US wind power reserves just trebled

Says a government lab, using conservative assumptions.

The estimate of potential US wind power has for a long time been based on 20-year-old research whose cautious assumptions about future technical progress have been outstripped by events. A 1991 study I cited in a  2007 post set the economic wind power potential of the 48 contiguous states at 3.7 terawatts, or 1.1 terawatts continuous equivalent based on a 30% load factor. A conservative new study from the National Renewable Energy Laboratory, based on raising the height of towers from 50 to the current standard 80 meters, increases the estimate to 10.5 terawatts installed or 3 terawatts continuous equivalent. It does not assume any breakthroughs such as kite ladders. [Update: there’s even more energy at 100m or 150m, which can be captured by bigger conventional rotors, but it becomes impracticable to move the blades by road; at sea, you float them out, so offshore turbines can be bigger. The 80m band is sensible onshore.]

Here is a screengrab of their map (the original is interactive and lets you get more detail by state) :

NERL updated map of US wind resource at 80m height


All areas from from orange to purple are good for wind. Roughly speaking, anywhere on the
Great Plains, a third of the continental United States, the siting problem for a wind turbine can be solved by walking out of the front door.

The NREL actually counted potential output in annual terawatt-hours. Though the continuous equivalent could be misleading – no source of energy is actually available without interruption, and wind is more variable than most – it is the more relevant number for a thought experiment. 3 terawatts continuous is over six times current US electricity consumption, which I make out to be 440 GW continuous equivalent (3,872 GW-hours/year / 8760 hours).

At which point some armchair genius will point out that wind power is intermittent, as if the fact had not occurred to the NREL, General Electric, the Danish government, Iberdrola and T. Boone Pickens. Yes, you will of course need to complement wind (and for that matter solar) electricity with some fossil or nuclear capacity – which you will be relegating to intermittent peak loppers, as wind electricity has strictly zero marginal cost – , and/or a lot of storage, and/or fancy load-shedding and time-shifting by consumers like you and me.  Another recent NREL study found that you could incorporate 20% wind into the East Coast electricity supply without appealing to technologies more sophisticated than an expanded transmission grid, and at a cost of under 0.5 cent a kilowatt/hr. The ultimate practical constraint on the share of intermittent renewables with foreseeable new technologies must be quite a lot higher. 100% is technically feasible simply by adding enough storage.

So, Dr. Chu: where is that nice new interstate grid?

Author: James Wimberley

James Wimberley (b. 1946, an Englishman raised in the Channel Islands. three adult children) is a former career international bureaucrat with the Council of Europe in Strasbourg. His main achievements there were the Lisbon Convention on recognition of qualifications and the Kosovo law on school education. He retired in 2006 to a little white house in Andalucia, His first wife Patricia Morris died in 2009 after a long illness. He remarried in 2011. to the former Brazilian TV actress Lu Mendonça. The cat overlords are now three. I suppose I've been invited to join real scholars on the list because my skills, acquired in a decade of technical assistance work in eastern Europe, include being able to ask faux-naïf questions like the exotic Persians and Chinese of eighteenth-century philosophical fiction. So I'm quite comfortable in the role of country-cousin blogger with a European perspective. The other specialised skill I learnt was making toasts with a moral in the course of drunken Caucasian banquets. I'm open to expenses-paid offers to retell Noah the great Armenian and Columbus, the orange, and university reform in Georgia. James Wimberley's occasional publications on the web

32 thoughts on “US wind power reserves just trebled”

  1. Note that if we had a lot more electrical generating capacity, we would want to convert a lot of existing fossil fuel-based energy use — gasoline for cars, heating oil for homes, etc. — to electric-based where possible.

  2. Or, if the grid has too much spillage from transmitting energy from the Empty Quarter to where people live, the Plains can become host to wind-powered ammonia factories, combining surface water, nitrogen in the air, and electricity into a transportable, burnable fuel.

  3. Current installed electric generating capacity is about 1 TW; a big power plant is about 1 GW. So 3 TW is a big number; the question is how much of that is economically feasible given the costs of transporting energy.

  4. > Yes, you will of course need to complement wind (and for that matter solar)

    > electricity with some fossil or nuclear capacity – which you will be

    > relegating to intermittent peak loppers, as wind electricity has

    > strictly zero marginal cost – , and/or a lot of storage

    There is a lot more to it than that. Start with the fact that in almost all of continental North America, wind energy is usable between 0200 and 0700 hours, and between 1800 and 2100 hours. You'll note that the entire period of peak usage doesn't fall into those buckets.

    Then be honest about that fact that other than pumped hydro, which has its own ecological problems, there is NO usable method of storing bulk electrical energy.

    Cranky

  5. Wind power isn't about being honest. It's about feeling good by doing something economically stupid. (Outside of a few very exceptional areas where the wind blows almost all the time.)

    Every KW of installed wind power requires backup by another, more reliable, (And generally cheaper!) source of power. Until wind becomes much cheaper than those other reliable power sources, the most economical thing to do is just skip building the windmills, and run the 'backup' plant full time.

    Zero marginal cost. What BS. Only if you flunk economics.

  6. The cost of Coal may be cheaper, depending on what cost you give to emitting CO2 into the atmosphere.

  7. But I'm not in favor of building coal plants. I think we should be building nuke plants. Lots and lots of nuke plants. Even if I have some doubts about how firmly grounded global warming really is, coal is really dirty, at all stages, kills a lot of miners, ruins the landscape… It really IS an all around nasty source of energy.

    Anyway, I think we all know what really tripled US wind capacity; Senator Kennedy croaked.

  8. OK Brett, I'll bite. What is wrong with my assertion about marginal cost? Rational grid managers ideally decide which power stations should be switched on or off based on a merit order reflecting their short-run (nearly instantaneous) system marginal cost. (Contracts may force them to make less efficient choices.) Since wind turbines rotate all the time, so even wear and tear is constant, my assertion that this cost is zero (as near as dammit) must be correct at the point of production. The short-run system marginal costs of transmission are simply power loss, proportional to distance, so wind may not always be at the top of the merit order everywhere – I don't think this qualification is that important; 10% of zero is zero. Nuclear power stations burn fuel so their short-run marginal cost, though very low, is a little higher: perhaps there are operational reasons for running them all the time, I wouldn't know. The important comparison however is with fossil power. Since the bulk of the current US capacity is fossil, the backup is already in place and does not need to be paid for. Shifting the first-choice load to renewables makes these plants less remunerative; tough luck, sunk costs are sunk.

    I love Cranky Observer's factoids on wind's work-to-rule in Garrison Keillor country. Pumped hydro is a feasible existing technology, like it or not, and there are plenty of valleys in the Appalachians. People are working on underground compressed air, hot salts (currently for solar), fuel cells, and mega-batteries. My observation was about technical not economic feasibility.

  9. "OK Brett, I’ll bite. What is wrong with my assertion about marginal cost?"

    Well, as a trivial matter, it could only be true if wind turbines weren't subject to wear, and didn't require maintainance.

    "Since wind turbines rotate all the time, so even wear and tear is constant,…"

    Speaking as an actual engineer, that claim strikes me as so obviously false as to be absurd. Let us leave aside the fact that the wind actually stops at times, so the only way the wind turbines could be turning all the time is if you ran them backwards as motors when the wind wasn't blowing. It is a simple fact that, in order to generate power, a wind turbine must resist the force of the wind. Work equals force through distance, after all… This means that a wind turbine, even under exactly the same wind conditions, will be loading it's bearings, and all other elements of it's structure, more when generating power, than when it's not. And loaded mechanisms wear faster than unloaded mechanisms. As they approach their design limits, much faster.

    I suppose it's just barely possible that wind turbines are, in practice, overbuilt to such an insane degree that their useful lifetime is dictated by rare accidents instead of running wear. This would be a different form of economic insanity.

    But my point had more to do with the marginal cost windmills introduce by requiring that baseline power plants be used as "intermittent peak loppers". Yes, it DOES cost more to run them that way, than to just keep them running in baseline mode. Without making the plant that's being used that way one bit cheaper, you're artificially depressing it's duty cycle. The capital cost of building it, the staff who must be present, everything about the plant gets amortized over fewer KWHs. They even wear out after fewer KWHs have been produced, because you're not running them all the time in the most efficient mode.

    This all must, realistically, be consider a marginal cost of deciding to feed that wind turbine power into the grid, knowing that it's not reliable.

    "Nuclear power stations burn fuel so their short-run marginal cost, though very low, is a little higher: perhaps there are operational reasons for running them all the time, I wouldn’t know."

    Yeah, actually, there are. The core of a nuclear reactor has huge thermal mass, and the power output can't be turned off on a dime. In the case of a nuclear plant designed for baseline power production, using it as an "intermittent peak lopper" would require throwing away energy you couldn't avoid generating. They're designed this way because so much of the cost of nuclear power is capital expense that it simply doesn't make economic sense to shut one down when it could be producing power.

    The situation is somewhat similar for coal plants, though not so extreme. They can't be throttled as fast as the wind varies, either.

    So, no, for a variety of reasons the marginal cost of wind power is NOT zero.

  10. Maintenance costs: a very technical win for Brett. The Danish Wind Industry Association has this to say on these costs:

    "For newer machines the estimates range around 1.5 to 2 per cent per year of the original turbine investment.

    Most of maintenance cost is a fixed amount per year for the regular service of the turbines, but some people prefer to use a fixed amount per kWh of output in their calculations, usually around 0.01 USD/kWh. The reasoning behind this method is that tear and wear [sic] on the turbine generally increases with increasing production."

    So the marginal wear-and-tear cost is somewhere between nothing and 1 cent per kw/hr. The latter looks to be an accounting practice not a true economic estimate of the fixed/variable breakdown. I should have said "near-zero".

    It is touching that libertarians now think capitalists should consider the losses to their competitors when evaluating investments.

  11. If you're going to force the utilities to shut down their baseline plants to make room for wind power, whenever the wind farms happen to produce it, they're not exactly competitors in any sense a libertarian would recognize.

  12. Read these two reports and then get back to us:

    http://www.nerc.com/files/IVGTF_Report_041609.pdf

    http://www.uwig.org/CEC-500-2007-081-APB.pdf

    In fact, read all the reports on this page and then get back to us:

    http://www.nrel.gov/wind/systemsintegration/relat

    Seriously people, I love wind, I'd love to see a greater percentage of total generation be wind. But to think that (a) building wind turbines is a magic wand that will make the issues of availability and our current utter dependence on coal go away (b) that there aren't serious problems with integrating a variable and stochastic resource into a system with a base expectation of 100.0% reliability and 1.00:1.00 capacity:demand ratio (c) that there aren't a lot of very knowledgeable, very experienced people working on these problems and that they aren't encountering some really knotty problems (d) that some of the big wind investors aren't trying to dump THEIR variability problem onto the rest of the grid at zero cost to themselves, is just silly.

    Cranky

  13. Brett: "If you’re going to force the utilities to shut down their baseline plants to make room for wind power.." Who said anything about forcing? This is just an efficient market at work. Since the wind turbine operators can offer marginal power at any price above 1c a kw/hour, their product should automatically displace fossil plant with higher short-run marginal costs. If there are long-term contracts preventing this, that's a market imperfection justifying government intervention to unwind, given the overwhelming public interest in cutting carbon emissions.

    Cranky: "a system with a base expectation of 100.0% reliability and 1.00:1.00 capacity:demand ratio": if you define problems as insoluble, of course they are. Why should we keep the expectations of the twentieth century in the twenty-first? We don't think it a disaster or even a problem when roads and parks operate at one-fifth capacity; and in the days of sail, ships were often stuck in port for days till the wind changed. I'll side with your "lot of very knowledgeable, very experienced people working on these problems" at the NREL who after have gone on record as stating that these are entirely soluble at the 20% level. Nobody knows how much higher the US can practically go, but luckily the Danes are going to find out for you.

    BTW, can either of you find anybody who has ever seriously proposed a 100% wind electricity supply, as opposed to putting forward the sort of thought experiment in my post? The standard radical green approach to zero-carbon electricity is SFIK a cocktail of different renewables (wind, solar thermal, solar PV, geothermal, biomass, and hydro), plus storage, demand management [update: increased trade] and conservation. Since we know that you could, at possibly substantial inconvenience and excess cost, easily cover current US demand from just one renewable technology, the conversion will be far less costly and inconvenient with an optimised cocktail.

    [Update: FWIW, here's my old back-of-an-envelope proposal for the UK electricity market, to meet an 80% reduction in carbon emissions: 75GW of wind capacity, 25GW of baseload nuclear (against 11 today), keeping the current 29GW of gas as the variable reserve, and scrapping coal and oil.]

  14. > Brett: “If you’re going to force the utilities to shut down their baseline

    > plants to make room for wind power..” Who said anything about forcing? This is

    > just an efficient market at work.

    It takes three days to shut down a baseload coal plant and then bring it back on line according to spec, and in doing so – even if you stay within technical limits – you consume a certain percentage of the plant's lifetime. Depending on size and design, gas turbines can be started in anywhere from 10 minutes to 12 hours, and again depending on design have a higher tolerance for multiple starts, but even so every start consumes a significant percentage of total lifetime. Other than hydro, which is pretty well built-out in North America modulo environmental acceptability, there is no known reasonably-priced, fast start, _dispatchable_, high-turndown-ratio electrical source that is also capable of regulating and providing VAR support. There just isn't, and very smart people have been looking very hard for 50 years to find one.

    Does that mean that (a) there never will be (b) we should stop trying to find one (c) there won't someday be totally unexpected innovations? By no means; I am neither a luddite nor a technical pessimist (not a total pessimist anyway). But the name of this blog is "Same Facts"; it does mean that those commenting on the wind situation need to really dig into the two reports I linked above (as well as other resources, including the docket on this question that FERC opened in February and which has 30 or so comments {Sunflower Energy of Kansas' comment, with wind generation in their service territory equal to 80% of their total load, are particularly interesting}) and really understand the massive engineering challenges that are involved in getting more wind on the continental grid (note that Denmark and Norway are primarily ocean-driven wind, which has a much different daily profile).

    And that's just pure engineering challenges, not even including those found at the boundary of engineering, economics, and politics which are also huge.

    Cranky

  15. By the way, if you are truly proposing to change the reliability and capacity of the US electrical grid so that it resembles India's (8 hours service/day, variable, no guarantees on voltage, frequency, or reactive power) or Myanmar's (service any randomly-selected 4 hours in a day; strict limits on amperage per connection, "what do you mean by voltage and frequency standards?") then you had darn well better get that in front of the US population in a referendum because trying to do it on the sly will result in unbelievable blowback.

    Cranky

  16. Mr. Observer,

    Using coal as peaker plants sounds infeasible. But what about natural gas – it's my understanding that those plant are more suitable for that use, and are less carbon intensive. Likewise, Hydro is often limited by the water in the reservoir. If wind allows that water to be released as needed, that would be good.

    The general point is how much CO2 emission can be displaced by wind power. If wind is most available at night, then that is when we should charge batteries for our cars. Or release less water from the reservoirs at night, more during the day instead of burning coal.

  17. > Using coal as peaker plants sounds infeasible. But what about

    > natural gas – it’s my understanding that those plant are more

    > suitable for that use, and are less carbon intensive.

    Yes, that's what is more-or-less being done; gas turbines are being used to back up variable resources. This is particularly true in Texas which has substantial nominal wind capacity.

    But gas turbines are significantly less efficient (= more expensive) than steam plants (whether those are fired with coal or gas), and to many both in the bulk electric supply industry and the electric dispatch service it doesn't appear that the variable generators[1] is absorbing anything close to the cost of backing up its inability to guarantee reliable dispatch. Just because it is a new market designed according to the best precepts of the Chicago School doesn't mean it does not contain unpriced externalities and hidden subsidies. Then there is the issue of making ourselves more dependent on one fossil fuel (gas) to incorporate a specific renewable (wind). Is there a net gain, and how long can we sustain that path? In part the FERC docket I referenced above is a start at exploring those questions.

    > Likewise, Hydro is often limited by the water in the reservoir. If wind allows that water to be

    > released as needed, that would be good.

    > The general point is how much CO2 emission can be displaced by wind power. If wind

    > is most available at night, then that is when we should charge batteries for our cars.

    > Or release less water from the reservoirs at night, more during the day instead of burning coal.

    Hydro resources have been dispatched that way since 1890, so you are dead on there! There just isn't enough hydro to go around, nor is it close enough to our largest load centers.

    Really, read the two reports I referenced above. They contain some technical language (VAR and "reactive power" are probably new concepts), but they are surprisingly readable with just a high school science background (the economics should be accessible to everyone who reads this blog). The issues are significant, and remain significant despite people who really know what they are doing working on them for quite a while (since 1970 in the case of California).

    Cranky

    [1] Primarily wind, as solar is so far only significant in CA and AZ and has different loading characteristics

  18. "Who said anything about forcing? This is just an efficient market at work. Since the wind turbine operators can offer marginal power at any price above 1c a kw/hour, their product should automatically displace fossil plant with higher short-run marginal costs. "

    It IS a forced purchase. As has been detailed above, having to purchase an unreliable source of power like wind has massive costs for a utility. It's doubtful they'd buy any significant amount of wind power voluntarily even if it were 'free'. Because it's not free once you take into account the effects of having to shut off the power plants they were relying on when the wind suddenly decided to blow.

    I find it hard to believe that anyone with even a passing acquaintance with wind power is unaware that utilities are being FORCED to buy it. It's not an 'efficient' market, it's a compelled market, which wouldn't exist in the absence of the legal compulsion. No legal obligation to buy wind power, no market.

    I would also point out that, while it might make economic sense, once you've got the windmill, to sell the power even if it's only at 1c/kwh, it makes economic sense only in a 'that way we go broke slower' sense. It's only sustainable if they're selling it at the real cost.

  19. The intermittent nature of wind is a small issue that is easily, and ecologically overcome. To say that "there is NO usable method of storing bulk electrical energy" is just wrong. It is old technology.

    http://en.wikipedia.org/wiki/Ludington_Pumped_Sto

    A compliment of pumping stations combined with a compliment of wind and other power sources could greatly increase the percentage of our national electrical grid from sources other than coal or imported fissle materials.

  20. It's not that there's no usable method, it's that there's no economically feasible method. To be economically feasible, the storage end of things has to be so incredibly cheap that the combination of wind and storage is cheaper than alternate sources of energy.

    As things presently stand with wind, that means that the storage mechanism would have to have negative cost…

  21. One problem, James – how are you going to get all that power from the major areas of wind production (the heartland of the country) to major population centers and towns without some obscene losses due to transmission distance?

  22. Please present empirical evidence that the combination of wind and storage is more expensive than alternatives.

  23. Brett number two, that was the "So, Dr. Chu: where is that nice new interstate grid?" part of the post.

    A high efficiency grid would be required to circulate the power, even out the regional variance.

    All sorts of ideas have been proposed, some of which are insanely expensive. DC superconducting wire, etc.

  24. "Please present empirical evidence that the combination of wind and storage is more expensive than alternatives."

    Note that we actually have laws on the books, mandating that utilities buy wind power. Lobbied for by wind farmers. Wouldn't need that if it was cheaper than the alternative. The utilities themselves would be racing to put up windmills.

    Other Brett, in all fairness, I wouldn't characterize long distance line losses as "obscene". They exist, but aren't of THAT scale. Smaller than the power lost in most storage schemes, for instance.

  25. Brett Bellmore,

    I see that when your lie is confronted, like any dishonest conservative you change the subject. I'll consider the matter closed.

  26. BTW, that map really, really s*cks. The scale goes from blue to red to purple to violet to red

    to brown to yellow to green.

  27. Benny Lava: now, now.

    Wikipedia http://en.wikipedia.org/wiki/High-voltage_direct_… cites transmission losses for high-voltage DC, as used by Quebec Hydro, at 3% per 1000 km; so 10% or so for mid-west wind power to either coast. Bearable, especially as there's no carbon emitted in the first place.

    Brett: if power generators are allowed to own transmission, they need to be forced to buy in power from any source if there is to be anything like a competitive generating market, so the laws you cite are not necessarily biased towards wind; though a bias toward renewables is good policy as carbon emission is a colossal externality, "the greatest market failure in history" in Stern's words. If you want a truly competitive electricity market, you should support full separation of grid and generators, as in Thatcher's successful privatisation, and, surprisingly, in GW Bush's Texas reform. Separation is now required by EU law. A modern grid is market-maker to hundreds of generators and millions of consumers.

  28. James, yeah, and if we're to have a competative food market, grocers need to be forced to buy cream of wheat laced with rat poison, too, if somebody offers them a good price on it. You've got a really weird notion of what constitutes a 'competative market'. Here's a clue: In a genuine competative market, you aren't FORCED to buy squat.

  29. I have difficulty with the idea of any market not relying on collective – primarily state – regulation to enforce fair dealing. Would Venetian traders have bothered showing up in Provins in 1200 without the gallows of the Counts of Champagne overlooking the market street? The less tangible and immediate the exchange, the more regulation is required. How do you think central counterparties in securities markets work?

Comments are closed.