One infrequently discussed but very real issue with pumped storage is biogenic methane. In some cases, reservoirs can make more greenhouse gas per kWh than coal(!). Earlier we thought this was just bad in tropics but more recent studies indicate it may be most strongly correlated simply to reservoir size. If these are even moderately accurate, pumped storage for significant energy storage is not an option for a carbon-free system.

[1] http://pubs.acs.org/doi/abs/10.1021/es401820p

[2] http://journals.plos.org/plosone/article?id=10.1371/journal....

That's really interesting. Methane is pretty bad, but at least biogenic methane doesn't increase the total amount of carbon in the atmosphere.

Methane is far worse than CO2 though, about 30x. A process that converts one to the other while staying carbon neutral isn't what we need.

Governments are hesitating to use this because pump storage has major environmental impact. It just needs _a lot_ of space. You also need mountains for that and not every country has those.

There is some research going into putting concrete spheres into the ocean and pumping water in and out of them. The principle is the same but it doesn't need hundreds of hectares of forest to be cleared.

I recently visited Cruachan ( https://www.visitcruachan.co.uk/ ) which does this, with four 120MW reversible turbines; it's great but requires very specific topology. Most countries don't have too many places they could build one.

Any hydro lake works as an energy storage system - it doesn't have to be pumped.

That's why you see energy being exported to hydro-rich Norway on windy winter days. It allows water to be retained in Norway's storage lakes for later use.

Even more efficient is to control the flow in a hydroelectric plant by stopping the generation when the demand is low. This removes the cost of pumping. If there is sufficient reserve water capacity it is done in places where the energy price depends on the market and economic incentives are there.

Hydro also has non-trivial enviromental impact - you are creating pretty big "lake" above the dam. And if you want to turn it on and off, you'll need one more to make the flow normal again.

And that reservoir can emit very significant biogenic methane. Large reservoirs for pumped storage are likely not a scalable option for climate change.

For the downvoters, here are scientific references [1,2]. I was shocked too.

[1] http://pubs.acs.org/doi/abs/10.1021/es401820p

[2] http://journals.plos.org/plosone/article?id=10.1371/journal....

One thing to note. While biogenic and thus "green", the methane is about 30 times as potent greenhouse gas (over 100 years), so we can not consider pure "carbon footprint" here. There are similar problems with restoring wetlands and undraining peatlands, which used to be carbon sinks, where biogenic methane production goes up for a while.

There are also problems with N2O emissions on the reservoirs, specially near the agricultural regions. And this is all without even talking about reducing biodiversity and destroying spawning areas for the fish by filling those with mud. So yes, efficiency is not always friendly.

This is true. High efficiency and low carbon emission are not always in the same boat.