OK, I'm not sure this is not a silly question, but here goes.

I've often wondered what the impact is of wind turbines on the weather. My question is, since wind turbines transform wind energy into electricity, this means that the winds lose in power. I'm well aware that for the effect to be noticeable, you'd need a pretty large wind turbine park. But I wondered how large? How would the effect scale with the size of the park?

Bonus question: what about tidal turbines?

Most closely related questions:

Is it possible that windmills can consume the wind?

What does the quantification of causes and effect look like for clouds in offshore wind turbine wakes?


Yes, you would need a lot of wind turbines to have a significant, noticeable effect on the weather. There are three potential impacts:

  1. Wind turbines will change the pattern of turbulence downwind from them.
  2. Wind turbines will take some energy out of the wind
  3. Wind turbines will typically displace thermal plant generation, which means that less heat and less steam / water vapour is released from thermal plants: this may have an effect on the local weather in the immediate vicinity of the thermal plant.

As far as I know, point 3 hasn't been studied yet. Points one and two have been looked at in various forms, with different types of micro, meso and macro modelling. The difficulty, as ever, is that we are modelling turbulence, with uncertain boundary conditions. Those who do meso and macro scale modelling are aware of the problems that this causes, but publish nevertheless. There are those who do do bottom-up modelling, starting with the micro modelling of individual turbines, and building up from there, either with processor-intensive area-wide micro-modelling, or through parameterisation. Professor Mark Z Jacobson of Stanford has published quite a lot in the field, and also contributes comments on others work in the field.

At the moment, the bottom line is that deployment is so low, as to give virtually no significant results, with one notable exception: wind turbines can increase the turbulence immediately above ground, in their wake, and this can be quite effective in preventing ground frost. This has been used by farmers, who put turbines on the edge of their field to help protect their winter crops from frosts.

From reference 1 (parameterisation of micro results):

Should wind supply the world’s energy needs, this parameterization estimates energy loss in the lowest 1 km of the atmosphere to be ~0.007%. This is an order of magnitude smaller than atmospheric energy loss from aerosol pollution and urbanization, and orders of magnitude less than the energy added to the atmosphere from doubling CO2. Also, the net heat added to the environment due to wind dissipation is much less than that added by thermal plants that the turbines displace.

There are various works in the field, and it can be helpful to do a quick sanity check on a paper's assumptions, to assess it:

  • Does the paper account for all vertical mixing within reasonable range of the turbine (good), or does it impose some arbitrary ceiling close to turbine height, and assume no mixing above it (bad)?
  • Does it take account of a mix of onshore and offshore wind ?
  • Do any extrapolations of deployed wind copy existing patterns (bad), or model deployment based on realistic forecasts of planned deployment (good)?


  1. Investigating the Effect of Large Wind Farms on Energy in the Atmosphere Magdalena R. V. Sta. Maria and Mark Z. Jacobson, DOI:10.3390/en20400816
  • 2
    $\begingroup$ +1 for a rare answer from someone who reads the literature on the particular subject. I read an article about this a while ago, which seems to simulate along the lines of your #2, earth-syst-dynam.net/2/1/2011/esd-2-1-2011.html. Your #3 point seems would be accomplished by simple algebra, but you should understand that radiative forcing from emissions is over $10^4$ more potent that the direct warming effects of power plants. The CO2 from concrete going into nuclear plants warms the planet more than its waste heat, no? $\endgroup$ – Alan Rominger Jun 18 '12 at 14:54
  • $\begingroup$ re #3: Yes: I was thinking of weather, rather than climate, (as per the OP): weather in the immediate vicinity of thermal plant. $\endgroup$ – 410 gone Jun 18 '12 at 15:12
  • $\begingroup$ re the paper you give: Miller et al, and Kirk-Davidoff are rather notorious for modelling turbulence just by altering the surface roughness length. It's a very coarse method that eliminates all of the information of micro-scale modelling, and their conclusions are as a result, hmmm how shall I say this nicely, speculative at best. See e.g. this and this. $\endgroup$ – 410 gone Jun 18 '12 at 15:21
  • $\begingroup$ Great reply. What is your background, EnergyNumbers? $\endgroup$ – Raskolnikov Jun 18 '12 at 21:39

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.