Dunkelflaute, a northern European weather pattern of heavy cloud cover and low wind demonstrates the fragility of power systems that rely heavily on variable renewable energy sources like Wind and Solar. This piece provides a back-of-napkin analysis on how big this problem is and what level of investment is required to fix it
It’s early January here in New Haven. The fog rolling off the Long Island Sound makes a grey, low ceiling over Yale’s Campus. My mind wanders as I look out the leaded glass windows of Sterling Memorial Library’s graduate studies room and I’m reminded of “Dunkelflaute”, a German word meaning “dark doldrums” introduced to me in a New York Times article from last week.
Besides providing a fun addition to the vocabulary, the article describes how Dunkelflaute weather patterns are creating problems for the European energy market- periods of heavy cloud cover and low wind across Northern Europe translate to low electricity production across the region that now relies on Solar and Wind to produce much of its power. In response to decreasing production, power is purchased from countries with more reliable sources of energy (Hydropower from the Nordics, Nuclear from France) driving up prices for consumers across the entire continent.
Dunkelflaute underscores Wind and Solar’s Intermittency problem. Today, Wind and Solar are the fastest growing segments of the global energy mix, making intermittency a grid problem the world over. In the EU, the combined output of wind and solar eclipsed fossil fuel electricity production in 2024. [1] Outside of Europe, the share of renewables in global electricity supply rose to 30% in 2023 and is projected to climb further to 35% in 2025. [2]
Increased exposure to intermittency can lead to surging prices like those discussed in the article. It also adds strain to the electricity grid as utility operators are challenged to balance energy supply and demand across the duck curve. These unintended consequences of the world’s push to decarbonize energy production are creating demand for grid-scale energy storage solutions. Short-term energy storage, like that provided by Battery Electric Storage Systems (BESS), holds energy for less than 12 hours and is a key tool for grid operators to balance the production/consumption discrepancies modeled in the duck curve. For example, in a region with high solar penetration, BESS could help hold energy created at peak output (noon) until it is demanded by households (sundown). Long-term energy storage, like that provided by Pump Storage Hydro (PSH), describes systems that hold large amounts of energy for longer than a day and in most cases, months to years. PSH represents the largest energy storage type on the US grid today. This is a rather elegant, old-school solution that was developed to help balance the production and consumption differences in Nuclear powered grids of the 20th century. PSH systems typically include dams and reservoirs of water at different elevations. When energy is abundant, PSH ships water up the hill. When energy is needed, locks open and to allow water to flow back down the hill through hydropower turbines. To strengthen Europe’s grid, increases in both short-term and long-term storage systems are necessary.
How big is the problem?
To my knowledge, there is not yet a golden rule for how increases in variable energy production should be matched with energy storage capacity. If we look to China, we can at least find a clue. By 2030, China is expected to reach 1000 GW of solar energy capacity and has begun mandating Energy Storage capacity increases as part of new energy production projects. [3] The most stringent storage requirements are in Shandong Zaozhuang province which mandate storage equal to 15%–30% of the installed PV rated capacity, with a duration time of 2–4hrs. [4] While Wind power likely requires a lower ratio of storage to production, let’s use the low end of this 15%-30% of production capacity figure for our back-of-napkin math.
In the EU, the current installed capacity of Wind and Solar is about 485 GW. [5] According to Bloomberg NEF, there are 15 GW of installed energy storage across the region as of November 2024. When we take a look at the current capacity of energy storage capacity to wind and solar capacity, the EU underperforms our Shandong ratio by a factor of 5. To meet this figure for today’s production capacity output, we would need to add an additional 60 GW of storage to the EU grid.
| Scenario | Energy Storage Capacity | Variable Energy Production Capacity | Storage/ Production | Gap |
| 2024 Estimates | 15 GW | 485 GW | 3% | – |
| 2024 @ 15% Coverage | 75 GW | 485 GW | 15% | 60 GW Storage |
60 additional GW of storage is a MASSIVE amount of storage. For reference, the largest Energy Storage system in Europe will be the GIGA Green Turtle installation in Belgium, will have a 600 MW capacity, or 60% of 1 GW. [6] Today’s largest energy storage system, China’s Fengning Pumped Storage Power station, has a capacity of 3.6 GW. [7] When it comes to 60GW standalone storage systems in 2025, our projections are in the realm of science fiction.
Yet 60 GW only covers today’s wind and solar capacity in Europe. The IEA projects up to 500 additional GW of Wind and Solar generation in Europe by 2028. Updating our table:
| Scenario | Energy Storage Capacity | Variable Energy Production Capacity | Storage/ Production | Gap |
| 2024 Estimates | 15 GW | 485 GW | 3% | – |
| 2024 @ 15% Coverage | 75 GW | 485 GW | 15% | 60 GW Storage |
| 2028 @ 15% Coverage | 150 GW | 985 GW | 15% | 135 GW Storage |
How much will this cost?
BESS systems have become the go-to solution for additional energy storage thanks to a relatively easy installation process and their ability to built-to-size; making these systems bigger or smaller is a function of adding/subtracting modules. Recent figures in the US show bulk-storage BESS costs about $526 per KWh to construct. [8] Translating this $526 price point to the scale of multiple gigawatts requires some significant math-ing:
Let’s say we’re building a 4hr short-term BESS system which is probably the smallest duration a developer would install today.
- For a 1GW/4GWh system, we’re looking at 4,000,0000 KWh * $526 per KWh= $2.104B
- To provide a backup for 15% of today’s European capacity (the 60GW figure above) we’re looking at a price tag of $126.24B.
- Providing 15% coverage for 2030’s projection of nearly a Terawatt of Wind and Solar implies a cost of $284.04B.
While this is an overly-simplistic model that assumes no economies of scale and consistent prices that will inevitably drop as BESS proliferates, these jarring dollar figures demonstrate how much investment is required to create a robust grid powered by Wind and Solar. Oh, and let’s not forget- our hypothetical $284.04B system only holds 4hrs worth of energy. This will not solve the Dunkelflaute problem.
Pump Hydro, like the Fengning system mentioned above, allows for cheaper capacity at scale and a significantly longer duration. However, creating a quick-and-dirty PSH cost estimate is more challenging than what we’ve modeled for BESS above. Cursory research surfaced a $1.87B contract for the Fengning plant’s power station, but this does little for us. [9] A few reasons:
- Pump Hydro at the scale of Fengning requires a massive public works project: dams need to be built, rivers need to be diverted etc
- China’s authoritarian government is immune to the sort of NIMBYism and Environmental Protections that are likely to plague a project in a Western democracy
- The $1.87B price figure quoted above is exclusive to the power plant component of Fengning. This does not incorporate the land engineering discussed in 1.
While Fengning does not do much to illustrate our cost estimate, PSH systems have become investment targets. Publicly disclosed deals can provide some guidance here.
- In 2016, Public Sector Investment Pension board (PSP Investments) acquired a portfolio of US-based hydro power assets for $1.2B that includes a 1.2GW PSH system [10]
- In February 2023, TransAlta announced the acquisition of 50% of an early-stage, 320MW PSH system in Alberta, Canada for up to $25M USD [11]
- An article from May 2023 describes a project to convert a traditional Scottish Hydropower plan to a PSH system, Coire-Glass. The 30GWh project has an estimated cost of £1.5 billion ($1.85B USD) price tag [12]
These 3 deals give a very blurry picture for what newly constructed PSH capacity to solve the EU’s 135GW storage gap for 2030. Acquiring aged assets as PSP Investments did is not skips the complexity of permitting new projects. Purchasing a stake in a relatively small project that is underway with a price estimate is not the same as a final price. And retrofitting existing traditional systems, while additional, does not account for the feasibility of breaking new ground for projects. Because of this, I’ll refrain from overfitting these price points similar to the above BESS exercise.
Wrapping Up
Energy Storage is becoming a huge problem for a world that increasingly relies on variable renewable energy sources like Wind and Solar to meet its growing energy demand. Solutions exist, namely BESS and PSH, that will require significant investment (hundreds of billions of USD) to meet just Europe’s 2030 energy storage needs. This may represent a generational opportunity for infrastructure investors and developers.
Citations
[1] IEA- Electricity Mid-Year Report, July 2024.
[2] Ibid.
[4] Ibid.
[5] European Commission, State of the Energy Union Report 2024
[6] PV Magazine, Europe’s largest battery storage project secures approval
[7] Wikipedia, Fengning Pumped Storage Power Station
[8] UtilityDrive, New York energy storage additions since 2018 approach 1 GW: report
[9] Wikipedia, Fengning Pumped Storage Power Station
[10] PSP Investments, PSP Investments to acquire 1.4GW hydroelectric assets in New England
[11] TransAlta, TransAlta Announces Acquisition of 50% Interest in Early-Stage Pumped Hydro Energy Storage Development Project
[12] Current News, Britain’s ‘largest hydro power plant’ to be converted for pumped hydro storage
Shane Wilson's Blog 