July 29, 2023: Wind Power
Good afternoon. Today we are going to take a look at some of the environmental issues surrounding wind power. There are no environmentally benign industries, and wind power is no exception. Some of the concerns include land use, embodied impacts from the materials that wind turbines require, security surrounding critical minerals, impacts on wildlife, end-of-life treatment for turbines, health effects, and visual disamenity. We will take a brief look at each of these.
Land Use
There have been quite a few studies assessing the land requirements for various energy systems, such as this one and this one. The latter, a recent study whose coauthors include several former colleagues at The Breakthrough Institute, is one I find to be most helpful.
The two entries for wind power, Wind- and Wind+, differ in how land use is accounted for. For the smaller value, Wind-, a footprint metric is used, which assesses the area occupied by the turbine bases and access roads in wind farms. For Wind+ values, land use is the entire perimeter of the wind site. The natural gas values, NG- and NG+, are distinguished in a similar manner. The median figures are 130 ha/TWh/yr for Wind- and 12,000 ha/TWh/year for Wind+.
The paper goes on to estimate the total land use requirements for various future energy scenarios. In all scenarios, the land use for electricity increases, and the authors assess that a 3-5-fold increase for total electricity generation may be needed under the electrification envisioned in a low-carbon energy system. Across scenarios, land use for electricity ranges from about 160 million hectares to 900 million hectares under the more expansive (spacing) accounting. If world electricity demand grows by a factor of 4 by midcentury, and that demand were to be met entirely by wind power, the land use would be about 15 million hectares under Wind- and 1.4 billion hectares under Wind+. These number contrast to about 5 billion hectares for agriculture and 130 million hectares of urbanized land at present.
Wind- is a more useful figure for understanding environmental impacts, as wind farms can coexist with other land uses or wildlife. Wind+ is a more useful figure if we are assessing potential for generation or the risk of land use conflicts. In any event, we observe that land use, while nontrivial, is not a showstopper for wind power under plausible scenarios.
The aforementioned paper, like all others on the subject that I am aware of, offers the caveat,
Our LUIE calculations do not include land that is occupied by the upstream manufacturing of electricity generating facilities (e.g., the land required to mine materials for solar panel or wind turbine production, or the materials that go into nuclear or coal power plants). We also exclude land required for electricity transmission infrastructure (e.g., high voltage transmission corridors), offshore area impacts (for wind farms and natural gas drilling), and underground impacts for geothermal, natural gas, and coal mining).
This is an unfortunate limitation. I do not have a good sense the land use impacts for mining, and I would very much like to. Lifecycle greenhouse gas emissions generally do take into account such upstream impacts. This paper from 2012 states that mining and quarrying occupy 40 million hectares (this recent paper gives a figure closer to 10 million hectares, but I am not sure how comprehensive it is). Most of that is not for the energy system, but certain minerals will see their demand vastly increase under a high-renewable energy system. If we take 40 million hectares as the amount of land, within an order of magnitude, needed for the mining to support renewable energy, then again it is a nontrivial but manageable impact.
This paper confirms findings from previous studies that nuclear power performs better, by a wide margin, than other electricity sources for land use. They find the median LUIE for nuclear is 7.1 ha/TWh/yr. While upstream impacts were not generally assessed, they find an additional 0.08 ha/TWh/yr for uranium mining and processing and 0.012 to 2.9 ha/TWh/yr for the now-cancelled Yucca Mountain waste repository. There is also an additional 3.9 ha/TWh/yr for the exclusion zones around Chernobyl and Fukushima, an impact that almost all nuclear plants would not require. Nuclear continues to perform very well when these upstream impacts are included. The bottom line is that if we are really concerned about land use, then agriculture, and in particular greenhouses and meat consumption, rather than renewable energy, would be the first place to look.
Embodied Impacts
Wind energy uses “a lot” of material, from which carbon dioxide is emitted in the manufacturing process. But as the saying goes, “a lot” is not a number. Lifecycle greenhouse gas emissions of electricity generation has been estimated as follows.
I don’t know what more to say about this. The differences between solar (PV and CSP), wind, geothermal, and nuclear energy are trivial in comparison to the differences between each of these and the fossil fuels: gas, oil, and coal. Yes, it is true that renewable energy generally requires more material than fossil fuels and nuclear power. But the impact of greater material needs is dwarfed by the impact of fossil fuel combustion.
There are other impacts of upstream mining to consider, such as water pollution (e.g. eutrophication), which I have not had a chance to look at.
Mineral Security
Since the energy crises of the 1970s, energy security has been a watchword and a motivator of energy policy. Would a switch to renewable energy merely swap energy security vulnerability for mineral security vulnerability? According to the IEA report on critical minerals, which I have discussed many times, 71% of world platinum reserves are in South Africa, 69% of cobalt reserves are in the Democratic Republic of the Congo, 63% of graphite reserves are in China, 60% of rare Earth mineral reserves are in China, and 52% of lithium reserves are in Australia. Each of these minerals (or classes of minerals as with rare Earths) is vital for renewable energy and/or battery storage.
One thing to keep in mind is that a sudden loss of critical minerals, unlike a sudden loss of natural gas, oil, or coal, would not have an immediate impact on energy availability. There would be much more time for an importing country to replace the source and mitigate the disruption. Two research papers assess that the risk of conflict over critical minerals would be much less than the risk of conflict over fossil fuels.
Geologic availability of supply is a concern as well, but not a well-founded concern in my view. Another recent paper, also including some former Breakthrough Institute colleagues, finds that the supply of minerals for renewable energy should be sufficient under plausible scenarios.
John Cochrane, critical of electric vehicle subsidies from a free market perspective, argues that resource availability might be a concern despite full awareness of how similar past prognostications of resource shortages have been far off the mark. He argues semi-plausibly (italics in the original),
Now any economist of a certain era like me remembers the resource limits arguments of the 1970s, and the fact that raw materials prices keep going down over time. How does this square with the above? Simple: over time energy prices go down, and a bit the cost of machinery goes down, so we can mine lower and lower quality ore at the same or lower price. This is how we overcome the laws of thermodynamics. As the thing you want is more and more needle in a haystack of rubble, it takes more and more energy to pull it out. We could make copper from sea water with enough free energy. We could make copper from other elements in particle accelerators with enough free energy. Energy abundance drives declining resource prices. And, of course, current policy is driving us to more and more expensive energy. So perhaps the resource limited crowd might be right after all, though only because of our self-imposed energy policies.
There are many good reasons to question the wisdom of EV subsidies, but I don’t think this is one of them.
Wildlife
What about birds and bats? The idea that wind turbines are a leading killer of birds is a myth. Orders of magnitude more birds die from cat predation, window collision, and car collision than as a result of wind turbines. Though it should be noted that wind turbines are disproportionately dangerous to larger birds of prey. On the other hand, emerging turbine technology can mitigate the problem.
End-of-life
When wind turbines reach the end of their lifespan in large numbers, won’t this cause a big problem? This issue too is, if you will pardon the pun, overblown. An estimate from a couple years ago is that turbine blade recycling should require no more than 2% of the energy that the turbine generates over the course of its life. The same paper estimates that by 2050, if turbine blades are landfilled, this should require 1% of remaining landfill capacity by volume and 0.2% of remaining capacity by mass. And “remaining landfill capacity” is by no means the limit if new landfills are commissioned.
Health and Aesthetics
“Wind turbine syndrome” refers to a cluster of alleged negative health impacts resulting from living close to wind turbines. There is no evidence to suggest that this effect is real, though there could be a “nocebo” effect; the opposite of a placebo, a nocebo is a situation in which the expectation of an adverse health effect causes that effect. This review considers infrasound (low frequently sound waves) and shadow flickers and finds no evidence of negative health impacts. Wind turbine syndrome should be placed in the same category as the MMR vaccine/autism thing or the notion that cell phones cause cancer.
Aesthetics are harder to judge objectively, though we can try. Some people will complain about the noise from turbines and that it causes annoyance. For people living near turbines, road noise is still greater than turbine noise, though maybe the low-frequency noise from turbines make them disproportionately annoying. Some people complain about appearance as well. A negative effect on the sales price of property near turbines has been observed, though this effect is inconsistent. In a review of electricity technology from a few years ago, Samadi finds a visual disamenity cost of wind turbines of 0.5 cents/kWh, which to me seems high, but that’s what we have (retail price of electricity in the United States is now around 12 cents/kWh). This paper finds a high cost in Germany, a median of around 6-7 cents/kWh, which is comparable to generation cost.
Conclusion
A review of the negative issues around wind power reveals that they are nontrivial but manageable. Aesthetic costs are hard to pin down rigorously but might be the most significant. One wonders though if these costs also reflect a “nocebo” effect as with health impacts. Critics of wind power do a great disservice when they make specious claims such as that there are significant health hazards, that they are a leading cause of bird mortality, or that mineral availability and end-of-life treatments are show-stoppers.
I don’t like to single out individuals who I know personally for criticism, and so I will refrain from doing so, but there are a few prominent advocates of nuclear power that think that a good way to advance their cause is to attack wind and solar power with bogus claims. This is especially distressing because nuclear energy has been greatly hampered by equally specious claims. Fearmongering is a negative-sum game. I see no evidence that anti-wind NIMBYism has helped nuclear plants be built; instead, it contributes to the political dysfunction that makes it difficult to build anything at all.
Quick Hits
Taking advantage of the interest surrounding the new movie, Cold Fusion published a video on Robert Oppenheimer. In it, he makes the claim that by August 1945, Japan was planning a biological attack on California in a desperate attempt to turn the war around, that U.S. intelligence was aware that this attack was being planned, and that knowledge of the impending attack was a consideration in the decision to use atomic bombs. Japan’s notorious Unit 731 managed their biological weapons program, and their atrocities rank with the things the Nazis did. They had indeed been planning a biological attack called Operation Cherry Blossoms at Night. The plan was nixed in early 1945, but revived as the situation got desperate for Japan and was planned for September 22, 1945. Whether the plan would have been executed, or whether this form of biological warfare would have been effective, are unclear. I haven’t been able to find any other discussion about whether the U.S. knew about the plan, and if so, if it influenced the decision to use the atomic bombs. If any readers can point me to information on the subject, I would appreciate it.
There is a new claim this week of a room temperature superconductor. While such an innovation would have many useful applications, most of the press’ reaction has been skeptical because many similar claims have come and gone over the years. One of them is from the researcher Ranga Dias, which was retracted. Dias later made a second claim which is in doubt. There was fraud from the German physicist Jan Hendrik Schön, who claimed, among other things, to have developed a superconductor with organic materials. All his claims were withdrawn. There was another claim in 2018, around which there are allegations (not proven as far as I know) of fraud. The same researchers released more data in 2019, and I am not aware of any activity since then.
There was a coup in Niger this week. Niger’s former president, Mohamed Bazoum, was the last pro-Western leader in West Africa after others had been deposed in coups. In their recent review of Salafi-jihadi activity, the Institute for the Study of War discusses how this event will be a setback for Western antiterrorist activity. Among the people happy about this are Yevgeny Prigozhin, who himself attempted to launch a coup in Russia last month. Prigozhin, according to the article, sees the coup as an opportunity to sell Wagner services in Niger.
In 2019, a now-notorious report from John Hopkins University (there is a 2021 edition too) assessed various countries’ preparedness for a pandemic. There is little correlation between the countries the report assessed were well prepared and the countries that had a low death toll from COVID-19. Last year, Omberg and Tabarrok assessed which factors did correlate; the only thing they found is that countries with a high general state capacity tended to have lower death tolls. See Tim Harford’s blog post for more on the subject. There are surely lessons for other groups concerned about risks from artificial intelligence or nuclear war.
Western Europe is endangering their security by continuing to neglect development and production of long-range missiles.
I find the Hudson Institute to be valuable resource on foreign policy. They recently discussed Xi Jinping’s ambitions in the Middle East and the challenge this poses to American foreign policy.