Thanks for the detailed reply. I see that you're skilled enough to calculate that your refrigerator-sized stack of car batteries could provide the power to pump several thousand gallons of water, demonstrating that batteries are a better way to store energy than lifting water is.

From your reported power usage, it sounds like you're probably single. If the rest of your figures are correct, we'd have roughly a refrigerator-sized stack of batteries _per_person_. Inverters are 0% efficient at no load (they waste 20 watts idling) to 90% at full load, so figure around 75% average efficiency, so 16-18 batteries per person rather than 12. Batteries lose capacity as they age. You don't want to replace your batteries every two years, but rather continue using them as their capacity decreases over five years, so we better go with 21 of those batteries.

The specific energy of a lead-acid battery is about 35 watt-hours per kilogram, or 64 pounds per Kwh. Our bank of 22 batteries is 193 Kwh at the battery when fresh, so you've got 12,352 pounds of lead and sulfuric acid to mine, then dispose of and replace every 5 years. Four billion pounds for the US. You're going to need a lot of large mines to get all that lead. You could do it, but it wouldn't be very good for the environment.

You asked me to show my work. Flood models are a bit complex, of course, but we can at least get a general idea by calculating the minimum and maximum possible, assuming ideal topography and worst-case topography. My original number was based on a calculation in the Proceedings of the National Academy of Sciences entitled "Powering the planet: Chemical challenges in solar energy utilization", but they might be wrong, so let's do a fresh calculation:

As per the Bureau of Land Reclamation (operators of Hoover Dam / Lake Mead) Hoover dam produces 4 billion kwh annually. Per EIA, the US uses 4,047 billion kwh. So we need 1,000 Hoover Dams. BLR says Hoover is 726 feet high and flooded 248 square miles. They also say the amount of water pushing on the dam would cover the entire state of Pennsylvania 1 foot deep.

The location of Hoover dam was of course chosen with some care - it's a good place for a dam, in a deep canyon. It's a place where you can build a dam 726 feet high, so the flows hundreds of feet down through the turbines, releasing a lot of energy. We won't find 1,000 to build dams over 700 feet high, but let's pretend we could in order to figure a MINIMUM possible amount of flooding. This is the minimum assuming ideal topography, where we have all the deep canyons we want. Hoover Dam times 1,000 is 248,000 square miles for the minimum. Dams go on rivers, of course, filling the river valley, so reservoirs tend to be long and thin, not square. If our reservoir is 10 miles across, it'll be 24,800 miles long. Oops, that's longer the distance around the earth. Let's make it 100 miles across, so it'll be 2,480 miles long, roughly the width of the United States. That's the minimum, pretending we have 1,000 deep canyons to fill hundreds of feet deep. A dam built in flat land will just create a shallow flood across the whole state. Worst case, assuming flat topography, would have the whole US under 70 feet of water. If you go into flood simulator software that's been loaded with the actual topography of the US and start placing dams on actual rivers and let it calculate the flooding based on real topography, you end up with about 80% flooded.