What if... we just made a bunch of stuff up?

The Gronk Gallery returns in the service of Gina.

Thanks to everyone who completed the survey last week! It really does help and we appreciate the feedback and chance to engage with readers. If you ever have a general suggestion or a topic or feedback (good and bad) please drop us an email, or chat comment, or pigeon carrier or just shout at us from the window of a moving car.

If you’d like to add your thoughts, the survey link is here.

Every couple of months Australia’s pre-eminent comedy organisation the Institute of Public Affairs asks the question, “What if instead of the dog eating our homework, we got the dog to author our homework?” The result of this bi-monthly process is a white paper, which smells so terrible that it genuinely makes you wonder why they bother.¹

The benefit of this approach presumably is that operating overheads are very low (Gina isn’t that kind with her money and Tableau licences aren’t cheap. Also I assume the canines haven’t unionised yet?), and the report conclusions map nicely to whatever the particular policy the coalition is currently proposing.²

The latest comedy bit research paper from the IPA claims that a third of agricultural land in Australia will be lost to renewables in order to meet the federal government’s net zero ambitions. This is so insane to be actually genuinely funny.³

The Guardian summed up the core problems with this paper pretty concisely, but I love a good comedy, so let’s dive a bit deeper and figure out how the IPA arrived at these fantastical numbers.

How much of Australia is agricultural land?

The Australian Bureau of Agricultural and Resource Economics and Sciences (ABARES) usefully provides data on the agricultural land usage in Australia.⁴

Source

Depending on your definition of land use, there are 62.95 million hectares of ‘modified’ agricultural land (including horticulture) or 346.32 million hectares in total, including land used for grazing.

The IPA’s Table 3 quotes 369 million hectares (maybe they thought it needed to be inflation adjusted), which is close enough that whatever, not going to argue that one. So far so good right?

What about the generation?

Turns out this is about the last number in the entire report within eyesight of reality.

The IPA outlined the four scenarios comprising their analysing in a dedicated section titled ‘Research Methodology’, which lol.

1. “All renewable energy produced in Australia to replace energy from coal, gas, and oil will be derived from solar power.”

2. “All renewable energy produced in Australia to replace energy from coal, gas, and oil, will be derived from wind power.”

3. “All renewable energy produced in Australia to replace energy from coal, gas, and oil is derived from wind and solar based on a solar- intensive renewable expansion.” (a generation ratio of 9:1 solar to wind).

4. “All renewable energy produced in Australia to replace energy from coal, gas, and oil is derived from wind and solar in equal proportion.” (i.e. 50-50 split of wind and solar).

If these scenarios sound about as sophisticated an analysis as a Friday night discussion at the pub, well yeah. Anyway, let’s carry on.

The core premise of the analysis was to look at how much energy Australia produces currently, including domestic usage and exports. To do this they looked at the 2023 Q3 numbers from the Department of Industry, Science and Resources. These numbers include all energy consumed and exported by Australia in terms of tonnes of coal, barrels of oil and volume of gas (collectively hydrocarbons).

The Research Methodology (chuckle) section helpfully informs us that they found some random tool online to convert barrels of oil and tonnes of coal into an energy figure.

Now, customarily, you would report the value of energy contained in a lump of coal as Joules (metric system) or calories (imperial system). This is the primary energy associated with a fuel source.

To produce electricity from coal or oil however we need to burn it, which involves significant losses through wasted heat. A coal-fired station has a thermal efficiency somewhere in the ballpark of 30-40%⁵ which means that only 40% of your primary energy is converted into usable electricity. This quantity of usable electricity is typically quoted in Watt-hours (Wh), and sometimes for the explicit avoidance of doubt, Wh-e.

Watt-hours (or kilowatt-hours, megawatt-hours, gigawatt-hours, terrawatt-hours) are of course just another unit of measurement of energy. And so it possible to directly convert Joules to Watt-hours… or tonnes of coal to Watt-hours.⁶

You may have guessed where this was going; the listed numbers in the IPA report are primary energy figures. If that’s not amateurish enough, in Table 1 on page 7 they include the energy associated with metallurgical coal exports.

We’ve talked about this before, but Australia exports two types of coal – thermal coal which is burnt in coal-fired power stations to produce power, and metallurgical coal which is burnt in blast furnaces to release carbon which is alloyed with iron to produce steel (and yes, the energy released from the burnt coal is used to heat the iron). Just to reiterate for those at the IPA, metallurgical coal is not burnt for electricity production. It is far too valuable for that.

Confusing the primary energy value of metallurgical coal with the value of electricity generated from something not burnt is just *mwah* chef’s kiss.

Anyway, after all this the IPA determines that Australia currently produces 5,025 TWh of energy from coal, oil and gas.

If you don’t believe me they that they have no fucking idea what they’re doing, and that they were intentionally looking to determine the primary energy value of Australia’s hydrocarbons; this paragraph is right beneath Table 1 (emphasis mine).

The energy production covers export as well as domestic use. Australia’s total energy production from natural gas, coal, and crude oil is equivalent to approximately 5,025 TWh. This amounts to roughly one-fifth of the world’s current entire electricity consumption.

lol.

The final piece of this analysis is to then project what this hydrocarbon consumption will look like in 2050 (because the entire point of this white paper is to demonstrate how unachievable Net Zero by 2050 will allegedly be). Leaving aside the fact that forecasting 2050 hydrocarbon exports in 2023 is patently silly, their methodology is to take the IEA’s electricity growth estimates of 3.5% per annum, and just scale up the primary energy figures for each category.

This yields a whopping 15,459 TWh of hydrocarbon primary energy “electricity usage” in 2050.

What are the actual numbers?

According to OpenNEM, Australia’s two largest grids (the NEM on the east coast, the WEM in south west WA) consumed 224.87 TWh in 2023. In fact this number has hovered around 220 TWh per annum for the last decade and a half.

There is of course electricity consumption outside of the NEM and WEM – WA’s north west interconnected system, NT’s Darwin-Katherine system, Mount Isa and various off-grid mine sites scattered around the country. Let’s be generous and say that Australia’s current total annual electricity consumption is comfortably less than 300 TWh.⁷

300 TWh is an entire order of magnitude less than the IPA’s wild estimate of 5,000 TWh.

But the point of the IPA paper is to discredit the entire idea of net zero by 2050, so looking at the current electricity consumption in our hydrocarbon-intensive world isn’t the point.

So what is a credible outlook? In December AEMO released the draft 2024 Integrated System Plan which is a good place to start.⁸ AEMO considers three scenarios:

1. Step Change, which fulfils Australia’s emission reduction commitments in a growing economy,

2. Progressive Change, which reflects slower economic growth and energy investment, and

3. Green Energy Exports, which sees very strong industrial decarbonisation and low-emission energy exports.

The Step Change scenario is often taken by those in industry to be the most likely path of development, (although the Progressive Change scenario is also a reasonable candidate particularly given the transition challenges). The Green Energy Exports scenario (which in previous ISPs was known as the Hydrogen Superpower scenario) is probably the closest analogue to what the IPA is try to muddle through.

Figure 5 on page 25 estimates the electricity consumption in the NEM will be 420 TWh in 2050 under the Step Change scenario. This number includes the demand met behind-the-meter from rooftop PV.

Source

The Draft ISP doesn’t have the numbers for the Green Energy Exports scenario, but we can find the numbers in the appendix (the generation and storage outlook file). This much more ambitious scenario estimates that in 2050 the total electricity consumption of Australia is estimated to be a little over 1,000 TWh, a lot of which is associated with green hydrogen production for export (something on the order of 550 TWh).

That’s a lot of electricity. But it’s still five times less than the IPA’s current estimate of how much electricity we’ll need to generate from wind and solar farms now, let alone in 30 years time.

How much land does a solar farm take up?

In the Research Methodology section (giggle) the IPA use a figure of 3,700 hectares per TWh of energy. Again, this number is plucked from a random website.

How they arrived at this number is genuinely amazing.

1. They found some random website which lists solar land use for a range of different locations.

2. The chart (below) estimates how much land is needed for solar necessary to power an average US household for a year (which is 10.72MWh per year, apparently). It varies by location because solar works better where there’s more sunlight. Despite Sydney being a location on the chart, they took the highest possible land usage figure (i.e. lowest irradiance) which corresponds to London.

3. The solar land usage in London is 3,926 sq-ft/US household. This is where they get the 3,700 hectares/TWh number from, although my calculations place 3,926 sq-ft/10.72 MWh as 3,402 ha/TWh, but whatever, at least it’s the same order of magnitude.

Research = random charts.

Using the random value from the chart of 1,185 sq-ft (per 10.72 MWh) for Sydney yields a somewhat more reasonable sounding 1,027 ha/TWh.

Without changing data sources, we’ve already cut the land required for solar by a third… using the same random dodgy reference.

Curiously the same reference lists a chart from Hannah Ritchie which lists the median solar land usage of 19 m² per MWh or 1,900 ha/TWh. Did the IPA deliberately go out of their way to find the highest possible number or are they just wildly incompetent? Hard to say honestly.

In reality the calculation for land usage of a solar farm is (obviously) more complex. For one, there is a big difference between the annual yields from consumer-grade residential or commercial rooftop PV and utility-scale farms.

Secondly, solar irradiance is a direct function of latitude, and Australia is a big country — the output of solar panels in northern Cairns is not the same as panels in Hobart.⁹

So, how much land does a solar farm take up?

On Page 6 the author asserts

Information on land area coverage and average historical or forecast generation output of wind farms and wind project are not readily accessible to the public.

If this information is “inaccessible”, it suggests the author may not have access to the internet. Or a computer.

With just a Small Amount of Effort^TM we can do this.

OpenNEM has facility-level data and is “readily accessible”.

OpenNEM informs us that Darlington Point, a 324 MW solar farm in NSW yielded 532 GWh and 542 GWh in 2022 and 2023, respectively.

The literal first Google search results provides Edify Energy’s Darlington Point page, which says the farm is built on 1,993 acres of land, which using our random online unit converter equates to 806 hectares or 1,515 ha/TWh (per annum).

Our yield of 1,027 ha/TWh for Sydney plucked from a random online source is not half bad actually!

Applying the same methodology:

• Ross River Solar Farm a 128 MW solar farm outside of Townsville, Queensland (one of the northernmost solar farms in the country) yields 748 ha/TWh per annum.

• Winton Solar Farm, in Ned Kelly country outside Wangaratta, Victoria (and the southernmost solar farm in the country) yields 1,612 ha/TWh per annum.

But just how much land?

Using our more reasonable number of 1,027 ha/TWh (still plucked from the random source) and a more sensible number of 250 TWh of electricity consumption in 2023, yields a total land area of 256,750 hectares (for our very silly calculation).

This is 0.07% of Australia’s total agricultural land, or 0.4% of Australia’s ‘prime’ modified agricultural land. In 2050, under the aggressive Green Energy Exports scenario, this number is 0.28% of Australian agricultural land.

Oh, while we’re doing stupid comparisons, our 256,750 hectares of solar farms is still only a quarter of the size of the area taken up by Melbourne. By 2050 we’ll need to cover the entire city and suburbs in solar farms in order to export hydrogen, I guess 🤷‍♂️.¹⁰

Speaking of city rooftops…

Even if you assume, like the IPA do in their scenario 1, that all of the energy demands are met by solar, they seem to somehow have forgotten that roughly 1 in 3 Australian households currently have solar on the roof (and plenty of businesses). And although solar installations have slowed a little in recent years, they’re still growing significantly.¹¹

Going back to the 2024 ISP, an estimated 23 TWh of electricity was generated from rooftop PV in 2023 and under the Step Change scenario is forecast to grow to 86 TWh in 2050. That’s around 10% of our current annual consumption and growing to 20% in 2050.

Anyway I’m not going to bother doing the same yield analysis for wind, because I think I’ve juiced this lemon enough. But hopefully it’s pretty clear that the numbers in the report are just utter dogshit.

The Point

I had a bit of fun writing this article because the paper is so bad, the numbers are so ludicrous and the research is threadbare. The general standard of work would be embarrassing for most high school students. It's very possible I put more work into this post than the author's did on their paper.

The Guardian reported that when challenged on the veracity of the numbers, the report’s author responded accusing the critics of “weasel words from the climate lobby” and that they had “grossly misrepresented the report”.

There’s a certain absurd quality to this defence. I like to imagine it lives in the same vein as avidly defending the merits of the 2019 artwork Comedian. Is duct taping a banana to the wall and informing people that it’s art really genius? Do you honestly think people see deep meaning in a fucking banana taped to the wall? Wasn’t it all just a big joke designed to reel in suckers?¹²

I’m inclined to think it’s the latter, because the paper has already served its purpose. Some scary bullshit numbers, ‘research methodology’ (for the last time, lol) and a bunch of references that no one will click on gives the thing a thin veneer of credibility allowing the IPA to put out an official looking publication so that mouth breathers can stand in front of TV cameras and wax lyrical about the dangers of renewable energy blah blah blah. The numbers don’t actually need to be even remotely solid, they’re just props for partisan rhetorical attacks.

Honestly the same thing could have been achieved by just stapling a ream of blank pages together and handing it to Gina to wave about and say “renewables = bad”.

Here’s the rub – the energy transition is challenging and there are plenty of legitimate criticisms and concerns over just how we’ll get there. And we should be having these frank conversations. But this report contributes nothing but shallow and trite partisan rubbish for conservative talking heads.

---

Things Happen

• Hertz is yeeting a third of its global EV fleet after apparently inking a bad deal with Tesla. This less of a bad omen for EVs and more Hertz miscalculated depreciation.

• Winter storm Heather had ERCOT jittery, but so far it’s a long way from Uri in 2021.

• Queensland smashed demand records on a disgustingly hot and sticky day which saw cloud cover limit the benefit of rooftop PV.

• Federal environment minister Tanya Plibersek vetoed a Victorian government plan to build an offshore wind terminal at the Port of Hastings. In terms of energy transition challenges, this one feels like the left and right hands probably could have done a bit more talking.

1. To be fair this process does waste Gina Reinhart’s money, and it is a little bit funny, so keep it up I guess?

2. Or is it the other way? Based on the current shadow cabinet of superstars, it seems more likely that the tail is wagging the dog these days.

3. For a bonus laugh, the accompanying media release titled Shards of Truth media release has some properly wild hallucinations about flying glass, sheep chewing exposed wires and toxic transformer leaks.

4. Future note to the IPA – graphics are a really strong way of presenting information, and showing just how much of the Australian continent is consumed by agricultural land is a powerful visualisation. One quick Google search and a screenshot does the trick 👌

5. Depending on the type of coal, station design etcetera, etcetera

6. FYI, you could also convert barrels of oil into Megaelectron volts (MeV), a unit of measure typically associated with the energy of sub-atomic particles, or kilotonnes of TNT, in case you were wondering how many cartoon sticks of dynamite Australians consume.

7. I’m being incredibly generous here in my assumption of what all of the various off-grid mining sites in Australia might use. I also assume that there is a more reliable number somewhere, but I couldn’t find anything useful in a quick search and I’m honestly not going down this rabbit-hole (this time).

8. The final report will be released in mid-2024.

9. Genuinely can’t believe I had to write this sentence, but here we are.

10. Maybe the IPA would be happier with this outcome? Although it covers not only the latte-sipping inner city elites but also the true blue battlers of the outer suburbs.

11. I guess you could have bee hives on the roof or a little market garden, but I don’t think the Australian government is counting people’s roofs as agricultural land.

12. He sold two versions of the banana for US$120k each, and ARE YOU FUCKING KIDDING ME.