Cold snap that broke a grid

A week of Siberian air rolled across northern Europe at the start of February, and the temperatures that resulted were not record-breaking in the historical sense, just low enough to matter. One regional grid operator, after a weekend of warnings, shed load for the first time in years, briefly cutting power to a district of heavy industry to keep the system from collapsing. The cold was the trigger, and the cause was the margin between demand and supply, which has been shrinking quietly for a decade and has now reached the size where a single weather event can expose it.

The arithmetic of the event is straightforward and uncomfortable. The cold raised demand as every electric heater and industrial process came on at once; the wind fell to a regional lull at exactly the wrong hour; and the interconnectors, the cables that let grids borrow power from their neighbours, were already saturated by the same weather affecting every neighbour at the same time. The operator’s reserves, which are supposed to cover the worst credible hour of the year, covered the second-worst hour and then ran out. The shed load lasted forty minutes and was managed cleanly, and the system recovered, which is what the operator’s press release said and what the engineers’ post-mortem will quietly qualify.

The deeper story is about how the margin got thin. Europe has spent a decade adding wind and solar at record pace, and the new capacity is cheap and clean and variable, and the flexible back-up, the gas plants, the storage, the demand response that a variable system requires, has grown far more slowly than the models assumed. The result is a grid that looks strong on average and weak at the edges, and the edges are exactly where February’s weather lives. The irony is that the same governments that celebrated the renewable build-out are now discovering that the boring part, the flexible capacity, is the expensive part, and that the bills for it arrive not as investment but as emergencies.

This event should be read as a normal part of the transition rather than a failure of it. No grid in the world has yet solved the problem of a cold, dark, still winter week with a mostly renewable fleet, and the countries that pretend otherwise are selling something. The useful response is the unglamorous one: more interconnector capacity, better demand response, storage that can last a week rather than an hour, and a market that pays for firmness rather than assuming it will appear. Those are procurement decisions, not policy speeches, and they are the decisions that determine whether the next cold snap is an inconvenience or an emergency.

The coldest weeks of February are still ahead, and the operator that shed load this weekend will spend them watching the weather with an attention that the summer months do not require. The grid will probably hold; the margin will be thinner than anyone wants to admit; and the industry will start planning, in earnest this time, for the version of this week that arrives with the wind calm and the interconnectors full. The cold snap that broke a grid did not break the system, and it did something more useful, which is to show exactly where the next one will bite.