By this argument, nothing is zero emissions and the phrase zero emissions is worthless.

The process of nuclear fission is zero-carbon (unlike combustion) and the lifecycle carbon emissions for nuclear are 12 gCO2-eq/kWh compared to 40 for solar, 490 for fracked natural gas, and 820 for coal.

It's easier and meaningful in my opinion to just call nuclear zero carbon.

I don't think that is a valid argument. Take the extreme case of that some energy source requires huge amount of CO2 to build, but produces no emissions during operation. Is that a zero emission source? Also for nuclear we need to include the uranium mining, and ideally we would account for waste storage as well (not sure if that goes into current estimates).

Now the OP argument disingenuous because he also included emmissions of other sources into wind and solar. However as many have already pointed out it would be much easier to run a grid on renewables only than at nuclear only, because for renewables you only need a large enough grid and overprovisioning. Nuclear on the other hand is very bad at handling variable loads, so you will essentially have to overprovision to the the peak demand and run your plants at that demand and burn the energy somehow. Note that Belgium used to do that by illuminating all their highways at night.

I think you're missing my point. I'm saying that nuclear fission is indeed a totally zero carbon process, but the life-cycle emits 12 gCO2-eq/kWh. The term life-cycle is a technical term from the academic concept of Life Cycle Assessment (LCA), which includes all the energy and emissions from the processes of mining, fabrication, construction, operation, and decommissioning.

The life-cycle numbers I give are sourced directly from an IPCC meta-analysis [2]. With nuclear, the entire life-cycle is extraordinarily low carbon even though it requires lots of concrete to build because you get about a billion Watts 90% of the time for 60 years given that much concrete.

As for what's easier, the common answer seems to be that variable renewables are ok for a while but at deep decarbonization, the situation is much easier and cheaper if you include some kind of low-carbon energy that can run 24/7, aka a "firm low-carbon source" [3].

[1] https://en.wikipedia.org/wiki/Life-cycle_assessment

[2] https://www.ipcc.ch/site/assets/uploads/2018/02/ipcc_wg3_ar5...

[3] https://doi.org/10.1016/j.joule.2018.08.006

By your definition then none of it is zero emissions. Including wind, solar, hydro, or geothermal. "Zero emissions" is a common, agreed upon term for those sources, and therefore so is nuclear because they all have comparable infrastructure related carbon emissions. Please don't be pedantic.