Coping with the Dunkelflaute: Power system implications of variable renewable energy droughts in Europe

TL;DR

The study addresses how Europe can maintain a fully renewable power system in the face of prolonged renewable droughts, or Dunkelflaute. It combines a drought-identification framework (VREDA) with a least-cost capacity-expansion model (DIETER) over 35 weather years and four interconnection settings to quantify long-duration storage needs and their dependence on cross-border balancing, nuclear presence, and hydrogen-based storage. Findings show that extreme droughts define the required scale of long-duration storage, that geographical balancing can substantially reduce these needs yet cannot fully eliminate them, and that even with nuclear or DACCS options, sizeable hydrogen storage remains essential. The results underscore the practical importance of rapidly expanding long-duration hydrogen storage and cross-border exchange to safeguard Europe’s renewable transition, while highlighting that weather-year selection and interaction with other flexibility options strongly shape planning outcomes.

Abstract

Coping with prolonged periods of low availability of wind and solar power, also referred to as renewable energy droughts or "Dunkelflaute", emerges as a key challenge for realizing decarbonized energy systems based on renewable energy sources. Here we investigate the role of long-duration electricity storage and geographical balancing in dealing with such events, combining a time series analysis of renewable availability with power sector modeling of 35 historical weather years. We find that extreme droughts define long-duration storage operation and investment. Assuming policy-relevant interconnection in our model, we find 351 TWh long-duration storage capacity or 7% of yearly electricity demand in the least-cost system that can cope with the most extreme event in Europe. While nuclear power can partially reduce storage needs, the storage-mitigating effect of fossil backup plants in combination with carbon removal is limited. Policymakers and system planners should prepare for a rapid expansion of long-duration storage to safeguard the renewable energy transition in Europe.

Figures (29)

• Figure 1: Simulated drought events, electricity demand, and least-cost state-of-charge of long-duration storage in winter 1996/97. For each region, the figure shows identified drought patterns lasting longer than 12 hours across all color-coded thresholds (upper panels) and the most extreme drought events occurring in winter (teal boxes). For the UK, where the most extreme drought throughout the year occurs in summer, this event is additionally shown (gray box). The panels below show exogenous smoothed demand profiles used in the optimization. The lower panels further show least-cost storage state-of-charge levels for two counterfactual scenarios, one with isolated countries and for a pan-European copperplate (CP) with unconstrained energy exchange across Europe.
• Figure 2: Correlation of the drought mass of most extreme winter drought events and normalized storage energy capacity. For comparison, we normalize the least-cost storage energy with the annual demand for electricity (including electrified heating) and hydrogen. For illustration, we exclude countries with least-cost storage energy below 5 TWh and countries with binding storage expansion potential constraints. We further include the pan-European copperplate scenario (CP). Figure \ref{['fig:regression_no_filter']} shows the unfiltered regression results.
• Figure 3: Demand seasonality across countries including the pan-European copperplate scenario (CP). The figure shows climatological mean demand as a bold line over all weather years using a moving average over a window of 168 hours (resulting in the blank first week) as a line, the standard deviation range as an area ($mean \pm std \ dev$, dark green), the difference between the climatological minimum and maximum as an area using a moving average over a window of 24 hours (light green), and the regional difference between the minimum and maximum climatological mean normalized by the maximum in parenthesis. Each vertical axis is scaled to show its range if demand seasonality in this region was as pronounced as in France. Demand seasonality is particularly pronounced in France both in terms of level but also variance during winter due to high shares of electrified heat.
• Figure 4: Least-cost long-duration storage energy capacity aggregated across all countries for all modeled weather years and interconnection scenarios. Each dot refers to one weather year, which is modeled independently of other weather years. The year with the highest long-duration storage need is 1996/97 (red). The year that benefits most from rising interconnection capacity in terms of decreasing long-duration storage investments is 1988/89 (green). The year that benefits the least from increasing interconnection is 1987/88 (orange). Figure \ref{['fig:confetti_h2_sto_e']} illustrates the impact of interconnection on the ranking of weather years in terms of least-cost long-duration storage energy.
• Figure 5: Simulated drought events, electricity demand, and least-cost state-of-charge of long-duration storage in winter 1996/97 in countries with highest long-duration storage energy capacities. For each region, the figure illustrates identified drought patterns lasting longer than 12 hours across all color-coded thresholds (upper panel) and the most extreme drought events occurring in winter (teal boxes). For the UK, where the most extreme drought throughout the year occurs in summer, this event is additionally shown (gray box). The lower panels show exogenous demand profiles, which have been smoothed to highlight demand seasonality. They further display least-cost storage state-of-charge levels for isolated countries modeled within the interconnection scenario (1), for policy-oriented interconnection levels in scenario (3), and the pan-European copperplate (CP) in scenario (4).