That's the whole land belonging to the site, most of which is nature, not the plant itself, which looks like less than 200 acres, including the car parks.

In any case, it's not close at all either for that specific example or in general, not least considering that a nuclear plant can produce 24/7 and that solar often does not reach peak production because of the weather.

"A solar PV facility must have an installed capacity of 3,300 MW and 5,400 MW to match a 1,000-MW nuclear facility’s output, requiring between 45 and 75 square miles. For comparison, the District of Columbia’s total land area is 68 square miles." [1]

[1] Nuclear Energy Institute @ https://www.nei.org/news/2015/land-needs-for-wind-solar-dwar...

Look through several nuclear locations and ~1000 acres is fairly typical. In theory they could be more compact, but security concerns etc means nobody puts 2GW of nuclear on a 200 acre site.

1,000 acres, https://en.wikipedia.org/wiki/Beaver_Valley_Nuclear_Power_St...

1,782-acre, https://en.wikipedia.org/wiki/Byron_Nuclear_Generating_Stati...

2,767 acres https://en.wikipedia.org/wiki/Callaway_Nuclear_Generating_St... for just 1,190 MW.

Some are significantly more compact for example 391-acre https://en.wikipedia.org/wiki/Catawba_Nuclear_Station, but that’s surrounded by water.

To put this in comparison, a solar farm would need 50,000 to 100,000 acres to match the net output of those nuclear plants. Plus the land used for energy storage, which would typically involve damming rivers which has its own set of ecological effects.

And it's possible to build much denser plants. The Shin-Kori facility in Korea is massive energy relative to the land it uses. Nuclear power is far more energy dense than solar or wind. This is just thermodynamic fact.

Try running the numbers for actual lifespan.

Callaway is 1,190 MW on 2,767 acres at 87.70% over it’s useful lifespan, but that’s ignoring permits + construction = 10 years and decommissioning which takes ~30 year. Even a generous 55 year operating lifespan is still reduced to 87.7 * (60 / (60 * 10 + 30)) = 48% capacity factor. Using a 20% capacity factor for solar (aka non tracking in a good but not great area).

That’s 1,190 MW * 48/20 = 2856 MW. A very good modern panel is hitting 220w/m2 add spacing, equipment etc, and 110w/m2 is a safe bet. That’s 2856 * 1000 * 1000 / 110 = 26,000,000m2 or 26 km2 or ~6,424 acres solar vs 2,767 acres nuclear. Lower efficiency panels bump that by 25% or so.

Clearly a win for nuclear, but not a 100,000 acre win.

PS: That said, this is largely a moot point as even with reprocessing we would quickly run out of fuel with large scale indoor farming.

30 year decommissioning is a very pessimistic assumption. Also most new nuclear plants have operating lifetimes closer to 80 years than 55.

You also have to adjust for the overproduction necessary to make intermittent sources a reliable primary provider of electricity. In far north or southern latitudes, angle of inclination is such that solar panels collect ~70-50% less energy than near the equator. This gets worse with seasonal fluctuations, which are more extreme the closer to the poles you get. Add weather on top of this and it can drop even further.

This might not seem like too big an issue, but keep in mind that most of the world's electricity consumption is in North America and Europe. These places don't have as good weather for solar.

100,000 acres is roughly the ratio that the article came up with, a factor of 75.

> PS: That said, this is largely a moot point as even with reprocessing we would quickly run out of fuel with large scale indoor farming.

No, we wouldn't: https://www.forbes.com/sites/jamesconca/2016/07/01/uranium-s...

30 year decommissioning is somewhat optimistic average based on current plans.

https://en.wikipedia.org/wiki/SAFSTOR “For nuclear power plants governed by the United States Nuclear Regulatory Commission, SAFSTOR (SAFe STORage) is one of the options for nuclear decommissioning of a shut down plant. During SAFSTOR the de-fuelled plant is monitored for up to sixty years before complete decontamination and dismantling of the site, to a condition where nuclear licensing is no longer required. During the storage interval, some of the radioactive contaminants of the reactor and power plant will decay, which will reduce the quantity of radioactive material to be removed during the final decontamination phase.”

For example, Crystal River 3 (Florida) “Duke Energy announced in Feb-2013 that the Crystal River NPP would be permanently shut down.” “Systems Removal & Building Remediation(2070–2072)“ and that’s if things go well. https://en.wikipedia.org/wiki/Nuclear_decommissioning

As to indoor farming.

“It’s not just that the 4 billion tons of uranium in seawater now would fuel a thousand 1,000-MW nuclear power plants for a 100,000.”

That sounds like a lot, but of you want ~100w of power per m2 that’s 0.1GW of power per km2. So your 1,000 GW power plants are only replacing 10,000km2 of farm land. Meanwhile agriculture takes 51,000,000km2 worth of land. In other words replace 20% of global farmland and you got ~100 years worth of uranium from all the worlds oceans, it’s replaced by rocks.

But, “And those rocks contain 100 trillion tons of uranium.” gives you 2,500 years which is not bad, but that’s not going to be replaced.

PS: You might be able to beat 100w/m2 indoors, but remember this is also for 20% of farmland.

> But, “And those rocks contain 100 trillion tons of uranium.” gives you 2,500 years which is not bad, but that’s not going to be replaced.

How did you arrive at these figures? I think you missed a conversion from pounds to tons.

Current global uranium consumption is in the hundreds of millions of pounds annually - hundreds of thousands of tons. And this is without reprocessing. Nuclear power already generates 10% electricity globally. Even if we assume a 200x increase in consumption from 200 million pounds to 20 billion pounds that still only 10 million tons of consumption annual. 100 trillion divided by 10 million is a lot more than 2,500.

But surely they don’t just own the “nature” land for the fun of it. I assume it could not be used for other things, like housing, or another nuclear power plant!

The difference with solar/wind space requirements is so big that it does not matter.

If the land is reserved and left to nature that is still a positive thing because that helps the environment and biodiversity. But I'm sure that they could put solar panels on it if they wanted.

The point is that this land is not required for production and in any case solar require much, much more land, so, no, it's not even close.

Winds density is off the charts as you can use the ground around it for crops or even solar panels. https://thumbs.dreamstime.com/b/wind-turbines-corn-fields-ae...

Wind still requires a huge area covered with wind turbines to match a single nuclear plant. Each of which takes space, needs to be accessible, and needs to be connected to the grid through some facility. I would not call that a density off the charts... (see link in my previous comment as well)

It doesn't really make sense to talk about land usage that just so happens to be owned by a nuclear plant, or even certain fixed land requirements for a nuclear plant. It only makes sense to talk about the limiting behavior of land usage for various types of power plants.

Solar has hugely larger requirements than nuclear however you look at it, contrary to what was suggested above.

How much land does rooftop solar require?

You're saying that a couple of 50 year old reactors when including a safety zone and adjusting for lifetime capacity outputs about twice the maximum power of commercial solar panels before adjusting for capacity factor.

When you adjust PV for capacity factor the difference ends up at over an order of magnitude.

Thought: Put PV inside the "safety zone" to increase output and use "unusable" space.

You’re skipping permits +construction ~10 years and decommissioning which seems to take 30+ years on average though few have actually finished the process. https://en.wikipedia.org/wiki/Nuclear_decommissioning

75% capacity factor * 50y / (50y + 10y construction + ~30y decommissioning) = 42% capacity factor which is higher than solar but not by that much.

But let’s assume you’re at a 1.5x capacity factor advantage. So 1.5 GW of solar = 1.0 GW of nuclear. 1.5GW / 220w/m2 = 2.9 square miles of panels plus panel spacing and whatever infrastructure is needed. Double it to be really pessimistic and your under 6 square miles.

>"So 1.5 GW of solar = 1.0 GW of nuclear."

That's not how it works. PV capacity factor is 11-12% to the 90-95% of nuclear meaning you need to install about an order of magnitude more PV effect not 1.5x.

Also one batch of PV panels equivalent to a reactor cost 6x as much and with a lifetime of 20 years to the 60 for a nuclear reactor so you will need to replace these at least twice. All 32.5 Million of them.

Replacing panels is just a cost question which dramatically favors solar. 2c/kWh is massively better than any nuclear reactor ever built or operated.

As we are talking land use, construction and decommissioning is a major hit to nuclear. But with solar you can operate continuously by just swapping panels and replacing wires etc as needed.

Basically, in steady state a 50 year nuclear power plant spends the first 30 years with the previous reactor being decommissioned, and it’s last 10 years with it’s replacement being built. Effectively you need 1.8 locations for a single power plant.

Also, I used a specific 75% capacity factor over it’s operating lifespan in that calculation. Some are higher, but the trend is down over time so you want to use an old reactor as your baseline. Here is 72.5% Decommission date 29 June 2020 https://en.wikipedia.org/wiki/Fessenheim_Nuclear_Power_Plant

Just no. Reactor buildings take up a small fraction of the total site area. Constructing additional reactors on existing sites has been done for about as long as nuclear power has existed and multiple reactors coexist simultaneously at the same site.

>"Replacing panels is just a cost question"

Sure, I bet you won't need more than 2 guys and a pickup truck to install, maintain and replace the 120 million panels it would take to create the equivalent to a site with 4 modern reactors. After all, that's only 35 000 panels to install every day if you want to build the site in 10 years. Everything is magical with solar.

>"2c/kWh"

That's a hypothetical figure for 2050.

There are actual current contracts that agree to sell solar at under 2c/kWh. Construction of multiple reactors on the same location is very common, constructing a new reactor next to a reactor being decommissioned isn’t.

https://en.wikipedia.org/wiki/Crystal_River_Nuclear_Plant https://en.wikipedia.org/wiki/Rancho_Seco_Nuclear_Generating... https://en.wikipedia.org/wiki/Shippingport_Atomic_Power_Stat... https://en.wikipedia.org/wiki/San_Onofre_Nuclear_Generating_...

Etc

I mean in theory sure it seems like it should happen, but in practice not so much.

Please link. I've read about some far future contracts relying heavily on public subsidies but even those were at a higher price. At that price, earnings would (depending largely on geography) be something like 160-400 USD / 1.8m² panel over a 20 year period. This needs to cover panel cost, inverters and electrical infrastructure, installation&decommission, land rent, loss due to some prematurely dead panels, administration etc.

I would presume it's a matter of very few new nuclear reactors being built. Some smaller sites will be decommissioned regardless though.

Subsides are common for all forms of electricity generation, but LA’s 2019 1.997c/kWh power purchase agreement is hardly the far future. https://www.pv-magazine.com/2019/06/28/los-angeles-seeks-rec...

Anyway to use a specific example Capacity factor 26.6% (average 2015–2018) https://en.wikipedia.org/wiki/Topaz_Solar_Farm.

Bulk panels are $0.29 per W or 290$ per kW. PS-P72-330W https://sunelec.com/

If 1kw is worth 0.01997 $/h * 24 * 0.266 * 365 * 20 = 930$. Of course they don’t instantly break exactly in year 20. They lose efficiency over time so averaging ~90% over 25 years which is ~100% over 22 years. https://sunelec.com/wp/wp-content/uploads/2020/09/P72_outlin...

I can’t find an exact breakdown of costs spent on panels vs interest, instillation, inverters etc. But, at most we are talking about fractions of a cent worth of subsides to hit 2c/kWh.

>"0.01997 $/h * 24 * 0.266 * 365 * 20 = 930$"

That's a calculation based on 1kW of installed capacity, which is about 3 of those 345W panels so the earning per panel over its lifetime (in sunny southern California btw) is about $310.

Also the $ sign goes in front of the number.

Also, here's a B.Sc. paper from my alma mater (English abstract in the paper) http://www.diva-portal.se/smash/get/diva2:1236618/FULLTEXT01... that puts the current all inclusive production cost of nuclear power in Sweden (incl. funding for permanent spent fuel storage) at $0.03/kWh. That's an actual cost for stuff that actually exists today.

The inability to replace nuclear plants is actually the driving factor behind it's poor public perception. If you keep running untested 50 year old power plants on the brink of failure why would anyone be surprised that some of them do end up failing? If you could just turn the old plants off and replace them all accidents would have been prevented.