Over the past few weeks I have been engaged in a discussion over the benefits of wind and nuclear with Dave Erickson at Re/Action on Climate Protection. Two of his recent posts, Replace coal with Wind and Cost of Wind vs. Cost of Nuclear to Replace Coal, deserve further discussion.
Dave argues that wind is more cost effective than nuclear. However, his post is about climate protection and he never mentions that nuclear power is the leading source of emission free electricity in the U.S. Nuclear plants already avoid almost 700 million metric tons of CO2 annually in the U.S., and more than 2 billion metric tons of CO2 annually worldwide.
Replace Coal with Wind? Probably not. The U.S. has the largest reserves of coal in the world. Replacing coal completely with wind simply doesn't make economic sense. While coal releases significant emissions, it is getting cleaner. Any new coal plant built has to meet many stringent air requirements. To comply, coal plants are fitted with scrubbers and flue gas desulphurization emission technologies as well as other emission control techniques.
While replacing coal with wind could be beneficial for our air, I asked Dave how much land would be required. I noted that just to replace all the nuclear plants in the U.S. with wind, we would need an area the size of the state of Wisconsin. Dave contends that if wind turbines were packed closely together, to replace all of coal we'd only need an area equivalent to the size of West Virginia.
But think -- is it realistic enough land would be available? We’re talking about areas the size of states. Sure, there’s plenty of land in the Dakotas, Oklahoma, Nebraska and Kansas, but the vast majority of future electricity demand will be somewhere else. What good would that do? What about offshore wind? Possibly, if everyone doesn’t mind looking out at the ocean and viewing a horizon of turbines. As we've seen, Americans are picky about our scenery.
I'll have more on Monday when we post Part II of my analysis.
Technorati tags: Nuclear Energy, Nuclear Power, Energy, Electricity, Environment, Cape Cod, Wind Power, New England, Riverkeeper, NIMBY
Dave argues that wind is more cost effective than nuclear. However, his post is about climate protection and he never mentions that nuclear power is the leading source of emission free electricity in the U.S. Nuclear plants already avoid almost 700 million metric tons of CO2 annually in the U.S., and more than 2 billion metric tons of CO2 annually worldwide.
Replace Coal with Wind? Probably not. The U.S. has the largest reserves of coal in the world. Replacing coal completely with wind simply doesn't make economic sense. While coal releases significant emissions, it is getting cleaner. Any new coal plant built has to meet many stringent air requirements. To comply, coal plants are fitted with scrubbers and flue gas desulphurization emission technologies as well as other emission control techniques.
While replacing coal with wind could be beneficial for our air, I asked Dave how much land would be required. I noted that just to replace all the nuclear plants in the U.S. with wind, we would need an area the size of the state of Wisconsin. Dave contends that if wind turbines were packed closely together, to replace all of coal we'd only need an area equivalent to the size of West Virginia.
But think -- is it realistic enough land would be available? We’re talking about areas the size of states. Sure, there’s plenty of land in the Dakotas, Oklahoma, Nebraska and Kansas, but the vast majority of future electricity demand will be somewhere else. What good would that do? What about offshore wind? Possibly, if everyone doesn’t mind looking out at the ocean and viewing a horizon of turbines. As we've seen, Americans are picky about our scenery.
I'll have more on Monday when we post Part II of my analysis.
Technorati tags: Nuclear Energy, Nuclear Power, Energy, Electricity, Environment, Cape Cod, Wind Power, New England, Riverkeeper, NIMBY
Comments
The biggest issues with wind is that capital costs are veru high (but one can build small modular units wich is good) and that the wind does not blow all the time. This means wind can not be base load power. Because of this wind and nuclear are not exactly competitors.
Also wind power is pretty expensive. This might be acceptable to households but is not acceptable to electricity intensive industries which compete on a global market.
So:
* High cost
* Not base load
To this one could add NIMBY (which contrary to popular opiniion is not a problem for nuclear power: http://www.nei.org/index.asp?catnum=4&catid=851 ).
One could also add lack of potential. While the global wind resource is absolutely massive, the wind might not always be where the consumers are. For example the Swedish wind resiurce is only 10 TWh while our yearly power consumption is 140 TWh.
(But in Denmark 15-20 % of all power (not of all capacity!) is wind.)
Wait till part 2. It will address much of the cost figures.
I am not trying to attack wind. I hope people can draw some conclusions that wind and nuclear are not really competitors like you mentioned. I'm just trying to make straight the analysis of the blogger in the article.
Wind and nuclear are just two sources that can provide emission free electricity in different ways.
Small amounts (~10 %) of wind power can be added to the grid, assuming the remaining 90% is rock solid generating capacity. The German wind power industry, regarded as the best of its kind, says as much in their 2005 year-end report:
http://www.eon-energie.de/bestellsystem/frameset_eng.php?choosenBu=eonenergie&choosenId=1725
But why bother? It's not as if the externalities associated with wind power are better than those for nuclear. Per kw-hr, the ExterneE study, which is extrememly conservative in its assessment of nuclear power externalities, finds only a slight edge in favor of wind power. Given the uncertainties associated with these calculations, there is no reason to believe that wind power is any better for the environment than nuclear power. It's a tie, more or less, even if you use absurdly pessismistic numbers for the nuclear power industry.
Wind power's dirty little secret is its poor industrial safety record. There's a 2002 article by Paul Gipe - one of wind power's most ardent supporters - detailing wind power accidents and calculating death rate per unit energy produced. Gipe found, to his surprise, that the numbers for wind power are little better than those for coal mining, emissions excluded of course.
http://www.wind-works.org/articles/BreathLife.html
Dave contends that if wind turbines were packed closely together, to replace all of coal we'd only need an area equivalent to the size of West Virginia.
As anyone who has ever spent much time in one of those fleets of sailboats that "anonymous" described will understand, devices that capture wind energy create a rather large "shadow" of disturbed air. When you put those devices too close together, the turbulent flow in that shadow reduces the efficiency of the device. The closer together they are, the less useful energy they can capture. In the limit, the output can go to almost zero, even if there is a good breeze.
I have been in the starting line fracas of a sailboat race on a windy day with flapping sails and moving nowhere.
I do not have the time to go back and review all of the calculations of wind proponents, but if they assume spacing of less than 3 times the rotor diameter (a thumb rule number I learned in my alternative energy engineering courses) without also beginning to reduce the efficiency numbers for their turbine, they are blowing smoke and will never reach their designed output.
In addition, the turbulence can cause some pretty significant reliability issues (those flapping sails that I experienced would not last very long if allowed to continue for too long.) For the techies in this conversation, here is a pretty interesting paper on the effects of turbulence in a large wind farm.
http://www.dewi.de/dewi_neu/deutsch/themen/bibli/pdf/seifert_kroening.pdf
Well you should be. I am sick and tired of nuclear advocates who preach technological 'coexistance' with wind. Stop calling for mediocre energy solutions. Wind is an awful technology and deserves to be critisized regardless of how personally the 'green' community might take it. I have withessed other such pandering-compromises from the past. No one wins.
http://www.otherpower.com/
But not for large, baseload supply for an industrialized society.
Truly I wish I had the money to waste - er, I mean 'invest' - on erecting a wind mill with battery backup for my own home. But it simply isn't cost effective and who wants to live without electricity for up to 10% of a year (and that's if you're lucky)?
Sorry, I don't mean to be nasty and sarcastic, but if wind were so great, then why doesn't humanity still use sailing ships to transport goods and people across the oceans?
Wind power has its place in supplying electricity for out of the way, isolated locations. Perhaps it has a place off Martha's Vineyard, but eco-green Senator Ted Kennedy and his Riverkeeper nephew RFK Jr. oppose that with the same vigor that they mutually oppose the Pilgrim Nuclear Power Plant in Massachusettes and the Indian Point Energy Center in NY.
http://www.eon-energie.de/bestellsystem/frameset_eng.php?choosenBu=eonenergie&choosenId=1725
See page 8 for a graphic example of the downside of wind. On Dec. 24, wind produced 6,000 MW. On Dec. 26, two days later, the winds had changed and their windmills only produced 24 MW.
That means that E.On needed to be able to call on the equivalent of 4 to 6 large nuclear or coal units, otherwise sitting idle in hot standby waiting for the winds to die off.
Where is the capital efficency in this, if one needs to pay for both wind AND nuclear?
I've never seen a figure this high: usually it's in the range of 60 acres per MW (or less, depending on terrain and turbine size) to prevent interference between windmills, with 5% or less actually occupied by the windmill and access roads, etc.
Where does the figure come from?
Your presumption is correct. It is an incorrect stat and I'm embarrassed and apologize for it.
To replace all of nuclear with wind, we would need an area the size of the state of West Virginia. West Virginia is about one third the size of Wisconsin. Big error on my part.
We used a stat from one of NRC's Impact Statement that said 1,000 MW of wind require 150,000 acres of land. What was not known is that wind's capacity factor is already calculated in the stat.
Before we knew that we took the 150,000 acres figure and used the capacity factor again and that's why we came up with such a high number. On AWEA's website they say 60 acres per MW (along the lines of what you assume).
About a month ago we did an analysis of how much land each generating technology requires and discovered the error then.
Why is it in this blog then? I simply forgot it was incorrect.
I hope readers will interpret this as a simple error and not try to accuse myself or the industry of being liars or deceivers or whatever anyone wants to call us.
Our intention is to convey accurate and up to date info as best as possible. We strive to be open and honest and we always try to own up to our mistakes. Sorry about the error. Thanks Nick.
David
We might also want to note, for the benefit of readers like Rod Adams, that the figure of 60 acres per MW is for spacing between windmills, not actually land used up by windmills and their infrastructure. The 60 acre figure is useful for estimates of potential wind resource, but if the intention is to analyze the land consumed by a generating technology, it may not be as helpful.
Estimates of actual wind land usage range from .25 to 3 acres per MW. At the 3 acre per MW figure, you would need 810,000 acres, or 1,266 square miles, or 5% of W. Virginia's 24,231 sq miles.
A useful comparison might be with farmland: 810k acres is .3% of the 275M acres of US farmland.
Does that make sense?
You're exactly right. Only 5% of the 60 acres per MW is actually occupied by the turbines.
While 60 acres per MW is the land space needed to appropriately cite the turbines for efficiency, 95% can be used for plenty of other things like farming and cattle grazing...
"Nuclear power is uneconomical and unnecessary." http://www.climateprotectionblog.com/cpc/nuclear_option/index.html
My original analysis that lead to this conclusion was based on the necessity to replace the largest point source emitters of carbon dioxide: coal-fired power plants. David is right, my primary interest is in what it will take to transition to a non-carbon-emitting economy. Coal-fired plants account for 81% of the greenhouse gas emissions of the total emissions for electric power generation.
I posed the question: What would it take to replace the existing generation capacity of the coal-fired power plants in the United States with a non-carbon-emitting technology? I wanted to get an "apples-to-apples" comparison of the costs and the magnitude of the project.
The two non-carbon emitting technologies I compared are nuclear power and large scale wind. I did not consider a carbon capture and sequestration retrofit to existing coal plants. Currently, this is completely economically unfeasible.
I based the choice of large scale wind in the comparison primarily on the work of Dr. Mark Jacobson at Stanford. http://www.stanford.edu/group/efmh/jacobson/ Jacobson has made, in my view, two fundamental observations about wind power: 1) when enough turbines are deployed over a geographically disparate area, and appropriately wheeled using the grid, the total contribution of wind power approximates baseload generation; 2) the availability of adequate wind resources to drive large wind turbines, i.e., 8m/s or greater, is much larger than previously thought, if measurements are made at hub height, i.e., 80m. Jacobson has extensively documented these two observations on his website and in publications in Science magazine. (and elsewhere not published, check with me for details).
The first question is: Is wind power adequate? In an email to me, Jacobson showed the following, and I verified it:
"...the global total energy demand (for all power sources) is 8.2-8.4 x 10^13 kWh/yr (from Archer C.L., and M.Z. Jacobson, Evaluation of global wind power, J. Geophys. Res., 110, D12110, doi:10.1029/2004JD005462, 2005, www.stanford.edu/group/efmh/winds/global_winds.html.).
...
About 16 million 1.5 MW turbines (about 2.66 million km^2) would be needed for all power demand. This area is about 0.5% of the Earth's total (land + ocean) surface area. The global wind evaluation paper above found that about 13% of land area
on Earth had winds 7 m/s or faster at hub height."
To restate: Jacobson asserts that all global energy demand from all sources (not just electricity) could be supplied by 16 million 2MW turbines occupying 0.5% of Earth's total area.
Thus, wind power is adequate to supply, not only the total current global electric power demand, but the entire global energy demand.
To summarize: large scale wind, (largest turbine available today- 5MW), massively deployed (for example, to replace coal) is more than adequate to supply global energy needs. Wind power has baseload generation characteristics if deployed over a geographically disparate area. But is it cost effective?
In order to answer this question, we need to look at an compare the two primary contenders for a non-carbon emitting, baseload power source, nuclear power plants and a large fleet of 5 MW wind turbines.
Here are the parameters of the comparison:
Total electric power generated by coal plants in the US in 2004: 1,979 billion kilowatt hours
Number of 5 MW wind turbines required to replace coal: 118,548 (more or less, calculation shown below)
Average capacity factor of nuclear plants in US since 1989 (EIA): 78%
Number of 1,000 MW nuclear plants to replace coal (at 78% capacity factor): 290
Cost to build wind turbine in US: $1024/kW
Cost to build nuclear plant in US:
$1894/kW
Estimated cost to build 118,548, 5 MW onshore wind turbines: $607 billion
Estimated cost to build 290, 1000 MW nuclear plants:
$548 billion
Number of 5 MW wind turbines that must be built per year to build all of them in 5 years: 24,000 (about 2000/month)
Number of people required to complete 24,000 wind turbine installations per year: 142,000 (crew of 12 builds 2 per year,estimated)
Total Annual O&M and fuel cost of all new nuclear in the coal replacement project ($63/kWe): $18.2 billion
Total annual O&M cost of all new wind turbines ($27/kWe): $16 billion
So far, we see the 10.6% higher cost of the wind project being paid back in about 26 years because of the lower O&M and fuel cost. However, we have not taken into account the cost of financing the two projects.
I argue that the reason that wind is more economical is because the cost of financing is lower due to the relative perceived risk and externalities of the two technologies. Today, in the US, it is "impossible" ( http://climateprotectioncampaign.typepad.com/cpc/2006/03/impossibility_o.html ) to secure financing for a new nuclear project. All new wind projects in the US are privately financed, and there was historic growth in the wind power sector last year ( http://climateprotectioncampaign.typepad.com/cpc/2006/03/global_wind_pow.html ). Clearly, investment dollars are being attracted to wind, and not to nuclear. How does this fact translate to cost?
In the "Projected Costs" study, levelized costs of different generation technologies are presented at two discount rates, 5% and 10%. In the US, which I argue must serve as the context for this example, the levelized cost of nuclear at the 5% discount rate is $30.1/MWh, and $46.5/MWh at the 10% discount rate.
For wind, the levelized costs are $31.1/MWh at 5% and $47.8/MWh at 10%. To compare equally, the two projects need to consider levelized cost, without subsidies, as the "Projected Costs" study does. I argue that, because of the greater perceived investment risk and externalities, the levelized cost of new nuclear will run closer to the 10% discount rate, or $46.5/MWh. Wind, because the investment risks and externalities are lower, will run closer to the $31.1/MWh.
Therefore, I conclude that, based on a project to replace all coal-fired power plants in the US with non-carbon-emitting generators, wind power provides all the necessary energy at a lower cost than nuclear. Not a single dollar needs to be spent on nuclear power, instead of wind.
Here is the calculation used to compute the number of wind turbines required:
E (in kWh) = 8760 (hours per year) * P (nominal output kW) * CF (capacity factor)
where CF = 0.087 * V - P/D^2
V is mean Rayleigh wind speed in meters per second
P is turbine rated power in kW
D is rotor diameter in meters
For the 5 MW, 126 m rotor diameter turbine with 8 m/s mean wind speed:
Capacity Factor= 0.38
Annual Energy = 17 million kWh
Why are you using a 78% CF for nuclear and a 38% CF for wind?
If you are going to say that a new wind turbine will operate that efficiently than you have to say that a new nuclear plant will be operating at a 90% CF. Not take what the average CF has been for the fleet over the past 15 years.
If you are going to do that, then do it for wind. From 1995-2005, the average CF for wind was 27%.
Another thing, you assume all 290 nuke plants will be risky projects. They won't be. Risk is assumed if the company building the plant has mediocre credit and doesn't have enough financial cushion to support an intensive capital project like nuclear.
Most companies that are looking to build nuclear right now are joining up together to build the projects. Therefore the risk is much lower than if a company were to do it alone.
Even when you calculate your numbers at the most optimistic of wind and pessimistic of nuclear, you still find that nuclear's 290 plants are cheaper than wind.
Now you know what I mean about making assumptions. Not everyone makes the same ones. But when you make them, you need to be consistent in your analysis.
Secondly, on nuclear capacity factor. The figure I used is the average of all the plants operating in the US since 1989 (the first year statistics become available). I think it is incorrect to assume that a new nuclear plant will immediately begin operating at maximum capacity factor. That is just not the history. The MIT report evaluates the cost of nuclear power using both 75% capacity factor and 85% capacity factor. They note that 90+% has only been achieved the last two out of three years.
Thirdly, the costs that I calculated for the two projects are about 10% different. I am not willing to say that this "back of the envelope calculation" is going to be 5% accurate. I challenge you to cost a project of this magnitude within 5%. I think a "fair minded" analyst would say that these two projects "cost the same" or are "very close".
I will say more about the investment risk. It has to do with the fact that, in the wind project, the investor sees a return on investment immediately. In this case, it would be to the tune of 10,000 MW of faceplate capacity coming online each month, for 5 years. With nuclear plants, on the other hand, the quantum of energy capacity is much larger, and you have to wait much longer before the project starts generating income. You typically won't see marketable power coming out of a nuclear plant for from 5-10 years. The risk is that a lot can happen during that time, such as construction delays and cost overruns. This has been historically what has happened with nuclear plant construction.
You may assert or proclaim that it is all different now. However, you have offered no data to back it up. Where's the data that would convince me as an investor that it was any less risking to go ahead with a new nuclear power plant today than it was 10 or 20 years ago? If I look at the history of nuclear power plant construction, I see a history of cost overruns and delays.
38% capacity factor on wind power is very speculative. The existing installations that often have actual capacity factors of 15% to 20% have theoretical capacity factors of 30% or more.
The concept of widely distributed wind generation simulating baseload generation is interesting but very hypothetical. It's impractical to transmit power over very large distances. What happens when the entire Midwest is becalmed or when a hurrican hits the southeast and all the generators have to protectively shut down due to excessive wind speeds. In practical terms I think the effective capacity factor takes another big hit.
Of course I don't have anything to back up new plants in the U.S. because they haven't been built yet.
Many people may not know this but there are two parts to the history of nuclear costs. Before Three Mile Island and after. The average costs before TMI were around $1,500 / kW. The costs after TMI rocketed due to mainly regulatory delays.
About half of the operating units came online before 1979.
The average construction period was 4-5 years. Whereas the other half took 10-15 years to build after 1979. The last half is what everyone including you seems to only see.
The licensing process back then was that a utility would receive a construction permit, build the plant and then receive a license to operate. It was this process which opened itself up to many delays.
The process now is to apply for a combined construction and operating license where you are not vulnerable to the delays. You receive the licenses before investors put any money down and then once you receive it you start constructing, testing and once ready, turn on the key and run. It should only take about 4-5 years for construction to running.
We're going to test this out in the next 10 years and demonstrate to Wall Street and investors that the process works.
At the same time we hope that environmentalists, anti-nukes or anyone else opposed can bring up their issues during licensing. It will help with the testing.
Read my blogs again about capacity factor and Watts Bar for example. You are incorrect about the assumptions you make.
Nuke plants are built to run 24/7. Not 60%, not 30%. The only time they go down are for refueling and occasionally a reactor trip. That's why their CF is 90% and not 100%.
The assumptions for nuclear in MIT are assumptions. If you look, they use the same figures for coal and gas. In 2004, coal's overall CF was around 70% and gas was around 40%. They are not based on actual performance data, they are based on how they see baseload plants operating which is between 75%-85% CF.
You say nuclear needs much backup capacity for when they refuel. You may not know this but nuke plants shutdown in the spring and fall when electricity demand is low. Therefore not requiring an excessive amount of capacity from other sources. As well, when demand is low, it means market prices are low and replacement power is cheap.
I'm not going after your calculations for wind or how the Spanish grid works because I'm not interested. I'm trying to inform you about how nuclear works which you said you were so eager to learn awhile ago. Obviously you're not interested because you're cherry picking everything to support your claims.
You're whole argument against nuclear being uneconomical is risk. Everything else, nuclear and wind are comparable. Well in the next 10, 20 to 30 years we'll demonstrate how risky nuclear is.
Thanks for discussing this issue David. Hope you learned something. I did.
Dave
Thank you for your contribution as well. I enjoyed the debate!
For these reasons, there are few, if any, energy experts who believe that wind will be able to do more than the percentages above. Thus, simply talking about wind’s per-Kw economics, as though the intermittantcy problem did not exist, and advertising it as a “baseload” option for replacing all coal generation is disingenuous, and not helpful with respect to educating the public in energy issues.
Other minor points:
There is little, if any chance that nuclear’s future capacity factors will ever (again) be much below 90%. This can be stated with confidence, and should be the basis of any analysis. The industry (as well as its regulators) has climbed the learning curve, and is now mature.
The cost of ~$1800/kW for the ~300 nuclear plants required to replace coal is too high. This cost estimate would only apply for the first few “demonstration” plants, as it represents first-of-a-kind engineering and learning-curve costs. Without these costs, the price will be measurably lower. Once the first few copies are made, the follow-on plants, which are simply carbon-copies of the demo plants, and which would be built in large numbers (i.e., ~300), would only cost ~$1200-$1500/kW. Some say the cost for a large number of follow on copies would be as low as ~$1000-$1200/kW, especially if such a large number were ordered (larger scale of production, i.e., a larger number of copies, will reduce unit costs). Heck, if knew we were building ~300 plants, we’d set up a nuclear plant assembly line factory, and REALLY reduce costs. Anyway, any significant reduction in the up-front capital cost will significantly impact the economic analysis.
A nuclear plant is NOT harder to site than an equivalent amount of wind turbines (based on annual generation). If anything, the reverse is true. This will be true mainly due to the vastly larger land area required for a given amount of wind generation. Even though wind is still less than 1% of overall generation, there are already surprisingly high levels of NIMBY resistance. In Europe, this is even more true. Imagine if we tried to get ~100 times as much power from wind…. Nuclear, on the other hand, has much smaller siting issues. It may even be possible to build at least ~100-200 plants at existing sites, most of which can handle many more reactors. At present, most of these sites have strong local support, where they are actually begging (and offering incentives) for utilities to build new plants there.
Finally, on a philosophical note, given that wind still represents less than 1% of overall generation, it is very premature to state with confidence that wind is able, right now, to replace 50% of our overall generation (coal), and that other options such as nuclear need not even be considered or planned for. Wind is still way to early in its learning curve, and has not gone through its growing pains yet. At less than 1% of generation, it would be accurate to state that wind has not really been tried (yet). Should we “try”? Yes!! But it’s not yet time to put all our eggs in that basket. In the meantime, other options need to be developed and applied as well.
What do you think would happen if we declared that all coal plants must close, but we then let the free market decide how their generation would be replaced. Do you really think that we would not build any nukes and that all we would do is build windfarms? All of us here are quite confident that an enormous number of nuclear plants (as well as windfarms) would be built.
What about the added cost of nuclear waste management?
The Hanford Site is a decade late and another decade from completion and grossly over budget (I believe that I heard it is projected at about $600 billion now). Also the Hanford Site has had a massive leakage plume that has nearly reached the Columbis River. How much to clean that up?
And what about Yucca Mtn.? When and how much?
I see Hanford and Yucca as temporary solutions. Eventually there just has to be a way fo dealing with it permanently. In any case, these are factors that directly affect the overall cost of the implementation of nuclear.
The only possible solution that I have heard mentioned is Accelerator Driven nuclear Reactors (subcritical reactors). What about that?
Again, I am not an expert in this field, so don't blast me.
Earthling
Why hasn't the US been recycling its nuclear byproducts like the rest of the planet has been. This would greatly reduce it's waste disposal costs, and greatly favor the use of Nuclear energy as a greener choice.