Pumped Hydro Energy Storage Is Having a Renaissance

In late November, in a quiet corner of Devon, England, workers began adding a secret, light brown powder to water as they mixed up a special fluid that can store energy. They blended it with the utmost care, like some kind of giant protein shake, over the course of multiple weeks. Their goal was to achieve a mixture 2.5 times denser than water.

“It’s quite a hands-on process. At bigger scales, we would automate it,” says Stephen Crosher, chief executive and co-founder of RheEnergise, a British energy-storage company. He emphasizes that the mineral-based fluid must flow easily. “You want it to be really runny.”

That’s because, in the company’s demonstration system, at a china clay mine near the city of Plymouth, this mystery liquid can now slosh down angled pipes connecting an upper container to a lower container 80 meters below. In the process, the fluid drives turbines to create electricity. Pumping the mixture back up at times when there’s excess energy on the grid resets the whole system. It’s a new take on an old energy-storage technology currently experiencing a renaissance: pumped hydro.

Pumped hydro first emerged in the late 19th century. During subsequent decades, countries including the US and UK built lots of large plants, though construction had waned by the 1990s. The tech was originally designed to complement fossil fuel power plants, making use of excess energy they produced. But today grid operators increasingly value pumped hydro plants as workhorses able to mediate highly variable wind and solar assets. They can fill in shortfalls in electricity generation or soak up surplus energy within minutes, and store it for short or long periods.

Currently, significant amounts of energy go to waste because there’s no way to consume it at the moment of generation. The UK, for one, has squandered more than £1 billion ($1.32 billion) this year alone by turning off wind turbines due to lack of energy demand.

Pumped hydro plants could help to solve this problem but building them has often been expensive and difficult. RheEnergise’s denser-than-water fluid means the firm can pack more potential energy into a smaller space and at lower elevations. To replicate the firm’s 500 kilowatt (kW) demonstrator with a water-based version, for example, you’d need more than twice the volume of liquid they’re using, and the upper reservoir would need to be elevated to 200 meters rather than 80.

Vast reservoirs and towering mountains, historically associated with pumped hydro projects, may no longer be essential. “In somewhere like the UK, we think there are probably 20–25 viable sites for traditional pumped hydro, whereas for us it’s 6,500,” says Crosher. There could be hundreds of thousands of potential locations for the company’s technology worldwide—if RheEnergise can prove it works as intended. The firm tells WIRED it generated the first power from the system this week. Should the results from these tests continue to satisfy, a commercial-scale 10 MW project could follow by 2028.

RheEnergise is just one example of how pumped hydro is evolving. “It’s a really exciting time,” says Rebecca Ellis, senior energy policy manager at the International Hydropower Association (IHA), a nonprofit membership organization. The IHA estimates that 600 GW of pumped hydro projects are in the global pipeline. 8.4 GW was installed in 2024, she adds. One project that helped boost that total was the 3.6 GW pumped hydro plant in Fengning, China. It’s the biggest such facility, in terms of power capacity, in the world.

Day in, day out, plants like this move water up hills and down hills on a gigantic scale—and with incredible power.

At Goldisthal in central Germany, an upper reservoir containing approximately 12 million cubic metres of water—enough to fill 4,800 Olympic swimming pools—is linked by a pumped hydro power station to a lower reservoir of nearly 19 million cubic meters. It has two 800-meter-long penstocks, slanting tubes connecting the reservoirs, and, at maximum power, a capacity of 1.06 GW.

“Goldisthal is our biggest power plant,” says René Kühne, head of energy company Vattenfall’s pumped hydro fleet in Germany. If required, the facility could provide 1.06 GW of electricity for between eight and nine hours. In just 90 seconds, Goldisthal can switch from standstill to full generation. Water floods down each penstock at 100 cubic metres per second. The plant can alternate between generation mode and pumping mode in a matter of minutes, which means that instead of supplying 1.06 GW of energy to the grid, it can absorb that much power.

In Vattenfall’s central control room, human operators aided by algorithms monitor the electricity grid, and the market, to judge whether their pumped hydro plants ought to be generating or pumping. Kühne adds that the frequency of alternating between these modes has gone up over time because of renewables’ variability.

If you can get the response right, however, you can make a lot of money. On its website, Vattenfall describes pumped hydro as “highly profitable.” A paper published last month estimated the effect of rising renewables in Spain between now and 2050. With gradually decreasing electricity prices, higher variability, and less need to import electricity overall, the authors found that energy storage would be utilized 12 percent more in the future—and that a system combining renewables with pumped hydro energy storage would see its profits rise.

Pumped hydro could, in principle, work in lots of places around the world, says Rosie Madge, a systems engineer at Energy Systems Catapult, a nonprofit research and innovation center: “Most countries in the world do have geographies that are suitable for it.”

A report by Madge and colleagues, published in October, scored 11 countries in terms of their suitability for pumped hydro and other long-term energy storage tech. Two notoriously flat nations, Denmark and the Netherlands, fared poorly. But the others were all extremely well-suited to conventional pumped hydro and a few, including the UK, Australia, and China, were very well-suited to the high-density version. The scores were based partly on how ready and willing each country was to deploy the tech, and also on market conditions.

But even in that analysis, it was conventional pumped hydro that appeared most deployable overall—when compared to multiple other long-duration storage technologies including high-density pumped hydro, hydrogen, ammonia, metal air batteries, compressed air, and non-pumped-hydro gravity storage.

If you want to get into the pumped hydro game, though, first you have to build your infrastructure. And doing so on a conventionally large scale is hard.

In Australia, a truly massive pumped hydro project called Snowy 2.0 is currently under construction. It is an expansion of an existing pumped hydro system that utilizes lakes in the Snowy Mountains of southern Australia. While Goldisthal can provide a total of 8.5 gigawatt hours (GWh) of energy—its 1.06 GW capacity delivered over a maximum of 8.5 hours—Snowy 2.0 will offer an astonishing 350 GWh when completed.

However, Snowy 2.0 has been beset by delays and cost overruns. Work on the project involves constructing huge tunnels, totalling 27 km in length. But one of the tunnel boring machines deployed to dig them was stuck for several months in unexpectedly soft rock. It later became stuck again. Plus, the company behind the project, Snowy Hydro, and some of its contractors have been fined over several alleged incidents of pollution caused by construction activities. A 2024 report by the nongovernmental National Parks Association of New South Wales (NPA) claimed its officials had allegedly observed roughly a dozen environmental compliance failures from early 2022 to mid-2023.

In 2024, in response to the NPA report on its purported environmental performance, the company said: “Snowy Hydro takes its environmental compliance obligations very seriously and is committed to ensuring that the construction and operation of the Snowy 2.0 Project proceeds in a manner that is compliant with all applicable laws and approvals.”

Snowy 2.0 is currently due to complete in late 2028. Funded by taxpayers, it will swallow up much more than the original estimate of AUD$2 billion ($1.29 billion). The final bill will be at least six times that, likely ending up in the region of AUD$15–18 bn, according to Andrew Blakers of Australian National University. He calculates, though, that even at that price, Snowy 2.0 will cost roughly one Australian cent, per Australian, over its projected 150-year lifespan.

“If you want serious storage, you want pumped hydro,” he says, adding that the UK, for one, is currently “not serious” about pumped hydro and won’t be until it constructs a 500 GWh system at Loch Ness in Scotland. Multiple firms are vying to install new pumped hydro infrastructure there, though these schemes have faced local opposition.

Construction headaches sometimes come up for pumped hydro projects, and other large-scale infrastructure schemes, partly because it’s so difficult to know what’s in the ground before you start digging, says Brian Minhinick, global hydropower practice leader at Mott MacDonald, an engineering consultancy. “I’d very much like it that it didn’t happen,” he says. “We try to do our best.” Planning for various scenarios can help, as can using 3D models of pumped hydro projects to plan construction in detail. Workers can adjust those digital plans should they find unexpectedly hard or soft rock, for instance.

New machines have helped, too. “You can actually get drilling rigs that have three computer-controlled drill arms [operating] at the same time—to speed up the work,” adds Minhinick. They can, for example, simultaneously excavate multiple holes for packing explosives into, as part of drill-and-blast operations where tunnel-boring machines aren’t the tool of choice.

The sheer scale of some pumped hydro projects is one of their key advantages. Blakers says that building really large pumped hydro projects is how a country can show it means business when it comes to energy storage.

But Crosher at RheEnergise makes another point. While large-scale, traditional pumped hydro has its place, the world needs climate solutions fast. His goal is to offer a version of pumped hydro that is much easier and quicker to deploy. “If you want solutions for the climate emergency and the energy transition, then [traditional] pumped hydro will do part of it, but they’re too slow to do it all,” he says.