the fact remains that each kWh of wind displaces a carbon-based kWh

Yes, but it doesn't necessarily displace the fuel consumption and the CO2 emission at the same degree. It all depends on how reliably wind farms can commit to a specified production for up to the start-up time of the back-up fossil fuel plants. Otherwise, the fossil fuel plants must remain in hot stand-by with a very high fuel consumption while doing nothing.

It looks like this is what happened to Texas. They got caught with their pants down, the wind on strike and no planning between wind producers and fossil-fuel producers.

Even the fastest combined cycles gas plants take more than four hours from a cold start to come on-line at their full rated power if you want to remain within somewhat acceptable NOx discharge levels and stress on the turbines (General Electric US patent 6978620B2 2004).

I genuinely have no clue what's the reality of committed power reliability for a wind farm so I'm not going to throw the book but that's something wind proponents need to address seriously with hard numbers.

---

Regarding nukes, when a serious modern plant aka Areva EPR or MHI APWR goes off line, it's 1,650 to 1,800 MW that go MIA, not 1,400 MW. Don't be demeaning to nukes :)

That being said, unplanned and unwarned outages for internal causes are really a rarity with nukes. They are very reliable on that respect.

Out-of-the-blue outages are in the largest majority from external causes like what happened in Florida few days ago: a glitch on the grid, load shedding, loss of grid and the reactor scrams for safety reasons.

But that's more a regulatory requirement and a legacy from the integrated operator days than a technical necessity. Nukes were built at a time when the grid was assumed to be extremely stable and operated under a single operator. Hence this type of event was considered very rare and addressable in a centralized manner. Hence the regulatory requirement for immediate scram on loss of grid and the fact there is no provision in the plant for load bypass.

A variable steam bypass on the secondary circuit around the turbines directly from the steam generators to additional high-temperature condensers would solve that issue with no additional cooling requirement in the case of air cooling towers. For sea-water or on-the-flow cooling, a larger cooling provision is required to maintain acceptable discharge temperature. If the load drops or goes away, the plant just pisses the unused heat away. A bit stupid but fast and not very problematic given the very low cost of nuclear fuel. The real issue would be changing the regulations to allow that.

Otherwise, most of the remaining unplanned outages are related to things like limits on water discharge temperature but that kind of outage takes places over hours or days and gives enough time to start back-up fossil-fuel plants from cold (US context) or ramp-up other nukes (France). The rest - mechanical breakages, false alarms and the like - are really rare past the shakedown period on a new plant.