Everything's a hypothesis in science, Ian, things are only proven in mathematics. Your idea of waiting to see if the worst happens before responding is not a tenable, precautionary approach.

Insofar as there being no empirical evidence, I disagree, but will not waste time going toe-to-toe when the arguments are well documented.

Re the article itself, I haven't responded because I don't quite follow this: "Generation plus storage capacity multipliers needed for given reliability standards, using 100% 'new' renewables as base-load and back-up, are shown in the chart below." I'd like the explanation and figuring teased out more. Somebody?

Is Geoff out there to help poor dolts like me?

Luciferase, this article is about what I have been waffling about for sometime.
To get 100% renewables you have to use a continent sized country so
that the wind/solar can take advantage of the weather differential in
different parts of the continent. ie no wind in the south, hope it is
blowing in the north. What Geoff is pointing out is that the more
uptime your performance standard requires the more wind and solar
installations you need.

You can get closer to knowing what amount solar & wind and where it
should be installed. You need a computer connected to AEMO to read in
the current demand continuously. Knowing the performance of wind farms
and solar farms for wind speed and solar radiation you can place
virtual wind farms and solar farms around the country, inputting
wind speeds from the BOM. Where wind data not available, install
cheap wx stations.

After a year or so you would have a pretty good idea whether the
number of wind/solar/batteries could be afforded.
As Geoff says it is very dependant on the reliability that you demand.

If you reduce reliability then all buildings above the third floor
will have to be abandoned. This is the nightmare that the greens
have been ignoring.

Sorry reduce reliability should be increase reliability.
I was thinking reduce the figure itself.
It appears that the reliability figure at present is .002%.

I believe Trumps decision to pull out of this farcical of a con, was the right thing to do.
Everyone should follow. With the technology we have today, we are nowhere near a renewable age.
The current technology is at best unreliable.
At worst incapable of doing what is required of it.
It will be some time before any real and affordable technology will appear, and even then, if we shoot every politician who dares go anywhere near it.
As long as the loony left and the garbage greens are allowed to breathe oxygen, we will all just keep on suffering and copping pretty much more of the same as we have been.

Bazz, thanks for trying, but I'm still no closer to understanding the graph, or its basis.

Take SA's 35% wind. I'm assuming if that were presumed to have sufficient "renewable backup" (an oxymoron if its batteries of any hue) to be guaranteed, then only need two times this, roughly, is needed in other guaranteed generation to complete SA's electricity needs, i.e to take generation to ~100%.

Geoff C?

Unfortunately Luciferase that 35% figure is the amount of power that
a turbine will typically produce in a year compared to its nameplate rating times 365x24 Mwhrs.
When I turned my computer on this morning Sth Aus was producing 28 Mwatts.
Their maximum is about 1200 Mwatts.
I do not know how many turbines they have installed so divide 28
by the number of turbines.
Because their output can be very low and all others be below their
rated output, you just cannot divide 100/35 = 2.85 and install three
times as many turbines and get 100% output.
Basically this is what Carmody's article is about.
He points out as you demand better reliability it becomes more and
more difficult to achieve 100% reliability.

You have to install more wind/solar farms over a physically larger
area to improve the chances of getting more wind.
Batteries would help buffer calming events on short time scales but
no one seems to understand that the batteries have to be recharged and
that takes away from the available power, probably the next day, if it
has been a night with not much wind.

I do not think anyone can work out theoretically how many wind and
solar farms would be needed to run the country with any reliability
percentage you can pick. I believe it could be done as I suggested
using realtime data for demand, wind speeds and time of year variations for wind and sun.

I suspect, just from a gut feeling, that the result would cost such
an enormous amount of money that we would need to build a printfarm
to house the printing machines to print the money.
I know, I know, it would all be done with pixel money and never be paid back.