This summer, record-breaking heatwaves in Europe led to extreme stress in the power system with high evening prices due to air conditioning demand, insufficient battery storage, and outages at thermal and nuclear power plants. Events such as these highlight the challenge of maintaining a reliable power supply during extreme weather.

The same goes for so-called compound events, when a combination of meteorological factors leads to extreme situations, as in late 2024, where periods of low wind and solar generation across Europe triggered sharp spikes in electricity prices and increased reliance on fossil fuels and imports.

In a new study published in Nature Sustainability (see fact box), researchers from the Technical University of Denmark, DTU, and Newcastle University modelled a future net-zero European power system and tested it against 80 years of historical weather data to understand how it would handle stress with periods of low wind and solar generation, combined with high demand.

They have developed new ways to assess how well future power systems can withstand these conditions, underscoring the growing need for reliable backup capacity as Europe expands renewable energy. Based on historical data, they conclude that the risk is greatest during winter when cold and wind-still conditions persist over several days - too long for batteries to compensate.

“Future energy systems will rely heavily on renewables, as they are the cheapest form of generating electricity and increase our energy independence. However, to cover the power deficit when demand outstrips production, we need back-up capacities,” says Marta Victoria, professor at DTU Wind.

“We also find that the capacities needed to keep the system secure during these periods are otherwise used rarely. They are more expensive than renewables or other technologies like hydropower, making them difficult to finance under current market conditions and leading to high electricity prices. ”

The study examines so-called system-defining events, periods when the power system faces particularly high strain and investment decisions are most affected. Using long-term weather data and detailed modelling of the European grid, the researchers identify when and how these situations arise and what they mean for planning future energy infrastructure.

To carry out the analysis, the researchers used an open-source energy system model and constructed a counterfactual on how a net-zero power system with 80 years of historical weather data (spanning from 1941 to 2021) would look. This makes it possible to capture both typical conditions and more extreme years that could shape future systems.

System-defining events are identified using a metric based on “shadow prices”, a modelling concept that reflects the stress level of the system to meet demand at a given moment. In practical terms, higher values indicate the need for higher capacities (generation, storage, transmission) and shape investments as well as the design of the energy system.

These events are often referred to as energy droughts, periods when wind and solar generation simultaneously fall short. Because Europe’s electricity grid is highly interconnected, periods like these can affect large parts of the continent at once.

For example, in 2021, unusually low wind speeds across northwestern and central Europe reduced wind power output for months, contributing to higher electricity prices. And a few years earlier, a cold spell in January 2017 combined low wind generation with high heating demand, increasing pressure on electricity systems in several countries.

“Our approach identifies when the system is under the most pressure, what drives those situations, and how the system reacts. The greatest risk occurs in the dark winter months when low wind speeds and cold temperatures coincide. So, extreme weather, as far as the energy system is concerned, doesn’t necessarily mean ‘bad weather’,” says Aleksander Grochowicz, postdoctoral researcher at DTU Wind and lead author of the study.

“We often use the German term dunkelflaute to describe these lulls. Under those conditions, we mostly lack power capacities, and we have energy available, but we cannot dispatch it.”

The study also distinguishes between short-term and long-term resilience challenges. Short-term resilience refers to the system’s ability to cope with events lasting hours or days, such as sudden drops in renewable generation. Long-term resilience concerns the overall design of the system, including how much generation and storage capacity should be built to withstand variability over years or decades.

“We show that back-up power capacities improve short-term resilience, whereas long-term resilience comes from the adequate size of the system and sufficient renewables,” says Aleksander Grochowicz.

A key finding is that supplying energy relies on many factors and a combination of renewables, other low-carbon sources, batteries, and storage, as well as transmission. Also, for short-term stress, backup capacities are needed.

The complexity of such systems can handle even extremes, but requires good coordination and planning. However, the researchers note that the economic viability of such back-up capacities remains uncertain, since extreme weather events are infrequent and unpredictable. Backup systems may therefore be underused for long periods, making it difficult to justify investments.

“The build-out of renewables needs to be sped up, as they are the best strategy to obtain cheap and reliable electricity for almost all of the year, reduce emissions, and increase energy security,” says Marta Victoria.

“We need to think more about resilience for the short amount of time when renewables are not sufficient, and how much we are willing to pay for it. Better planning leads to investments and higher energy security that pays off. What we have done is identify when the system is under the most stress and how often it happens, providing a clearer basis for planning a resilient system."