Fact check: Harvard study misses real-world facts about wind power
A recent study errs in its assessment of potential wind energy resources by ignoring real-world data and experience and instead relying on crude theoretical modeling techniques. In reality, wind project developers and investors work closely with atmospheric scientists and other experts to make sure that their projects will produce as much as expected, and real-world data from large-scale wind installations in the US and Europe confirms that they do. Regardless of who is correct, the inescapable fact is that America's developable wind energy resources are many times greater than our country's energy needs.
Real-world data confirms large wind installations live up to expectations
The question posed by this research has already been answered with real-world experience, and the answer is that wind works.
While the study is interesting from a theoretical perspective, it wasn’t necessary to use models when real-world data is available from large-scale wind installations that have been operating and producing as expected for some time.
In areas like Sweetwater, Texas; the Columbia River Gorge in the Pacific Northwest, and Buffalo Ridge in Minnesota, thousands of utility-scale wind turbines have been installed in relatively densely clustered areas, and wind energy output has not suffered significantly.
As documented in great detail on pages 41-46 here, the output of wind projects in these and other areas has remained high, despite temporary transmission constraints causing curtailment of some wind generation and also causing wind developers to build in locations with lower wind speeds. As long-needed grid upgrades are completed over the next several years, wind project capacity factors are likely to continue their long-term growth.
Experts already take wind farm interactions into account
For a typical wind project, hundreds of millions of dollars in revenue hinge on accurately predicting what the wind output of a potential project will be. As a result, wind project developers and investors work extensively with atmospheric scientists and other experts to develop very precise estimates of wind project output.
In fact, the American Wind Energy Association (AWEA) hosts a major conference every year devoted entirely to the topic of wind project resource assessment, and it is attended by hundreds of experts who are working on wind project resource assessment. Concerns about some wind turbines potentially reducing the output of other turbines are analyzed carefully in the wind resource assessment process before the project is built, and any concerns are typically addressed by choosing a different placement for the wind turbines.
The wind industry continues to move to larger wind projects, driven by the economies of scale associated with larger wind projects. Would wind project developers and their investors be risking hundreds of millions of dollars if their experts thought there was any chance that these large wind projects would suffer from significantly lower wind output?
Other analyses and real-world experience confirm wind is an important clean energy solution
Other theoretical assessments of global wind energy resources have accounted for wind turbines potentially reducing wind speeds at other turbines, and still found that there are sufficient wind energy resources to meet humanity’s energy needs many times over. The new paper cites two such analyses (see http://www.pnas.org/content/109/39/15679.abstract and http://fsi.stanford.edu/publications/geophysical_limits_to_global_wind_power).
A potential source of error in the new paper is that its atmospheric model is based on coarse grid cells that measure 10 km by 10 km, which the paper notes tends to underestimate actual wind plant output by: 1) Ignoring that wind turbines will be sited at optimal locations within that 10 km by 10 km cell, and 2) Calculating the sum of the wind energy production for the cell based simply on the average wind speed for the entire cell, when in reality the sum of the wind energy production for the cell would be higher--because wind output is proportional to the cube of wind speed, wind turbines exposed to above-average wind speeds would produce disproportionately more than turbines exposed to below average wind speeds.
As a hypothetical example to illustrate this second factor:
Their method: 8 meters/second average wind speed = 8cubed = 512 energy output
More realistic: Half of turbines at 6 meters/second, half of turbines at 10 meters/second = (6cubed + 10cubed)/2 = 608 energy output
Even if the analysis in the new paper is correct and the hundreds of experts and real-world data are not, there would still be enough wind energy resources to meet all of humanity’s energy needs. Analysis by the Department of Energy and National Renewable Energy Laboratory conservatively calculates that the U.S. has enough developable, high-quality wind energy resources to meet U.S. electricity needs more than a dozen times over. Even if this conservative calculation were actually optimistic by a factor of several times over, U.S. wind energy resources would still be sufficient to meet all of our energy needs.
European countries like Denmark, Germany, Spain, Portugal, and Ireland now obtain 10-30% of their electricity from wind energy, without experiencing significant declines in wind output. Most of these countries began with wind energy resources that are 20-50% less productive than the average wind resources in the U.S., and with total wind resources that are dozens if not hundreds of times smaller than the total wind resources in the U.S. These countries currently obtain 3-8 times more of their electricity from wind than the U.S. does, and they are in the process of developing new wind projects that will push their wind use even higher. With the far superior wind resources in the U.S., the sky is the limit.
Readers should be careful not to draw erroneous conclusions from the paper’s analysis of how massive deployments of wind turbines could slightly affect the mixing of air in the atmosphere. The paper makes it very clear that this localized movement of air “is very different than warming due to greenhouse gases, in that the warming is primarily local, depends on the stability of the atmosphere, and has a finite limit locally in magnitude due to the depth of mixing occurring.”
Most importantly, readers should not confuse the simple act of moving air around, which has no impact on the heat balance of the Earth, with global warming that is caused by the introduction of long-lasting greenhouse gases that continually alter the Earth’s energy balance. It is also important to keep in perspective that nearly all human activities, such as planting crops, building cities, managing forests, and even operating nuclear power plants, can have localized impacts on movements of air.
Regarding land use, DOE has conducted detailed analysis that found that obtaining 20% of U.S. electricity from wind energy would use less land than is currently occupied by the city of Anchorage, Alaska. In fact, the report noted that the 100,000 to 250,000 hectares of land that would be used to obtain 20% of our electricity from wind is significantly less than the 400,000 hectares of new land that is disturbed every year in the U.S. to mine coal for electricity production.
Fact check: Flawed science journalism on wind energy, April 30, 2012