The Liberal Party was voted out in 2022 by Labor, which ran on a platform that included action on climate and promoting renewable energy. Today, as a private individual, Turnbull’s business interests include developing two large-scale pumped hydro energy storage (PHES) projects, along with a wind farm, through his company Turnbull Renewables.
His speech this morning focused largely on the need for long-duration energy storage (LDES) for the Australian grid, arguing that PHES and batteries are technologies that both have an important role to play, not just in decarbonisation but also in providing security of energy supply and lowering costs of running the system.
Below, you can read a transcript of his prepared remarks at the event in Sydney today.
‘Challenge has been very clear for some time’
Let me cut right to the chase. We know that renewables are the cheapest form of new generation, particularly in Australia, and they will play the leading role in our energy systems into the 2030s and beyond, and we’re moving into a world where most of our primary generation is going to be either solar or wind.
And I think it’s becoming increasingly obvious that the mixture, as between solar and wind, is going to have more solar than we originally anticipated, because of the challenges of building wind, and that, of course, because of the correlation of solar to when the sun is shining, we’re going to have a need for even more storage than we’d originally anticipated, and that’s what I’m going to talk to you about today.
Now, the update to the Electricity Statement of Opportunities (EESO), AEMO as you know, has forecast gaps in all mainland NEM regions in the next decade. That challenge has been very clear for some time, but it’s worth repeating. We need to firm up the various variable renewables with storage when the wind doesn’t blow and the sun doesn’t shine. Variable renewable energy supported by storage: batteries, pumped hydro, green hydrogen, can take us to a zero emission energy world. It’s absolutely doable.
The level of storage in AEMO’s draft 2024 ISP forecasts an additional six gigawatts of utility storage will be needed to be added to the NEM between 2030 and 2035, but these Step Change scenario forecasts in our view are at risk of underestimating the long-duration storage requirements of the NEM due to four key factors.
Firstly, the ISP Step Change scenario for distributed storage, and this is essentially home batteries, EVS and so forth, is underpinned in part by an unrealised assumption that the Commonwealth will establish a national household battery rebate that subsidises 50% of the capital cost. Now that rebate does not exist, but that’s factored into the ISPs assumptions.
Secondly, the lengthy planning assessment processes for wind farms may see utility-scale and rooftop solar PV grow at a faster pace than wind which further increases storage requirements, as I said earlier. From 2016 to 2024 capacity additions in the NEM comprised 50 wind projects, 9.6GW capacity, compared to 111 solar projects, 13GW, and 17GW of rooftop solar.
Now, the assumption, the third fallacy or flawed assumption, is that the gas market is endlessly flexible, and that assumption is not realistic. For the older people in the room, this reminds me of the what I used to call in Lotus 1-2-3 days the backslash-copy School of Modeling, where you just take an assumption and assume it’s going to be good forever. And I’ll come to some work that deals with that. But treating gas as the sort of limitless corrective that’s going to solve gaps in generation is not viable.
Recent research suggests that the pipeline capacities and gas storage capacity constraints will limit the gas supply needed to adequately meet demand during the winter months.
And finally, slower than expected transmission rollout could increase energy storage requirements by two to five times. Now, one of the problems AEMO has, is because it’s a creature of governments, for the most part, they’re not really in a position to say, let’s assume that all of the government policies that have been announced, fail to deliver as proposed. You can imagine the stakeholders will be peeved by that, but there has to be a more realistic approach to the modelling.
All those four factors mean we need much more long-duration storage that’s been forecast. Paul Simhauser, the CEO of Powerlink which is the big transmission utility in Queensland, and Associate Professor Joel Gilmore, Griffith University have recently done some work and published on the Griffith University energy policy research site, which suggests that a dispatchable long-duration storage target of 162GWh of storage spread across the NEM with a capacity of at least 9GW by 2035, is more appropriate, and that’s in addition to the existing pumped hydro projects under construction, Snowy 2.0 and the much smaller Kidston.
‘Much more needs to be done to incentivise long-duration’
Now, the Capacity Investment Scheme is a welcome policy that will provide greater investment certainty to a range of important renewable projects, but much more needs to be done if we’re going to incentivise the development of long-duration storage that we need to maintain reliability.
Short-duration batteries played a role in firming, wind and solar but we cannot afford to put all our eggs in one technology basket. It’s not a choice between batteries and pumped hydro, we need both.
And as more and more coal comes out, we’re going to need much longer duration storage than we’ve been able to operate with today. In terms of cost, the CSIRO’s own work, as you know, shows that as of next year, 2025, the levelised cost of storage in an 8-hour pumped hydro project is forecast to be about 30% cheaper per megawatt-hour than an 8-hour lithium-ion battery.
Snowing 2.0, much criticised, now with a headline cost of AU$12 billion nonetheless offers storage energy of 350GWh at a cost of approximately AU$34 per kilowatt-hour, which is far below the equivalent cost of batteries.
In addition to being more competitive on cost, pumped hydro has a lifespan of 80 to 100 years. Properly maintained, the civil infrastructure, which is where the bulk of the money is, such as the dams and tunnels, have an indefinite economic life. The electromechanical part of the plant can be operated for many decades, 50,60 years and more if properly maintained. So pumped hydro is absolutely a long-term infrastructure.
It can also defer or reduce the need for network upgrades by balancing load profiles and absorbing excess energy, thereby optimising and increasing the effective carrying capacity of existing infrastructure, that assists by minimising the costly and often unpopular transmission upgrades.
I’ll make another point which is very pertinent. You know, when we talk about, a Future Made in Australia, the investment in pumped hydro is overwhelmingly in Australia. I mean, it’s in the civil works. It is in labour and materials sourced locally and while batteries, whether short-duration or long-duration, are almost entirely imported. So if we’re serious about having investment in infrastructure made in Australia, learning to be able to build and rollout extensive pumped hydro effectively is going to be a big part of that.
This issue, by the way is central to the Indian government’s focus on pumped hydro which some of you may be familiar with. Now, there are currently around 292GWh publicly-announced large-scale pumped hydro projects under development in Australia, and this is in addition to Snowy and Kidston, which are [now] being built. Several are looking to repurpose existing mine voids as reservoirs, but most like our projects involving either two new reservoirs being built, or more often, one new reservoir being built on a hill next to an existing one, which is the case with our projects.
‘We need to look further ahead than 2030’
Now, regrettably, Moore’s law does not apply to digging holes and so we need to drive step change in the pace of delivery of pumped hydro projects and we believe that there are four things that need to be done to the Capacity Investment Scheme. I’ll just run through these and I’m very happy to take some questions and discuss them later.
Firstly, we must set the 2035 target for clean dispatchable capacity. Now, the 2030 target is fine, but we have to be looking further than that, and beyond that. The 2035 dispatchable capacity target sends a clear investment signal to long-duration storage developers who are making investment decisions now.
There should be separate tenders for short to medium storage, that is to say less than eight hours and long duration more than eight hours. Some people would argue really long-duration should be 12 hours but let’s leave that debate to another time. In the absence of a capacity market, the current Capacity Investment Scheme designed inherently drives developers to invest in short-duration projects as these represent the strongest business case. They’re commercially attractive as they arbitrage the morning and evening peaks from the middle of the day solar prices.
But as Australia moves to a NEM dominated by renewable energy generation, stressful periods will emerge that are characterised by lengthy winter periods of low solar and wind availability, as a couple of weeks ago, and can get a lot worse than that.
Long-duration energy storage can be supported if the Capacity Investment Scheme would run clean dispatchable CIS tenders for storage with a minimum duration of at least eight hours.
Thirdly, we believe we need to encourage the most competitive long-duration storage bids via support for pre-bid costs for contractual contingencies. Pumped hydro has a lower levelised cost of storage once operational, but it involves significant upfront costs, time and risk to get to the stage where investors can apply to a CIS tender compared to other types of storage projects.
The planning and geotechnical works are complex, development costs will run into the tens of millions of dollars before a developer can be successful in bidding. A large risk taken before a competitive bid can be submitted has the consequence of deterring or slowing pumped hydro developers who are proceeding more cautiously.
Finally, and fourth, align the Capacity Investment Scheme Agreement (CISA) or support to the technology type. The CISA is designed with a contract term of 15 years. Pumped hydro has a lifespan of 100 years. The New South Wales LTESA by contrast has a contract length of 40 years, which is more appropriate.
The UK’s Cap and Floor scheme, that contract scheme is based on the project lifetime depending on the technology. So extending the contract term will provide revenue certainty in the later years of the project life.
‘Every resource in abundance except one’
Now, I believe that unless we get cracking with building the pumped hydro we need, more of it, the energy transition will fail. Storage is the critical requirement, but there seems to be a breezy assumption which Paul Simhauser has exploded with his work that either gas will fill the build, an endless supply of gas peakers, which of course is burning fossil fuel, which is what I thought we were trying to stop doing. Or and/or that batteries will miraculously solve all the problems.
We have the capacity, as Andrew Blaker has demonstrated in his work, that I’m sure you’re all familiar with, and if you’re not, check out his Pumped Hydro Atlas. So, work that was done with ANU where Andy is a professor, that we have so many buildable sites in Australia, that’s not the problem.
We don’t need a lot of water, but the problem is you’ve got to get cracking and build it and it takes time to get the approvals and do the geotechnical work and so forth.
So, we’ve got every resource to deal with this energy transition, every single resource in abundance, except one, and that’s time.
Thank you very much.
Transcribed with minor edits by Andy Colthorpe.
