Editorial note: I received an interesting note in my inbox from Dr. Benjamin Sovacool.
Hi Rod/Atomic Insights, I believe you already know me as I see
multiple posts attacking my research. One of the most recent ones is
from Paul Lorenzini. Given that this post has a number of factual
errors in it (and it misrepresents both my data and that of others),
would you permit me to write a response? I realize we’re on
different “sides” of this issue, if you can even call it that, but
hopefully that won’t prevent you from letting me correct misstatements.
The post to which he is referring is Nukes kill more birds than wind?.
Benjamin’s note and the subsequent email exchange led to my acceptance of the following guest post. I have published it without modification in hopes of generating continued worthwhile conversation. All opinions expressed in this post belong to the author, not to me or Atomic Insights. End editorial note.
By Benjamin Sovacool
Nukes, Birds, and Better Arguments
As the author of these two studies http://dx.doi.org/10.1016/j.enpol.2009.02.011 and http://dx.doi.org/10.1080/1943815X.2012.746993, I wanted to the engage the discussion taking place on Atomic Insights—especially to correct some errors and half-truths in Paul’s piece, to explain the methods used in mine, and to point everyone to newer analysis that reflects some of shortcomings inherent in my initial 2009 article.
First, and most important: Paul has misstated the actual conclusion from my original study. It never advances the conclusion, as he claims, that “nuclear power causes more bird kills than wind.” Instead, the article has three conclusions, one primary, and two secondary.
The primary conclusion was to highlight the need for better data and analysis on the issue of avian deaths and energy technology—inclusive of not only nuclear energy but also wind turbines and fossil fuels. This conclusion is something I think Paul and most readers would agree with. Here is one of the paragraphs straight from that section of the article:
Far more detailed, rigorous, and sophisticated analysis is called for that takes into account the complexities of the wind, fossil-fueled, and nuclear energy fuel cycles. The shortcomings of this preliminary study are as obvious as they are numerous: a focus on bird deaths but not bird births; a small sample size for wind, coal, and nuclear facilities that may not be representative; a focus on individual species such as the wood thrush or waterfowl to produce overall estimates of avian mortality that are definitely not representative (and undoubtedly conservative); a presumption that coal was only mined using mountaintop removal (thereby excluding the impacts from other types of coal mining); fatalities that happened on particular days and weeks that were then presumed to be the only ones throughout the year (also resulting in conservative estimates); an assumption that only carbon dioxide emissions from power plants contribute to climate change (again conservative for excluding other greenhouse gases); highly uncertain deaths attributed to climate change that may be prevented if future greenhouse gas emissions are significantly reduced.
Despite the unnecessarily vitriolic nature of Paul’s commentary, my own study was intended to start a discussion and lead to better research.
So imagine my surprise when Paul argues we need better numbers and analysis, when this is exactly the point I make after collecting hundreds of studies in the academic literature on the topic and finding most of them deficient in some way. It is also somewhat disingenuous to attack me for the “numbers being wrong” when the study already tells you the numbers are very rough estimates that need to be improved. I even explicitly state this, as well, in the conclusion: “the rudimentary numbers presented here are intended to provoke further research and discussion,” in the abstract “this paper should be respected as a preliminary assessment,” and in the title of the study, which has the word “preliminary” in it.
One secondary conclusion is that visible impacts for a given energy system may not be the most important—we can see birds crashing into wind turbines, but we often don’t see their nests degraded by the manufacturing plant that made the blades, witness them crashing into the fossil-fueled smokestack, or picture them drinking contaminated water near a uranium mine or mill. This challenges us to make “visible” previously “invisible” parts of our energy infrastructure. While we can disagree about the precise numbers—and I hedge this by underscoring that the numbers in my study are preliminary, first-order guesses that need followed through with other research—those impacts are nonetheless there.
A final secondary conclusion is that if there is a real “bird killer,” it is neither wind energy nor nuclear power but coal and fossil fuels, especially if you factor in climate change. As I state in the conclusion, again, “fossil-fueled facilities are about 17 times more dangerous to birds on a per GWh basis than wind and nuclear power stations.” I’m grouping in nuclear and renewables here together; I’m not saying we need to phase out one in lieu of the other just for the sake of birds. This secondary conclusion is also consistent with how I rank preferred energy options: I believe first in energy efficiency, then with renewables, then nuclear, moving up (to lower preferences) to natural gas, oil, and coal. This is very similar to the idea of “climate stabilization wedges” advocated here http://dx.doi.org/10.1126/science.1100103.
So, contrary to those who label me some fierce anti-nuclear zealot, my position is more complex than simply “pro” or “anti,” which readers would know if they take the time to read my recent book from MIT Press entitled Global Energy Security (which also argues in favor of a moderate role for nuclear power, you can order it here http://mitpress.mit.edu/books/climate-change-and-global-energy-security). I am not entirely against all forms of nuclear energy in all circumstances, either, just critical of doing it in centralized power plants susceptible to cost overruns, in ways that threaten the energy security of countries (think Fukushima or Chernobyl), and in ways that fail to adequately handle the issue of nuclear waste and future generations. In this way, I am also critical of poorly sited wind farms (see http://dx.doi.org/10.1016/j.tej.2011.09.019), of placing hydroelectric dams in rainforests (see http://dx.doi.org/10.1016/j.enpol.2011.06.035), and of large-scale solar energy projects that displace nomadic herdsman (see http://dx.doi.org/10.1016/j.renene.2012.07.027 and http://dx.doi.org/10.1016/j.renene.2012.07.039).
The biggest criticism I therefore have of Paul’s piece is that it misstates my findings, its motives and intentions, and the way that it contextualizes its numbers as highly uncertain.
However, there are a few other problems with Paul’s article.
As one example, he writes that that I confuse a uranium mining operator in Colorado with one in Wyoming. This is false: on p. 2245, I clearly state that the Canon City mill is in Colorado: “Uranium milling and mining can poison and kill hundreds of birds per facility per year. Indeed, in early 2008 the Cotter Corporation was fined $40,000 for the death of 40 geese and ducks at the Canon City Uranium Mill in Colorado.”
Furthermore, as for the claim about mines in Wyoming, Paul slightly misrepresents what both my study and the U.S. Fish & Wildlife Service (FWS) state. I never claimed Berkeley pit was a uranium mine, I instead said that open pit mines, similar to those at Berkeley, can kill ~300 birds per year because selenium can leech into water supplies. The US Fish & Wildlife Service clearly makes this argument linking mines like Berkeley with uranium mines, as readers can check for themselves at http://www.fws.gov/mountain-prairie/contaminants/contaminants8.html. Here’s a most relevant paragraph:
Abandoned open pit uranium mines in Wyoming also have formed pit lakes. Uranium-bearing formations are usually associated with strata containing high concentrations of selenium. Pit lakes formed in open pit uranium mines can contain very high levels of selenium in the water. One pit lake in Wyoming contains over 100 parts per million (microgram per Liter [µg/L]) of selenium. Waterborne selenium concentrations greater than 2µg/L are known to impair the reproduction and survival of aquatic birds due to the high potential for dietary toxicity through food chain bioaccumulation.
Now, it is true the ~300 number is a rough guess and a proxy, but I admit this in the article—and I was unaware of any other data looking at bird deaths and open-pit mines.
Moreover, readers should not presume that uranium mines in the USA always strictly adhere to environmental, air, and water regulations, nor should they pretend that mines in places like Namibia or Kazakhstan adhere to strict regulations at all (so globally, there is some evidence that the frontend of the nuclear fuelcycle is worse for birds than in places like the United States).
Now, why did I produce different numbers in 2012? This is admittedly a legitimate question from Paul. The answer is that feedback from the peer-reviewers—yes, it was peer-reviewed by three anonymous referees for the Journal of Integrative Environmental Science—suggested that the Canon City mill estimate was a “one-off” event, an accident, whereas the issue of water contamination at mines (noted in the FWS article above) was more chronic. So they instructed me to remove the Canon City estimate, which left me only with Wyoming.
Upon reflection, for my 2009 study, I probably should have added the numbers from Colorado and Wyoming together because they reflect different parts of the nuclear fuel cycle—mining, and milling/enrichment—instead of taking the average between them. So in that regard, both of my studies underestimate potential deaths from the frontend of the nuclear lifecycle, even though, again, I admit the numbers are very rough.
As for the issue with Florida Power Corporation’s Crystal River Generating Facility, here I admit the data is sketchy and thank Paul and Rod for pointing out a likely error: Florida Power Corporation’s Crystal River Generating Facility did indeed used to be (or perhaps still is) a mix of coal and nuclear units, and when I read that 3,000 bird deaths occurred from colliding with smoke stacks, I presumed that it was the same as if they were hitting nuclear cooling towers. After reading what Paul and Rod have written, this looks to be false. I was working with the best data I could find from the library at the National University of Singapore, and I’ve never visited the Florida facility, so this was an honest mistake, but it was a mistake. These bird deaths should, instead, be attributed to fossil-fuels, and subtracted from the nuclear numbers.
Same with the issue that a reader—Laurence Aurbach—pointed out indicating that Biewald gives different numbers later in his testimony. This was not in the printed version of the testimony I had available to me—that version ends five pages earlier than the one that Aurbach was able to find online. The new figures from Biewald should also be averaged with his earlier, higher numbers, again, lowering the ultimate numbers for nuclear power.
However, this does not mean that cooling towers pose no threat to birds. The attributes of smoke stacks that draw birds to them hold true for cooling towers: both are tall and large (the smoke stacks in question were 152 to 183 meters tall, whereas nuclear cooling towers are 100 to 200 meters tall), and far more important than height is lighting. The Maehr study from Florida, for instance, argues that “lighting appears to attract migrating birds under overcast conditions,” thus contradicting Biewald’s claim that lighting allows birds to avoid towers. Crystal River’s nuclear cooling towers would have lighting, as would practically every other unit in the USA, and most towers are even wider than the narrow smokestacks, so it’s fair to say they could attract birds in the same manner.
Turning to the “critique” from Willis et al., and their number of 1.46 deaths per GWh (let’s not add bats into the mix, because none of my estimates for fossil fuels or nuclear include bats—we need to compare apples to apples), Paul ignores some serious flaws with their analysis, which have also be rebutted in the peer-reviewed literature by me here http://dx.doi.org/10.1016/j.enpol.2009.08.052. (He must have missed this rebuttal even though it was published side-by-side the Willis et al. article). Here is an excerpt from the rebuttal:
[Willis et al.] present adjusted fatality estimates for wind energy but not for nuclear power and fossil fuels, [and] confuse installed capacity with actual electricity generation … the adjusted fatality estimates for wind energy offered by Willis et al. create a distorted picture when compared with the unadjusted fatality estimates for these other sources.
For instance, Willis et al. confuse MW (installed capacity) with MWh (energy), which means they do not take into account the improved capacity factor of wind turbines, which is why their number cannot be trusted. They also, among other mistakes, ignore possible flaws with carcass reinsertion—when predators around a wind farm wishing to shelter there can bring birds into it that were killed elsewhere.
Also, and highly relevant here, is that newer research that I have undertaken with Donald McCubbin using modeling tools from the U.S. EPA, and looking at wind farms in California and Idaho, shows that bird deaths are not static (see here http://dx.doi.org/10.1016/j.enpol.2012.11.004). They change over time. For instance, looking at the Altamont pass in California, assuming production of roughly 1,000 GWh/year, we posit a bird mortality rate due to collisions at 2.5 to 13.8 deaths/GWh per year during the period 1987-2006. These numbers are much higher than even Willis et al.’s and my earlier studies, which used data from 6 wind farms but none in California. But from 2012 to 2031, we predict a range of 0.5 to 1.0 deaths/GWh per year for Altamont, lower figures due to the improved efficiency of newer wind turbines. And in Sawtooth, Idaho, the range is 0.1 to 0.5 deaths/GWh, due to smaller turbines and even better siting.
Thus, neither my original studies nor the new one with McCubbin are “junk science.” If they were, they never would have passed peer-review. They are, instead, admittedly preliminary estimates, and they are upfront about bracketing their findings with a number of caveats and calling for better research. In that regard, this discussion on Atomic Insights is proof that I have succeeded, though perhaps not in the way I had imagined.
But I am after the truth here: I did my initial study in 2009 from a desk in Singapore. If readers have better data on the impacts of uranium mining, enrichment, construction, operation, etc. and birds and bats, send it to me. Let’s improve the numbers so we can all make better informed decisions about energy options, which is what the piece was intended to contribute towards.
In closing, however, I find it ironic that Paul attacks me for being lazy and having “ulterior” motives, but then commits a number of distortions that, while I would not call them “malicious” and “dishonest,” are certainly selective in how they portray my article and its findings. Could it be that because Paul works (or did work) for a nuclear energy company, he also has an incentive to skew the numbers?
I am also struck by both the ferocious tone and inaccuracy of the assertions presented in a number of comments in the section below the article. Here is a small sample of false assertions:
- that “UVM must be proud” (I work for the Vermont Law School, which isn’t part of UVM);
- that nobody could confirm if the article was peer-reviewed (you can easily confirm this by emailing the author or the editor);
- that I am funded by some shadowy group of “501(c)(3) organizations” (I’m not);
- that I am a “fraud” (this would require me being convicted of some type of scientific misconduct, something that absolutely hasn’t happened);
- that if I “get my way, any surviving raptors will find the planet nuclear-free, but having to dodge the industrial (sic) sprawl from legions of lethal diffuse-energy harvesting structures” (where do either of the studies ever say anything close to this?);
- that my greenhouse gas numbers from another study are baseless (I can point to two follow up peer-reviewed studies, available http://dx.doi.org/10.1016/j.enpol.2009.06.073 and http://dx.doi.org/10.1111/j.1530-9290.2012.00472.x, that give mean numbers almost identical to mine).
These statements from Atomic Insights readers are all untrue, and could easily have been checked. It is sardonic that so many readers throw around baseless claims in an article from Paul trying to prove somebody else is making baseless claims. I ask that you each hold yourselves to the same standard you held my article to: having high standards means they have to cut both ways. Think, and check the facts, before you post.
In the future, if anybody, Paul included, has questions about this study or any study of mine, email me – my email address is right there in both articles, and my profile is readily available for all to see on the Vermont Law School faculty page. Perhaps if I had emailed Paul or Rod before publishing my article, I would have caught the mistakes relating to Crystal River and Biewald. And if Paul or Rod had emailed me, I could have explained how I was using the FWS numbers, pointed out the flaws with the Willis et al. study, and directed them to more careful, follow-up research. Sometimes collaboration is most important with those you disagree with.
Ultimately, maybe the central lesson here is that we need to start working together—you “nuclear” folks and us “renewables” folks—if we’re not only to catch errors and mistakes, but also to tackle the culprit we all seem to be after, which is fossil fuels.
Associate Professor of Law
Vermont Law School