The way I look at it, if you consider the air mass from ground level up to the top of the atmosphere, the bottom will always have a roughly zero speed since it

The way I look at it, if you consider the air mass from ground level up to the top of the atmosphere, the bottom will always have a roughly zero speed since it is down in the terrain and the weeds. A carpet of windmills a hundred meters tall could at best move that zero speed edge a little higher. The kilometers of air above will be relatively unaffected. To illustrate, consider https://embed.windy.com/embed2.html?lat=41.327&lon=-88.033&zoom=8&level=100m&overlay=wind&marker=true&type=map&metricWind=kt https://embed.windy.com/embed2.html?lat=41.327&lon=-88.033&z... which if I've got that URL right will show you winds at a windmill farm in Illinois USA (nope, it misses slightly for some reason, odd, it is right when I copy it). Look in the lower right of the screen and find the slider with an airplane at the top and a mountain at the bottom. You can use this to change altitudes and explore the wind speed above ground level. Lots more windspeed up there even if the density goes down. Consider the 5500 meters up, about half the air pressure, but twice the wind speed today, so that works out to twice the energy per square foot (on edge).