Wow, that's crazy amazing. It's so weird that you can jam a bunch of air into a pit and call it energy... of course you can, it just seems nuts to think about.

It's not just intermittency. It's also off peak supply.

Wind is coming on line quite rapidly. Wind blows 24 hours a day (somewhere).

As we build more wind we end up with power we don't really need at night. Storing some of that energy to be used to smooth out daytime/peak production makes a lot of sense to me.

It means that we would need to build a lot less wind and solar to meet our needs.

It also means that we can build a lot more wind generation where the wind blows hard and where few people live without having to build as robust a grid.

Feed power to storage during the night on a less stressed grid. Then pull it back out closer to home when demand is high.

I understand a great deal of energy is lost as heat compressing the gas. In fact the diagram shows a heat exhaust.

Call me simplistic, but can't you recover some of that heat for useful purposes. (e.g. district heating)

OTOH when the gas is pumped back up and decompresses - you have a district cooling system.

I sometimes think some of the proposed renewable sytems are as guilty of waste and inefficiency as the fossils.

I agree completely with your assessment, that energy storage is important but not as important as it is often made out to be. Still, this is cool. How much space per unit energy is needed, and what is the efficiency of storage and retrieval?

I don't think that efficiency of storage is an issue. Night time energy generation is energy wasted. It makes sense to use that energy, even in an inefficient manner.

The Swiss have been doing something similar for years. They use surplus generation to pump water uphill into dams where it is released again through turbines during peak usage hours.

Producing hydrogen through electrolysis using surplus energy is another good strategy.

What if there's a storage leak, how will they clean up that air?

How is energy storage not as important as it is made out to be? David MacKay points to data that indicates that wind output, like internet traffic, is bursty even as you increase scale (that is, take more and more farms over a wider and wider area.)

This is a real concern. MacKay points to pumped-water storage facilities, which already exist at places where there are steep slopes and large water supplies (often the ocean itself). These pay for themselves because electrical demand changes, allowing pumping electricity to be bought when cheap and turbine generated electricity to be sold when expensive.

A large increase in the number of these facilities should coincide with increase in wind turbines, and, of course, grid augmentation.

Very Brilliant! Since Off peak>Peak energy

Leak would not be a problem, since it is you know - air. And it gets back to the atmosphere sooner or later when harnessed. But the storage needs to be gelogically solid. Anyway, if you have acces to caves or other such formations, this is a great way to store clean energy. But the best consept so far is to use the same principle with water (has better efficiency since water is just pumped and no energy is wasted on compression). But to do that You need an elevated reservoir, so again, geology matters. I think that united states may have several places that would be suited for that?
Anyway, GB has such hydro installation for peak demand levelling and if I remember correctly, it had a response time of 23 seconds - the fasest of any power plant type. That is impressive. And when You consider that the electrcity used to pump the water to the reservoir during the night would be wasted otherwise, it is a very good solution (even though it has losses)
So depending on location this comressed air storage can be quite affordable and clean storage system.

Gas formations that may be used for compressed air storage have already proven themselves leak free for eons and during decades of production. As long as the compressed air pressure is kept at or below levels that the formation has shown to hold leakage shouldn't be a problem. Natural Gas and Propane has been stored in salt mine caverns for decades. The gas leaks that cause the explosions at Hutchinson, KS a few years back, where leaks that could have been detected and repaired. Admitting I don't know what kind of pressure is needed to operate a turbine, I don't believe leaks will be something that can't be handle. True, compressing air heats up the air, but that's a result of the energy used to compress the air, I'm not sure additional energy goes into heating the air. Perhaps the heat produced by compressing air could be added to other renewable heat sources to produce steam to run the turbines?