Thread by @yestiseye

@simonahac @ZimmermanErik But I am saying there'll be arbitrage opportunity, initially. In fact we can see it already exists in South Australia, especially if you can hold the energy for several days to exploit the largest margin.
This is a non-trivial issue though, so `lil tweet storm… 1/×

@simonahac @ZimmermanErik First worth noting that it clearly makes sense in such arbitrage heavy scenarios, a combined solar or wind (or gas!) plant with storage would be optimum for the individual market players.*

Now look at the current, existing data in SA or California. 2/×

@simonahac @ZimmermanErik Short, sharp energy trades like you'd see on computer-trading in major stock markets. Because that's how you actually make arbitrage really pay - high frequency micro margins.
This is obviously the opposite of the large chunks of time-shifted energy we need practically. 3/×

@simonahac @ZimmermanErik Worth pausing too at this point whether it really is a good idea for us to turn the electricity grid into a weather-induced stock market. Extra degrees of difficulty aren't a positive when we're attempting the Hardest Thing Ever aka "deep decarbonisation". 4/×

@simonahac @ZimmermanErik Now are we seeing stock market behaviour out of the HPR facility in SA for example because the MWh(or indeed GWh) available to it just aren't large enough?
Maybe. But the implications of larger storage installed is daunting. The first, say 10% of the stored energy = profit. 5/×

@simonahac @ZimmermanErik That's easy. The last 10% of your Big Energy Storage unit will hardly ever get used. So less profit, if any. LIFO type problem with costs incurred.

This is how arbitrage, with its zero sum nature, begins to in effect backfire. 6/×

So now Snowy 2.0 (see * previously).

@simonahac @ZimmermanErik A large, standalone storage facility, especially with 350GW at its disposal will not do wonders for the profitability of utility solar or wind farms. In fact I'd use words like 'Brutalise' when I consider the financial implications. To make Snowy 2.0 work and do so… 7/×

@simonahac @ZimmermanErik … *without* coal will require it to do so. Peversely the storage required to make VRE a firm(-ish) energy source, is more economically sound when supplied (at least in part) by already firm energy sources.
Think about that for a moment. It's important. 8/×

@simonahac @ZimmermanErik * 350GWh

@simonahac @ZimmermanErik [aside: I'd be watching for batteries being whacked onto the side of gas plants, in combination with solar or wind for the fuel saving benefits - especially at mining operations.
Already the first signs of this happening. But the gas/diesel plants are staying.] 9/×

@simonahac @ZimmermanErik Two other points that could easily be their own threadlines of thought.
There's been discussion of running a system with VRE deliberately curtailed in part because this minimises storage requirements (hi @DavidOsmond8!). The pricing effects on solar/wind farms? Not great. 10/×

@simonahac @ZimmermanErik @DavidOsmond8 The extent to which this "not great" effect impacts, especially in a quantified $/MW (/MWh, /time, who knows the right way to do this? I don't… not yet) is hard to say at this point. But it'll take a big chunk of work to figure out.
Worth a PhD thesis or 3 I suspect. 11/×

@simonahac @ZimmermanErik @DavidOsmond8 The intellectual exercise is certainly interesting; it'd be more fun if the big picture implications weren't so dire.
Second extra point: the game theory of OCGTs vs CCGTs. A GE gas plant in California will close early, site to become storage facility… but it's not a peaker.12/×

@simonahac @ZimmermanErik @DavidOsmond8 South Australia has already made this mistake that California *could* be heading down. (Or it may only be related to GE's mistakes.) Storage needs to be affecting gas peaker capacity, not undermining CCGTs. Otherwise that could become very ugly in a whole heap of ways. 13/13