A popular discussion point for the 100 % renewables set over the last three or four years, particular in Australia it seems, has been the declaration that the need for baseload power is a “myth“.
It goes something like this. We don’t actually need a system based on electricity generators that can supply electricity 24/7, 365 along with peaking plant and enough back-up for when things goes wrong (“contingency events” being the lingo). That’s baseload, and that’s so passé.
The most important thing is to have enough electricity generators with supply that is “dispatchable”, meaning the power can be sent out in response to demand. With enough different types of dispatchable supply in the mix, supply can meet demand at all times. The dispatchable supplies can be wound up or wound down to accommodate the non-dispatchable suppliers. To put that last bit in lay terms, when the wind is blowing and/or the sun is shining on PV panels, we can turn the other stuff down (or off). When the wind is not blowing and/or the sun is not shining, we can turn the other stuff up. Ergo, “baseload” is a myth.
From there, it is possible (and in some circles it seems, encouraged) to get a little conspiratorial about baseload. “Baseload” is really just a ruse to maintain centralised power generation. Consumers are lulled into providing load for electricity by cheap over-night prices.
So that’s the baseload myth. Its chief purveyor in Australia is probably Dr Mark Diesendorf of the University of New South Wales (though honourable mention goes to Prof Ian Lowe). You may have enjoyed Mark and I having a bit of argy-bargy on these matters on the recent episode of “Awaken” on NITV. It’s largely the work of Mark and his team that I will be referring to from here, from their first in a series of related papers, Simulations of scenarios with 100% renewable electricity in the Australian National Electricity Market. I’ll be saying EDM (Elliston, Diesendorf, MacGill) for short.
So… is baseload a myth? Or is the myth of baseload a myth?
I’m going to start with those key terms: “baseload” and “dispatchable”.
I agree: the most valuable characteristic for an electricity generator (forgetting greenhouse emissions for a moment) is reliable dispatchability. The ability to respond to demand in the grid is absolutely essential. The inclusion, to date, of supplies in Australia that do not respond to demand (such as wind and solar PV) succeeds on the back of an established, large, mature system of generators that can and do.
Here’s the thing though: what would you do with a generator that was so “dispatchable”, it could dispatch power at close to its full rated capacity for almost 24/7, 365 (barring planned or unplanned outages)? Even those not particularly familiar with energy systems would agree, the smart thing to do would be to run it in exactly that way to contribute to meeting demand in a modern society where, daft conspiracy notions aside, there is a demand for electricity every second of every year.
That’s something we call “baseload”.
To illustrate, here’s an example of electricity demand from Victoria for a day this month, brought to my attention by The Actinide Age (whom you should read, by the way).
AMENDMENT: Not so fast Mr Heard. This is actually supply from Victoria to the National Electricity Market (NEM), as well-spotted by @BurkeKB, not demand within Victoria. Demand in Victoria is currently hitting lows of about 3,800 MW daily, as shown by AEMO. The Victorian plants, being the cheapest baseload supplier, run full-tilt to supply the NEM, which is what is seen below. To see the baseload contribution from black coal to the NEM for example, go to Empower.me and look at the profiles from NSW and QLD. Back to the article, now with a couple of amendments, and thanks to @BurkeKB. End aside.
So, that big slab of brown you see is Victoria’s contribution to the baseload of the National Electricity Market, as in, the minimum load that was demanded for that 24 hour period. It is all being provided by brown coal, and merged with the black coal from NSW and QLD and the brown coal and gas from SA, the emissions for the NEM are among the highest in the world per unit electricity. It is beyond obvious that baseload, as a concept for designing and managing electricity supply in Australia, is and will remain utterly relevant. The only problem with the graph above is that the baseload supplier is brown coal. Were that a zero-carbon source of any kind, we ought not care. It should also be beyond obvious that if you have power generators that can provide that big horizontal slab of demand every day, it would be economically stupid not to build a system around running them at all times, and let other generators deliver the intermediate and peak load above it.
The baseload myth argues that what I have just said is utter rubbish. We don’t need those generators providing that big horizontal line of demand. We can use dispatchability to do the job with a combination of dispatchable generators and non-dispatchable generators. Well, let’s see that what looks like. Below is a week of supply in winter as published by EDM. This is how EDM make supply meet demand for a week in winter for the NEM, based upon actual demand from 2010 and simulated supply from renewables based on meteorological data, combined with an assumed “copper plate” network (meaning, the authors have assumed unconstrained transmission of electricity across the entire National Electricity Market. This, of course, is not the case at all). As far as methodologies go, is a perfectly sensible one to explore this challenge, provided one remains aware of the limitations.
For me, the figure above makes pretty ordinary modern art, but scares the pants off me in terms of trying to respond urgently to the decarbonisation challenge. Let’s look at some key characteristics.
Firstly, the “baseload” for this week in winter is about 18,500 MW- 19,000 MW. That’s the minimum level of electricity demand for this seven-day period. That’s not a myth, it’s a matter of public record. That demand could be met by about 19,000 MW of any technology with that ultimate level of dispatchability I discussed above.
Secondly, the yellow in the graph is the concentrating solar thermal, or CST. This is the principal technology upon which the much hyped “dispatchability, not baseload is what matters” line is based. Yes, CST is dispatchable. You can see it is sitting to the right of the blue in the graph each day, which is non-dispatchable solar PV. So the solar resource caught by the CST is held and dispatched later in the day to match the evening peak. Then, it goes away, because there is no more energy than that in winter time. So yeeeeeeeeeessss… it’s dispatchable. It’s also unable to dispatch for any more than a few hours per day in mid-winter (without being dramatically expanded in size).
Thirdly we see the mountain-like profile in green, which is electricity from wind. This, like PV, is non-dispatchable, so when it’s coming, everything else has to make way. In the course of seven days, supply varies from over 15,000 MW on June 28 to near-zero across the entire NEM at about midday on July 1. This is nothing whatsoever to do with demand, and everything to do with the weather. In this “baseload myth” world, Australia needs to design and manage an electricity supply system to accommodate that level of change not as an unforeseen contingency, but as a part of regular operations.
Finally, we see the light-brown on top. Such an appropriate shading, as this would approximate the colour of the smoke from combusting the biomass that is required to back-up almost the entire system. This “baseload myth” system builds in 24,000 MW of biomass generating capacity. The essential purpose is to provide cover for the darkest, stillest weeks in winter. Note the night of July 1, where biomass and nearly Australia’s entire hydro capacity is providing almost all demand in the NEM. The additional reality is that it provides 14 % of the total annual supply in the simulation. EDM may have wanted to minimise it because, as they acknowledge, biomass is not desirable from a sustainability perspective. But I’m afraid 14 % is not small beer. We also have no idea what level of back-up may be required in any other year, like, for example, a year of terrible drought where the hydro capacity is low.
The other reality is that these generators could, in principal, perform in exactly the same way as the coal generators shown in the Victorian example if we wanted them to do so. So to be clear, EDM have not designed a system without baseload. They have designed a system where 24,000 MW of baseload-capable generators are deployed to the minimum extent possible… on purpose. So, as shown in the image below, for a week in January, that 24,000 MW is virtually idle.
And to think people accuse the environmental set of being economically irresponsible…
Let me say this: having now reviewed 15 studies, globally, of systems for purportedly supplying 100 % renewable electricity and energy, the work from EDM is among the best. The methodology makes sense and lets us draw relevant and useful conclusions, and the paper is written in a way that makes it very easy to understand the proposed system (at least, compared to much of the other literature). I have no problem with the work. I take serious issue with the dramatic over-reaching that Diesendorf, in particular, does with the findings.
One of the inescapable conclusions from the global literature is that the studies serve to reinforce, not undermine, the relevance of baseload. The studies maximise hydro where available, exploit CST where available, and time and again must draw on substantial biomass, all in the name of supply meeting demand, all in an attempt to replicate, as closely as possible, that horizontal line we see in the first figure.
But Australia’s electricity system isn’t a puzzle waiting to be solved. It should not be treated like an academic hobby. It does not need to be torn down and re-built according to radically different rules, using only some types of energy. The system gives Australia what it needs, which is lots of reliable electricity, at a scandalous price in terms of our greenhouse gas emissions. We need to take the greenhouse emissions out. That’s it. Nothing more complicated than that. The less complicated the better, because that will dramatically improve our chance of success.
What the “baseload myth” demonstrates is one of the great flaws in our electricity discussions: the belief that demonstrating we could run on renewables-only is equivalent to establishing that we should. This perspective is foolish in the extreme, elevating an obsession with certain technologies, and an irrational dislike of one in particular, above an urgent need for decarbonisation. Even the relative outsider to electricity supply can see that a straight replacement of the horizontal line of supply shown in the first figure with something zero-carbon is the lowest risk, highest certainty, lowest disruption path to getting the job done. Imagine a grid that had done exactly that for over 13 million people…
Like what you see here? Please subscribe to the blog, Like Decarbonise SA on Facebook and follow @BenThinkClimate on Twitter. Read more about the potential for nuclear power in Australia at Zero Carbon Options.