Engineers at a 400 MW North Sea wind farm could not explain the power they were missing, and a lidar on the platform traced the thinned wind back to another company’s turbines 34 miles upwind

The optical head turns on its mount, stops, and turns again.

Its housing is furred white with salt after months bolted to a deck a hundred miles from land.

What it throws out is invisible, a pulse of light that comes back carrying the speed of the air it passed through.

What the lidar changed is the conversation about what an operator in a crowded sea can reasonably expect to own of the wind that arrives at its fence line.

Point it at the horizon and sweep, and you get a cross section of the wind for miles around.

The turbines behind it were making less power than the forecast said they should.

The wind arriving at them had already been used.

How a rotor casts a shadow across open water

A turbine takes kinetic energy out of moving air. The air does not snap back to full speed the moment it clears the blades.

What forms instead is a long corridor of slower, more turbulent flow trailing downwind.

Inside a single farm that corridor is designed around. Rows are spaced to let the mess dissolve before the next rank of machines.

Between separately owned farms nobody plans for it at all.

Whether the corridor survives across open sea comes down to the air itself. On calm, clear days the North Sea settles into layers that barely mix vertically.

In that stable state a wake cannot spread upward and dilute. It travels sideways, more or less intact, for far longer than churning air would allow.

Which means the loveliest weather is exactly when a downwind farm is most exposed.

What the instrument saw when it looked upwind

A conventional met mast reads one fixed point at one height.

This is a long range Doppler unit that scans, mapping incoming wind across a wide arc of sea at once.

The picture it returned was not subtle. On specific days a river of slower air was arriving from the direction of the upwind clusters, in a band you could trace back to its source.

That slower air was reaching the rotors before they had any chance to work with full speed wind.

The instrument was not alone in saying so.

Satellite radar caught the same bands from orbit, and the farm’s own control system logged the power drop underneath them.

The distances and the deficits

Two upwind clusters mattered. One sits about 16 miles to the southwest with 802 megawatts of turbines in it. The other sits about 34 miles to the southeast with 914 megawatts.

Both left a mark.

In one lidar scan the nearer cluster’s wake showed a wind speed deficit of 40.5 percent. Averaged over the field it came out at 24.7 percent. The farther cluster produced 26.4 percent in the lidar data.

The campaign ran from the summer of 2018, and the analysis appeared in a wind energy journal in 2020.

A separate satellite study published in the spring of 2026 put the phenomenon in context, working through 7,122 radar images of more than 60 offshore farms across Europe and Asia between 2020 and 2022.

Its average was a 12.4 percent reduction, and it found wake signatures persisting beyond 60 miles.

Why the averages and the peaks are not the same number

The 12.4 percent figure needs a caveat the headlines drop. It was measured at 33 feet above the sea, not at hub height, and the authors say plainly that conditions up where the blades turn can differ.

The lidar peaks need one too. A single scan is a moment, not a season, and the farm does not lose 40 percent of anything on an annual basis.

What makes even modest deficits expensive is the arithmetic of a rotor. Power scales with the cube of wind speed, so shaving a tenth off the speed takes closer to a quarter off the output.

The effect also needs its weather. Stable and weakly unstable air is the condition, and mixed turbulent days break the corridor up.

Planning models have their own trouble here, and they already underestimate wakes inside a single farm.

The full measurement record is public, which is more than can be said for most disputes about missing revenue.

The problem no lease boundary resolves

The deeper issue is ownership.

Each farm was permitted, designed and financed on the assumption that its own wind resource was independent.

No boundary on a seabed chart stops moving air from crossing into the next concession block, and no agreement compels an upwind operator to change its layout for a farm it has no contract with.

The satellite authors say it directly. No government or industry body has yet established any settlement or legislation to compensate downwind farms for revenue lost to upwind wakes, and they expect measurements like theirs to accelerate that.

Wakes are visible from the air as well, in the six mile ghost tails pilots have been reporting off Denmark for years.

This North Sea wind farm did not pick its neighbors, and the neighbors did nothing wrong.

What the lidar changed is the conversation about what an operator in a crowded sea can reasonably expect to own of the wind that arrives at its fence line.