Wind Dreams
The optimal amount of practical wind power in the global energy mix is greater than zero. It is also much less than 100%. Today I argue why the proportion of wind power in the global electricity generation mix is always going to be closer to zero than to 100%. That doesn’t mean that wind power is not of value or useful, but it does mean that wind power is not going to drive a global energy transformation, or even be a big part of any such transformation. The sooner we realize that, the better for energy and climate policies.
This post gives explains three reasons why wind will always be niche — low density, low capacity, the age effect — and why costs are not among those reasons.
Wind Power is Low Density
The figure above comes from a recent study by a research group in Norway of annual energy density (measured as terawatt-hours per square kilometer) for 11 different technologies of electricity production.
A first thing to notice is that the top panel — with a linear scale — shows that wind power, either onshore or offshore, does not even appear on the graph when scaled against nuclear and natural gas.
Onshore wind requires about ~370x the land area for the production of the equivalent amount of energy as nuclear (offshore requires ~200x the area of nuclear). According to this study, to meet today’s total global electricity demands using wind would require land area equal to about Brazil — or about 3% of all global land area (and this included places like Antarctica, the Sahara, and the Amazon).
The point here is not that the world would ever try to meet all of its need with wind — of course not. But the implied land (or sea) requirements alone necessary to meet any significant portion of global energy demand imply that wind can only be a niche technology. Wind power makes sense where there is ample space for deployment, ready connections to where power is needed, and integrates well with other technologies, especially necessary back-up generation.
Wind Capacity Factors are Limited
The wind does not always blow and when it does blow, turbines are not always in operation. The proportion of energy generated by a wind turbine, as compared to its nameplate capacity is called its “capacity factor.” Over time, with the extensive deployment of wind technologies around the world, capacity factors have not increased much at all, even as wind technologies have improved.
The four-paneled figure below shows different ways to look at wind energy capacity and capacity factors from 2000 to 2020.
The data show that the even as U.S. wind capacity expanded dramatically over 2000 to 2020, overall capacity factors remained fairly constant. More recent data shows that in 2023 wind capacity factors declined to an 8-year low, and early data from EIA for 2024 shows further year-on-year declines.
The United States is also not representative of wind power deployments around the world. The table below — sorted by onshore capacity factors, high to low — from Xu et al. 2023, shows that the United States is among the world’s best performers for wind capacity, both onshore and offshore.
The low capacity factors of wind necessarily mean that wind power needs back up. That back up might include:
Higher capacity energy technologies, like nuclear or natural gas;
Energy storage, such as in batteries or pumped hydro;
Overbuilding of wind technology and easy transport of electricity from place to place.
These back-up options all raise issues — With nuclear, wind is not needed, with gas, considerable carbon dioxide emissions may still result even if wind displaces some of them, energy storage at the massive scale needed are not presently possible, and overbuilding wind and the grid seems fanciful, given the massive deployment challenges.
Once again, these numbers suggest that while wind has a future role in the global energy mix, that future will remain niche.
Wind Power Technologies Age Quickly
The figure below, also from Xu et al. 2023, shows that deployed wind turbines age quickly in terms of capacity factors that start decreasing at most over a few years after initial deployment.
Nuclear and natural gas power plants of course also have aging and maintenance issues, but have demonstrated nothing like the longer-term declines in capacity shown by wind turbines. The aging effect means that wind turbines may need greater back-up (than the prodigious amount already needed), need to be replaced frequently, and require extensive upkeep and maintenance.
Show me an entire U.S. state or a medium-sized country powered by wind without the issues raised above about necessary back-up, and I’ll be happy reconsider these concerns.
Notice I Haven’t Even Raised Costs
That is right. In my view, costs are not the limiting factor for the expansion of wind beyond its niche. Wind power could be completely free and it would still be low density, low capacity, and experience age effects.
Think of this analogy — walking from place to place is pretty much free. But is is also low density (i.e., it takes a long time compared to alternatives) and it is low capacity (e.g., it is difficult to ship goods by walking). Speaking for myself, there is also an age effect.
Does walking have a role to play in global movement? Sure it does? Is it capable of assuming a major role in the global energy mix? Unlikely.
Energy realism means prioritizing math, physics, politics (yes), and the societal demand for reliable and accessible electricity (and also of course, economics!). With the world committing to accelerating decarbonization — which I support — real progress can only be made by looking to energy sources that are reliable and accessible, which means more dense, higher capacity, and here for the long term.
That is not wind energy.
❤️Click the heart if you think that wind blows!
Thanks for reading! I welcome your critique, comments, counter-arguments, and pointers to other research which may contradict those presented here. THB is a reader-supported project — Please consider supporting with a subscription! Learn more about THB here.
"Nonsense, you're guessing."
I'm not guessing. I'm looking at current and likely future trends in PV technologies (particularly the trend towards roll-to-roll production of perovskites) and current and likely future trends in automotive and battery technologies.
Autonomous vehicles will promote mobility-as-a-service (MaaS), and use of electric vehicles by fleet owners, who will then integrate the batteries in their vehicles (including vehicles whose charge level has dropped below 70-80%) with the grid, for the benefit of the fleet owners and the grid. (The fleet owners will benefit from charging their vehicles when there is a surplus of PV electricity, and discharging to the grid when electricity is needed by the grid, and therefore prices are higher.)
I called this accurately ***more than a decade ago*** on my blog. It's a simple fact, easily verified:
https://markbahner.typepad.com/random_thoughts/2013/01/the-future-of-transportation.html
From that blog post:
"January 18, 2013
The Future of Transportation
"I think artificial intelligence is going to have a huge impact on transportation (and everything) in the next few decades.
"My guesstimate for time-frames are:
1) Early 2020s: First fully automated computer-driven cars.
2) Early 2030s: Virtually all new cars are computer-driven.
3) Early 2040s: All vehicles on the road are computer-driven."
I then predicted 11 likely outcome from all vehicles on the road being computer-driven by the early 2040s. You can read them all, but here are a few:
"2) People will obtain cars like we now get airline tickets, except the cars will often be ordered hours or even minutes in advance, rather than months or weeks in advance. One will state one's destination and time of departure into a smart phone or personal assistant, and options will come up, with different prices and different arrival times. Non-stop short arrival times will be more expensive, and multi-stop trips with long arrival times will be less expensive. Also, fancier and less-fuel-efficient cars will be more expensive, and plainer, more-fuel-efficienct cars will be less expensive. The car will take you from the door at one location, to the door at the other location. No parking in parking lots."
"4) Many cars will become incredibly small. Even Smart Cars and Cooper Minis have room for two passengers. In the 2040s, single-seat cars will be extremely common. They will likely be powered by batteries, rather than gasoline. This is because there will be no need for "range anxiety". That is, no car will come to you unless it knows it has enough battery life to get you to your destination and then to a charging station."
"6) Transportation will be much more electrified. Virtually all short-range trips will be by battery-power. Total U.S. gasoline usage would cut by more than half."
"8) Traffic congestion would be completely eliminated. Instead of traffic lights or stop signs, cars at intersections would simply pass within feet of one another at right angles, at full speed, because the computer controls between the cars would communicate which car crossed first. Speeds on freeways, even within cities, would be 70+ mph, at all times. Cars and buses would travel at these speeds within feet of one another."
I also came very close to accurately predicting global CO2 emissions throughout the 21st century...***way back in January 2005*** when I predicted that global CO2 emissions would peak in 2030, would decline by about 5 percent from the 2030 peak by 2050, and more than 20 percent from the peak by 2070. My peak was a little low (only 8.8 gigatons as carbon, or GtC, in 2030). And my predicted rate of decline from that peak was probably too shallow, especially after 2050.
So overall, my predicted total global CO2 emissions from industry (i.e. not counting land use emissions) during the 21st century was 712 GtC, which I think will end up VERY close to the actual value, because my prediction is between the RCP 4.5 prediction of 828 GtC, and the RCP 2.6 prediction of 437 GtC...and closer to the RCP 4.5 value. Again, this is a matter of record:
https://markbahner.typepad.com/random_thoughts/2017/07/mark-bahner-vs-wigley-and-raper-science-2001-vs-ipcc-rcps.html
"Been living off the electrical engineering skillset for the last 20 years."
And did any of that "living off the electrical engineering skillset" involve doing anything that would lead you to any particular expertise in matters related to the U.S. electrical grid, and technologies involved therein, now and in the future?
For example, work for GE-Hitachi or Westinghouse? An electric utility? An RTO? FERC? EEI?