That really depends on what percentage of water's potential energy is expended in small creeks and the like vs big rivers.

I actually have no idea how to estimate that. Anyone have a good estimate on that? Intuitively, I'm thinking that most of those rivers are fed by many small streams, so it would be approximately equal. I'm entirely unsure of that though so treat it as having huge error bars.

The problem becomes one of how spread out the energy is.

Sure the water in that creek might be falling hundreds of meters, or even kilometers, but without a dam the height differential at your generator is only going to he a few metres at best. Enough to power the lighting, fridge, and washing machine in the houses that the creeks happen to run by, but not much else.

With the advent of long lasting perovskites and (hopefully) some less polluting battery technologies you're not even better off cost-wise.

You might be able to do something with a long pipe parallel to the creek to get more head, but then you're going to spend a lot of your energy on wall friction unless it's very wide.

> Sure the water in that creek might be falling hundreds of meters, or even kilometers, but without a dam the height differential at your generator is only going to he a few metres at best. Enough to power the lighting, fridge, and washing machine in the houses that the creeks happen to run by, but not much else.

Could be, but that's a bunch of stuff you don't have to power with the existing system. Given how many creeks there are, that seems like it could really add up.

> With the advent of long lasting perovskites and (hopefully) some less polluting battery technologies you're not even better off cost-wise.

That's fair. I've learned in my life that I don't have a good intuition for what the costs of large infrastructure projects are.

It would also have a tremendously negative impact on the local environment. Even tiny dams cause big disruptions to the riparian ecosystem.