Keynote Address to Waterpower Week 2022
“Why Pumped Storage is the Ignored Crisis within the Energy Crisis”
The Honourable Malcolm Turnbull AC
6 April 2022
Barely a day passes without another reminder of the urgency of the energy transition. Whether it is fire and floods in Australia, this week’s latest grim warning from the IPCC or, indeed, Putin’s brutal invasion of Ukraine.
There is a lot of lazy, glib talk about how the transition to net zero requires new technologies, innovations yet to spring from the imagination of a new generation of inventors.
Don’t believe a word of it. We have the tools to do this job, right now. Variable renewable energy - wind and solar - supported by storage - batteries, pumped storage and green hydrogen - can take us to a zero emission energy world.
Of course new technologies, and refinements to existing ones, will emerge - whether it is bigger and better batteries and electrolysers, smarter turbines, or the creation of virtual power plants to name just a few.
But right now, the single most important priority should be the planning and construction of Long Duration Electricity Storage (LDES) in the form of pumped hydro.
In most of the world,including the United States and Australia, this is not yet happening at anything like the pace we need.
This is the ignored crisis within the energy crisis.
Today pumped hydro provides about 95% of the world’s electricity storage..
Solar and wind are already, in almost all markets, cheaper than new thermal generation, and getting cheaper. Because they are zero marginal cost generators, as they are deployed coal and gas fired generation is closing down - when the sun is shining and the wind is blowing the prices are so low the coal fired generators are all too often operating at a loss.
But the wind doesn’t always blow and the sun doesn’t always shine…we know that. Neither does it rain all the time. And we worked out how to store water a long time ago.
So the answer is the same.
The only way to make renewables reliable in the zero emission energy system we need is with storage and that is going to include long duration storage for 8, 12, 24 and more hours. How much and how long will depend on the circumstances of each grid connected system. And it will be pumped storage that will do the heavy lifting.
The IEA recently forecast that pumped storage will provide 42% of global expansion of electricity storage, of all durations, with at least 40GW in the next five years reaching a total of 200 GW by 2026. IRENA forecasts that by 2050 the world will have trebled today’s pumped storage of 160GW to 420 GW.
Now when I was Prime Minister of Australia we were faced with a premonition of what can go wrong if we don’t carefully plan the clean energy transition.
In 2016 South Australia experienced a massive blackout.
More than any other Australian state South Australia had been building out industrial scale wind and solar as well as enormous amounts of residential rooftop solar - in which Australia is the world leader. This was destroying the business model of continuous coal burning thermal generation which was beginning to close.
The South Australians had relied for their back up on long transmission lines to thermal generation in a neighboring state. A storm had knocked some of them out as well as triggering a shut down of several large wind farms.
There were several factors, as there always are, but one really registered with me. We were building more and more variable renewable energy and not even thinking about the storage we would need to back it up. There seemed to be a breezy assumption that gas peakers would always be available at $4 GJ - then but a happy memory.
Over the Australian summer of 2016/17 I researched the history and operation of pumped storage. We didn't have much of it in Australia and hadn’t built any for more than thirty years.
We had to get cracking. I had learned over the years in other projects that Moore’s law does not apply to digging holes, whether they are trenches in the street, 30 hectare turkeys' nest dams or vast penstocks and turbine halls deep under the ground.
The short point wás that we could roll out solar and wind in months, but to build pumped storage required years. Batteries were coming onto the scene but were nowhere near being credible as long duration storage technologies.
So I gave a speech in early 2017 which highlighted the importance of pumped storage.
Australia is a large, dry and mostly flat continent. So we don’t have a lot of hydro - compared to other parts of the world. We have two big hydro schemes - Snowy Hydro in the mountains between NSW and Victoria and Hydro Tasmania.
I followed my speech up with calls to both utilities. The result is that Snowy Hydro 2.0 is under construction, it will be a 2000 MW scheme linking two large dams, Tantangra and Talbingo, about 27 kms apart and 700 metres difference in elevation. It should be generating in 2025 and will be able to run for 7 and a half days without repumping; storing 350 GWh of electricity.
Magnificent though Snowy 2.0 will be, it is not sufficient for Australia’s LDES needs.
We need much more long duration storage. AEMO is the Australian Energy Market Operator and it manages our National Electricity Market (NEM) which covers all of Australia with the exception of the Northern Territory and Western Australia (whose energy market it also operates). At 5,000 kilometres from top to bottom the NEM is the longest connected electricity market in the world.
AEMO has forecast we will need at least another 45 GW of long duration storage in our NEM alone. The global figure is close to 900 GW.
So what should it look like?
Critics of pumped storage will say every project is unique and if you focus on the big ones, like Snowy 2.0, that’s probably right. But you could also say every building is unique.
The key is that we need to build out the pumped storage the world desperately needs in a format that is repeatable. The standard technique which has been used around the world in years past is worth reviving - build a turkey’s nest dam high on a ridge next to an existing reservoir and establish a closed loop system.
Another is to build two new dams, both off river. As they are a closed loop system they can be filled once and topped up over time to make up for any net evaporation. Andrew Blakers’ Global Pumped Storage Hydro Atlas identifies over 600,000 such sites around the world all in environmentally low impact locations.
So how do we get it done?
First each energy market operator has to assess how much long duration storage it needs, and at what durations. In doing so we need to recognise that in almost all markets without nuclear power, the end of thermal generation means the end of continuous generation.
So we will need storage to deliver power over night, especially over low wind nights, and we will need even more to deal with periods, as in winter, of low radiation often coupled with low wind - what the Germans call dunkelflaute.
All of this can be done, but it will require leadership from Governments to make it happen. And we don’t have any time to waste.
Snowy 2.0 is an Australian government project, so I suppose one solution is for Governments to do it all, but that’s neither repeatable or desirable, at least in free market economies like ours.
Clearly we need a competitive market mechanism that incentivises developers to build storage capacity some of which will not be cycled every day, like batteries, but held in reserve for those cloudy windless days or weeks when renewable production is low. And at the lowest cost to protect the interests of consumers.
The leading example of such a mechanism can be found in Australia’s largest state, New South Wales.
In 2018, Water NSW which owns and operates many dams across the State started calling for expressions of interest to establish pumped hydro schemes adjacent to them.
There are hundreds of potential locations including in areas where coal fired generation and mining is going to close. Whether it is Australia’s Hunter Valley or America’s West Virginia, pumped storage can provide new clean energy jobs for communities where coal is shutting down. And most importantly it can do so in locations which take advantage of existing transmission.
Water NSW was overwhelmed with the level of interest shown by developers. Almost all of the proposals involved building turkey’s nest dams on top of a hill adjacent to the dam, many involved very little tunnelling and as closed loop systems minimal environmental impact.
However financiers were still holding back - they understood the many revenues that could be derived from pumped storage - arbitrage, grid stability, frequency maintenance and so on. But it was easier, they said, to see the business case for batteries with only a few hours storage; ideal to take advantage of the early evening peak demand spike.
But the NSW Government advised by AEMO recognised that long duration storage, 8 hours plus, was going to be needed and especially in a world without continuous thermal generation.
So as part of the State’s Energy Roadmap, Energy Minister Matt Kean has proposed LTESAs (long term energy supply agreements) which would see the State grant a series of options to the developer of long ( 8 hours plus) duration storage, which would almost certainly be a pumped storage project, to require the State to guarantee a level of income to the project. This is not referenced to a price per MWh delivered or stored.
Revenues actually received would be offset against the guaranteed amount. Projects which were net recipients of state support when options were exercised would have to share future revenues which exceeded a revenue threshold. These options will be granted in a competitive process - those seeking the least support for the most long duration capacity will be preferred.
This scheme, still in development and to be finalised by the end of this year, is not like a contract for difference or a floor price for energy. And that makes sense when we talk about storage, because the policy objective is not to secure the generation and despatch of renewable electricity, but rather to make sure there is a quantity of stored electricity able to be despatched as and when required.
The storage LTESA proposal will deliver, at least cost to NSW electricity customers, the amount of long duration storage the system will need. It has the potential to essentially underwrite a level of income for a project that would cover all or much of the project’s debt financing obligations.
All of you will be much more familiar with the policy environment here in Washington than me. So I will venture into local affairs with some trepidation.
Pumped storage, and the need for it, is not a secret. One of the world’s largest pumped storage schemes is only 200 miles from here in Bath County Virginia. But it was opened 36 years ago. More recently the International Hydropower Association and your Department of Energy collaborated on an International Forum on Pumped Storage Hydropower. Its chairs were myself and Asst Secretary for Energy Kelly Speakes Backman.
It identified the single greatest challenge to be securing financing for pumped storage in a rapidly changing and hence uncertain energy transition.
Regrettably, however, pumped storage is a long way from getting the attention it needs at the political level. President Biden’s clean energy agenda fact sheet does not appear to mention it at all. All too often people say “there are no suitable sites” which - as Andrew Blaker’s ANU study* of over 600,000 off-river sites shows - is manifestly untrue, or “its too expensive” which begs the question of how you otherwise keep the lights on overnight without burning fossil fuels? And if you think pumped storage is expensive, how about nuclear?
Some argue that hydropower is environmentally and socially harmful, and in the past it has been if not managed well. Pumped storage however does not require rivers to be dammed, storages can be relatively small and operate as a closed loop
And in any event, now there is a Hydropower Sustainability Standard agreed between companies, NGOs and governments, which IHA CEO Eddie Rich will talk about later.
Going forward, the only acceptable hydropower is sustainable hydropower.
Perhaps Joe Manchin would not have helped kill the Build Back Better Bill if there had been more opportunities for pumped storage in his state, although even there, where it would make such a difference socially, economically and environmentally, it has had difficulties securing public support.
To conclude with three observations.
The first is a penetrating glimpse of the obvious. We have to stop burning coal and gas. We have the renewables to generate unlimited power, but it is intermittent. To make it reliable we need long duration storage and the best solution to that end is pumped storage.
The second is that we need to get cracking and build it. Not just because it takes more time than building a solar farm but because we need to show it can be done repeatedly and at reasonable cost. Seeing will be believing. There was a time when people didn’t believe solar PV would work at scale either.
And the third is a call to action. We have to put this issue on the agenda everywhere.
Advocacy is needed, more than ever –like the new global campaign by IHA, NHA and other partners around the world, ‘We can, #withhydropower’.
Without pumped storage, the energy transition is in danger of stalling just as it should be accelerating. The energy transition, the achievement of net zero, containing global warming to 2 degrees all depends on it, depends, in other words, on us.