Intermittency Trap

Unpleasant Energy Transition Arithmetic

“No willpower in the world can override the basic rules of physics, not even Dr. Robert Habeck.” — Swedish Deputy Prime Minister Ebba Busch.

With abundant hydrocarbons and enough hydropower for 95% of domestic use, few countries in the world have a better energy endowment than Norway. Nonetheless, on December 12th, prices rocketed to twenty times their normal level.

That day’s cost surge extended across borders. In southern Sweden, prices were 190 times higher than in the north of the country.

In the Netherlands, electricity tariffs broke records, hitting 120 cents/KWh. (Normal prices are around 15-20 cents). 350,000 households in the country have dynamic contracts and paid the full spot price.

In Spain, the grid operator cut power to heavy industrial users for the fourth time this year due to insufficient supply.

The cause of all this was a crash in intermittent power generation. Europe faced what Germans call a dunkelflaute — a weather phenomenon where there is both no wind and no sun.

The dunkelflaute is not a rare occurrence. According to Li et al. (2021), during the winter in the North Sea region there are between 50 to 100 hours of dunkelflaute per month, with most events lasting between 12 and 24 hours. Outliers can last up to five days.¹

During those dunkelflaute hours, almost no wind or solar power can be generated.

In December 2023, solar and wind were responsible for 41.5% of Germany’s power. During the dunkelflaute last week that percentage dropped to 4.4%.

In Spain, wind power dropped to just 8% of normal output. Solar fell to 7.7%. Officials chose between keeping homes lit and keeping factories running.

On average the EU uses around 7400 GWh of power every day for heating, light and industry during the winter. When there is no wind and no sun and a renewable-heavy mix, prices skyrocket.

As conventional baseload power comes offline and more intermittent renewable energy comes online, these short, sharp peaks are becoming more pronounced.

Solutions

There were supposed to be solutions. As Europe’s transmission capacity improves, regions with (temporarily) low power should be able draw on those with more resources. Spain may have sun even when Germany has its doldrums; transporting the power is a way to avoid regional strains.

Unfortunately, Europe is currently behind when it comes to transmission. Draghi estimates that Europe needs 500 billion dollars of additional investment in transmission by 2030 to lay down the lines in order to make it work.² The organisation for European transmission system operators estimates cross-border transmission needs to double in the next 7 years. Though better than America, over thirty percent of Europe’s significant cross-border electricity projects face permitting-related delays. Just one transmission project — Germany’s North Sea Ultranet — requires 13,500 permits.

The second problem for transmission is political. Norway's electricity shock was not due to its own production, but demand from elsewhere in Europe. After last week's dunkelflaute, both coalition parties in Norway are now against renewing interconnection with Denmark. The right-wing Progress party, likely to take power after the next elections, wants to reassess links with Germany and the UK as well. As prices spike, countries with extensive dispatchable power will be hesitant to tranfer it to those without.

Storage offers another way out. As Casey Handmer notes, power demand has much more temporal variation (the volumes change by 2x daily) than spatial variation (nearby regions often have correlated demand patterns). In principle, one would prefer to arbitrage power prices over time (batteries) rather than across space (transmission).

But as with transmission, it is unlikely that storage will arrive on time for Europe to cover the lengths and severities of dunkelflaute. According to the 2023 Ten-Year-Network-Development-Plan, Germany should have 52 GWh of battery storage by 2030.³ During the winter, Germans on average consume more than 60 GWh every single hour. If it relied on batteries alone, to be comfortable for a 48-hour dunkelflaute Germany would need over 25x its most optimistic 2030 scenario: 2800 GWh of batteries. It is not a question of will: procuring that many batteries means buying 40% of all lithium-ion battery production capacity in the world in 2030.

Germany pins its other storage hopes on hydrogen. But only three pilot projects totaling 2 GWh have been finalized, and new facilities require around eleven years to build.⁴ A recent study by Frontier Economics, a research group, estimated that the gap between projected hydrogen demand and capacity that needs to be bridged is between 13,300 and 16,300 GWh — 8000 times the storage currently underway.⁵

If neither storage nor transmission are in place, the solution will have to come in the form of dispatchable capacity. This option is appealing. A single one-gigawatt nuclear plant over the course of a 96-hour dunkelflaute generates 96 GWh of power, providing close to twice as much power as all the battery storage Germany is projected to build by 2030.⁶

The obvious path is recommissioning nuclear plants. Germany can cheaply and easily bringing back 10+ GW of clean power. According to a recent report by Radiant Energy, two plants with a combined capacity of 2,7 GW would require an investment of just €1 billion per plant, and could be brought back online within one to three years. A larger group of seven plants with a total capacity of 9,4 GW could be restarted within the next four to eight years.

While able to blunt some of the downside, recommissioned plants are not enough. Dealing with dunkelflaute in Europe means building additional dispatchable power. The major objection is that nuclear power won't be there in time. That is a policy choice. While building a plant took on average 127 months in France, it takes just 56 months in Korea, and 46 months in Japan.

Of course, in reality there is no plan to surge more nuclear power, or even recommission plants that were recently closed. Is Europe resigning itself to rationing and shocks and high energy prices? The surprise after the dunkelflaute, the reaction of Norwegians, and distress of the Europeans all imply that perhaps not as much thought went into intermittency as one would have hoped.

The other outcome will be Europe burning a lot more hydrocarbons as a way to prevent rationing and shortages. Germany’s coalition long aimed to phase-out coal by 2030. Problems with intermittency have forced Green Leader Robert Habeck to concede that they will be stuck with 2038.

This option is bad for the climate targets. It is also very inefficient. Making 85-90% of the grid renewable but keeping a second grid's worth of hydrocarbon capacity in reserve in case of changing weather is extremely costly. Baseload plants would need to recoup their capital costs during the few hours of surge operation — guaranteeing soaring prices.

The problems with intermittency don’t mean that Europeans need to be against the energy transition. Solar is becoming extraordinarily cheap and efficient. Engineer and founder Casey Handmer (blog highly recommended) calls solar panels "panes of glass on the ground that print wealth." The cost-curves make it near-certain that 15 years from now photovoltaics will have transformed the world.

But here is a reasonable rule: if power needs to be rationed or price records are broken, policy has failed. The current approach seems to be heading towards the intermittency trap. Avoiding it will mean surging baseload, or conjuring a miracle with storage.

1. Li, Bowen, Sukanta Basu, Simon J. Watson, and Herman W. J. Russchenberg. "A Brief Climatology of Dunkelflaute Events over and Surrounding the North and Baltic Sea Areas." Energies 14, no. 20 (2021): 6508.

2. Draghi, Mario. 2024b. "EU Competitiveness: In-depth analysis and recommendations" EU Commission. https://commission.europa.eu/topics/strengthening-european-competitiveness/eu-competitiveness-looking-ahead_en, 15.

3. Frontier Economics. "Wert von Großbatteriespeichern im deutschen Stromsystem." Study commissioned by BayWa r.e. AG, ECO STOR GmbH, enspired GmbH, Fluence Energy GmbH, and Kyon Energy Solutions GmbH, December 2023, 3.

4. Bothe, David, Matthias Janssen, Jasmina Biller, and Anna Lane. "Finanzierungsmechanismus für den Aufbau von Wasserstoffspeichern." Frontier Economics, Study commissioned by BDEW, August 2024, 5.

5. Bothe, David, Matthias Janssen, Jasmina Biller, and Anna Lane. "Finanzierungsmechanismus für den Aufbau von Wasserstoffspeichern." Frontier Economics, Study commissioned by BDEW, August 2024, 5.

6. Of course, you want more peak output than just 1 GW