of lead are worse.
> my state directly subsidizes nuclear plants because the operational expenses and capital costs make the electricity produced more expensive than the open market will pay.
This didn't stop Germanny bringing in solar. I've heard a rumour that German solar and wind are triple the cost of French nuclear, last half as long.
Easier, potentially, but the question is also of relative volumes. You need an awful lot of cadmium telluride to produce the same amount of power as a fingernail size pellet of uranium.
Per TWh of generated electricity you end up with an order of magnitude more deaths from solar power. It’s always a balance.
You might be surprised how little cadmium telluride is required. With a 2.5 micron CdTe cell thickness [1] and 17.6% beginning-of-life module efficiency, 25 years operation with linear power degradation [2], that's a lifetime average module efficiency of 16.45%. If it's illuminated to achieve 20% capacity factor -- which is realistic for utility scale solar projects [3] -- a square meter of panel generates 7205 kWh over its lifetime. A square meter of panel contains 16 grams of CdTe. That's 3150 kWh of electricity generated per 7 grams of cadmium telluride over the module's lifetime.
If a 7 gram uranium fuel pellet releases as much thermal energy over its lifetime as a tonne of coal [4], that's 8141 kWh. Assuming a thermal efficiency of 33% for conversion to electricity [5], that's 2687 kWh of electricity from one 7 gram fuel pellet.
A breeder reactor could extract many times as much energy from one fuel pellet's worth of uranium, but the world currently operates hundreds of power reactors and only one breeder power reactor, the Russian BN-800.
AFAIK photovoltaic cells with the electricity producing junction made of pure silicon do not exist. Don't know how much of non-silicon is there, but arsenic from GaAs cells is not something you want in your soil either.
As for WEEE, I hope most of the cells are returned and recycled or somehow properly disposed of, but not all will be collected, and there will be breakage, maybe even leakage of acid-rain digested cells. Also, even now, the collected e-waste is not really processed in a environmentally friendly way.
The cells are made of purified, doped silicon. That means very pure crystalline silicon modified with tiny additions of boron and phosphorus. A chemist would consider them "pure" silicon -- they are more than 99.99% silicon. About 95% of solar module manufacturing uses silicon cells. The remaining ~5% is split between thin film cadmium telluride modules and an even smaller volume of copper indium gallium diselenide (CIGS) thin film modules.
GaAs cells are far too expensive for terrestrial use.
The EOL PV products are just as dangerous as radiation; lead and cadmium poisoning are nasty businesses.
If they were held to the same safety standards as nuclear waste it would require containment; the risk of harm is greater (especially after accounting for volume). PV waste isn't held to that silly standard because then it would be impossible to produce solar panels.
If nuclear waste were contained using the same standard of acceptable harm as solar panels the cost would be a rounding error.
"In an effort to reduce costs, the Energy Department developed a plan to ship nearly 75 percent of the fissile materials in Building 3019, as is, to a landfill at the Nevada Nuclear Security Site by the end of 2014. Because such disposal would violate the agency’s formal safeguards and radioactive waste disposal requirements, the Energy Department changed those rules, which it can do without public notification or comment. Never before has the agency or its predecessors taken steps to deliberately dump a large amount of highly concentrated fissile material in a landfill, an action that violates international standards and norms."
Only because of political opposition. It's been consistently shown to be a very safe long term storage plan. Ironically, efforts to stop the Yucca proposal because of "safety" issues make it so that we must store much of our nuclear waste in metal barrels at the same nuclear power plants they were produced in.
Anti-nuclear activists are right when they say that nuclear energy is politically impossible but only because they're the ones making it politically impossible.
As usual experts do not judge it as simple as that.
Rust and corrosion will attack the strongest container – all they need are the right conditions and enough time to work. Not only that, but metals behave differently (and chemical reactions proceed more quickly) at higher temperatures – such as those produced by the decay of fission products. So the thermal effects also have to be factored in when designing the things.
So here’s the bad news about long-term disposal of spent reactor fuel – and the containers meant to hold it. Nobody knows how a container is going to hold up over even 100,000 years, let alone a million years (the time span required by EPA).
Who cares if the waste leaks out of the metal barrels? It's still going to be in Nevada, hundreds of miles from any human settlement for thousands of years.
> No one knows the actual cost of French Nuclear. France has yet to dismantle one of its end of life reactor.
This.
The sad thing is that the article is probably right about nuclear being preferrable over fossil fuels due to relative priorities of global warming vs. nuclear accidents and garbage disposal.
But this cost comparison against renewables is just comically miscalculated.
In 1995 Germany estimated 3 - 5 billion Euro to dismantle one of the old DDR nuclear plants that was decommissioned in 1990 by 2012. The work has not finished yet. About 1000 people work on this project.
Long-term waste management is another ticking bomb.
Moreover if there is a serious glitch (Chernobyl, Fukushima...), are bets are off. You can obtain an insurance policy for anything, AFAIK even for a space trip, but no insurance company covers such nuke risk.
Bad project management does not equate bad technology. German government also messed up Brandenburg Willy Brandt Airport, it's 7 years late and massively over budget. In UK it takes us 30 years to build a train line.
Those cases don't prove that trains and plains are uneconomical.
There's a common feature of massive complex projects the world over - they pretty much always come late and massively over budget. Then as infrastructure mostly end up sticking around for a century or more, sometimes limiting future choices.
Rail, tube and tram lines will still be there and graded 100 years later - after track replacements, roads are not dissimilar, reservoirs and storage caverns far longer with a turbine replacement or two. Nuclear on the other hand has a much more limited life in which to absorb those overruns.
What has this to do with bad project management? I do not see anything particularly bad happening in this project, while the BER is indeed a clusterfuck that keeps giving.
They are still within their financial budget for this project, will maybe overshoot it by a bit in the end. But that's OK, since such a dismantling had no precedence whatsoever, and it was just an estimate and everybody knew that. They will exceed their time budget by a lot, which is also OK, since again such a dismantling had no precedence whatsoever, it doesn't really matter if this thing takes a decade longer to dismantle, and the reason they take longer than initially estimated is that they had to overcome a ton of problems nobody could have known about before and went about their work carefully and systematically. There were a lot of lessons learned from this, which should help with being faster when dismantling other nuclear plants, but and a bit cheaper (maybe at 2/3rds the cost in the future).
With BER on the other hand the date of completion did matter, both politically and economically, and the exploding time and financial budgets were a result of gross incompetence in particular from the politicians that personally kept meddling in the project (it was "prestige" project), complete failure in contractor oversight (which resulted in a ton of defects and also some corruption and embezzlement) and a completely useless project leadership (of mostly political appointments, aside from the politicians themselves, that e.g. resulted in them building that thing without working fire safety).
But they should increase your bayesian estimate of typical project costs for these types of projects. If we had a way of reliably preventing bad project management such things wouldn't happen.
That some other projects were late does not mean much for dismantling a nuclear power plant, which is a very different/ unique problem with different actors.
Wind and solar projects are not coming in at 3-4x their costs anywhere.
So the fact that they are simply simpler projects which rarely (never?) have massive overruns due to "bad project management" is a huge financial point in their favor that is rarely factored in.
Indeed. A huge and complicated project is a much wonderful playground for involved people (from the planners to the decommissioners). They benefit from it (on all accounts, including intellectual, financial, reputation, tolerance for error...) much more than they would with simple and transparent ways.
Agreed - the lifespan argument is really not portrayed in an accurate way. Besides, France is basically screwed - it hasn’t been able to expand its energy supply beyond population demands (ie it doesn’t and couldn’t support a data center economy). More than likely they will be buying energy rather than generating it in the future — unless of course someone can get the salt reactor commercialized.
Huh? France exports energy, and it has a massive nuclear and hydroelectric industry. Why on earth would it be more screwed than anywhere else in Europe?
Germany is number 1 because of the wind... there are days when they generate more than they can take. Exporting is not really a good measure of energy security — for example, how much do you think Germany can charge for the wind it cannot use? Hint: it’s a negative number. I am a proponent of nuclear - but I wish to see newer more efficient and safer designs (ie salt). When I look at France through the same lens I look at Norway, I see one country electrifying it’s entire car fleet, all of its homes, its entire North Sea oil production, is investing in massive hectare size data centers and still has plenty left over and then I see France who will be lucky to replace its aging nuclear infrastructure just get back to what it has today. I would love to see them continue to invest in nuclear but it’s own stated goal is to drop nuclear to 50% of the overall rather than the current 75%. If the article’s author is correct, that aging nuclear designs are superior to other forms of energy, why then isn’t France keeping its production at 75%?
Today they produce a lot of subsidized power that when everything operating correctly produces an excess of power. However we are not talking about today, we are talking about the future. The average plant age is 30 years. Some plants are so old that frances neighbors accuse it of risking another Fukushima incident.. Germany closed its plants for precisely that reason. It will like cost close to 100b if not more to decommission and untold amounts to replace whatever they take offline. My point is that just for France to unwind it’s aging nuclear it’s going to have to build and spend a lot and that is just to get back to even... much less expand to cover new sources of electricity like electrification of the car fleet. Germany in contrast has been building new transmission lines and production for a decade and still has a decade to go. Norway actually has a true surplus which is expected to be a multiple of france’s entire output, etc etc. My point was that the author was saying nuclear has an expected age of 40-80 years... I call bs on that lifespan and point to France as an example: there’s just no way we see them continue their existing plants for 80 years when their existing plants are already showing signs they may be at risk or not be able to operate much past 35-40...
They are doing less sorting. They just get a bigger cave and put the entire facility into it. There is room in the large cave for all US nuclear plants.
Plants have been dismantled in the past, not in France though. I think closing Fessenheim is a bad decision, but at least it could shut down this bad argument by showing that costs of decommisioning a nuclear power plants are not 3 years of GDP.
Quotes don't matter when it doesn't help the argument I'm making on the internet.* Meanwhile projections are used extensively in major engineering projects. The fact error rates exist in probabilities doesn't make them useless.
The parents point is sound in that estimates only have validity if they are about well understood or frequently undertaken activities. There is no basis for certainty about decommissioning of nuclear; older plants especially have unique problems. There have been enormous cost overruns for any plant that has had an incident, and many plants will have problems only discovered upon disassembly.
Problems unique to older plants have no relevance to the estimates for newer plants. They're not even relevant to the choice whether to keep existing plants in operation, except insofar as pushing the decommissioning date further into the future lowers its net present cost due to the time value of money.
Ah but they do! As we discover those unique problems and how to solve them and how long it takes and the cost, our a priori cost estimate improves. The total numbers are so small it is really hard to say what the reliablity is. Newer plants tend to have more commonality and few parts, but we are still to discover all the many fantastic failure mechanisms. There are only 7 active Gen III reactors in the world, so it will be some time before we gain statistical confidence -- well after future designs are finalised.
> Ah but they do! As we discover those unique problems and how to solve them and how long it takes and the cost, our a priori cost estimate improves.
But it's not just our cost estimate that improves. Gen III reactors were designed with more data on how Gen I reactors were decommissioned than was available when Gen II reactors were being designed, and so on.
Finding a new decommissioning snag affects how you design the next reactor you build, but it doesn't really affect its predicted decommissioning cost because by that point you're aware of the issue and take steps to prevent it from occurring for the generation now being constructed.
Not to mention the cost of permanently evacuating an area when the impossible event of a meltdown manages to happen anyway. There's a reason why countries like to put their reactors near the border, that way part of that area is someone else's problem.
Exactly. The real costs of France's national nuclear program can be hidden inside of the accounting of the entire nation, and fudged beyond belief. It's a point of national pride at this point, they would definitely lie and take to the grave anything that might suggest it's not an overwhelming success. And given how fudged financing can be easy to do and compound into massive errors, it could take a massive team of financial forensic analysts to uncover it.
I'm a very big fan of nuclear and I believe it should be subsidized at high cost as necessary. However, it is clear that even France has extreme inbound cost problems related to its nuclear industry -
"French power utility EDF estimates it would cost at least 46 billion euros ($51 billion) to build six of its latest generation EPR nuclear reactors if the government decides to build them"
"the Flamanville EPR reactor under construction in northern France has been plagued by cost overruns and a series of technical problems resulting in years of delays. EDF, in which the state has an 84% stake, said in October the project which began in 2006 would cost 1.5 billion euros more than previously expected, raising the total cost to 12.4 billion euros."
France did and does heavily subsidize nuclear, it pretty much paid for the whole fleet out of government pockets when it was originally built in the 70s [0]
And because that was expensive, there isn't enough money for decommissioning left, which needs equally massive subsidies [1].
What EDF estimates is meaningless without a strike price and decommissioning cost.
The UK's National Audit Office puts the price of one EDF reactor around that once you include the excess price paid on every MWh over the 50 year projected life, and the decommissioning (which is supposedly built into that MWh price). It remains to be seen if it will be correct. From Wikipedia:
The National Audit Office estimates the additional cost to consumers (above the estimated market price of electricity) under the "strike price" will be £50 billion, which "will continue to vary as the outlook for wholesale market prices shifts"
That's $64bn on top of £20bn ($25bn) latest build estimate. A build with such poor worker conditions that it has gained a reputation for suicides.
$89bn would build a ton of pumped storage to underpin offshore wind at a quarter the MWh price.
The Hinckley point nuclear deal is comically badly managed, and two EDF executives were so opposed to it, they resigned before the contract was signed.
All of it is down to the creative accounting of the conservative government to demonstrate that they eliminated the deficit.
Instead of financing the construction directly and purchasing the plant as "turn-key" asset, they forced EDF to take a multi-decade loan and agreed a price of electricity decades in the future that would enable EDF to pay back that loan. This doubled or triples the cost.
What's 'a ton' of pumped storage? Thats doesn't mean much. Can it can back up at least 16GWh, the equivalent of a single nuclear reactor's output over night.
I have never heard of a project of this magnitude that wasn't just on the drawing board.
The reason Hinkley was structured as it was is because every other nuclear decomissioning results in socialising the cost to the state. This was intended to ensure that was not the case - despite the massive state subsidy. Which rather brings us to the normality of nuclear - the costs are always indicative of comical mismanagement, overrun and very often completely ignoring the decommissioning entirely. So it never, ever brings electricity at the alleged cost it can in the developed world - decommissioning makes a complete mockery of the figures, every damn time.
Considering Dinorwig cost £425m in the 1980s to get around 10GWh, even allowing for inflation and construction costs rising much in excess, there should be easily enough for multiple times 16GWh from $89bn even with major overruns. Hence "a ton". Probably enough to derive most generation from wind with more than adequate storage underlying it.
Clearly that needs suitable geography for several sites, but considering there were two backup sites within 10mi of Dinorwig should the main site prove unsuitable, and given the geography of the UK those should not be lacking. I know where I would spend my money given the outlook for renewable costs - absolutely not on nuclear. I was once quite keen in my naive twenties.
That is assuming you have the suitable terrain and don't have a problem with forcibly moving the inhabitants - its not the 50/60's the locals might not just pack up and move.
In some or even many cases neither need necessarily apply.
It doesn't always have to be a populated valley that gets dammed. There's a decent selection of disused quarries, maybe some of the Highland lochs that have potential to create something like Dinorwig at minimal disruption to environment once complete. What I don't know is how those constructions compare in today's money to alternatives. Still, £70bn brings an awful lot of choices, especially when lifespan is indefinite, just periodic turbine replacements.
If the geography isn't available there's the Netherlands project that's pumped storage on entirely flat land. Involves building an artificial lake, with a cavern and turbines below it. Suitable rock for the cavern is quite a long way down, which adds to the cost, but also adds to the head so it should end up with significant capacity. As far as I know it is the first such scheme, so costs are projection only at this stage, but it looks pretty interesting if it turns out to be viable economically: https://o-pac.nl
Or the UK company that has a plan involving winching weights (thousands of tonnes) up and down disused mine shafts. The open question of course is how many of the world's thousands of disused mine shafts are going to be viable. If viable the advantage seems to be very low surface profile.
All these large-scale energy storage projects are in their infancy. I haven't heard of anyone breaking ground on a project multi-GWh capacilty. Until that happens, it's all hypothetical.
Also, I am no expert on the subject, but digging out underground reservoir sounds expensive.
Dinorwig is 10GWh. There were two backup sites chosen in the same immediate area to cover the case the main site had proved impractical. Many years have passed so it's unlikely they're still there, but geography is not lacking if the political will were there.
Given rule changes even without devolved government repeats of the Liverpool reservoir would be much harder. Dinorwig used already existing disused quarries - not flooding a pristine valley.
In the case of Scotland it's been the Scottish government looking at using lochs. No idea what the case is in Wales.
That seems to be the plan - enough capacity to store the whole topside lake, which seems to be planned at 500m x 500m (depth 10m). No idea how feasible or economic it will be compared to caverns from hollowing out a mountainside.
Everyone know that the provisioned costs don't make sense. It was pointed by the national court of auditors in 2012 and 2014 [1] and again in a report from the national assembly in 2017 [2]. We are not even taking about puny mistakes. Even the actualisation ratio used is garbage.
[1] https://www.ccomptes.fr/sites/default/files/EzPublish/201405...
[2] http://www2.assemblee-nationale.fr/documents/notice/14/rap-i...