Key words :
greenhouse gas emissions
Nuclear power: False climate change prophet?
21 Jul, 2008 09:31 am
A new study reveals that nuclear power is not as clean as the industry claims.
Which side is right?
One new study published in the August 2008 issue of the peer-reviewed journal Energy Policy attempts to answer this question. It screened 103 lifecycle studies of greenhouse gas equivalent emissions for nuclear power plants to identify a subset of the most current, original, and methodologically rigorous studies. The study found that while the range of emissions for nuclear energy over the lifetime of a plant reported was from 1.4 grams of carbon dioxide equivalent per kWh (gCO2e/kWh) to 288 gCO2e/kWh, the mean value was 66 gCO2e/kWh.
The frontend component of the nuclear fuel cycle (uranium mining, milling, and enrichment) is responsible for 38 percent of equivalent emissions. Decommissioning and plant operation, including the use of fossil-fueled generators to backup nuclear plants when they offline for servicing, account for 35 percent. The backend of the fuel cycle, which includes storing spent fuel and fuel conditioning, account for 15 percent of the emissions, and plant construction is responsible for 12 percent.
This average—66 grams of carbon dioxide for every kWh—is shockingly high compared to what the nuclear industry has reported. It also shows, conclusively, that nuclear energy is in no way “carbon free” or “emissions free,” and that nuclear power is worse than the equivalent carbon emissions over the lifecycle of renewable and small scale distributed generators (although it is an improvement over oil-, coal-, and natural gas-fired generators).
To provide just a rough estimate of how much equivalent carbon dioxide nuclear plants emit over the course of their lifecycle, a 1,000 MW reactor operating at a 90 percent capacity factor will emit the equivalent of 1,427 tons of carbon dioxide every day, or 522,323 metric tons of carbon dioxide every year. Nuclear facilities were responsible for emitting the equivalent of some 183 million metric tons of carbon dioxide in 2005. Assuming a carbon tax of $24 per ton—nothing too extreme—and that 1,000 MW nuclear plant would have to pay almost $12.6 million per year for its carbon-equivalent emissions. For the global nuclear power industry, this equates to approximately $4.4 billion in carbon taxes per year.
Researchers in the United Kingdom conducted lifecycle analyses for 15 separate distributed generation and renewable energy technologies found that all but one, solar photovoltaics (PV), emitted much less gCO2e/kWh than the mean reported for nuclear plants. In an analysis using updated data on solar PV, researchers in the United States found that current estimates on the greenhouse gas emissions for typical solar PV systems range from 29 to 35 gCO2/kWh.
This has two very important insights for the current debate about nuclear power and climate change.
First, nuclear power plants would not benefit directly from a global carbon tax or a carbon cap-and-trade system. While the nuclear industry would be penalized less than fossil-fueled generators, the carbon equivalent emissions from uranium mining operations, enrichment facilities, plant construction, decommissioning, and spent fuel storage are significant. Any type of extra cost for carbon-equivalent would increase, absolutely, the price of these elements of the nuclear fuel cycle, and would thus make nuclear power more expensive.
Second, while it may be unfair to compare baseload sources such as nuclear to intermittent or non-dispatchable sources such as wind and solar PV, if these numbers are correct, then offshore wind power has less than one-seventh the carbon equivalent emissions of nuclear plants; large-scale hydropower, onshore wind, and biogas, about one-sixth the emissions; small-scale hydroelectric and solar thermal one-fifth. This makes these renewable energy technologies seven-, six-, and five-times more effective on a per kWh basis at fighting climate change.
Put simply, investments in nuclear power are much worse at fighting climate change than pursuing wind, solar, and other small-scale power generators. Policymakers would be wise to embrace these more environmentally friendly technologies if they are serious about producing electricity and mitigating climate change.
For further reading:
Barnaby, Frank and James Kemp. 2007. Secure Energy? Civil Nuclear Power, Security, and Global Warming (Oxford: Oxford Research Group, March, 2007).
Fthenakis, V.M., Kim, H.C., and Alsema, M. 2008. “Emissions from Photovoltaic Life Cycles.” Environmental Science and Technology 42, 2168-2174.
Pehnt, Marin. 2006. “Dynamic Lifecycle Assessment of Renewable Energy Technologies.” Renewable Energy 31 (2006), pp. 55-71.
Sovacool, Benjamin K. 2008. “Valuing the Greenhouse Gas Emissions from Nuclear Power: A Critical Survey,” Energy Policy 36 (8) (August), pp. 2940-2953.
Sovacool, Benjamin K. 2008. “Nuclear Power is a False Solution to Climate Change,” The Jakarta Post (July 15), p. 6.
Key words :
greenhouse gas emissions
Â“Peak OilÂ” is NonsenseÂ… Because ThereÂ’s Enough Gas to Last 250 Years.
Threat of Population Surge to "10 Billion" Espoused in London Theatre.
Current Commentary: Energy from Nuclear Fusion Â– Realities, Prospects and Fantasies?
The Oil Industry's Deceitful Promise of American Energy Independence
Shaky Foundations for Offshore Wind Farms
The first mistake is to treat all estimates as valid. It ought to be clear that if the range of estimates is 1.4 grams to 288 grams, they can't all be valid. As we scan the list of sources, we see that the high-end estimates come from highly biased sources. Tokimatsu is an advocate for fusion energy. He also is biased against renewable energy sources; in his paper he says that photovoltaic sources generate more CO2 than nuclear fission plants, despite his vastly exaggerated estimate for fission. Oxford Research Group, with its self-aggrandizing name, is a political group opposed to nuclear energy. Dones changed his mind between 2004 and 2005 but the 2004 estimate is included anyway. More cyber-ink has been spilled over Storm's miscalculations than I'd ever care to contemplate.
The second mistake is weighting all the estimates equally. Doing so automatically over-weights the high-end estimates. Take the simple case as an example. Averaging 1.4 and 288 yields 144.7. This overweights the result to the point where it is meaningless. Both extreme numbers can't be right and for the analysis to be valid it's necessary to determine which one is right and which one is wrong. That's what's missing here.
Third, this comparison doesn't treat nuclear and renewables in the same way. The overstated results here, based on disparate estimates from sources of varying degrees of validity, are compared to a single source for renewable-energy emissions.
Fourth, this paper, like so many anti-nuke polemics, simply ignores the fact that renewables don't work without backup. The CO2 emissions of the backup also have to be included.
Finally, this study doesn't pass the sniff test. Here's the final paragraph in your blog article: "Put simply, investments in nuclear power are much worse at fighting climate change than pursuing wind, solar, and other small-scale power generators. Policymakers would be wise to embrace these more environmentally friendly technologies if they are serious about producing electricity and mitigating climate change." It's pretty clear that you started with the conclusion and cobbled together a paper to get you to the desired conclusion. Having reached it, you weren't concerned over the process that got you to it.
I suppose the problem is that the low estimates are equally invalid. They look at only one or two reactors; presume operating capacities in excess of reality; look at technologies, such as Generation IV reactors, that do not yet exist; or presume the use of only high grade uranium ore.
Given that both "sides" of lifecycle estimates have their flaws, it seems to me the best way, or perhaps even only way, to get a handle on what the true lifecycle emissions for nuclear power plants are is to look at as many studies as possible (in this case 103), exclude ones that are too old or inaccessible, and then take the mean from the rest, no matter what it might be. In this case, it is about 66 g/kWh. Such an approach can indeed be critiqued, but keep in mind that it?s more statistically accurate than any study in isolation. In other words, the number of 66 may have its flaws?heck, half of the Energy Policy article is spent talking about how all of the lifecycle studies differ?but it?s probably the best we?ve got for now.
The question of overweighting doesn?t apply because the analysis presented in the Energy Policy article accounted for the means from each study. This is precisely why I didn?t arrive at 144.7 (1.4 plus 288, divided by 2), and instead arrived at 66. Of course the extreme numbers can?t be right, which is why my study looked at the mean for each component of the fuel cycle, rather than taking the range.
Your claim that renewables don?t work without backup is simply wrong. Hydroelectric, geothermal, and biomass resources frequently operate as base-load generators. Solar and wind can be coupled to compressed air energy storage, pumped hydro storage, or attached to biomass facilities to make them base-load generators as well.
Finally, as for your ?sniff? test, if I truly started with my conclusion and worked backwards?which I assure you I did not?it would not have passed peer review, and I would most certainly have not included the numerous studies with low estimates.
Remember that it was you all that started this debate: you and Charles cited ridiculously low numbers of the nuclear lifecycle. Looking at more than just your select studies, nuclear plants are magnitudes of order worse at fighting climate change than renewables. I?m sorry, but the data disagree with you.
Your first paragraph invites comment. It's not clear why the results for one or two reactors would be different from the results for many. Existing reactors are more or less the same with the same fuel cycle. The capacity factors for all reactors in the US are running at 90% or slightly better, so it's impossible to overstate them by more than 10%, so they couldn't be far off. It's fair to say that estimates for Gen IV reactors would be dodgy, since they don't exist, but the designs are complete so the estimates wouldn't be far off enough to disqualify them. Anyway, the comparison is more appropriate since what the world will be building will be new designs, not old ones. Actually, there's enough high-grade ore to last for over six hundred years and many thousands of years with advanced fuel cycles, so calculations based on high-grade ore are valid.
The problem with averaging estimates you don't trust is that the results depend too much on which studies are included. Here's an example. Say you select five studies that yield 100 grams and five that yield 10 grams. They average out to 55. But then you add five more that yield 100 and the average changes to 70. Or instead you add five more that yield 10 and the average changes to 40. The number resulting changes from 70 to 40. The actual emissions haven't changed, just the calculated average. That's not analysis, it's scorekeeping.
If you're not equipped to do the analysis yourself, then find an analysis you trust. For a start, reject any studies from biased sources. For example, if NEI commissioned Dewey, Cheatham and Howe to do the study, it would be reasonable to reject it. The results could be right, but the source is suspect. Similarly, anything done by ORG or Storm or any of the leading opponents of nuclear energy or exponents of alternative energy should be rejected. What we're looking for is objective work done by qualified analysts who don't have an agenda to promote.
Most discussions don't include hydro as renewable because its environmental consequences are huge and because there aren't any additional hydro sites available. Geothermal is not renewable because the fields are exhausted after some years of operation and have to be left idle for many years to recover. Also, the pollution from them exceeds what most people consider reasonable. Biofuel is a loser; since it takes a gallon of fuel to produce a gallon of fuel, and if it required no fuel to produce fuel there still wouldn't be enough land to produce more that a fraction of the motor fuel required. There is no possible way to store enough wind energy or solar energy to run a full-time economy; please look here for details.
Sorry, Benjamin, your arguments against nuclear energy have been so enthusiastic and so poorly founded that you can never pretend to be unbiased. We know before reading your articles exactly what the conclusion will be and we know you'll twist both logic and fact to reach it, just as you did in this article. When your conclusion is stated the way this one's was, it never will pass the sniff test.
Ironically, you've done the same thing again in your last comment. You can't (or at least haven't) shown any reason to doubt the low numbers Mr. Barton and I have cited, which come from professional analysts who don't have a financial or ideological connection to nuclear energy. Nonetheless, you choose to characterize them as "ridiculously low" in order to protect your pre-established conclusion.
While I recognize Red?s appreciation for nuanced argument, I?m still unconvinced. The one point I do agree with Red on is that a more detailed discussion of the carbon-equivalent emissions from the lifecycles of renewable power technologies is called for. I cite a few numbers in the end of my article, but I would invite some shrewd thinkers with a little more time than me to conduct the same type of analysis I did with nuclear power plants for wind, solar, etc. Taking solar, for instance, some studies show its emissions are quite low (below 40) while others put it above 100. I went with the most recent estimate?one from VM Fthenakis at Brookhaven National Laboratory?but closer analysis may be warranted.
However, that is where my agreement ends. As the EP article shows (and Charles, I did send Red a copy, and would put one online if Elsevier?s copyright policy permitted me), the performance of reactors throughout the world has been very different than what almost all of the lifecycle estimates at the low end of the spectrum assume. Look at just one example: capacity factor. The studies you mention with low estimates presume capacity factors in the mid to high 90s, while the actual average capacity factor for all reactors is?get this?about 66 percent. Those estimates in no way match the reality of currently operating reactors.
Let?s also not forget how the idea for the article in Energy Policy began: with the two of you citing carbon-equivalent estimates in the 1 to 4 grams/kWh range from completely biased sources (one including the Nuclear Energy Institute and the World Nuclear Association, if I remember correctly). Despite what you may think, I did not start out with any type of preconceived number when writing my EP piece. I really just wanted to find out what the studies said, which is also why 9 out of the 17 estimates I include as ?qualified? have estimates below 30 grams/kWh. If I was truly looking only for a scorecard, why in the world would have I included 9 low estimates? Just for kicks?
No, I truly believe the analysis presented in the EP article is sound?and I also admit in the article, quite clearly, that I think nuclear plants are better than fossil fueled plants (from a global warming perspective) (This is proof, yet again, that I am not an advocate of ?coal? despite Charles accusing me of being so). I just don?t think nuclear plants are better than renewables or energy efficiency, which is where we disagree. And I also have no financial or ideological commitment to nuclear or renewables, other than to pursue the technologies that I think have the most benefits for society. No hidden stock options, no concealed payments, no personal relationships, no other interests: just analysis.
A final but important correction: I?d check your facts about renewables. The environmental consequences from virtually every renewable power resource, save large hydro, are far less than nuclear and fossil plants. And this analysis doesn?t come from me, but from about 40 different studies from authors around the world. Unfortunately, most of these studies are not online, but a good starting point is Thomas Sundqvist and Patrik Soderholm, ?Valuing the Environmental Impacts of Electricity Generation: A Critical Survey,? Journal of Energy Literature 8(2) (2002), pp. 1-18; Thomas Sundqvist, ?What Causes the Disparity of Electricity Externality Estimates?? Energy Policy 32 (2004), pp. 1753-1766.
Charles, finding copies of these articles may necessitate actually visiting a library, but the studies may force you
The CO2 emissions I quoted came from P.J. Meier, Life-Cycle Assessment of Electricity Generation Systems and Applications for Climate Change Policy Analysis, Ph.D. Dissertation, University of Wisconsin ? Madison, 2002 [source] It shows
Combined-cycle natural gas 469
Nuclear fission 15
DT fusion 9
I don't think you contrived to leave out any particular studies because they didn't give the desired results. Rather, I think you jumbled together a lot of results from a number of sources, good and bad, and averaged them. The resulting average told you what you wanted so you published it. If the results had turned out to favor nuclear energy, which they could easily have done if by chance you selected a different set of studies, you just wouldn't have published it.
I'd rather not try to track down every study on environmental effects you've found that you approve of. Instead, let me refer you to the most comprehensive one done. According to the European Commission , the total external cost of nuclear is higher than for wind and about the same as solar. However, the difference between nuclear and wind isn't worth bragging about, as both are a small fraction of the costs for coal and oil. In any case, the external costs of backup for wind or solar will drive both of them out of competition if the backup is either fossil-fuel or storage. I don't mean to imply, however, that storage could ever be practical; only that if it were attempted the external cost would follow direct cost into oblivion, as shown here.
The 66 percent capacity factor is what the World Nuclear Association reported in 2005 as the global average for reactor performance for the lifetime of all reactors currently in operation. (Thus, it does include mostly Generation II reactors but it also provides a picture to how well the world's operating nuclear plants have performed, as a whole and over their entire lifetime, to date).
I believe the study you mention by the EC is the Externe study, a good one that has since been revised and published a few times. The other "big" externality studies I am familiar with are from ORNL/RFF and the Pace University Law School. The beauty of the Thomas Sundqvist and Patrik Soderholm pieces (both based on the same study) are that they looked at more than 60 externality studies - including the EC one you mention - and 132 different estimates for individual power generators. Concerned that one could tweak the numbers, so to speak, by only quoting a few studies, they thought it best to simply look at them all, and average the results, much like I attempted to do with my nuclear piece. The results are that each power generator produced the following amount of currently unpriced negative externalities:
Coal, 14.9 cents/kWh;
They provide their estimates in 1998$, and conclude that solar and wind have the least negative externalities of all generators. Definitely worth a read, and a more complete picture of externalities than any individual study.
Agreed that the Journal of Energy Literature is a bit hard to find, but the journal for the second source - Energy Policy - is not. Online versions appear to be at http://ideas.repec.org/a/eee/enepol/v32y2004i15p1753-1766.html, linkinghub.elsevier.com/retrieve/pii/S0301421503001654, and http://econpapers.repec.org/article/eeeenepol/v_3A32_3Ay_3A2004_3Ai_3A15_3Ap_3A1753-1766.htm.
Given that all studies have their own methodology and varying ideologies, how else, other than averaging for all studies, would one ever attempt to wade through them?
Another problem your conclusion is that it assumes a once through fuel cycle with thermal neutron spectrum reactors. A transition to fast neutron spectrum reactors would reduce the CO2/KWh value by a factor of 10 or better. While fast reactors have not yet been commercialized, the technology is known and it does not differ greatly from thermal reactors. It also does not matter that it may takes 25 years to commercialize fast reactors, since they will be able to utilize the spent fuel from that accumulate from thermal reactors.
I'm afraid that the energy sector as a whole, and the electric utility sector specifically, is so politicized that you won't find agreement on numbers for ANYTHING. Whether it's the capital cost for building new generators or assessing the acid rain damage from coal plants or determining the levelized cost of electricity for power systems or the avoided cost of energy efficiency, the studies all differ my magnitudes of order. One has either the option of simply ignoring studies one disagrees with, or attempting to find a way to assess them all evenly.
As for Generation IV reactors, they are completely theoretical and, as you point out, at least 20 to 30 years away. We need carbon reducing electricity technologies NOW, not later, and even if you are correct that Gen IV reactors will be very efficient, it doesnt justify the nuclear industry claiming today that existing reactors are emissions free.
However, the Sundqvist papers confirm Mike's remarks. In fact, the "What causes the disparity of electricity externality estimates?" paper says the same thing in more words. Here's the conclusion. I'm including all of the key paragraph to allay suspicions that I'm editing part out.
"The analysis shows that there are several reasonable
(systematical) explanations to the discrepancies among
studies. For example, it has been shown that the choice
of methodological approach tends to affect the esti-
mates, and that there are significant differences among
fuel externalities. Furthermore, the completeness of
approach tends to affect the size of estimates. It cannot
be shown that the disparity of results arises due to site
specificity. Consequently, the results indicate that the
disparity arises due to methodological reasons and due
to problems in the application of these methods
(especially abatement cost and top-down approaches).
Thus, the results produced by the previous externality
costing studies may be problematic to interpret and use
in some reasonable way. The results also show that the
bottom-up approach produces the lowest external cost
estimates. Whether this is due to some problem in the
application of the method or due to the fact that the
other methodologies produce too high estimates is,
however, beyond the scope of this study."
You cite this paper as justification of your methodology, but you ignore the conclusion. In fact, you look past the charts which show that the external costs of wind, solar, and nuclear overlap each other. Then you go so far as to say in note  that Sundqvist calculates external costs by averaging the results of different studies. As the paper makes clear, he only does the averaging to prove the results are unreliable.
Benjamin, this isn't trivial at all. You're putting your reputation at risk by the shady manipulation of information as you did here. To endanger your career for the sake of such a misguided cause as antinuclearism is a huge personal mistake. Please reconsider.
I appreciate your sudden concern for my career, but I think you may be misinterpreting Sundquvist's conclusion. He does say that trying to incorporate so many divergent studies is difficult, and that attempting to find any "true" cost of power is impossible, especially since the methodologies behind each individual study differ so greatly.
But in addition to raising the point you make, he also concludes (in the paper written with Soderholm, p. 36) that ?this does not imply that valuation efforts have been in vain. Previous studies have taught us a lot about the environmental impacts of power generation (in particular health effects), and even if much of this knowledge cannot be transferred directly into a tax or a regulation, it should be able to impact upon the focus of the political debate and ultimately on policy decisions.?
In other words, comparing many studies with different foci and methodologies has inescapable flaws, but in the end such comparisons should affect and influence policy. This connects to what I was writing to Mike: when confronted with hundreds of different studies, we can either ignore some or try and find a way to assess them all. Peter Novick once compared compiling numbers and statistics to nailing jelly to the wall. And, despite the fact that pretty much every effort to statistically measure something in essence nails jelly to the wall, we all still do it ? we all use economics and tools and numbers in our discussions about energy policy.
Finally, note that I make a similar conclusion to Sundqvist explicitly in my study:
?The second (and perhaps more obvious) conclusion is that lifecycle studies of greenhouse gas emissions associated with the nuclear fuel cycle need to become more accurate, transparent, accountable, and comprehensive. Thirty-nine percent of lifecycle studies reviewed were more than ten years old. Nine percent, while cited in the literature, are inaccessible. Thirty-four percent did not explain their research methodology, relied completely on secondary sources, or were not explicit about the distribution of carbon-equivalent emissions over the different stages of the nuclear fuel cycle. All in all, this meant that 81 percent of studies had methodological shortcomings that justified excluding them from the assessment conducted here. No identifiable industry standard provides guidance for utilities and companies operating nuclear facilities concerning how to report their carbon-equivalent emissions. Regulators, utilities, and operators should consider developing formal standardization and reporting criteria for the greenhouse gas emissions associated with nuclear lifecycles similar to those that provide general guidance for environmental management and lifecycle assessment, such as ISO 14040 and 14044, but adapted exclusively to the nuclear industry.
Of the remaining nineteen percent of studies that were relatively up to date, accessible, and methodologically explicit, they varied greatly in their comprehensiveness, some counting just construction and decommissioning as part of the fuel cycle, and others including mining, milling, enrichment, conversion, construction, operation, processing, waste storage, and decommissioning. Adding even more variation, studies differed in whether they assessed future emissions for a few individual reactors or past emissions for the global nuclear fleet; assumed existing technologies or those under development; and presumed whether the electricity needed for mining and enrichment came from fossil fuels, other nuclear plants, renewable energy technologies, or a combination thereof.
Furthermore, the specific reactors studied differ greatly themselves. Some utilize relatively high-quality uranium ore located close to the reactor site; others require the importation of low-quality ore from thousands of kilometers away. A nuclear plant in Canada may receive its fuel from open-pit uranium mines enriched at a gaseous diffusion facility, whereas a reactor in Egypt may receive its fuel from an underground mine enriched through centrifuge. A nuclear facility in France may operate with a load factor of 83 percent for 40 years on a closed fuel cycle relying on reprocessed fuel, whereas a light water reactor in the United States may operate with a load factor of 81 percent for 25 years on a once-through fuel cycle that generates significant amounts of spent nuclear fuel.
Rather than detail the complexity and variation inherent in the greenhouse gas emissions associated with the nuclear lifecycle, most studies obscure it; especially those motivated on both sides of the nuclear debate attempting to make nuclear energy look cleaner or dirtier than it really is.?
Keep in mind that at least I?along with Sundqvist?make an attempt to be very transparent about research methods so that we can have these types of discussions. And that is more than the bulk of authors reporting on lifecycle emissions from nuclear power plants can say.
So you and I agree, if I'm reading your comment correctly, that the numbers are meaningless. But you maintain we should base policy decisions on them anyway. Okay, I'm totally baffled. And exactly why policymakers should favor renewables over nuclear, as your article concluded, becomes even more deeply enshrouded in incomprehensibility.
Clearly, we're not going to agree. I'm more convinced than before that it makes more sense to depend on a really thorough, non-political study like ExternE than to average a collection of biased and incomplete exercises.
You definitely get cudos for a good sense of humor.
I wouldn't say the numbers are meaningless, just that there are many flaws with ever trying to reduce something as complex as a nuclear fuel cycle or the levelized cost of a generator to a single number. But if one is going to do it - and quantification has certainly become pervasive in energy policy - I still maintain that looking at many studies is better than looking at just one. Otherwise, we're stuck in a battle over individual studies that benefits no one (since each study can be used to disprove the other).
I read your paper and find it quite interesting. You put a lot of work into it. I have a few questions and comments.
1 You should delete all references to Leeuwen. I exchanged several e-mails with him and it is clear that he is not engaged in a search for the truth. If you are interested I can copy them into a file and forward them.
2 Is you goal to produce a paper on; (A) The world?s historical emissions of CO2 from nuclear power plants, or (B) CO2 emissions from future Gen III reactors built in the U.S.?
Your calculation of capacity factor is consistent with A. To make the cost estimate consistent with A, average the actual construction cost of all plants built so far. I would expect a number around $1.00/watt.
Your use of U.S. Gen III construction cost estimates, the highest in the world, makes me believe that your objective is a paper that will be useful to policy makers deciding the future of nuclear power in the U.S., therefore your goal should be B.
To be consistent you should estimate the capacity factor of future Gen III reactors in the U.S.. Gen II reactors in the U.S. have ramped up from 50% in the 70?s to about 90% in recent years in spite of the fact that
Gen II plants are handicapped by an old decaying grid that experiences occasional outages requiring nuclear plants to throttle back or shutdown. Experts agree that we need to overhaul the grid to increase capacity and reliability, regardless of the energy source mix.
Gen III plants are Gen II plants that incorporate the lessons learned over the last 40 years. They have reduced complexity, inherently safe design features and vastly improved instrumentation and control systems, making them more reliable. With these improvements the most probable capacity factor for U.S. Gen III reactors is well over 90%, not 81 %.
3 U.S. Gen II plants were designed for 40 year lifetimes. Almost half have received license extensions to 60 years.
Gen III plants are designed for 60 years with possible extension to 80 or more years. The assumption of 30-40 year lifespan for future Gen III reactors is not appropriate.
You would not evaluate the future performance of wind and solar based on 1950-1980 windmill and solar cell designs. Nuclear power plant design has been frozen at an immature level for several decades, roughly equivalent to the DC-3 in aviation, but the DC-3 had the advantage of being factory mass produced. There is enormous room for evolution in nuclear power plant design and construction.
4 By far the biggest problem is the assumption that the energy mix does not change over the life of the plant. Most wind and solar emissions come before the first watt hour is produced, whereas half of the nuclear emissions are released after 20-40 years of operation. What are the odds that coal will be generating 50% of our electricity 20, 40, 60, 80 years from now? There is rapidly growing resistance to more coal in the U.S. and many existing plants are nearing end of life. Dr Hansen (NASA) believes we must get off coal soon. Of course in 80 years global cooling might be the big issue, but for now the up front CO2 loading of wind and solar is a disadvantage nobody is talking about.
The transportation mix is going to shift away from oil into natural gas, electric and biofuel, reducing fossil carbon/ton mile substantially. Do you distinguish between fossil carbon and recycled atmospheric carbon?
Accounting for these changes over the life of the plant will dramatically reduce average fossil CO2/kWh for nuclear plants, less so for other options with shorter life spans and higher up front emissions.
An easier method, yet still reasonable for comparison purposes, would be to assume that all electrical inputs are from the technology being evaluated.
Underground uranium mines are largely electric, open pit mines will shift toward natural gas, electric and biofuel, sea water uranium can eliminate mining.
Milling and enrichment are electric, cold war diffusion enrichment plants are going away. The U.S. is building two centrifuge enrichment plants and two more are in planning.
5 I do not support Yucca Mountain. 85-90% of all mined uranium is sitting at enrichment facilities ready for use in advanced reactors.
An 80 year lifetime supply of electricity from fission in second and third generation reactors using a once through fuel cycle leaves less than 10 pounds of heavy metal in spent fuel. The mass of heavy metal in existing spent fuel is tiny compared to the amount humans will use if nothing better than fission is discovered.
The time, man-hours, money and political capital required to implement reprocessing at this point is far too high a price to warrant saving this material. We should bury it under the deep seabed.
This solution is simple, very safe and requires little energy to implement.
6 If the U.S. ramped up to 80% nuclear and 20% wind/solar, the fossil CO2/kWh from nuclear power would be a small fraction of the paper?s estimate.
If the U.S. ramped up to 80% wind/solar and 20% nuclear, the fossil CO2/kWh from nuclear power would be a small fraction of the paper?s estimate.
Greetings?thanks for letting me know about your post. For some reason, the automated Scitizen system (which is supposed to email me whenever posts are made) didn?t. Allow me to engage a few of your responses in order:
1. The Leeuwen issue is indeed interesting. I?ve read his work, obviously, and have also read the many critiques of it and his response to those critiques. To be honest, I?m not sure who?s right, but for my study his work certainly deserved to be included since it met all of the criteria I noted (transparency, recency, etc.) in my paper. If we start excluding authors who are controversial, I?m not sure what we?d be left with, but the issue of Leeuwen certainly deserves raising.
2. Point well taken. Really the paper was not meant to be either A or B?I just wanted to see what the literature said about GHG emissions from nuclear plants?but in the end I suppose it ended mixing A and B up. This is because many of the studies analyzed mixed them up, with some looking at historical emissions in places like the US, and others looking at future emissions in places like Japan or Sweden. I think a more careful paper that does either A or B would be very useful, and if it did B, it would need to account for the high capacity factor of US nuclear plants that you point out.
3. The 40-60 year lifetime for newer plants is also a good point, and this is the first I?ve heard of it (much of the literature I?ve read says 20-40 years). Naturally, the longer nuclear plants operate, the lower their emissions per kWh from construction and decommissioning will be. Scarcer supplies of uranium could offset this improvement if more GHG are emitted to mine and enrich the uranium, but your point is valid.
4. I hope you?re right about coal, and as I?ve told many others on this website nuclear plants are far superior from coal plants for a variety of reasons. The amazing thing is that electric utilities in the US are still talking about adding huge amounts of coal and natural gas capacity in the coming years. Both the EIA and IEA, for example, project that by 2030 and 2040 fossil fuels will provide the SAME mix of energy services that they do today, if not more. So while I agree many of the shifts you talk about would indeed be welcome (and more efficient), I?m sceptical that they will occur.
5. Reprocessing opens up a whole other debate. Even if it is technically sound, there is always the issue of political acceptability and proliferation, as well as the money the DOE is already funnelling into Yucca Mountain. I would be very surprised if the US started reprocessing in the next decade.
6. If the US ramped up to 80% nuclear and 20% renewable, GHG emissions would greatly drop and society would improve. No argument there. But if the US ramped up to 80% renewable and stayed at 20% nuclear, the GHG savings would be even greater.
Regarding Leeuwen, perhaps two more criterions are in order, fair and impartial.
As an example, consider his analysis on extracting uranium from sea water, page 57 of this October 2007 report.
In this analysis the most recent work considered was published in 2001 by Sugo on polymer adsorption of uranium from seawater.
A? He ignores the more recent 2006 report, [Confirming Cost Estimations of Uranium Collection from Seawater- Assessing High Function Metal Collectors for Seawater Uranium ]
It describes the experimental results from testing improved technology. It concludes;
?The lowest cost attainable now is 25,000 yen ($280/kg) with 4g-U/kg-adsorbent used in the sea area of Okinawa, with 18 repetition uses.?
B? The latest technology uses braided rope like adsorbents that can be deployed in single strands or long loops for continuous processing. Leeuwen prefers to analyze the older technology of the 2001 study that packs the adsorbent in metal cages that are responsible for a large fraction of the cost and weight. The study describes three techniques to deploy the cages, one is much cheaper than the other two. Leeuwen focuses his discussion on the two most expensive options.
He assumes that the cages will be shipped to a distant reprocessing plant, running up the transportation emissions. In reality the cages would be reloaded at sea, as in the crab industry.
C? By eliminating the cages and moving the extraction process onboard the adsorbent deployment ship, the transportable product mass is reduced by a factor of a few thousand making those transportation emissions negligible.
D? Leeuwen likes big numbers. He describes a single plant that would generate 1/7 of the world?s uranium consumption.
E? After beginning his analysis of Sugo 2001 he introduces a great deal of superfluous information from old reports evaluating a different technology, for example.
?Estimates of the cost of deriving uranium from seawater range between approximately $1000 and $25000 per kg uranium.?
F?He does not complete his cost estimate of the Sugo technology, but includes the authors estimate of $280 / kg in a table with the older/higher estimates saying, ?The cost estimates by Sugo et al. may be low by a factor of at least 10.?
He ignores statements by the author indicating the potential for improvement, for example.
?The calculated stoichiometric chemical uranium adsorption capacity of this adsorbent is 500 g per kg adsorbent based upon the concentration of amidoxime groups?
?It is clear from experiments that metal adsorption rate increases roughly 3-fold above 10?C.?
Most new technologies improve with time and his own table, D29, shows that the estimated cost of sea water uranium has dropped an order of magnitude in 30 years.
G? He ignores the other valuable products that could be extracted along with uranium, offsetting a portion of the cost.
H? The adsorbent concentrates the uranium from 3.3 parts per billion in seawater to 4 parts per thousand in the adsorbent, a concentration factor of 1.2 million.
On page 55 Leeuwen has a flow diagram showing 5 waste streams. He writes;
?A five-stage process with an assumed yield of 80% of each stage, would have an overall yield of 33%. If each stage has a individual yield of 70%, the overall yield would be 17%. A rough estimate of the overall yield of a five-stage extraction process, excluding the first stage (adsorption from seawater), may be in the range of 20-30%.?
Any chemical process engineer who recommends a process that throws away 70%-80% of the product AFTER concentrating it by a factor of 1.2 million would be laughed out of the building, pink slip in hand. The so called waste streams would be recycled.
I? Most importantly he does not explain that at a few hundred dollars / kg the uranium cost per kWh is lower than our cheapest fossil fuel, coal. Even at 10 times Sugo?s estimate, the cost of uranium per kWh would be about the same as natural gas, but the cost trend for natural gas is going up while that for sea water uranium is going down. See two comments here.
J? Gen 4 reactors will reduce uranium requirements / kWh by a factor of 60-100. Gen 4 plants using sea water cooling could extract all their fuel directly from the condenser cooling water.
Leeuwen?s strategy is to create a straw man based on irrational assumptions that will result in the desired analytical conclusion, and then applying that conclusion to all possible designs, declaring them all impractical.
Here is a review of Leeuwen?s work by Roberto Dones that does an excellent job of pointing out flaws in his calculation of CO2 emissions.
Energy and climate change are not high school football games or practical jokes. Hundreds of millions of lives are at stake, and energy decisions should not be based on the caliber of Leeuwen?s analysis. Knowledgeable people understand this, and it reflects on people who treat his work as if it is fair and impartial.
Thanks for commenting. You raise some good questions about Leeuwen?s methodology and I certainly don?t have the expertise to validate or counter them. Note, too, that many of the critiques you raise motivated me to write my study on nuclear power and greenhouse gas emissions.
Since many individual studies have methodological flaws, the best strategy for determining the carbon footprint of the nuclear fuel cycle, given the limited resources available to me, seemed to be to sample as many studies as possible. What I find interesting, here, is that many studies give estimates in the same range as Leeuwen but use different methodologies.
Consider, first of all, a recent study done by Mark Jacobson after my own article, available at http://www.rsc.org/delivery/_ArticleLinking/DisplayHTMLArticleforfree.cfm?JournalCode=EE&Year=2009&ManuscriptID=b809990c. Table 3 calculates emissions from the nuclear fuel cycle at 68?180.1 grams of CO2/kWh.
Also consider, from my own study, that more than a dozen of the studies examined had an upper range of emissions for nuclear power plants above 60 grams of CO2/kWh.
I can understand Leeuwen possibly making mistakes, but what are we to make of these other studies? Are they all making the same errors?
?You raise some good questions about Leeuwen?s methodology and I certainly don?t have the expertise to validate or counter them.?
Yes you do. No expert knowledge is required to evaluate many of his decisions, such as;
A? He picked the most expensive option from the 2001 study to evaluate and did not mention the least expensive.
B? He did not complete his cost estimate of that technique. Instead he introduces some hand waving and an old figure based on different technology.
C? He ignored the 2006 experimental data based on a technique that eliminated the biggest problems with the 2001 technique and confirms a relatively low cost.
D? He claimed the estimate could be off by a factor of ten, but he did not provide evidence of that or mention that even if it is a factor of ten higher, it is still affordable.
E? He ignores the author?s evidence of potential for substantial improvement.
F? He ignores the fact that his own data shows that the price is coming down with improved technology.
Benjamin, you are qualified to evaluate all of these things.
Regarding his CO2 emissions calculation, the Roberto Dones study
provides a number of points that do not require specialized knowledge to evaluate.
A? The use of outdated references.
B? No consideration of co-production of minerals.
C? Use of energy intensive cold war diffusion plants powered by coal. They will soon be phased out. Not appropriate for discussions of future energy supplies.
D? The use of old centrifuge performance standard, 290 kWh/SWU instead of more recent performance, 35-62 kWh/SWU,
These are not errors or mistakes. They are meditated choices designed to produce the worst possible result for nuclear power that might pass inspection from a non technical reader.
Benjamin, you are qualified to evaluate these choices.
Knowledgeable reviewers know that Leeuwen?s agenda is anti nuclear. They know that people who reference his work are either anti nuclear themselves or have been duped by the scientific appearance of his work.
People who promote biased science may enjoy fame in the short run, but in the long run it always turns to infamy.
" 1? ?If the US ramped up to 80% nuclear and 20% renewable, GHG emissions would greatly drop and society would improve. No argument there. But if the US ramped up to 80% renewable and stayed at 20% nuclear, the GHG savings would be even greater.? "
Do you have an analysis to back that up? Corn ethanol?s only advantage is a slight reduction in oil imports, but from a CO2 standpoint, that is largely canceled by the added consumption of natural gas and coal in the process, so it is nearly fossil carbon oil equivalent. Cellulosic biofuel will be somewhat better, but they will still have substantial fossil carbon content, and the percentage of our energy needs they can meet is limited.
If we stopped burning fossil fuel completely, the fossil CO2 per kWh for nuclear power and all other surviving energy sources would be near zero. Fission is the only proven non fossil technology that can be scaled up to produce essentially unlimited quantities of reliable dependable baseload power, at an affordable price.
If nuclear plants are built in large numbers they will displace coal and natural gas electrical production, thereby reducing the fossil carbon content of all electricity, including the electricity that builds and supports nuclear power.
2? ?Consider, first of all, a recent study done by Mark Jacobson after my own article...
Table 3 calculates emissions from the nuclear fuel cycle at 68?180.1 grams of CO2/kWh.
Also consider, from my own study, that more than a dozen of the studies examined had an upper range of emissions for nuclear power plants above 60 grams of CO2/kWh.
I can understand Leeuwen possibly making mistakes, but what are we to make of these other studies? Are they all making the same errors??
In many cases yes, they are making the same errors. There is a lot of tail chasing going on. Jacobson references your study for example.
3? ?Since many individual studies have methodological flaws, the best strategy for determining the carbon footprint of the nuclear fuel cycle, given the limited resources available to me, seemed to be to sample as many studies as possible.?
Studies that are free of flaws will produce results that are clustered around the correct answer. The huge range of the study results included in your review, (1.4-288 gms/kWh), proves that at best only a few are in the right ballpark, free of large flaws.
By leaving so many deeply flawed and wide ranging results in your analysis, the results of any good studies present are watered down into oblivion by the results of the flawed papers. The resulting number you published, 66 gms/kWh, may be far from the cluster of correct answers. The obvious first step is to identify and eliminate the many individual studies with flawed assumptions and boundary conditions before averaging the results.
An alternative viewpoint is to see each study as the correct answer to a different question, depending on the boundary conditions and assumptions it is based on.
From this perspective, the first step is to decide which question we want to answer. The most important question is the one your paper is most often claimed to have answered.
?If we build new Generation III nuclear power plants in large numbers, how much CO2 / kWh will that release??
Each of the calculations in your study should be evaluated to see if it answers this question. For example, do they account for;
A? The fact that the fossil carbon content of electricity will go down dramatically over the next 60-80 years. Over 70% of our electricity comes from fossil fuel now. If we replace the fossil plants with a large number of Gen III nuclear plants, the CO2 per kWh will drop by a huge factor, and that will feedback into a further reduction of nuclear CO2 per kWh. If we do not build large numbers of nuclear plants the CO2 content of nuclear kWh's is irrelevant because it will have a minor impact on our problems.
B? Capacity factors above 0.9
C? 60+ year lifetimes.
D? Continuing modest improvements in fuel design with gradually increasing energy yield per ton.
E? Continuing modest improvement in decommissioning techniques including the use of advanced robotic technology likely to be available in 60-80 years when Gen III plants begin reaching end of life.
F? The fact that cold war diffusion enrichment is going away soon and centrifuge technology will continue to improve at a modest rate over the next 60 years. Laser enrichment may reduce enrichment cost further but need not be considered at this stage.
G? A rational approach to spent fuel. Recycling into Generation IV reactors or a simple, safe, easy, low energy consumption solution like deep seabed disposal.
After weeding out all the studies that do not meet these criteria you will be left with a small number of studies with results that are clustered within a narrow range of the correct number. Average those numbers and you will have a valuable result. I would expect it to be near the low end of the results you reviewed.
If none of the reports meets all of the criteria to answer the most important question, you are in a perfect position to create the first analysis that does.
As an added bonus Charles and Red can sleep soundly knowing that your career is secure.
When I said the US could ramp up to 80 percent renewables and 20 percent nuclear, corn ethanol wasn?t what I had in mind. Plentiful, cheap, abundant hydro, geothermal, wind, solar, and biomass are more like it. As for transportation fuels, I like PHEVs and a possible V2G future that relies on electricity, or second generation biofuels like, yes, cellulosic ethanol and also algal fuels.
If we maximized investments in energy efficiency, stopped burning fossil fuels completely, and shifted to commercially available renewables, I can show you a slew of studies that say that GDP would actually improve (along with public health), jobs would be created, and the economy would be more volatile energy markets. And nuclear?s ?affordable? price still involves exposing future generations to long lived radioactive waste and almost complete dependence on government subsidies.
Let?s not be disingenuous about Jacobson?s article. Yes, he cites my study but he doesn?t even use it when making his calculations about the 68?180.1 grams of CO2/kWh, nor does he rely on Leeuwen. There are also noticeable errors in a number of the conservative studies underestimating greenhouse gas emissions from the nuclear fuel cycle. Most of them?underestimating and overestimating?are no good, something I forcefully state in the conclusion of my article.
I don?t think the correct question to answer is ?If we build Generation III nuclear plants ?? Those plants may never be built, given the recent increases in the capital cost for nuclear power plant construction, public resistance towards siting and the transportation of nuclear waste, and the risk of proliferation and accident (and no matter how many times we go back and forth about these issues in Scitizen, people will still believe what they want to believe). The better question, for me, is ?what are the greenhouse gas emissions associated with the current lifecycle,? the plants that will be operating for the next few years, the ones that are competing against existing generators. And here, I see a number of advantages in favor of wind, solar, etc.
I?m all for living in peace and harmony with Charles and Red. Perhaps when they embrace the beauty of wind energy?
I noticed you did not comment on Leeuwen?s anti nuclear assumptions and omissions, any thoughts?
I count 13 interesting issues raised in your last response, I have thoughts on many of them, but the subject of the essay at the top of this page is CO2 from nuclear power, so let?s stick with that.
1? True or False. Studies that are free of significant flaws will produce results that are clustered around the correct answer. The huge range of the study results included in your review, (1.4-288 gms/kWh), a ratio of 206:1, proves that at best only a few are in the right ballpark, free of significant flaws.
2? True or False. By leaving so many deeply flawed and wide ranging results in your analysis, the results of any good studies present are watered down into oblivion by the results of the flawed papers. The resulting number you published, 66 gms/kWh, may be far from the cluster of correct answers.
3? True or False. If we stopped burning fossil fuel completely, the fossil CO2 per kWh of nuclear power would be near zero.
4? You want your preferred technologies judged on their anticipated future performance. But you compare that with the past performance of old nuclear power technology operating in a fossil dominated world. Why the double standard?
Consider these two questions.
A? What are the CO2 emissions associated with the current lifecycle of existing Gen II nuclear plants?
B? If we replace our fossil power plants with Gen III nuclear plants, what are the CO2 emissions / kWh associated with the lifecycle of those new plants.
5? Which of these questions is most important to the future of the human race? Our Gen II reactors were designed in the 60?s and built in the 70?s ? 80?s. We are not going to build more of these reactors, or more Titanics or more Model T Fords. Do you compare the performance of 60?s reactor technology with the performance of 60?s model windmills, solar energy systems, geothermal and biomass technology? Almost none of them are still working, and the comparison would be meaningless for the future.
6? True of False. We are not going to build more Gen II reactors. For purposes of selecting future energy policy, emissions from Gen II reactors are as irrelevant as the emissions from old second generation windmills. The decision to build new Gen III reactors should not be based on the performance of old reactor technology operating in the old fossil dominated world.
7? True of False. The decision to build new Gen III reactors should consider their most likely lifecycle emissions, taking into consideration the most likely environment in which they will operate over their lifetime.
8? Benjamin, you have candidly acknowledged that A is the, ?better question, for me?. For those of us looking for the answer to B, will you acknowledge that your report does not answer question B?
9? If you do not answer ?yes? to question 8, what do you estimate is the probability that 66 gms / kWh or more is the correct answer to question B, and what is your basis for that probability?
Thanks for another thoughtful post. I didn?t comment on Leeuwen?s assumptions because I would need to double check all of the claims that you made, and I simply don?t have the time. I suspect you could make a similar laundry list for almost any of the lifecycle studies of nuclear power. They all have certain assumptions and exclude certain data.
To your questions, though:
1. Not sure that studies free of ?flaws? produce clustered results. Sometimes, ranges really do just exist. This is precisely the case with nuclear power: some of the most efficient plants in places like Sweden do have very low emissions, while some of the least efficient plants in China have very high emissions. It?s also the same with climate change; negative damages in some countries will be in the millions of dollars, while in others billions of dollars. Such range and complexity are part of life, and I suspect the range I give in my article is much closer to the truth than the numbers from any one study.
2. We haven?t established that most of the studies I have analyzed in my analysis are flawed. We?ve established that there are some questions about Leeuwen, but as I point out in the article there are questionable assumptions made by almost every single study, including the ones with lost estimates that you and Charles keep citing in your blogs.
3. If we stopped burning fossil fuels completely, and even used nuclear power plants or renewable power plants to create the electricity needed to enrich uranium etc., I do agree the c02 per kWh for nuclear would decline. But I suspect it would still be much higher than the c02 per kWh from other sources such as energy efficiency or renewables.
4. Let?s be careful about the past versus future discussion. The best metric I always try to use, so we compare apples with apples, are marginal costs, the cost or emissions profile of building the next power plant. Marginal costs are influenced both by past costs and future performance, so they provide a nice blend of both. Now, I am willing to admit that the marginal lifecycle c02 emissions for a new Generation III nuclear power plant would probably be lower than the 66 g/kWh that I report, since my study relied on a sample that included older less efficient reactors. But they would also increase over time as uranium becomes more energy intensive to mine, and my sample also included some estimates that had future Generation IV reactors that were presumed to be highly efficient (but do not yet exist). So I don?t think the 66 number is that far off, but I agree more data would be useful. As an FYI I?ve had discussions with the IEA and IAEA about doing a study that would either (a) take existing reactor performance for every reactor to determine exactly what its emissions profile was last year, so we could get an average number and also identify best practices or the cleanest nuclear power plants or (b) estimating the emissions from new nuclear reactors just about to get built but using a large enough data set to account for the changes in mining, enrichment, etc. world wide. Both times they said either study would be very useful, but they have neither the funds nor the interest to complete it. So we?re left with either studies that make a ton of assumptions and then produce estimates based on a very small sample size (giving great ranges in estimates), or my study which looked at them all to get an average.
I agree. This is exactly what I meant by;
?An alternative viewpoint is to see each study as the correct answer to a different question, depending on the boundary conditions and assumptions it is based on.?
Consider two questions. A? What is the mass of planet Mercury? B? What is the mass of planet Jupiter? If we are asked question C? What is the mass of earth, we would get the wrong answer if we averaged the answers to A and B. Averaging the answers to many different questions does not provide the answer to yet another different question. So I will refine my question.
1? True or False. Studies that are free of significant flaws, and answer the SAME question, will produce results that are clustered around the correct answer to THAT question. The huge range of the study results included in your review, (1.4-288 gms/kWh), a ratio of 206:1, proves that at best only a few are correct answers to the following key question.
?If we replace our fossil power plants with Gen III nuclear plants, what are the CO2 emissions / kWh associated with the lifecycle of those new plants.?
Vattenfall generates electricity in Sweden. It gets 37.5 % of its electricity from hydro and 61.7 % from nuclear. Only about 1/4 % comes from fossil fuel. The very low fossil carbon content of Vattenfall?s electricity makes the fossil content of its nuclear kWh's very low.
The lifecycle CO2 emissions of Vattenfall nuclear power is only 3.5 gms CO2 per kWh, 5% of your reports number. The lifecycle CO2 emissions of Vattenfall wind power is 10.5 gms CO2 per kWh, three times higher than nuclear.
If the U.S., or any country, replaces its fossil power plants with Gen III nuclear plants, the fossil carbon content of its kWh's will also be very low. Combine that with the effects of longer life, higher capacity factor, improved construction techniques and more efficient enrichment capacity, and the CO2 per kWh of nuclear power will be lower than it is in Sweden now.
3. If we stopped burning fossil fuels completely, and even used nuclear power plants or renewable power plants to create the electricity needed to enrich uranium etc., I do agree the c02 per kWh for nuclear would decline. But I suspect it would still be much higher than the c02 per kWh from other sources such as energy efficiency or renewables.
2? Why is Vattenfall wind power CO2 per kWh three times higher than nuclear? Windmills use much more steel and concrete per kWh than nuclear, and those emissions are almost all up front, before the first kWh is generated.
The question was ?True or False. If we stopped burning fossil fuel completely, the fossil CO2 per kWh of nuclear power would be near zero.?
The answer is obviously true, Vattenfall has already gone a long way towards this goal. If we leave the fossil carbon atoms in the ground in the form of coal, oil and natural gas deposits, the fossil CO2 emissions of nuclear power, and any other surviving energy source, would be approximately zero. If necessary we can even make nuclear power carbon negative by using some of the energy to extract CO2 from the atmosphere.
Now, I am willing to admit that the marginal lifecycle c02 emissions for a new Generation III nuclear power plant would probably be lower than the 66 g/kWh that I report, since my study relied on a sample that included older less efficient reactors. But they would also increase over time as uranium becomes more energy intensive to mine,
That conclusion assumes that the fossil carbon content of energy used for mining remains the same as it is now. As the fossil carbon content of energy goes down, as it has in Sweden, the fossil carbon content of uranium will go down.
The nuclear fuel cycle is largely electrically powered now and the electrical fraction is likely to increase. Those aspects that cannot be converted directly to electricity can be converted to nuclear generated hydrogen or low fossil biofuel.
Extracting uranium from seawater using ships anchored in the Gulf Stream and Black Current, powered by water turbines in the current, can provide fossil carbon free uranium for hundreds of years using Gen III reactors, and billions of years with Gen IV reactors.
People might ask, ?why not use the water turbine power directly on the grid?? The answer is that the uranium will generate several orders of magnitude more power than the ships will use extracting the uranium from the adsorbent.
I?ve had discussions with the IEA and IAEA about doing a study that would either (a) take existing reactor performance for every reactor to determine exactly what its emissions profile was last year, so we could get an average number and also identify best practices or the cleanest nuclear power plants or (b) estimating the emissions from new nuclear reactors just about to get built
Benjamin, I admire your brass. It takes balls to ask the nuclear industry for money given its unfavorable treatment by your papers. You should write a fair and balanced report that includes all the points that Red, Charles and myself have made. I think they would scrape up some cash for that. Of course you would cut Red, Charles and myself in for a piece of the action as coauthors, right?
We are still living in the age of fossil fuel. When we get serious about leaving fossil fuel in the ground, the fossil carbon emissions associated with nuclear power will be near zero.
One of the really excellent properties of nuclear is that is is very dense, requiring almost no footprint (on a per-megawatt basis) compared to anything else we know about. With nuclear, we can generate around 2500MW in hundreds of acres of land. All but a very few acres (on the order of 10's) are safe for native wildlife. Think how starkly different it is for wind, solar, and hydro where the enormous footprint is inviable for wildlife habitat (or human life). Even if nuclear produced marginally more CO2 than wind or hydro (I'm struggling with the assertions made on this) the environmental impact and safety of human life should greatly outweigh the difference in CO2 production. Look at Bureau of Labor Statistics data, the statistics are on the order of 1 lost time accident per 200,000 hours. Office workers have that low a rate. Nuclear power is simply safer to humans than the alternatives (for example, agriculture is one of the most injurious and deadly occupations making biomass a bad choice for human safety).
Finally, think about France where almost all electrical generation is nuclear. They are not fighting wars (another indirect source of fatalities and injury) to preserve a fossil fuel supply. They have clean air, effectively no CO2 production. Their mass transit system doesn't spew out particulates, SOX, NOX, etc., because they have a clean source. Just think about how much more pleasant (and less deadly) Los Angeles would be if people were using a mass tranist system run on electricity from a nucelar power plant.
1)How useful is it to take the average emission of nuclear plants around the world? Would it not be better looking at the average emissions for geographic regions and then making a comparison. i.e. It would be unfair to compare the emission from solar in California to nuclear emissions in Finland. One should be comparing solar emissions in Finland to nuclear emissions in Finland. Or solar vs nuclear in California, in my opinion this is many times more informative than world averages.
2)Are emissions for enrichment included? To my knowledge there is no need for fossil fuels to be used in the enrichment of uranium, electricity from a nuclear plant could be used for enrichment. Thus I think the proper method to factor in enrichment in CO2/kWh analysis would be to subtracted the energy needed for enrichment from the energy created by the power plant. Instead of assuming that the enrichment facility's electricity comes from coal and adding its portions of CO2 emissions to the total emissions from the power plant.
3) Also you mention the mean emissions from nuclear in your article, would it not be more informative to tell us the medium emissions in this article instead?