A study published this week in Nature Energy models what happens to Western European power grids when heatwaves raise demand, derate power plants and cut the amount of current transmission lines can carry, all at once. Its main finding is that national outcomes differ sharply. The Spanish grid shows temperature-induced bottlenecks and load shedding in the simulations, while the German grid stays comparatively resilient.
The paper lands months after the June and July 2026 heatwaves tested those same systems. It gives planners a way to see risks that standard power-flow models miss, and it comes with caveats that matter for how far the results can be pushed.
What the study did
The authors built a single framework that combines temperature-dependent demand, generator derating and line-by-line thermal limits inside an alternating-current optimal power-flow model. According to the abstract, they applied it to the Western European system using open-access grid data, high-resolution weather reanalysis and demand data. They simulated hundreds of projected heatwave scenarios for 2026 to 2030.
The paper argues that conventional AC-OPF formulations and simplified thermal models overlook bottlenecks because they do not fully capture the coupled effects of cooling demand, generation availability and weather-dependent transmission capacity. In its grouping, France, Spain and Italy show higher load-shedding risk, while Germany, the UK and several northern systems remain comparatively resilient.
Two further results are worth noting. Demand growth amplifies load shedding in the Spanish case, whereas a higher initial storage state of charge gives partial relief. Cross-border help is also uneven: it depends on whether neighbouring grids still have deliverable surplus capacity when they are in a heatwave themselves.
Why the mechanism matters
Heat acts on three parts of the system together. Cooling raises peak demand, generators lose output (French nuclear plants are limited by river temperatures; wind drops under a stagnant heat dome), and overhead conductors that heat up sag toward their clearance limits, so operators have to reduce current.
Static line ratings are usually set on near-worst-case weather. One academic study of the German grid describes the typical assumption as 40 °C ambient temperature and 0.6 m/s wind. Ember reported that successive heatwaves in June and July 2026 pushed temperatures above 40 °C in many parts of Europe.
A likely reading is that on the hottest days, a static rating stops being conservative in some places, and the network’s weak points move around as power is rerouted. The Enlit trade site makes a related point: lines that were lightly used the day before can become critical bottlenecks. That is a commentary view, not a measurement from the Nature paper.
What the 2026 heatwaves showed
The measured data comes from a different source and a different method, so it should not be read as a test of the model. Ember’s analysis covers France, Spain, Italy and Hungary. It found that daily power demand rose by as much as 28% in Italy, 23% in Hungary, 14% in France and 13% in Spain on heatwave days, compared with the week of 13 to 19 June.
Peak demand rose by less, and Ember notes that peak levels in France and Hungary stayed below their winter highs. Average daily prices during the late-June and early-July heatwaves were 119% higher in Hungary, 44% higher in France, 13% higher in Italy and 7% higher in Spain than in the preceding week, according to the same report.


| Country | Nature Energy risk grouping | Daily demand rise (%) | Peak demand rise (%) | Daily price change (%) |
|---|---|---|---|---|
| Spain | Higher load-shedding risk | 13 | 11 | 7 |
| France | Higher load-shedding risk | 14 | 9 | 44 |
| Italy | Higher load-shedding risk | 28 | 14 | 13 |
| Hungary | Not named in the grouping | 23 | 17 | 119 |
| Germany | Comparatively resilient | Not covered | Not covered | Not covered |
| UK | Comparatively resilient | Not covered | Not covered | Not covered |
Ember’s figures compare heatwave days with a reference week (demand and price) and are measured data. The risk grouping comes from the Nature Energy simulations of projected 2026 to 2030 scenarios. Ember also revised its demand figures after publication, so the numbers above are the corrected ones.
Supply was constrained too. In the Ember report’s own assessment, heat might have affected up to 18% of French nuclear capacity on 12 July, against an estimated 15% in the June 2025 heatwave. Analysts also warned that low water in Italy’s Po river could limit up to 32% of the country’s thermal fleet.
Solar was the one major source that did better than usual. According to Ember, average daily solar output on heatwave days was 17% higher in France and Hungary and 5% higher in Spain, with no change in Italy. EU solar generation reached a record 52 TWh in June and 55 TWh in July, about 25% of EU electricity in both months, Ember says.
The evening problem
Daytime solar covered much of the extra load. Prices and scarcity were worst in the evening, when solar fades and cooling demand stays high. Chris Rosslowe, a senior analyst at Ember, put it this way: “the real challenge starts after sundown”.
Ember’s recommendation is more battery storage, demand response and stronger interconnection. Its analyst Beatrice Petrovich has called the extreme price spikes a signal for regulatory changes that raise flexibility.
Storage addresses energy shifted in time. The Nature Energy paper’s concern is partly different: where on the network a thermal limit binds. A battery in the wrong node cannot relieve a constrained corridor, and the study’s finding on storage state of charge applies to its Spanish case only. A likely reading is that the two analyses point to complementary fixes, flexibility for the evening energy gap and network measures for location-specific limits.
A live example in Britain
Great Britain shows what happens when operating forecasts miss. Fintan Slye, chief executive of the National Energy System Operator (NESO), told an FT conference that the operator underestimated demand for air conditioning during the June heat, adding that managing the system is becoming “significantly more complex” with the shift to renewables, according to the FT as relayed by Electricity Info.
A Times report, as quoted on a Substack blog, put NESO’s summer outlook for peak daily requirement at 29.7 GW, against a peak of 35 GW on 23 June according to the software firm Kraken. I could not check those figures against NESO’s own documents. The same account says the grid ran below its normal operating threshold for a record 26 minutes.
The governance dispute is contested. Claire Coutinho, then shadow energy secretary, relayed whistleblower allegations of concealment. An investigation by the law firm Eversheds Sutherland did not substantiate the allegations set out by Coutinho, Solar Power Portal reports. Slye has said no customers lost power and that frequency stayed within statutory limits, according to GB News.
The analyst Kathryn Porter, quoted by the Telegraph via Yahoo, notes that a high risk of demand control (HRDR) notice, which NESO describes as issued only in rare circumstances, was considered on 23 and 25 June. The commentator at Watt-Logic says the Eversheds report also recorded five network constraints breached with no available mitigation. That reading differs from the FT’s summary, which says the investigation did not support claims that NESO breached operational limits. Ofgem’s separate technical investigation is still running.
The Nature Energy paper groups the UK among the comparatively resilient systems. June’s events show that a resilient network can still be strained by forecasting and operational margins, which a power-flow study does not model.
Who gains, who loses
Developers of battery storage and demand-response providers gain a stronger case in the high-risk markets, particularly where evening scarcity prices recur. Grid operators gain a planning argument for monitoring equipment. Consumers in Hungary, where average daily prices were more than double, and in France bore the visible cost in 2026. Thermal generators that run in the evening peak benefit from high prices, but they are also exposed to heat and water limits.
The losers in a high-risk scenario are customers on constrained corridors. In the paper’s Spanish simulations, they face load shedding rather than only higher bills.
Dynamic ratings: a partial remedy
Dynamic line rating (DLR) replaces worst-case assumptions with measured and forecast weather. The IRENA technology brief cites simulation studies in which DLR forecasts lead to an average increase in transmission capacity of 10 to 15%. CleanTechnica, summarising Idaho National Laboratory case studies, reports that Austria’s APG deployed DLR across about 15% of its network, with peak capacity gains around 10% on monitored lines and congestion savings of about EUR 12 million per year.
Heat limits what DLR can do. Its biggest gains come when air is cool or windy, which are not heatwave conditions. A US preprint modelling the Western Interconnection (not yet peer reviewed, as far as I could determine) estimates that static ratings lose 1 to 3% of transmission capacity by mid-century, and that advanced ratings remove most of the adaptation-driven capacity need. A likely reading is that DLR reduces the cost of adapting but does not remove the need for new lines or flexibility in the hottest corridors.
Risks and open questions
- Model scope. The paper’s results apply to the represented grid configurations built from open-access data. It does not forecast actual outages, and national results depend on how detailed each public grid model is.
- Short horizon, large uncertainty. Scenarios for 2026 to 2030 are built from projected weather, so the results show exposure rather than probability.
- Coincident heat. Cross-border relief fails when neighbours are hot at the same time, which is what the 2026 summer looked like.
- Market signals. Prices did not track the model’s risk ranking: Spain, in the high-risk group, had the smallest average price rise in Ember’s data. A likely reading is that system-wide prices can hide local network limits.
- Operator forecasting. Forecast error, as in GB, sits outside the transmission-physics story but compounds it.
What to watch
Ofgem’s technical investigation into the June heatwave and NESO’s final findings will show whether forecasting and constraint management change. Watch also for whether transmission operators in Spain, France and Italy move from pilot DLR to wider rollout, and whether regulators there lower barriers to grid-scale storage ahead of next summer.
Planners will test the Nature Energy framework against actual network data. Operators hold line-level information that the open-access models lack, and their results could confirm or reverse the country rankings.
Sources
- Source: Heatwave-induced capacity bottlenecks in European electricity grids (Nature Energy)
- Source: Solar helps Europe’s grid withstand extreme heat (Ember)
- Source: Solar output in European countries boosted by up to 17% on hot days (Ember, 13 August 2026)
- Source: Solar helps Europe’s grid withstand extreme heat (report PDF, Qualenergia)
- Source: Ember: Solar Helps Europe Through Heatwaves; Storage Needed (TaiyangNews)
- Source: How solar saved the grid during Europe’s heatwave (Enlit)
- Source: Europe’s heatwave is creating quadruple whammy for the grid (Enlit)
- Source: Dynamic Line Rating: Innovation Landscape Brief (IRENA)
- Source: Leveraging the existing German transmission grid with dynamic line rating (ScienceDirect)
- Source: Unlocking Existing Grid Capacity With Dynamic Line Rating (CleanTechnica)
- Source: Dynamic Line Ratings Counteract Transmission Investment Needs Driven by Climate Change (arXiv preprint)
- Source: NESO (Electricity Info, relaying Financial Times reporting)
- Source: Independent investigation finds no evidence of NESO misconduct (Solar Power Portal)
- Source: Review into Britain’s feared energy blackout (GB News)
- Source: NESO admits constraints breached and large forecast errors in interim grid security report (Watt-Logic)
- Source: Grid bosses cling on as investigation lays bare blackout panic (Yahoo News)
- Source: Neso’s modelling failed to anticipate June heatwave (Substack blog quoting The Times)

