China installed 217 gigawatts of solar power and 76 gigawatts of wind power in 2023. These are massive numbers and rather more than the 3.7 and 0.7 gigawatts of wind and solar power installed in Australia that year. The Chinese numbers are often quoted as some kind of proof that wind and solar can decarbonise the global electricity system rapidly and cheaply. So who is deploying wind and solar faster, Australia or China?
You need to divide by the populations to put these numbers in perspective. When you do that you see that Australia is going faster. If China was intending to decarbonise with wind and solar, the chart at the end of this article shows that it would take about 120 years, and yet China intends to be net-zero by 2060. Clearly, it has other plans. It's building many kinds of nuclear reactors, including one designed specifically to connect to a coal plant turbine, this will enable a rapid decarbonisation of all its recent coal plants without wasting the infrastructure of turbines, generators and grid connections. It may also be intending to mass produce smaller reactors just as it mass produces everything else. One thing is certain. China isn't following Australia in thinking it can rely on wind and sunshine. So why do we think we can decarbonise with wind and solar while the rest of the world has realised that there is no net zero without nuclear? You only have to look at how desperately slow we are going relative to what is required to understand why everybody else (except the Germans!) has woken up.
Let's see where we are and where we need to be.
Firstly: Can we build the ISP’s gas plants?
The Australian Energy Market Operator (AEMO) ISP 2024 edition is Australia’s plan to partially decarbonise electricity production. I say partially because the plan still includes a substantial amount of gas use, and will require about 9 gigawatts of new gas fired generators to be built over the next 15 to 20 years. That’s roughly 9 large scale nuclear plants worth of gas.
Our ISP is unique
We can put the ISP rollout speed in context by comparing it with other rollouts of energy technology around the world. Many countries are also engaged in such plans but in most other countries, they aren’t trying to do what we are doing. We are one of very few countries trying to build a system without a well established low carbon firming technology; either hydro, biomass or nuclear. Sweden has hydro and nuclear, Canada similarly, in Japan it is just nuclear, in the US it is nuclear and hydro, in Germany it is currently coal, gas and biomass, but the targets in Germany are no longer as clear as they once were.
The ISP’s reliance on building nearly two electricity systems isn’t well understood. The plan is for plenty of wind farms together with solar and batteries. These will be backed up by more and dirtier (higher carbon) gas turbines than we currently have, but which will typically only be running 10-20% of the time. In 2040 under the ISP, our gas plants will be running 10 percent of the time, then some will be retired, and those that are left will be running 20% of the time.
Building something truly independent of the weather and only running it 10-20% of the time isn’t silly, the trick will be to put them in the right places. A generator’s location matters for all kinds of reasons.
It is hoped that this will allow us to shut down our coal plants.
The first deals to extend the life of some of these coal plants have already been struck.
Here’s an indication of how ambitious the ISP is with respect to the fastest wind and solar deployment so far recorded; Finland.
The chart shows the ISP 2024 spreadsheet has Australia rolling out wind and solar 2-3 times faster than anybody has ever done it. More details later.
Building infrastructure takes a team
Many who lament the low ambition and glacial pace of Australia’s decarbonisation have never built anything, except perhaps a website. Nor is Australia noted for its manufacturing skills. We don’t tend to build much; just buy everything and perhaps assemble it. So our plans are utterly dependent on global supply chains and geopolitics.
Photovoltaic (PV) panels from China. Wind turbines from China. Gas turbines from Germany, China, Korea, among others. Transformers from … Australia! That’s right, these are one of the few things we make locally; kudos to those involved, Wilson and Tyree, among others. You are unlikely to have heard of them; transformers are one of the unsung tech heroes of any grid. Go for a walk around your local area and you’ll see them on a surprising number of poles. Rectangular boxes with fins are the most common but there are various designs. Really large transformers are astonishing devices weighing about the size of a nuclear reactor pressure vessel; 400-500 tonnes. The world’s largest transformer is double that size. We make some here but may opt to buy from elsewhere depending on a range of factors. The copper in these big monsters is wound by hand. This is one of the few tasks robots are yet to master; but probably not for long.
There is a global shortage of transformers at present; with waiting times of around 18 months for the big ones. Transformers are made with a special kind of steel (GOES, grain-oriented electrical steel) and a shortage of this steel, together with skilled workers has produced a global industrial supply chain blockage and price hike.
The range and depth of the skills required to build national grid scale infrastructure is truly awe inspiring. It’s rather more difficult than that first step of building a spreadsheet with numbers telling you what you want and when you want it.
International rollout speeds
Each year the Energy Institute publishes is Statistical Review of World Energy (formerly done by BP). I run some analytic code on the data to find who has added the most of what energy source over some specified period. The following chart was done with the recently released 2024 data.
It shows the top rollout speeds over a decade for various electricity technologies. You’d expect some new entries each year. In this case new entries will be for the period 2013 to 2023. We appear, along with the UAE, Finland, Denmark and Netherlands. But most entries are for past decades. Here are the top 30.
As I said above, the ISP envisages we can go 2-3 times faster than any country has gone before with wind+solara.
Did you notice how many small countries, by population, were in the above chart?
France is the biggest, with a population of 56 million in 1989 at the end of its fastest nuclear growth decade. Where is Japan? The US? China? India?
It’s much easier for small countries to grow anything at a high per capita rate than for big ones. Slovenia, with just 2 million people, only needed a single (smallish) nuclear reactor to make the top 30. Iceland gets in twice with a handful of geothermal and hydro projects.
The chart also shows the per capita clean electricity in each country with a notional clean electricity target of double the current level. This is the International Energy Agency (IEA) target for 2050. Australia’s current level of electricity use is just over 10 MWh/person, of which about 3.5 MWh/person was clean (in 2023), so getting to 21 MWh/person will take over 7 decades at our current rollout speed of adding 2.3 MWh/yr per person. If you are used to looking at data on OpenNem, you’ll notice that the total electricity numbers which the Energy Institute uses are a little higher than those in openNem. Some electricity usage doesn’t go through the market, it is “behind the meter” on mining sites and the like. The notional figure of electricity doubling shouldn’t be taken as gospel, it’s an IEA global estimate which will vary considerably between countries.
What about bigger countries?
At 26 million, Australia isn’t small, but nor is it in the big league. We have reached the stage where a high per person rollout speed won’t be trivial. If we select only countries with more than 20 million people in 2023, then all the little countries disappear and Australia’s recent wind and solar effort ends us in 3rd place and also in 7th for solar alone.
We also start to see the nuclear rollouts in Germany, Ukraine, the US, Japan and Spain. We also see the US’s more recent wind and solar expansion. China’s wind and solar rollout appears, but at less than half the rate of Australia’s. Charts of global renewable expansion frequently focus on totals without informing people that the figures are dominated by one country; China. They also typically don’t tell you the per-capita rate.
Lastly, the previous charts have excluded fossil fuel technologies. What happens if we allow them?
As you can see, wind and solar all but vanish; with Australia and Germany being the only representatives. China’s peak coal rollout finished back in 2011 and is double the rate it is achieving with wind and solar today. China’s nuclear buildout doesn’t make the table. Why not? China, with 1,400 million people, is working on 5-6 reactors in parallel. Back at the peak of its nuclear build, Japan was working on 8-9 in parallel. China simply doesn’t have the skilled workers for current nuclear technologies. Perhaps it will develop simpler and cheaper technologies. That will be truly game changing. We’ll see.
The bottom line, based on the above chart, is that there is something making wind and solar particularly slow everywhere, relative to fossil fuels or nuclear at its best in a technologically advanced country.
Bottlenecks and targets
What is it that makes wind and solar so slow to build everywhere?
Over a decade ago the World Wildlife Fund proposed that covering “just” 1 percent of the planet with solar panels would generate all the energy we needed. My response was “81,000 Truckers for solar” pointing out that nobody, even Australia, which has rather a lot of big trucks, has enough trucks to carry the panels to their resting places, assuming we had the panels in the first place. Even more modest proposals for solar development run into similar trucking hurdles, and mining hurdles.
But I’d reckon it is the land footprint coupled with the requirements for the slowest technology of any to build; transmission. Transmission is slow not because of its total footprint, but the number of boundaries it crosses. Put a road through 100 farms and see how long it takes to get approvals. Add in a few towns, cross a few railway lines and the complexity rises still further.
There’s nothing intrinsically slow about building a wind or solar farm; once you have the transmission in place, and the land environmental impact assessments and the grid approvals that demonstrate you won’t destabilise the grid by putting that particular level of generation in that particular place. Compare this with nuclear power. Getting site approval is typically a nightmare, but once you have it, you can build a source of electricity bigger than tens of thousands of hectares of wind or solar. You also minimise the transmission requirements. The time consuming part about big projects isn’t the big project.
As land issues go, Australia thinks of itself as a vast empty country. It isn’t. Nor does anybody really want to build anything in the areas where approvals might be fast. Everybody wants to build where there is a supportive population and infrastructure.
But for global decarbonisation, materials for renewables will be an increasing impediment to getting from 1-2 MWh/person per year, to the 5 that France got with nuclear or the 6 that the ISP envisages.
The ISP is pretty much just a spreadsheet tweaked to fit Government policy. It isn’t a plan. Under the National Electricity Rules, AEMO has to obey the National Electricity Objective, which requires it, simultaneously with promoting investment to make electricity cheap and reliable, to promote investment to:
“…the achievement of targets set by a participating jurisdiction— for reducing Australia’s greenhouse gas emissions; or that are likely to contribute to reducing Australia’s greenhouse gas emissions.”
In other words it is required to do what it has done; mock up a spreadsheet which will deliver the Government of the day the targets sets. It doesn’t matter if they can’t be met. By the time anybody wakes up to this, somebody else will be in charge. Giving a Government full and frank advice about its targets probably isn’t a good career move.
An actual plan, on the other hand, considers supply chains. A plan has an idea of who is doing what, coupled with where and when. It can do critical path analysis to identify and fix bottlenecks. That’s not the ISP.
If you really wanted a plan to do something that had never been done before, you’d need to detail all the supply chain requirements and have a plan for each and every one.
If it was only Australia rolling out wind and solar technology, we could (assuming we could find the cash) just buy our way to the head of the queue and let somebody else stick with coal for a bit longer. That’s what Germany did when Russia invaded Ukraine. Germany rapidly built some Liquified Natural Gas (LNG) import terminals and outbid poorer countries for the world’s LNG, leaving the global south to reduce their consumption and burn more coal or suffer blackouts, as in Pakistan. Is that what Australia should do to ensure we meet our targets? The climate doesn’t care who reduces their emissions. Displacing coal in China is just as good as displacing coal in Australia; and you don’t need to ship the panels as far to get the job done.
It isn’t all about us.
Stupidity with an Australian flair
There is a uniquely Australian kind of stupidity in our ISP that exceeds that of similar plans; namely that because we have so much sun we are well placed for rooftop solar. This creates an unbeatable competitor for anybody building utility scale wind or solar. The 2023 headline of grid watching software maker and blogger WattClarity said it all: Rooftop Solar is crushing* the returns for Large-Scale VRE. Utility scale suppliers will face an ever shrinking market and need to raise prices to compensate.
We’ve chosen stupidity over good policy for a couple of decades and are now rushing to meet an impossible target using the ultimate Australian industrial skill; the spreadsheet.
If you really want to know how likely we are to meet this target, you need to be watching the global minerals industry; with particular focus on minerals processing, especially in China. It’s incredibly slow and expensive to establish a new mine, hence miners like to make very sure of global markets before committing. Currently, mines are shutting rather than ramping up because mineral prices are falling. A hiccup in the electric vehicle (EV) sales growth looks like the main driver, but you’d best look to people with more knowledge about causal chains in the critical minerals business than me.
Does Australia have the smarts to build wind and solar 2-3 times faster than anybody else has done over the past 20 years? Or are we simply good at spreadsheets?
Comments