Bringing light into the Dunkelflaute
In November and December, central and northern Europe experienced a weather phenomenon that has a major impact on renewable electricity generation: Dunkelflaute. With it, critics of the energy transition began to question the security of supply for industrial nations such as Germany. It is therefore high time to shed some light and facts on this phenomenon, so great that the word has been adapted into the English language.
What exactly is 'Dunkelflaute?'
Unfortunately, there is no widely accepted definition of the exact scope of this situation, making it difficult to pinpoint in both expert and public discourse. Dunkelflaute translates as 'dark lull' and is supposed to describe a period of time when PV and wind turbines do not generate electricity due to the sole influence of weather conditions. Scientists (e.g. Li et al., 2016, Kasper et al., 2019) and energy experts take different approaches to define it. One way is to use the capacity factor. This factor describes the electricity output over time in relation to the maximum possible output. An example: A 3 MW wind turbine can theoretically generate 3MWh per hour, 72 MWh per day, 61,320 MWh per year. The same would apply to a solar farm. The annual CF in Germany in 2019 is on average 33% for offshore wind, almost 20% for onshore wind and about 13% for PV. (cf. renewables.ninja).
Now there has to be a difference between the annual capacity factor and the hourly CF. Including solar capacity at night in an average would misrepresent the picture of Dunkelflaute, as in the general solar average there is 0% CF of solar power during the night, regardless of the weather. However, the result of 14 hours of sun or 14 hours of cloud is clearly different. Li et al. thus use a threshold to define times of Dunkelflaute. It is 20% of the hourly capacity factor (cf. p.5). In other words: If in the middle of the day both the sun and the wind produce less than 20% of their maximum power, the scientists call it a Dunkelflaute event.
How often does Dunkelflaute occur?
Li et al. analysed the hourly CF for several countries around the North and Baltic Seas in 2016. Of course, the probability of Dunkelflaute is higher in winter, with peaks in November and January. Here, on average, there are 50-100 hours of Dunkelflaute per month, and as many as 150 hours in Sweden. In terms of occurrence, they “conclude that substantial periods of Dunkelflaute lasting for at least a day occur each year. For example, there were approximately 2–10 events (duration of longer than one day) each year in three of the exampled countries, Germany, Norway, and the UK, and similar frequencies were also found in the other eight countries” (p.12). A few cases in the analysis were longer, lasting three and even five days.
Energy supply and prizes during Dunkelflaute
During periods of high renewable supply, electricity markets show low prices because supply is high and PV and wind are the cheapest forms of electricity generation. When the amount of electricity from these sources is low, fossil fuel power plants take on a greater role or nuclear power plants increase their output. Based on the means of generation alone, prices naturally rise.
In addition, much of the electricity is traded a day ahead or during the day. In simple terms, this is a free market in central and northern Europe that also involves buying more electricity than is needed and selling it on at certain times. The value of electricity is therefore not only determined physically, but also in a market comparable to a normal stock exchange. This market is limited by the interconnection capacity between countries, which is not large enough to cover the entire supply of a neighbouring country in the event of a Dunkelflaute. The second constraint of this market is the total possible supply of electricity within a country.
However, in the case of Dunkelflaute on 12 December 2024, when prizes in the day-ahead market skyrocketed, something strange happened in Germany. There was a reserve of 10 GW of coal-fired power plants that could be brought online. On the one hand, this disproves the myth that Germany could not supply itself in times of Dunkelflaute. On the other hand it raises the question of whether prizes would be lower with this supply and why it was not used. This is now being investigated by the German Cartel Office to clear it up. (cf. br.de)
Even though there is no threat in the supply of energy, the influence of Dunkelflaute my rise in the future and the prizes already bother industries who need to stop production in situations like this. There are however solitions that need to be build.
Solution for Dunkelflaute #1: Grid Interconnections
Especially since renewable energy generation in a grid is so decentral, for Europe even all over the center of the continent, there is never zero generation of renewable energy. While in Norway there could be Dunkelflaute, Belgium could have a sunny and windy day. Li et al. found that the correlation of Dunkelflaute events in neighboring countries is just moderate (coefficients of 0.3 – 0.4) (cf. p12). And the CF of sun and wind electricity of interconnected grids is less likely to fall below the Dunkelflaute threshold: ”The mean frequency of Dunkelflaute (marked as a black horizontal line) is decreased from 3–9% for the individual countries to approximately 3.5% when interconnected.” (Li et al., 2016, p.9)
Thus: Interconnected grids are a great tool to make weather dependent generation less relevant. The EU has a goal for 2030 for countries to interconnect 15% of their power generation capacity with neighbors. Better grid connections from the north of Europe to the South would have helped many regions and countries on December 12.
Solution for Dunkelflaute #2: Batteries & Pumped Hydro Storage
Batteries are a great tool to shift energy from times with higher CF to times with lower CF and keep up the grid stability as well as smooth out prize peaks. With more battery capacity, the prize peaks in Dunkelflauten will become smaller. As batteries usually are designed for a maximum 4 hours of storing energy, it cannot bridge completely over the few events per year that are a day or longer. Here, pumped hydro storage comes into play, which can cover great capacity over longer periods and complement batteries.
Solution for Dunkelflaute #3: Thermal Energy Storage
In the future, the biggest part of industrial energy supply will be electrified heat. Thermal energy storage bridges large time periods. In Kraftblock’s case, additional capacity can be bought very cheap and then before the Dunkelflaute period, additional capacity can be charged up with longer charging times. This allows cost reduction and the transportation of large amounts of energy not only over hours but even covering a day or two, thus being able to cover almost all Dunkelflaute events. This even helps right before the event.
By charging Kraftblock in the night hours, the peak prices from after 6am are avoided and after 9pm the system can be recharged. This saves thousands if not hundreds of thousand euros for large factories.
Solution for Dunkelflaute #4: Backup Power Plants
Regions that do not have the luxury of large hydro capacity need a back-up solution. What that backup looks like is an issue worthy of great debate. While Germany has a law proposing planning for natural gas power plants, using LNG may not be a cleaner solution than coal, according to Cornell University (see Howard, 2024). It seems that hydrogen would be a good solution for the Dunkelflaute events, but only the feasibility is a problem, making it an unrealistic solution. CCS also needs to be proven to work. This is not an easy debate. Specific plants like thermal storages with repurposed thermal power plants and pumped hydro storage are the greenest solution in times of Dunkelflaute.
Solution for Dunkelflaute #5: Grid Expansion
As weather and thus Dunkelflaute can be regional, grid connections within a country are important as well. Sweden’s energy rich north has not a lot connections to the South, the same problem as in Germany’s north and south. Connecting larger capacities eases the regional weather influence on electricity generation.
Solution for Dunkelflaute #6: Electricity Market Design
The bidding zones of countries are important for optimising prices and the market. While small countries like Belgium can manage with one bidding zone because they do not have big regional differences or long distances to transport electricity, long and large countries like Sweden or Germany have big differences. That is why Sweden has bidding zones. In Germany, the north has a completely different profile than the south, mainly because of the difference between wind and solar. In general, a division would be beneficial, according to energy experts (cf. cleanenergywire.com). More details on that (in German) by Prof. Lion Hirth on LinkedIn here.
Sources
br.de, 2024. Online: https://www.br.de/nachrichten/meldung/kartellamt-will-hohe-strompreise-bei-dunkelflaute-ueberpruefen%2C3006ee317
Kaspar, F., Borsche, M., Pfeifroth, U., Trentmann, J.,Drücke, J., and Becker, P.: A climatological assessment of balancing effects and shortfall risks of photovoltaics and wind energy in Germany and Europe, Adv. Sci. Res., 16, 119–128, https://doi.org/10.5194/asr-16-119-2019, 2019.
Li, Bowen & Basu, Sukanta & Watson, Simon & Russchenberg, Herman. (2021). A Brief Climatology of Dunkelflaute Events over and Surrounding the North and Baltic Sea Areas. Energies. 14. 6508. 10.3390/en14206508. Online: https://www.researchgate.net/publication/355173603_A_Brief_Climatology_of_Dunkelflaute_Events_over_and_Surrounding_the_North_and_Baltic_Sea_Areas