The effects of a heatwave on the electricity market

Anyone who manages a refrigeration installation knows that a heatwave drives up electricity demand. But that rise in demand alone does not explain why prices can suddenly swing to extremes during such a period. That only happens when heat coincides with low wind, and solar output falls away at the end of the day while cooling is still running at full tilt. There is now a term for that coincidence: Hitzeflaute. Below: what exactly happens, in which sectors you see it on the meter, and which line of your bill it lands on.

Who actually consumes more during a heatwave

"People switch on the air conditioning" covers only a small part of the extra demand. In business terms, the increase sits mainly in installations that have to work against a warmer environment, and that have no choice in the matter:

• Food retail and fresh distribution. Refrigeration and freezing are often the majority of consumption here. At high outdoor temperatures the same refrigeration unit has to work against a higher condensing temperature. The same cooling capacity then costs more kWh, without a single extra door being opened.
• Refrigerated logistics, cold and freezer stores. The same effect, plus heat load through docks and doors. The item that quietly rises here is not the kWh but the measured peak capacity.
• Hospitality. Walk-in coolers, ice machines and climate installations run hardest during the evening service, precisely the hours in which the day-ahead price peaks during a heatwave.
• Offices and let real estate. Cooling, rather than heating, is the peak-determining load here. Where charge points are also installed, an unmanaged charging peak comes on top of it.
• Data centres and server rooms. At high wet-bulb temperatures, free cooling drops out and the chillers have to take over. The cooling load rises while the IT load stays the same.
• Greenhouse horticulture and agriculture. Irrigation and sprinkler pumps, screening and cooling; in dairy farming, extra milk and tank cooling on top of an already flat, continuous profile.
• Healthcare. Climate installations with strict requirements, 24/7. Consumption does rise here, but the room to respond to it is minimal. For this reader a heatwave is a price question, not a flexibility question.

What these examples have in common: the extra demand sits in capacity that has to run at fixed moments, not in consumption you can shift to the night. A heatwave therefore hits two items at once. The supply price of the expensive hours, and the peak component in your transport costs.

Hitzeflaute: heat without wind is the real price risk

The term derives from Dunkelflaute, the winter combination of little sun and little wind. A Hitzeflaute is the summer equivalent: high temperatures with almost no wind, usually under a persistent high-pressure area that sits in place for days.

The price effect does not come from demand, but from the merit order. The price on the day-ahead market is set by the last unit needed to cover demand; all producers receive that price. During the day, full solar output pushes that marginal unit far down, and the midday hours can be cheap or even negative. Around sunset that solar output disappears entirely within a few hours. Cooling demand does not: buildings, cold stores and processes are still carrying the heat of the whole day. If there is no wind either, that entire drop has to be covered by gas-fired capacity and peaking units. Those have the highest marginal costs, so they set the price for every hour in which they are needed.

That is where the reversal lies. A Hitzeflaute is not a demand problem but a simultaneity problem. It is not the level of demand that makes the evening expensive, but the loss of cheap supply at the moment demand is still high. It also explains why a heatwave with a solid sea breeze produces almost no price swing, while the same temperature in still air does.

What the Dutch day-ahead market showed in June 2026

In June 2026 this pattern occurred in the Netherlands in pronounced form. During a heatwave under a high-pressure area with barely any wind, the day-ahead price for delivery in the evening hours of Wednesday 24 June rose to over € 900/MWh (Dutch day-ahead market; see the market reporting on this). The midday hours of that same day were a multiple lower, while the sun was still producing at full output.

Two things stand out, and both are more useful than the record figure itself. First, the peak sat in a handful of evening hours, not in the day as a whole. Second, the spread was within a single day, not between days. Anyone tracking energy costs on a monthly average barely sees such an event at all, while a business whose cooling peak falls in exactly those hours pays for it in full.

What it means for you depends on your contract type, and those differences are large. On a dynamic contract you settle those hours one-for-one. On a variable contract it returns with a delay in the tariff. On a fixed price or a click-and-portfolio contract, such an evening does not touch your current supply price, but it does affect the level at which you still have to buy your open volumes. And a grower with a CHP unit sees the spark spread improve in those very hours: for that reader this is not a cost but a dispatch opportunity. The same market movement is therefore bad news for one party and good news for another.

Solar output in the heat: the "from 25 degrees" misconception

It is often said that solar panels start yielding less from 25 degrees Celsius. As it is usually meant, that is not correct. Those 25 °C are the cell temperature at which the panel's rated output is measured, standard test conditions, and not an outdoor temperature above which yield collapses.

What is correct: above that reference temperature the delivered output falls, and by how much is stated as the temperature coefficient on the datasheet of your own panels, in percent per degree. For crystalline silicon it is in the order of a few tenths of a percent per degree of cell temperature. Because a panel in full sun becomes considerably warmer than the surrounding air, the loss on a hot day is real. But it is a derating of a few percent, not an outage. If you want to know what it is for your installation, the datasheet of your panels is the only source that states it exactly.

That distinction determines where the problem actually sits. During a Hitzeflaute, solar is not falling short: under a cloudless sky it produces abundantly, and that is precisely why the midday hours are cheap. The bottleneck is the hour in which the sun sets and the wind is absent.

In the heat, the grid is tighter than it looks on paper

The transport capacity of cables and transformers is not a fixed number. It is limited by heat: the higher the ambient temperature, the less room there is to shed their own losses. During a heatwave the highest load therefore coincides with the smallest margin to carry that load. It is one of the reasons grid operators look more closely at their load data in periods like these.

What that means for you is emphatically local. Grid congestion is not a national figure: it differs per grid level, per area, and between offtake and feed-in. A business can have offtake congestion without feed-in congestion, and vice versa. If you have an expansion, a charging hub or an additional cooling installation planned, the status of your own grid level is not background information but a precondition. Request it from your grid operator before the design is fixed, not at the application stage.

What this asks of you

A heatwave is not an incident you can wait for, but a recurring pattern for which your profile is either prepared or not. Four steps you can take with data you already have:

• Lay your quarter-hour values from a hot day over the day-ahead profile of that same day. A single day profile shows whether your peak falls inside the expensive evening hours or outside them. That is the whole diagnosis, and the data already exists.
• Determine which part of your cooling load has thermal mass. Cold and freezer rooms can be pre-cooled within the limits of the product, taking capacity out of the most expensive hour. That does not apply to process-bound cooling or to healthcare climate control, acknowledge that difference rather than working around it.
• Look at the peak in kW, not only at the kWh. A single quarter-hour in which cooling, climate control and charging are all on at once can lock in your transport component for a longer period. You keep paying those costs in the months when it is not hot.
• Test your contract type against this pattern rather than against an annual average. If you cannot shift your evening peak, a dynamic contract mainly buys you risk. If you can shift it, it buys you value.

What needs to be checked for your situation

How hard a Hitzeflaute hits differs per consumption profile, connection and contract type. A fixed supply price protects you against the hourly prices of such an evening, but not against the peak component in your transport costs: that is set and passed on independently of the market. And managing cooling without validated measurement data is guesswork, without quarter-hour insight you do not know which hour you are actually shifting, or whether shifting it gains you anything.

COMCAM places your quarter-hour values alongside market prices and alongside your connection data, so that it becomes visible in which hours your costs actually arise and which part of them can be shifted. Would you like to know how your profile behaves in a hot, windless week? Contact us.