Energy density is a physical argument that is pretty sound. If you get more energy from less material and land, that can lead directly to a smaller land/material/waste/emissions footprint and therefore a smaller overall footprint. Nuclear being borderline magical here due to E=MC² is the key argument in support of splitting atoms for civilian power. The large land and material footprint of low-density renewables is only barely beginning to be noticed, though they still produce less than 5% of the world's energy. At scale it will be serious, mark my words.

EROI for renewables, especially as it decreases substantially with added storage at scale, is by some accounts (the paper linked) troublingly low. If you have to put in roughly the same amount of energy into a system as it produces in its full lifecycle, you run into issues. This is another good physical argument for nuclear.

One can state they're failed arguments, as anti-nuclear propagandists always do, but that alone is not, as David Deutsch would say, a "good argument".

Energy density is only important insofar as it affects cost. So, if we have cost data (which we do), pointing instead to energy density is pointless (and is likely a deliberate attempt to distract from cost numbers that the person making the argument does not like.)

> If you have to put in roughly the same amount of energy into a system as it produces in its full lifecycle, you run into issues.

Renewables are nowhere close to that. Claims otherwise are lying. For example, they pretend that massive storage is needed and then poorly optimize the storage system so building it requires lots of energy.