Why are nuclear power construction costs so high?

A project should garner favorable financing arrangements for its merits, not for its risks.

You're not wrong, but I'm not sure what's a better alternative.

Think of it in a different scope: government contracts are also expensive because the have to be open to competition. A lot of the red tape could be reduced with no-bid contracts, but people understand the corruption risk tradeoff is generally not worth it.

In your example, it seems like the NIMBYism is the root cause, not the process by which NIMBYism is wielded.

Huh. I'd argue that they should be state-sponsored because that is what the state is for-- investing in long-term infrastructure where the value is widely diffused and thus hard to capture via market mechanisms, i.e. investments which are foundational to the more focused investments which business is so well suited to maximize.

similar to things like interstate highways or internets or public schooling. I'd add "healthcare" to the list, but some countries haven't realized this yet.

Reactors may sometimes be shut down at sea to test backup generators and run reactor restarting drills, but it would not be done otherwise. Aircraft carrier propulsion comes from steam from the reactors - if the reactors are offline the ship cannot move. Moreover, the backup generators burn jet fuel, which is convenient because aircraft carriers already have a store of jet fuel for the planes and so don’t have to carry extra fuel for the backup generators, but it’s very expensive, and not something that would be done outside of an emergency or testing emergency preparedness.

This article is very light on the basics of nuclear fission and could use some help there. Here's a good source on background:

https://world-nuclear.org/information-library/nuclear-fuel-c...

To summarize, light water reactors rely on the production of uranium fuel rods, which hold uranium enriched to about 3% U-235 relative to 97% U-238. Naturally occuring uranium ores are about 0.7% U-235 and there is a large variation in the percentage of uranium in a given ore by total rock mass, with a few deposits being as much as 18% uranium ranging down to about 0.1%, which most sources describe as the economically recoverable limit. This will affect the cost of refueling a LWR (which has to be done every ~3 three years).

Fission in LWRs is due to slow thermal neutrons, which can only fission U-235 and Pu-239 (odd-numbered isotopes. Fast neutrons are a different story, see above source for that.) These slow neutron/U-235 events generate fission fragments (atomic masses in the range ~80-110 and ~130-150 with peaks at 95 and 135), gamma rays, and free neutrons. The initial energy distribution for heating the circulating fluid (water) is about 85% fission fragment kinetic energy, and about 15% gamma ray and neutron kinetic energy.

Some of the neutrons are captured by U-239, forming plutonium-239 (and other transuranics) - which is also subject to fission, and over the lifetime of the fuel, about 66% of this formed Pu-239 is itself fissioned, adding to the total energy output.

However, the actual heat produced by the reactor is also due to long-term decay of the fission fragments inside the fuel rods (about 6% of the total). This latter 6% is important because even if you halt the initial fission process, the reactor won't just go to zero, it still has to be cooled to prevent overheating and meltdown, as do the 'spent' fuel rods. It takes about ten years for used fuel rods to go from 10 kW decay heat/ton to 1 kW decay heat/ton. Storage of used fuel rods adds significantly to the operational costs of the reactor over time.

Long-term storage of used fuel is necessary almost entirely because of the transuranics, which are alpha-radiation emitters with half-lives of thousands of years. Most of the fission fragments appear to decay via faster beta/gamma processes.

Finally, there are the activation products to consider. Tritium is formed in the primary circulating water coolant loop, and is highly radioactive which is why this loop has to be isolated and its heat transferred to the secondary coolant loop which drives the steam turbines. Note here that reactors have to use a lot of water, at least as much as a coal-fired power plant, and these systems need to be highy engineered to prevent breakdowns, which would lead to meltdowns (Fukushima failure mode). The other activation products form in the reactor itself - carbon-14, cobalt-60, iron-55, nickel-63. This significantly adds to the cost of nuclear reactor decommissioning as the entire reactor body has to be treated as high-level waste.

These factors explain why nuclear reactors have to be overengineered relative to traditional fossil fuel power plants, oil refineries, etc. which regularly suffer major accidents and fires - but those accidents don't lead to 100-year+ exclusion zones around the accident sites, so it's deemed acceptable. Not to belabor the point, but wind/solar/storage also has much lower costs for these reasons. Security vis-a-vis terrorism, cyberattack, military conflict, etc. is also a major related cost.