Point an instrument at the open sea and there is nothing there to measure.
No hills, no buildings, no terrain to bend the air around.
Which is the reason people put wind farms out there in the first place.
So when the air upwind of one turns out to be moving slower, there is nothing to blame it on.
Nothing except the farm itself.
Sitting miles downwind, pushing back against its own supply.
How a row of turbines pushes on the air ahead of it
A turbine does not slice through wind. It takes momentum out of it, and Newton insists that the air pushes back just as hard.
That reaction force is thrust, and it goes upwind. A single rotor slows the air slightly in front of itself, which engineers have known and modeled for a century.
The open question was what happens when eighty of them stand in a block.
All that individual thrust adds up into a raised pressure zone ahead of the array, and the incoming air starts responding to it long before it arrives.
Some of it slows down. Some of it slides sideways or lifts over the top rather than pushing through.
This is not the wake effect, which is the slowed air a farm leaves behind it. This is the opposite end of the machine, and it is called global blockage.
What is actually standing out there
The site is a German array roughly 110 miles off the coast, in about 130 feet of water.
It is not floating. Each of its 80 machines sits on a steel tripod driven into the seabed, which is the standard approach at that depth.
The turbines are 5 megawatt units with rotors about 380 feet across, giving 400 megawatts across the block.
The instrument was a scanning long range Doppler lidar, which fires a beam out over the water and reads the speed of the air from what comes back.
It was set up to look upwind, along the water rather than up into the sky.
The beam reached out past three miles in front of the first row.
What the beam found, and when it did not
Air approaching the array was running about 4 percent slower than it should have been, out to roughly 3.4 miles ahead of the first turbines.
The team put the uncertainty on that at 2 to 6 percent, which is a wide band around a small number and they say so.
The condition attached to it is the part that matters. The slowdown appeared under stable atmospheric layering, where warm air sits over cooler sea and vertical mixing is suppressed.
In unstable conditions, with air churning up and down through the layer, there was no measurable blockage at all.
The second condition is mechanical. It showed up when the turbines were running at high thrust, above 0.8 on the coefficient scale, and vanished at 0.3 or below.
Which is not the same as running flat out. Thrust peaks in the middle of the power curve and falls away above rated wind speed, so the effect is strongest at moderate winds rather than at full output.
Why the size of it is still argued about
One campaign at one farm is not a general law, and the researchers frame it as evidence rather than proof.
A separate study on how blockage translates into lost power came out much smaller, on the order of one percent for a row averaged across a wide inflow sector, under idealized conditions.
So the air slows measurably while the effect on the meter is small and still contested, and those two facts sit together uncomfortably.
Blockage also gets confused with the other thing a farm does to wind, which is the shadow it throws downstream, and one North Sea array is already taking wind from another 34 miles away.
Those are different mechanisms with different fixes, and treating them as one number is how a yield forecast goes wrong.
The most vivid evidence is not a chart. Photographs over a Danish farm caught shallow sea fog thinning in the slowed air upwind of the turbines, the effect made visible.
What it changes for the next block of steel
Front row turbines are supposed to see clean wind. If they do not, the reference case every layout is measured against is slightly wrong.
That matters most where arrays are packed close, and offshore zones are being carved up into exactly that arrangement.
Spacing, orientation and blade pitch are all on the table as ways to bleed off the pressure, and none is free.
The money is the sharper edge. Project financing rests on production forecasts, and a systematic overestimate at the front of an offshore wind farm propagates through every number behind it.
Neighbors make it worse, which is why the dispute over one array costing another real money is going to become routine.
The correction is small, and it is measured, and it is published, which is more than most of the assumptions in a wind resource model can claim.
Read the whole thing?
Get the week's signal, not the noise
Our sharpest reporting on energy, climate and nature — free, once a week.
