In pumped hydro storage there is a tried-and-tested technology that not only fits the bill for future energy needs but has been doing the job for well over a century. Andrew Wade reports
It’s often assumed that to mitigate the worst effects of climate change we need an energy revolution incorporating a raft of new ideas. However, the core technologies needed to decarbonise the power sector already exist in the form of renewables, with wind and solar now the fastest-growing sources of new generation.
One major problem, as critics are quick to point out, is intermittency. If that problem could be solved, the energy sector could be transformed within a few short decades. And, as it happens, several countries are betting that we already have the answer.

In pumped hydro storage there is a tried-and-tested technology that not only fits the bill but has been doing the job for well over a century. First employed in the 1890s, pumped hydro makes up 97 per cent of energy storage worldwide, with around 168GW currently installed. Excess off-peak grid power – theoretically from renewables – is used to pump water uphill, where it’s stored as gravitational potential energy. When electricity demand is high, water is released downhill to power turbines, the elevated reservoirs essentially acting as giant batteries to help balance the grid.
Currently, the UK has four operational plants. When the youngest – Dinorwig in Snowdonia – was built over 30 years ago, it was the largest civil engineering contract ever awarded by the government. Today, just up the road in Glyn Rhonwy, a new pumped storage venture is being developed by private company Snowdonia Pumped Hydro (SPH). With 99.9MW of output and 700MWh of storage capacity, it won’t solve the climate change problem singlehanded. However, the process behind its selection also threw up hundreds of other viable UK sites. By exploiting just a handful of these, pumped hydro could be the last missing piece of the UK’s clean energy puzzle.
“We basically started with a few databases of lakes, mines, quarries and so on, and we wrote a little algorithm that would compare them, to see how close they were and try and make matches,” explained Dave Holmes, managing director of Quarry Battery Company (QBC), the parent firm behind SPH.
“We got down to maybe 1,000 sites, or potential pairs, and then we cruised around the country on Google Earth having a look at them all and trying to figure out which ones were the best.”
The dual bodies of water ideally need to be similar in size, with a suitable disparity of altitude, or head. As well as identifying sites that were technically desirable, the team had to consider other factors such as planning risk, proximity to the grid and nearby areas of conservation. Glyn Rhonwy’s two disused slate mines ticked a lot of boxes.
“There were a few things in its favour,” said Holmes. “It’s pretty close to the National Grid transmission system. The length of the tunnel that you need in comparison to the head; it’s a very steep site, which means the tunnel we’re building isn’t too long. The gradient is actually quite a challenge. It’s 19 per cent… and we’re looking to go in a single slope all the way to the top.”
The site also happened to be owned by the local council, which had plans to develop the lower reservoir into an industrial estate. Accommodating SHP’s powerhouse would not be a huge departure.
“We see it as recycling the land rather than using some new land for the site,” said Holmes.
Glyn Rhonwy gained approval for a 49.9MW plant several years ago, but changes in the energy market prompted SPH to rethink the design. The revised plan for the 99.9MW facility was greenlit in March 2017. According to Holmes, the only significant change was the inclusion of two 49.9MW turbines instead of the original pair of 24.95MW devices.

“We found there was more capacity there than we first understood, and it would have been a waste of that capacity to not have a more powerful turbine on it,” he explained.
The National Grid’s capacity mechanism also helped incentivise the change. Despite having 600MWh of storage in the original plans, SPH would have been rewarded only for its 49.9MW output. By doubling that to 99.9MW, the project will double the income it earns via the mechanism. The revised 700MWh capacity will be enough to supply 200,000 homes with electricity for seven hours a day. Although construction costs will increase from £140m to £160m, Holmes believes the payback period on the investment should be halved to under 10 years – a fraction of the plant’s projected 125-year operational lifetime.
“It’s a very exciting time because we can see a light at the end of the tunnel,” he said. “We’re hoping to break ground sometime next year, but it could be early the year after.”
As construction gets under way in Glyn Rhonwy, a pumped hydro plant of an altogether different breed will be nearing completion some 8,000km away. Located just to the north of Beijing, the Fengning Pumped Storage Power Station will have an installed capacity of 3,600MW, making it the largest pumped hydro facility in the world. Its generators will be fed by an upper reservoir holding nearly 49 million cubic metres of water – enough to fill about 19,500 Olympic pools. By comparison, each of Glyn Rhonwy’s reservoirs will hold just over 1 million cubic metres.
The $1.87bn (£1.42bn) Fengning plant will be equipped with 12 x 300MW pump turbines in a single cavern. Ten of these will be fixed-speed units contracted from local suppliers. The final two will be variable-speed pump turbines provided by Andritz, the Austrian engineering giant with a long history in hydro power.
“The major difference is in a completely different generator design,” Alexander Schwab, Andritz SVP of market management, explained to The Engineer. “With conventional synchronous generators, the number of poles and the frequency of the grid are defining the rotational speed of the unit. That defines the design speed of the turbine/pump. Consequently, for pump turbines it is necessary to find a compromise between optimal turbine and optimal pump.”
According to Schwab, the variable turbines will allow operation ranges for pumping and generating to be designed individually. This should enable Fengning to better manage its output and provide more grid flexibility.

Andritz has previously installed the technology at Germany’s Goldisthal pumped storage facility, but this is the first time variable speed units will be used in China’s ever-expanding hydro sector. “Taking all the facts into consideration, it will be a project needing the utmost attention from the Andritz side,” said Schwab.
When Fengning comes online, China will have three of the four biggest pumped storage facilities in the world. The technology is seen as increasingly vital if the country is to wean itself off the coal that has powered the industrial progress of recent decades. Air pollution across China, particularly around Beijing, has reached such worrying levels that it is no longer simply a public health issue. Speaking recently, the country’s environment protection minister, Li Ganjie, admitted that air quality had become a matter of social stability. Although it remains the world’s biggest polluter, China has committed to reducing CO2 emissions by 60-65 per cent before 2030. If this is to be achieved, massive expansion of renewables, coupled with pumped storage, will need to play a key role.
Back in the UK, plans are afoot for another major project. While not on the same scale as Fengning, the Coire Glas pumped hydro plant will boast an output of 1,500MW. It will lie to the east of Fort William, in the Great Glen of the Scottish Highlands. Using Loch Lochy as the lower body of water, developer SSE plans to build an upper reservoir 500m above, with a relatively steep incline delivering good energy efficiency.
The project received planning permission several years ago for a smaller 600MW plant but, similarly to Glyn Rhonwy, those plans have had to be scaled up in a bid to make the project more commercially attractive. In May 2017 SSE submitted a scoping report for the revised scheme, and a decision from the Scottish government is expected in the near future. Importantly, Coire Glas will have the capacity to store a massive amount of energy – around 30GWh – more than all existing UK storage capacity combined, according to SSE. The firm says this will help provide the flexibility required for integrating more renewables, as well as the changing usage patterns expected from the growth in electric vehicles.
“SSE considers there to be clear benefits in the delivery of new pumped storage projects in the UK,” Coire Glas project manager Andy Gregory told The Engineer. “It’s a proven technology that can be deployed at the tens of GWh scale, with a short response time and the ability to provide a number of important services to the system.”
Across the UK there are hundreds of potential pumped hydro sites, with the low-hanging fruit adding up to around 50GWh. According to QBC’s Holmes, these are mainly in Scotland and Wales due to topography.
“Our best guess at how much is economic already or will be economic soon is around 50GWh,” he said.
Some of these sites are similar in nature to Glyn Rhonwy and Coire Glas, where dual bodies of fresh water would work in tandem. Other sites, however, are less conventional. QBC has been awarded funding to explore the viability of non-standard sites around the UK, looking at seawater, drinking water and lower-head opportunities.
Seawater, in particular, presents both a unique opportunity as well as unique challenges. Using the sea as the lower ‘reservoir’ unlocks a host of locations that would previously have been overlooked as, in theory, all that’s required is elevated land close to the coast. But seawater brings complications such as fouling and corrosion, and using the open sea can impact on marine life, and have unwanted effects on the coastal profile.
To date, just a single seawater project has been commissioned – Japan’s Okinawa Yanbaru Seawater Pumped Storage Power Station. Built in 1999, it featured a striking octagonal upper reservoir set into the lush coastal vegetation. Instead of steel, the penstock was made of fibre-reinforced plastic to avoid corrosion and barnacle fouling, and the pump turbine included stainless steel to improve seawater resistance. Although widely hailed as an engineering success, the plant was shut down in 2016 as it no longer fitted local power demands.
Despite just one pilot project, seawater pumped storage is being explored in several countries, including the UK. South Australia’s proposed Cultana plant could generate 225MW of electricity with eight hours’ storage. The region already supports renewables, but arid conditions make traditional pumped hydro tricky. Combining existing solar and wind generation with coastal storage plants promises to be an innovative solution.
Elsewhere in Australia the planned Snowy Hydro 2.0 project will look to add pumped storage to the long-established hydro generation already in place. If completed, it will add generation capacity of 2,000MW and a massive 350GWh of storage. This year, a government-backed study by the Australian National University (ANU) into other viable locations identified 22,000 sites, the bulk of which are on the more densely populated east coast. Combined, these could provide around 67,000GWh of storage.
“We need to build only about one or two dozen to support a 100 per cent renewable electricity grid,” Andrew Blakers, a professor of engineering at ANU and one of the report’s authors, said recently.
“Pumped hydro, high-voltage DC interconnectors between the states, wind, batteries and demand management can do the whole job. Not just for electricity, but the whole job for energy: electrifying land transport, electrifying heating and cooling. You can have 75 per
cent cuts in Australia’s greenhouse gas emissions by electrifying nearly everything, and I think this is what’s going to happen over the next 15 to 20 years.”
The UK’s topography may not be as rich as Australia’s in terms of potential sites but there are ample resources to exploit. If tidal lagoons finally get the green light, they could also include a pumped hydro element, as well as help balance the grid via their natural geographic dispersion. And, with wind and solar costs still tumbling, there’s no reason why the UK couldn’t fully decarbonise over the next two decades. The fact that the Aussies are already in the race can only bring added incentive.
For more stories on renewable energy and its application, click here
How about using fresh water instead of sea water in those schemes that have the lower storage reservoir at sea level? The water could be kept in good condition by a regular top up from a fresh water source.
It makes no difference as the density is but very slightly increased by solutes in seawater. The only use of the pumped water is to run back to the lower elevation (generator). It would be very poor stewardship of fresh water resources to co-mingle fresh water storage with seawater on the down-stream side of the generator.
Yes I’ve pondered the possibilities to use fresh water if doing so helped to win government support from ministers and their advisers who might be concerned about the risks of polluting the Scottish Highland water table and aquifers with salt water from “leaks” that critics will allege are a risk when moving vast quantities of water from the coast many miles inland, as my published Strathdearn Pumped-Storage Hydro scheme concept proposes.
Estuary water is already brackish because fresh river water outflow mixes with salty sea water but it seems to me that it should be possible to restrict such mixing by engineering the river estuary to drain to the sea through a newly engineered estuary barrier, draining through flap valves to prevent return flow from the sea at high tide.
A lock for boats could be built into the estuary barrier to facilitate estuary navigation.
However for every environmentalist who would be pleased there would be no risk from salt water being pumped in-land there may be another who is annoyed that the new estuary barrier had made estuary water too fresh for the expectations of estuary fish and obstructed them swimming in from and out to sea.
Alternatively, to avoid the estuary barrier, one might construct an artificial fresh water reservoir in the estuary with some kind of lagoon wall? Perhaps even a giant fresh-water bladder might work?
I’d be confident of delivering either solution – sea-water or fresh-water. The technical differences and cost differences between the fresh / saltwater solutions are as nothing compared to the technical and cost difference between building Strathdearn pumped-storage hydro and not building it at all.
I think a feasibility study for Strathdearn should look at all the possible solutions, seawater and freshwater and I’d be happy to lead up that feasibility study if commissioned by the governments and / or industry to do so.
If , hopefully, a very much bigger percentage of our electricity is to come from renewable sources , pretty obviously that would result in surplus energy when the wind blows or the sun shines but with a corresponding shortage when they don’t. Hence storage becomes a much more critical element in the equation. With claimed efficiencies of 70-80%, combined with extremely long operating life, this would seem a very obvious route.
And has done for many years. So why hasn’t it happened?
Money. Pumped-storage is very expensive (£160m for a small scheme – see article). Using batteries or flywheels can be done with lots of smaller installations.
The other problem is that they often need to be in scenic locations i.e. National Parks and AONB’s, which makes planning permission difficult.
I visited Dinorwig many years ago: it is an amazing achievement, especially bearing in mind it is right next to a major tourist attraction (Mt Snowdon).
They sell the electricity to make the money back which they could do in under 10 years (also in the article). Hydro plants are good for well over 100 years. It’s lack of political will and incentives, somebody with a unified vision on energy needs to step forward and say: ‘This is what we’re going to do!’ Politicians know nothing about science and technology (only lobbyists) and should be sidelined from the real decision making process.
This is what we are going to do.
Scottish Scientist
Independent Scientific Adviser for Scotland
https://scottishscientist.wordpress.com/
Economics, and scale. The example above (700MWh capacity for 160 Mil GBP up-front cost) = 4 Wh/GBP, about 50x the cost of Lithium batteries c 2019. And as has been remarked in other comments, that facility gives 70,000 homes a 10KWh buffer. You would need a few hundred of those to scale up to address the needs of the whole of the UK.
Here in the states, back in the 1950’s, a similar project was built between Ludington and Pentwater Michigan along the eastern shore of Lake Michigan. Although older technology and perhaps turbine designs, this generating facility has been in use ever since it was built. And it makes for a site-seeing location with a grand view over looking Lake Michigan. It also is quite a steep hike up to the reservoir but worth the work out for those able to make it to the top.
The large number of wind blown unreliable power sources make storage almost essential if these are to be used in the future.
Cruachan is well worth a visit for anyone interested in pumped storage. novel approaches such as cliff-side pump stations could be very economic related to the offshore units.
The over-hyped title of the article is part of the on-going fear-campaign by the media and unworthy of the engineer; even if COP 23 is entertaining a lot of climate activists. Storage could make an valuable contribution to the energy mix if it is economic.
Yes ! Get on with it — stop talking to save the planet -This needs non political war footing planning with ultimate power given to a clean energy task force that has long term non interference power to make a difference.
Just to put some context behind these numbers: the average UK demand over the last 24 hours was about 35GW. (see http://www.gridwatch.templar.co.uk/ for more details). So even if all the “low hanging fruit” is built, giving 50MWh, there would only be 1.5hours of supply. This isn’t going to solve the problem of energy storage from renewables on its own.
Last time I checked, 35 GW would chew up 50 MW-hr in about 5.14 seconds.
It says 50GWh 🙂
“low-hanging fruit adding up to around 50GWh”
I wonder how many of the UK sites, as well as acres of pv/wind turbines, could be built for the £20bn cost of Hinckley Point C? And within the ‘projected’ 10 year build time.
Slight typo in my post, it should have been 50GWh, as Tom pointed out. However, the 1.5hours of supply is correct, and therefore still the same conclusion.
There are many more potential pumped storage sites than simply those identified by QBC with two suitable reservoirs in the UK. With vastly reduced transmission costs into the National Grid.
In Stuttgart , Germany the world’s tallest onshore wind turbine at 246.5 metres includes a 40 metre pumped storage tower base creating suitable head so only one reservoir is required. Salt or fresh water.
For those UK communities which choose to install onshore turbines possibly in devolved Wales & Scotland rather than areas where house prices and nimby considerations are more important this offers a real advantage. Due to reduced wind shear and stronger wind speeds as tower height increases energy performance for the blade swept area increases between 0.5 to 1% per metre. Additionally longer blades generate more power due to a greater swept area are possible as tower height increases. Double the swept area, double the power potential from the wind.
https://www.windpowermonthly.com/article/1448548/max-bogl-installs-record-breaking-2465-metre-turbine?utm_source=trending&utm_medium=trending&utm_campaign=trending
STRATHDEARN PUMPED-STORAGE HYDRO SCHEME (up to 180 GW / 6,800 GWh)
World’s biggest-ever pumped-storage hydro-scheme, for Scotland?
“The map shows how and where the biggest-ever pumped-storage hydro-scheme could be built – Strathdearn in the Scottish Highlands.
Energy storage capacity
The scheme requires a massive dam about 300 metres high and 2,000 metres long to impound about 4.4 billion metres-cubed of water in the upper glen of the River Findhorn. The surface elevation of the reservoir so impounded would be as much as 650 metres when full and the surface area would be as much as 40 square-kilometres.
The maximum potential energy which could be stored by such a scheme is colossal – about 6800 Gigawatt-hours – or 283 Gigawatt-days – enough capacity to balance and back-up the intermittent renewable energy generators such as wind and solar power now in use for the whole of Europe!”
Power
There would need to be two pumping and turbine generating stations at different locations – one by the sea at Inverness which pumps sea-water uphill via pressurised pipes to 300 metres of elevation to a water well head which feeds an unpressurised canal in which water flows to and from the other pumping and turbine generating station at the base of the dam which pumps water up into the reservoir impounded by the dam.
To fill or empty the reservoir in a day would require a flow rate of 51,000 metres-cubed per second, the equivalent of the discharge flow from the Congo River, only surpassed by the Amazon!
The power capacity emptying at such a flow rate could be equally colossal. When nearly empty and powering only the lower turbines by the sea, then about 132 GW could be produced. When nearly full and the upper turbines at the base of the dam fully powered too then about 264 GW could be produced.
Modelling of a wind turbine power and pumped-storage hydro system recommends –
• store energy capacity = 1.5 days x peak demand power
suggesting that a store energy capacity of 283 GW-days would be sufficient to serve a peak demand power of 283 / 1.5 = 189 GW, though this could only be produced from reservoir heads of at least 430 metres, at least 8% of energy capacity, assuming a flow rate of 51,000 m3/s. To supply 189 GW from the lowest operational head of 300 metres would require increasing the flow capacity to 73,000 m3/s.
This represents many times more power and energy-storage capacity than is needed to serve all of Britain’s electrical grid storage needs for backing-up and balancing intermittent renewable-energy electricity generators, such as wind turbines and solar photo-voltaic arrays for the foreseeable future, opening up the possibility to provide grid energy storage services to Europe as well.
The highest UK power demands occur over winter at typically 55 GW. This peak, unfortunately, corresponds with what is commonly a low wind period of over a week in Europe: a toxic combination if we do not have reliable generation in place. It is unlikely that storage can ever cover these situations economically, so that an adequate supply of reliable generators is essential.
The advocates of renewables and mass storage forget about economics and the fact that the UK industrial electricity prices are the second highest in Europe. It is essential to the UK’s economic wellbeing that we have low cost electricity
In the modelling of a UK wind-powered grid that I published, the “toxic combination” or worst-case scenario that I found was using Gridwatch UK wind and demand data from September 2014 when the wind was extremely low, making for a tougher challenge to keep the lights on than in the depths of winter, at peak demand, when the wind power is higher.
I modelled a wide range of system configurations of wind, storage and back-up from dispatchable generators.
Modelling of wind and pumped-storage power
https://scottishscientist.wordpress.com/2015/04/03/scientific-computer-modelling-of-wind-pumped-storage-hydro/
I also have produced an on-line calculator to help engineers to design wind-powered grid system configurations to supply renewable energy 24/7/52.
For example, this link generates a “Grid Watch Demand Focus Table” with system configuration rows, A to H, to meet a peak demand of 55GW.
Wind, storage and back-up system designer
http://scottish.scienceontheweb.net/Wind%20power%20storage%20back-up%20calculator.htm?peak=55&units=gw#grid
Thanks for the links to your web-site, all very interesting, but based on Scotland’s demands: do you know something about the neverederum that we don’t??? The assessment of El Hierro’s problems seems to tie up well with what Euan Mearns site reports; but the further investment needed in wind turbines is large.
As usual over-pessimism about climate and over-optimism about renewables from the romanticists. France already has a grid without fossil fuels and using pumped storage. The crux is that they needed an 80% baseline supply from nuclear power. Pumped storage is great for peak loads but – as pointed out – useless beyond 1.5 hours of a supply shortfall. The numbers were all worked out by Dr David McKay in his book “…without the hot air”. His conclusion? Either widespread nuclear power or CCS. Too many folk just prefer to avoid dealing with the basic arithmetic.
The reality is that the UK could drop from its lowly 2% to 0% CO2 and all it does is make a near-bankrupt country even poorer with zero effect on climate. Apparently we are providing leadership; except that nobody is following us off the cliff.
Unfortunately, the economic sites for tidal generation do nothing to reduce the peakiness of their output, as despite their geographic dispersion, they turn out to be more or less synchronous – and subject to substantial output variation between spring and neap tides every lunar month. There’s a nice summary of work done at the Proudman Institute that shows this:
http://euanmearns.com/green-mythology-tidal-base-load-power-in-the-uk/
“Adding pumped storage hydro provides an economical and practical way to smooth out daily fluctuations” – It’s not very economical as there are bound to be losses in both pumping and regeneration. After-generator electricity storage is wasteful, always and for ever. Before-Generator Energy Storage transforms the economics of electricity distribution by eliminating the fluctuations of tidal and smoothing the variability of wind and wave. All three can share their BGES facilities or function independently, depending on location. e.g. A reservoir built at 200m above Port Talbot could take the pumped output of the Swansea Bay Tidal Lagoon – no electricity involved! An holistic design would integrate “all of the above” in a single system whenever it’s feasible.
PHS is hardly practical. If you don’t have suitable geography you’d have to pay others for the service. Norway is self-sufficient in zero carbon electricity but converting their hydro to pumped storage would be a big investment. The NorthConnect cable is costed at £1.75bn and the owners would want to make a profit!
“Large-scale deployment of tidal stations will modify coastlines that deploying renewable energy is supposed to prevent.” – A little disingenuous, since sea level rise is already guaranteed and the Bristol Channel Barrage from Minehead to Aberthaw would in fact be a great investment in both flood prevention and 25TWh of dispatchable electricity. Moreover, it will be generating it cheaper for at least three times the life of Hinkley Point C.
The beauty of BGES innovation is that it will integrate all marine renewables by sharing a common storage capacity. This sea change in design removes generators from the harsh environment under water and inaccessible locations such as the nacelle of an offshore HAWT.
There is also tidal barrage, like the Rance Barrage in France. See wikipedia
Theoretically capturing high tide of water and releasing it via turbines would be a low generation option for local use .
Stand by for a US DOE Water Power Technologies pumped storage valuation and technologies report. Preparation of the report is by a multilab team lead by argonne National Lab.