This special interest group is for professionals to connect and discuss all types of carbon-free power alternatives, including nuclear, renewable, tidal and more.

Post

Millions of Gigawatt-hours of cheap energy storage to support solar and wind

image credit: Global off-river pumped hydro atlas, ANU

Andrew Blakers's picture
Professor of Engineering Australian National university

Andrew Blakers is Professor of Engineering at the Australian National University. He founded the solar PV research group at ANU. In the 1980s and 1990s he was responsible for the design and...

  • Member since 2021
  • 29 items added with 1,333 views

President Biden plans to halve greenhouse emissions by 2030. This requires lots more solar and wind generation, and lots of cheap energy storage. Fortunately, there is a vast, low-cost, off-the-shelf, environmentally benign storage solution, namely pumped hydro.

There is a twist: nearly all good pumped hydro sites are not on rivers - they are off-river sites, located away from any river.

Remarkably, pumped storage is virtually ignored in reports and analyses because people wrongly equate pumped hydro with new dams on rivers.

Your access to Member Features is limited.

Pumped hydro energy storage constitutes 99% of global storage energy. However, river-based pumped hydro is constrained by geographical availability and opposition to new dams on rivers.

An off-river pumped hydro system comprises a pair of reservoirs located away from any river and spaced several kilometres apart with an altitude difference of 200-800 m and connected with pipes or tunnels. The reservoirs can be purpose-built or can utilise old mining sites or existing reservoirs. On sunny/windy days, water is pumped to the upper reservoir. At night-time, the water flows back down through the turbines to recover the stored energy.

A pair of 100-hectare reservoirs with an altitude difference of 600 metres and 20 metre depth can store 24 Gigawatt-hours of energy, which means that the system could operate at a power of 1 Gigawatt for 24 hours. This is enough storage for a city of a million people relying mostly on solar and wind.

Off-river pumped hydro means that new dams on rivers and bespoke engineering in remote river valleys are avoided. Also, minimal provision needs to be made for flood control. Very little land and water is required: about 3 m2 per person and 3 litres per person per day respectively (averaged over 50 years) to support a 100% renewable electricity system.

The global pumped hydro storage atlas lists 616,000 potential sites with an enormous 23 million Gigawatt-hours of combined storage. The Atlas allows you to view the 3-D shape of any one of the hypothetical reservoirs here. Zoom in to see the tunnel route and the dam walls. Click on tunnel or reservoir for detailed information pop-ups.

The USA has 35,000 good potential sites outside national parks with combined storage of 1.4 million Gigawatt-hours. About 1% of these are needed to support a 100% renewable electricity system, which allows utilities to pick and choose the best sites. The sunny and windy states in the south eastern Rockies have thousands of sites, which allows them to export steady 24/7 solar and wind electricity to other states.

Little pumped storage has been built in the USA in recent years because there wasn’t much need for storage. However, much more storage will be needed when solar and wind generate 50-100% of electricity.

Australia (with 8% of the US population) is installing solar and wind 3 times faster per capita than the USA and is facing up to the need for mass storage right now. It has three existing pumped hydro systems built several decades ago. Two pumped hydro systems are under construction (Snowy 2.0 and Kidston) with combined energy and power capacity of 350 GWh and 2.3 GW respectively. Construction time is 3-5 years. Another dozen are under serious consideration. None involve new dams on rivers. Several GW of 1-3 hour batteries are also being deployed.

Pumped hydro energy storage and batteries complement each other. Batteries are best for short term (minutes to hours) high-power storage, while pumped hydro is best for overnight and longer duration storage (which is essential for a solar/wind dominated system).

The capital cost of a Gigawatt-rated off-river pumped hydro storage system with 24 hours of storage in a good site is $1-2 billion for a system that has a working lifetime of 50-100 years, low operating costs and whose working fluid is water rather than electro chemicals.

As the US rapidly deploys solar and wind to reduce emissions, it has effectively unlimited options for no-new-dam, low-cost, off-river, pumped hydro storage.

Andrew Blakers's picture
Thank Andrew for the Post!
Energy Central contributors share their experience and insights for the benefit of other Members (like you). Please show them your appreciation by leaving a comment, 'liking' this post, or following this Member.
More posts from this member

Discussions

Spell checking: Press the CTRL or COMMAND key then click on the underlined misspelled word.
Matt Chester's picture
Matt Chester on May 12, 2021

The USA has 35,000 good potential sites outside national parks with combined storage of 1.4 million Gigawatt-hours. About 1% of these are needed to support a 100% renewable electricity system, which allows utilities to pick and choose the best sites.

Wow, those are some striking numbers! It leads to the obvious question-- what's the hold up? Why is all the focus on R&D for battery tech if we already have the tech to make this happen? And down the road, will pumped hydro and batteries be more competitors or complementary tools? 

Andrew Blakers's picture
Andrew Blakers on May 12, 2021

Because the US has relatively little variable solar and wind so far there hasn't been a strong focus on supporting storage. Batteries solve local problems better than pumped hydro. BUT: as lots more solar and wind is installed then utility-scale storage is needed, and pumped hydro enters the picture.

Many people wrongly equate pumped hydro with more dams on rivers. However, 99% of the USA is not near a river, and nearly all good pumped hydro sites are off-river.

Matt Chester's picture
Matt Chester on May 13, 2021

It seems, then, that it might be a bit of a chicken and egg scenario-- if there was readily available storage, wouldn't more solar and wind be built at a quicker pace? 

nearly all good pumped hydro sites are off-river.

With this in mind, how restrictive is it to find a good pumped hydro site? If you were to randomly select a spot on the map, how likely would it be that there's somewhere nearby pumped hydro could be built if it was needed/desired? 

Andrew Blakers's picture
Andrew Blakers on May 13, 2021

Storage doesn't have to be co-located with a load. It just has to have adequate connection to a load.

Storage needs to be built at a steady rate, along with transmission, as the solar & wind fraction rises.

Texas (and adjoining states to the west) is a great place for storage because of great wind, sun and off-river hydro potential. It just needs to adequately connect itself to the rest of the country.

Bob Meinetz's picture
Bob Meinetz on May 12, 2021

Andrew, the devil is in the details. The system you describe:

"....could operate at a power of 1 Gigawatt for 24 hours. This is enough storage for a city of a million people relying mostly on solar and wind...."

What happens after two days of cloudy, calm weather...all the lights go out in this city of one million people? All businesses and schools close their doors?

This is the same renewables pixie dust we've been hearing about for half a century, the same myths that have brought us to the desperate straits we're at now. Enough is enough.

Andrew Blakers's picture
Andrew Blakers on May 13, 2021

The USA needs about 1GW and 24 GWh of storage per million people to support 100% renewable electricity, with enhanced interstate transmission to allow strong geographical smoothing of weather and demand at national scale.

Australia is similar to the USA in many ways except with 8% of the population and 3X faster per capita deployment rate of solar & wind. The renewable energy fraction (6-month moving average) is at 30% and heading to 50% in 2025. Prices on the wholesale spot market (both current and 2024 Futures) are sitting at $35/MWh. Electricity sector emissions are falling fast. The grid easily meets availability benchmarks. What is not to like?

The USA has great sun, wind, off-river pumped hydro potential and technical capability. There is no reason that the USA can't rapidly implement solar, wind, transmission & storage. Indeed, its strictly necessary to reach President Biden's goal of halving emissions by 2030.

Bob Meinetz's picture
Bob Meinetz on May 14, 2021

"The USA needs about 1GW and 24 GWh of storage per million people to support 100% renewable electricity..."

No, by your figures the USA needs 1GW and 24GWh of storage per million people to support 100% renewable electricity for one day. In Australia, the "enhanced interstate transmission" you describe only allows SA to import coal and gas generation from NSW and Victoria to help keep the lights on. One day of reliable electricity, it seems, isn't good enough.

Many experts have explored the possibilities for 100% renewable grids in greater detail than you; none share your boundless optimism. The Intergovernmental Panel on Climate Change (IPCC), in its 2014 AR5 Synthesis Report, includes no 100%-renewable scenarios among its viable options for mitigating climate change. There's no indication when AR6 is released next year that will change. Clack et al's Evaluation of a proposal for reliable low-cost grid power with 100% wind, water, and solar only considers one for the purpose of revealing its "errors, inappropriate methods, and implausible assumptions." That's a compliment - the foundation on which most 100%-renewables proposals rest would be better described as "imagination, wishful thinking, and voodoo."

If you want to make a case for a 100% renewable energy-powered grid the first step would be to provide an example of one, but I can save you the effort - there aren't any. Without a working prototype, betting the farm on someone's outlandish dream is a prospect best consigned to the Dustheap of Bad Ideas. There's no more time to waste.

Andrew Blakers's picture
Andrew Blakers on May 14, 2021

Dear Bob: why do you dislike solar & wind so much? They are generation technologies that happen to comprise 75% of global net generation capacity additions (and 99% in Australia).

The IPCC 2014 report was vastly wrong on solar & wind. An evaluation of its failings is here.

South Australia reached 70% solar & wind (6 month moving average) over the summer, heading for 100% in 2024. It achieved 100% for 7 separate calendar days (24 hour periods).

South Australia has a skinny interconnect (capacity = quarter of peak load) to the eastern states. That interconnect might get doubled in the next few years which will allow the state to push way beyond 100% through large-scale exports to the east. The wholesale spot price in the state is $35/MWh both now and in 2024 (Futures Market), which is the same as in the National Electricity Market.

Bob Meinetz's picture
Bob Meinetz on May 16, 2021

Andrew, I don't dislike solar and wind at all. I dislike the way their contribution to progress in the fight against climate change is being exaggerated - at a time when accurate assessment is critical.

You don't get off to a good start by claiming

"The IPCC 2014 report was vastly wrong on solar & wind. An evaluation of its failings is here,"

then referring to a paper you co-authored. Self-citations are generally considered acceptable if two conditions are met: 1) Your work supports or augments the work of other references cited in your claim, and 2) and a source is provided inline or as a footnote. Neither condition is met here.

The paper you reference reveals many examples of confirmation bias:

• Authors claim "...photovoltaics (PV) is now a mature technology, which is ready to deploy at the multi-terawatt scale" without any indication such deployment is practical or even possible;

• Authors claim "it [adoption of solar] shows a very steep progression" accompanied by a graph with adoption graphed exponentially to exaggerate solar's progress;

• Authors' claim that "solar provides the cheapest electricity in many parts of the world" disingenuously compares an intermittent, unpredictable source to ones which can generate electricity on demand, thus trivializing the value of availability and reliability;

• "Conversely, researchers in the PV community estimate that in 2050 solar electricity could contribute 41 to 96 PWh/year." I understand researchers in the coal community present the future of coal-fired generation in similarly-glowing terms.

Other examples abound, in both your paper and the works of others "in the PV community." They are case studies of confirmation bias: the work of groups with a pre-existing agenda. Whether their agenda is driven by financial or ideological considerations is unimportant.

Andrew Blakers's picture
Andrew Blakers on May 18, 2021

Solar & wind comprise 3/4 of global net capacity additions. Many people are collectively laying out $multi-100-billions in the well-founded believe that solar & wind are the cheapest energy sources.

Jim Stack's picture
Jim Stack on May 15, 2021

Great article. Hydro is a great resource. We even get close to 10% of our area power from hydro in the Arizona desert. 

Andrew Blakers's picture
Andrew Blakers on May 18, 2021

Pumped hydro in Arizona, Texas etc can facilitate a vast industry exporting solar & wind to the east & west coasts, supported by pumped hydro to ensure high load factors on the transmission lines ("send solar down the lines in the middle of the night").

Michael Keller's picture
Michael Keller on May 17, 2021

Good luck trying to navigate through all the regulations that stymie deploying energy projects that impact the environment. As a passing remark, dozens of large and small dams have been removed.

Pumped storage requires reservoirs with a noticeable elevation difference. A fair amount of water is required and the environment is affected.

Please note, I do not object to pumped storage, but the hurdles that must be overcome are significant. The realistic potential is limited when considering the extensive regulations in the US, with even more regulations being unleashed by the bureaucrats.

Andrew Blakers's picture
Andrew Blakers on May 18, 2021

Regulations can of course be used to inhibit solar & wind, but its hard to stop them in an open market because they have a compelling economic advantage.

Rooftop solar is a great way to start. Little Australia has 13 GW of rooftop solar (500 Watts/person) equivalent to about 170 GW in the USA. Offshore wind could really take off under the Biden administration.

Pumped hydro in Arizona, Texas etc can create a vast industry exporting solar & wind to the east & west coasts, supported by pumped hydro to ensure high load factors on the transmission lines ("send solar down the lines in the middle of the night").

The amount of water needed is small: less than 1 Gigalitre per GWh, which goes round and round a circle for the next 100 years. Rainfall exceeds evaporation if evaporation suppressors are used (floating plastic balls).

The amount of land needed is small: A pair of 100-hectare reservoirs with an altitude difference of 600 metres and 20 metre depth can store 24 Gigawatt-hours of energy, which means that the system could operate at a power of 1 Gigawatt for 24 hours. This is enough storage for a city of a million people relying mostly on solar and wind.

Matt Chester's picture
Matt Chester on May 18, 2021

The amount of water needed is small: less than 1 Gigalitre per GWh, which goes round and round a circle for the next 100 years. Rainfall exceeds evaporation if evaporation suppressors are used (floating plastic balls).

My attention is always caught when I see new types of units I haven't gotten to play with before! Having never worked with a Gigaliter before, how does that compare with other types of water needs-- what'st he Gigaliter per unit energy for hydrogenation (if different than pumped hydrostorage), for example? Or what are the levels of water use needed per GWh for centralized power plants that notoriously use high levels of water, like nuclear? 

Thanks for the informative discussion, Andrew!

Andrew Blakers's picture
Andrew Blakers on May 19, 2021

The US power industry abstracts 180,000 GL per year of water (https://www.usgs.gov/mission-areas/water-resources/science/thermoelectric-power-water-use?qt-science_center_objects=0#qt-science_center_objects). Much of this is returned to reservoirs and some is lost to evaporation. US agriculture also extracts a similar amount.

The US needs 10-20 TWh of storage to support 100% renewables. This requires 10000-20000 GL of water which is preserved in the system for 100 years. 5-10% percent is lost to evaporation each year which is offset by rainfall on the reservoirs and reduced by evaporation suppressors. A PV/wind/pumped hydro system uses vastly less water than a coal/nuclear system because of the lack of cooling towers.

Michael Keller's picture
Michael Keller on May 19, 2021

Small point, pumped hydro requires water and that is a real problem in much of the parched US West.  Recycling water in the pumped storage facility removes that water from other uses. The "bath tub" ring behind many of the reservoirs in the West provides points to an ominous future.

Also, the pumping of water to re-load the reservoir requires energy, typically at night. That can be a major problem if the grid excessively relies on renewable energy.

Andrew Blakers's picture
Andrew Blakers on May 19, 2021

Water supply really is not a problem. The water requirements are vastly less than coal power stations or agriculture. See reply above.

Pumping in a solar/wind systems happens whenever solar/wind is abundant and prices are low - typically in the middle of the day and when the wind is blowing strongly.

Michael Keller's picture
Michael Keller on May 20, 2021

Brings up the law-of-unintended consequences. Thermal power stations are now more-or-less required to use cooling towers. In the past, cooling water was withdrawn from a river or ocean, used to condense steam, and then returned to the river or ocean. While the cooling water was obviously heated up, the impact on the aquatic environment was a function of how much water was temporarily withdrawn. 

Cooling towers require vastly more water than traditional methods because the water is evaporated and lost into the air. Obviously unhelpful for conserving water, compliments of mindless application of regulations.

Recycling water from one place to another causes impacts to both locations.

My tangential point: pumped hydro is likely to be wrapped-around-the-axle largely by myopic environmental regulations, in spite of the many benefits the technology can provide.

Peter Key's picture
Peter Key on May 21, 2021

The biggest pumped storage facility in the world, at least according to this article, is located in Bath County, Virginia. Although I've read articles, which I was looking for when I found the one I just linked to, that say pumped storage facilities would be a good way to make use of old coal mines in Appalachia, this facility apparently is just two man-made lakes and the equipment needed to shuttle water between them and generate power. It's in the news now because one of its owners filed comments in a Virginia hearing involving Appalachian Power Co.'s plan to comply with a state law saying Appalachian's generation has to be all renewable by 2050. In the comments, the owner said the facility is available if Appalachian needs it to comply with the law.

Get Published - Build a Following

The Energy Central Power Industry Network is based on one core idea - power industry professionals helping each other and advancing the industry by sharing and learning from each other.

If you have an experience or insight to share or have learned something from a conference or seminar, your peers and colleagues on Energy Central want to hear about it. It's also easy to share a link to an article you've liked or an industry resource that you think would be helpful.

                 Learn more about posting on Energy Central »