Editor's note: The following essay is reprinted with permission from The Conversation, an online publication covering the latest research.
Wind turbines take energy from the atmosphere and turn it into electricity: so we know they must have some impact on the atmosphere’s flow. With industrial grade turbines being built at a terrific rate, scientists have been trying to assess exactly what the effects are both at local and at global levels.
New research using a finely resolved model which focuses on Europe indicates that the effects of doubling the present wind capacity by 2020 are at the level of a couple of percent changes in rain and a few tenths of a degree of temperature. The study by Vautard and colleagues shows that these effects include a mix of more rain and less rain, and warming and cooling, depending on where you are around Europe. But a key policy question left unresolved by this paper is: to what should these changes be compared?
There are two main ways of estimating the climate impact of wind turbines. One is to compare the real climate of a region before and after wind farms are installed there. This approach is limited by the availability of observations – modern wind farms are a relatively new phenomenon so there is not much data to compare the climate after wind farms are installed with the years before. Ruling out other reasons for the climate changes we observe is also a challenge.
Using this observational approach, researchers have found that the climate around a large wind farm in Texas was affected by the presence of the turbines. Taking the ground temperatures measured by satellites, they detected a warming of 0.5°C at night in the region directly under the farm. This warming effect was local and small – the nighttime warming did not extend beyond the farm’s immediate neighborhood.
But not many people live right among the wind turbines. To look at smaller effects far from the wind farms, another approach is to simulate the earth’s climate with computer models. Using climate history data, projections are made that account for motion, thermodynamics, chemistry and radiation in the atmosphere, land and sea. Years of simulations are run with and without wind farms installed.
This method has the advantage of being able to detect very small climate change signals, since the model can be run for a long time, and natural variability gets smaller as you average over more and more years. Also, all factors besides wind that might cause climate changes can be held constant. The disadvantage is that the model of climate behavior may not correspond exactly to what happens in reality.
This approach was first attempted by David Keith and colleagues in 2004. When they compared the climate in their model with and without extremely large wind farms (large enough to generate about twice the world’s total present electrical demand), they found that in addition to climate effects in the immediate vicinity of the wind farms, there were changes in climate all around the world.
There were regions of warming and cooling of about 0.5°C, and increases and decreases in precipitation by a few percent. A followup paper showed that these changes were mostly a result of changes in wind direction caused by the wind turbines. The model winds tended to shy away from the wind farms a little, so that downstream from the farms there would be regions of extra wind from the south, that would tend to be warmer, and regions of extra wind from the north, that would tend to be colder.
Vautard’s study agrees with this earlier work, in finding that the climate impacts of wind farms extend beyond the farms themselves and are caused by changes in the flow of the atmosphere that bring warming and cooling to different regions around the wind farms.
What’s the right comparison?
So how should we understand these results? Keith and colleagues compared the climate changes caused by wind turbines with the climate changes that would have been caused by fossil fuel burned to make the same amount of electricity. But because carbon dioxide lasts for centuries or millennia in the atmosphere (unless we suck it out somehow), it’s hard to know over how many years to calculate the avoided carbon dioxide emissions. For the 20-year life of the turbines? Or for the few hundred years of supply of fossil fuels that could be burnt?
Now that the impacts of wind turbines on climate are becoming better understood, more comprehensive studies of complete future energy systems are needed. We need to ask, what combination of wind power, solar power, nuclear power and fossil fuel power, together with what combination of measures to remove carbon from the atmosphere, will result in the lowest overall environmental and social costs.
Daniel Kirk-Davidoff works for a company (MDA Information Systems LLC) that provides wind generation forecasts for energy industry clients including wind farm owners.