Ah, the old "it hasn't happened yet, therefore it can't happen" argument. I'm sure if you honestly think about it you can realize why that argument is nonsense.

> (3 out of the 4 can be intermittent/dependant on environmental conditions and are not reliable).

We're talking here about power for hydrogen production. Intermittency doesn't matter much for that (it requires the electrolysers be cheap, but their costs have plummeted recently, so that's no longer an obstacle.)

> Perhaps I'm unaware of some technological breakthrough which can produce hydrogen using little energy?

Perhaps you could stop using strawman arguments? Hydrogen doesn't have to be produced with little energy in order for hydrogen storage to be viable. The energy that is used just needs to be sufficiently cheap. You know, like that $0.013/kWh power from large scale PV in the UAE? And that's only the latest in a constant string of record low bids.

> Nuclear is great once up-front costs are paid.

Yes, if you can get the Nuclear Fairy to wave a magic wand and make the power plant appear for free, it's much less expensive (but even then, the operating costs will be much higher than renewables). In the real world where real money has to be paid to build nuclear power plants, not so much.

> I have pondered whether it would be possible to generate hydrogen more cheaply via concentrating solar or geothermal.

Neither of these are competitive with PV now. Hell, it's now cheaper to make domestic solar hot water with PV and a resistive heater than it would be to use solar thermal collectors (although a heat pump water heater would be cheaper still). PV has gotten THAT cheap.