‘Energy storage is essential for electricity grids, particularly as the generation of renewable energy continues to rise,’ argues the Oxford Institute of Energy Studies (OEIS) in a paper issued in July. Other rival storage technologies such as batteries, compressed air and hydrogen are emerging, but pumped hydro has the virtue of being tried and tested as the kind of ‘long-duration’ backup that electricity grids will need.

‘Large-capacity, long-duration energy storage enhances grid resilience when used in combination with intermittent or highly variable generation,’ summarises engineering consultancy Arup, which is working on projects in Britain, including design and project management for the first large-scale battery energy storage system (BESS) in Northern Ireland, at Dunore Point for Northern Ireland Water, to be connected at a higher distribution voltage of 33 kV. ‘Stored energy can be dispatched when needed, improving grid reliability, especially during periods of high demand or during outages. Energy storage also reduces the need for curtailment of renewable energy during periods of high output and low demand,’ says the consultancy.

Essentially, a pumped hydro system (PHS) creates instant energy by sending huge volumes of water up and down pipes at high speed. According to one definition: ‘A PHS consists of two linked reservoirs, one upper and one lower. When energy demand is low, water is pumped from the lower reservoir to the higher one. When demand is high, water is released through pumps which act as turbines to generate electricity that can be released onto the grid in seconds.’

Globally, pumped hydro capacity is growing rapidly. According to the latest report from the International Hydro Association (IHA), 8.6 GW of storage was added around the world in 2024, almost a third of all added hydropower capacity. One example is the Hatta station, located near the Hatta Dam, in the mountainous region of the United Arab Emirates (UAE), that will go into full operation during 2025, storing up to 1,500 MWh.

There’s a further 600 GW in the pipeline. Some of this is supported by Brussels. ‘EU and national policy measures drive momentum for pumped storage,’ notes the IHA, but most of it will happen in China. ‘China continues to dominate global hydropower development, with 14.4 GW of new capacity added in 2024, including 7.75 GW of pumped storage,’ according to its 2025 World Hydropower Outlook.

Big bath
Pumped storage has been around for a long time, with plenty of benefits to learn. One of the world’s biggest stations, Bath County in America’s rugged Allegheny Mountains, was built before anybody was even talking about intermittency. Opened in 1985, it can produce 24 GWh. Absolutely vital for the region, the station balances the electricity needs of millions of homes and businesses across six regions by pumping up to 13.5mn gallons/min of water through six turbine generators.

Luxembourg’s Vianden station is even older. It opened between 1962 and 1976, with electrical output of 1,291 MW when operating in turbine mode, according to plant operator RWE. While, in an example of retrofit, Switzerland’s Linth-Limmern Dam added a PHS a decade ago, giving it an electrical capacity of 1.520 MW (up from 520 MW with the original power plants).

Australia is joining the race with its 2.2 GW Snowy 2.0 project. One of the country’s biggest industrial schemes, the station will enter commercial mode in 2028. There may be more on the way. Arup has completed a pumped hydro roadmap for the New South Wales government that identifies no less than 98,000 potential sites with ‘reservoir pairs’ that allow water to be pumped up and down.

One scheme that will not go ahead is an innovative salt water-based project in South Australia that would have produced the equivalent energy of 60,000 home battery storage systems at a third of the cost. This is despite the state being hit by a major energy blackout in September 2016 when extreme weather shut down the grid and cut power to 850,000 homes. Known as the Cultana scheme, it was floated around 2019 as one of a handful of seawater-based projects globally (97% of the world’s pumped hydro uses fresh water) and would have generated 225 MW with eight hours of storage from a reservoir of 3.5 gigalitres.

As Melbourne University’s Lawrence Molloy, Principal of Long-duration Energy Strategies, explained in a 2025 paper, Cultana was one of six pumped hydro projects evaluated in South Australia, a dry region, over the last 20 years but it fell foul of objections from the defence forces in the area and, oddly, from an improvement in connectivity with neighbouring state Victoria that has allayed fears of another blackout. Yet, argues Molloy, although ‘pumped hydro is stumbling for the moment in Australia, it will remain a major part of the energy storage paradigm given its high efficiencies, ability to provide synchronous inertia and balance substantial amounts of variable, inverter-based wind and solar’.

After a dearth of projects, Britain is returning to pumped hydro with at least three on the table, all in Scotland. Coire Glas is a potential 1.3 GW scheme. Glenmuckloch is in pre-construction and slated for commissioning in 2027. But the expansion of the Cruachan station looks to be in abeyance, pending financing.  

It’s now over 40 years since the Dinorwig station in Eryri (Snowdonia), Wales, started up, initially coal-fuelled, to help balance a hard-worked grid. It boasts a maximum power of 1,728 MW and a power output of over 9 GWh. And, like Bath County and other stations, Dinorwig has saved the grid time and again. It can be wound up to full load within 12 seconds when the pressure is on, as it often has been.  

The energy required by pumped hydro is prodigious. At peak output Dinorwig pumps enough water to fill a 25-metre swimming pool every single second.

‘China continues to dominate global hydropower development, with 14.4 GW of new capacity added in 2024, including 7.75 GW of pumped storage.’ – International Hydro Association

Environmental headaches
Pumped hydro schemes keep running into environmental opposition. There was a ruckus in the 1970s over Dinorwig, even though it was built in an abandoned slate quarry, as locals protested about the risk of desecration of Eryri National Park. The eventual solution was to locate the station deep inside tunnels and caverns in the mountain and to bury 10 km of cables underground rather than run the power through pylons.

Plans for a station at Exmoor National Park in south-west England never made it past the proposal stage and a long-running proposal for a 100 MW storage facility in disused slate pits at Glyn Rhonwy in Wales that was originally planned to start up in 2026 also appears to have stalled.

Yet the storage of what are essentially emergency volumes of energy has become essential in a more complex sector. The OEIS explains: ‘This complexity arises from the unpredictable output of weather-dependent generators such as solar panels and wind turbines. To mitigate this issue, various strategies can be employed, including the implementation of energy storage systems, optimisation of demand patterns, and enhancement of flexibility and connectivity between different energy grids at a regional level.’

The renewable energy sector is also unpredictable. In Germany, periods of low sunlight and calm winds are known as Dunkelflaute (literally a ‘dark lull’) that can last for as long as three weeks. Inevitably, consumers then resort to gas.

Energy from the air
Rival technologies are making progress, for instance in the shape of grid-scale batteries that, according to 2022 figures, can reach 100 GWh and rising steadily. There’s also hydrogen and compressed air, both of which are the beneficiaries of serious science and money.

Take compressed air energy storage (CAES), which is seen as a rival to pumped hydro because it’s another long-duration alternative with a lifespan up to 40 years. It is also highly efficient, as the compressed air is released instantly to drive the turbine. However, the low volumetric density of CAES means it requires vast storage areas such as disused mine shafts.

In the meantime, fossil-fuel storage in the form of natural gas continues to dominate. The OIES lists no less than 76 gas storage projects under construction around the world, with another 99 in the planning phase. ‘These statistics clearly indicate that the existing capacity for storing energy in molecular forms far exceeds that of other electricity storage methods, with a ratio of more than 100:1,’ it concludes.

Back to pumped hydro, with its virtues and drawbacks. On the plus side its operating efficiency pushes 85%, meaning that most of the energy expended during the pumping (or charging) phase is recovered when it’s in turbine (or discharging) mode. It is extremely durable – stations can run for up to 80 years, which is exactly what the UAE expects from the Hatta project. And according to German researchers, a well-built station is good for 14,000 storage cycles.

On the minus side, the technology needs mountains – ‘substantial elevation differences’ – so the water can be held in a higher reservoir as a natural battery. This is why the Scottish Highlands features in the most favoured sites. Also, 12 hours is about the maximum time the stations can deliver energy under current technology, although Snowy 2.0 is aiming for a breakthrough 20 hours.

Notably, environmental concerns will always figure high. Hydro engineers point out that natural life can actually be enhanced in these grand schemes. The recreation area at Bath County, for example, offers fishing, sailing and non-powered boating in large man-made lakes.

This is why retrofits, such as at Switzerland’s Linth-Limmern, look increasingly attractive because these dams are already half-way to pumped storage. As the IHA has pointed out, only 12% of the world’s existing hydroelectric assets do pumped storage. But many can be upgraded through a powerhouse and upper reservoir. One option, say engineers, is to install pumping systems to a series of cascade reservoirs that push the water back upstream.

‘The retrofit approach could offer opportunities to increase the value of hydroelectric assets as generation and storage facilities, while the use of existing reservoirs may help limit environmental impacts of pumped hydro and overcome some of the development challenges associated with new infrastructure projects’, explains Arup.  

However, in all the debate about pumped hydro nobody is agitating for any of the existing stations to be dismantled, because their contribution to the grid has become indispensable.  

• Further reading: ‘Unlocking pumped storage hydro for net zero Britain’. The global energy network requires a very large build-out of long-duration energy storage to reach net zero emissions. While pumped storage hydro is a mature and reliable technology that can meet a large portion of this need, the business model it would be applied to is new. In the UK, the government is exploring a cap and floor scheme to unlock investment into PSH development. Its implementation could distort the market. Alterations of that scheme, which could overcome those problems, are proposed by Roderick MacLeod, Director, Glen Earrach Energy (GEE).
• The UK government announced in October 2024 plans to implement a ‘cap and floor’ investment framework to support the deployment of long-duration energy storage (LDES) projects in the UK. Find more about the recently published Technical Decision Document (TDD) confirming key details of the scheme.