It does not, in fact, need "very favourable natural geography". It needs an elevation difference, something found almost everywhere, Holland and Kansas excepted. Kansas has deep underground cavities, which also suffice.
Building dikes is cheap and low tech. Dikes predate written language.
On the depth thing, I really only meant to mean you need lots and lots of water. You could havw that shallow, and take over a huge area of land, or or smaller amount of land if you have deeper water.
Perhaps your and my definitions of favourable geography are different. You need a water source, due to evaporation, you need a big resovoir at elevation, and another lower. If you don't want to build it all yourself you need valleys that you can close off at one end. You need ground that is not permeable, so the water you have stays where you want it.
Perhaps I am misreading you, but you seem a bit agitated with your in fact writing style. We're all friends here. I am not anti pumped storage. Far from it. If you can make it work economically, it's a great solution. It will definitely be part of the mix. But there's reasons it's not being rolled out everywhere, and a lot of this is due to cost and unfavorable geography.
The main reason it is not being rolled out much is that it is not time yet to roll it out. You need enough spare renewable generating capacity to charge it from, first, which we are very far from, most places. In the meantime the right place to spend is on generating capacity. Later, we will know better which storage methods are best.
My understanding is that it only becomes competitive with other options at a very large scale. A 10 m deep football sized reservoir, at an altitude of 1km above the generator would have approx 13Mwh of energy storage at 100% efficiency (if my math is correct). Battery prices are around $140 per kWh, so a 13 Mwh battery installation is going to be in the vicinity of 2 million dollars.
I would love to see some costs involved in building two man made ten meter deep football field size dams, a large 1km length (it will be quite a lot longer due to it running on a slope) with all the required engineering to run it down a steep incline. The add to that the and generating equipment, and pumps.
Once you've done that, we could compare the operating costs of the two options.
As for nukes, stable costs make them proportionally less competitive by the day, in the face of cheapening competition.
The the size and rainfall season matters for how long they can run. Cape Town has a winter-rain season that runs from roughly May to September.
https://en.m.wikipedia.org/wiki/Steenbras_Power_Station
The capacity of the smaller of the two dams is 3,560 megalitres. I think a 1km X 350m X 10m deep is this capacity. So "small" is fairly subjective term, but you are right that it is not gargantuan. It is still not going to be easy to place such a setup just anywhere.
I would be interested in how much it cost to build and run to obtain the 180mwh capacity. A similarly size gridscale battery would cost in the vicinity of 25-30 million USD (using the $140 per kWh cost).