Unintended Anti-Nuclear Consequences Lurking in the EPA Clean Power Plan

By Remy Devoe

The Environmental Protection Agency’s proposed Clean Power Plan has gained favor with some nuclear energy advocates. An extensive analysis of the proposal, however, reveals that current nuclear generating capacity would largely suffer under the new carbon rules. In fact, the results of an evaluation performed by my fellow graduate student Justin Knowles and myself show that 15 states are actually incentivized to shut down all of their nuclear units and replace them with natural gas combined cycle (NGCC) generation. In effect, this plan allows for increasing carbon emissions; a far cry from the stated goals of the Clean Power Plan.

We conducted our analysis after learning about the plan in July while participating in the Nuclear Engineering Student Delegation in Washington, DC. While in Washington, EPA representatives explained to us how nuclear energy was considered in the rule, but admitted that only a small portion of current nuclear energy generation would be credited in a state’s emissions rate. The EPA contacts we met with encouraged us to submit a comment of what changes we would make and analyses to support these recommendations. We have been working to understand this plan since then, and intend to submit our analyses with our comment.

Reading the entire 130-page rule is a daunting task, but the root of our concern can be found in the section titled “New and Preserved Nuclear Capacity” (page 34870 of the Federal Register), which states that current nuclear generation is given 5.8 percent credit for replacing fossil-fuel energy. While seemingly arbitrary, this figure comes from an Energy Information Administration (EIA) report that states that six reactors in the United States (equivalent to 5.8 percent of U.S. nuclear generation) are at risk of being shut down. The EPA recognizes that keeping current nuclear generation is the only way this plan will be able to achieve its goal, but erroneously attributes only 5.8 percent of the energy produced from nuclear plants to calculating a state’s emissions reduction goal.

The current rule regulates emissions through a state’s carbon intensity in lbs/MWh using an equation developed by the EPA for this specific purpose. (For example, the EPA provides a sample calculation for Ohio). As you can see, only fossil fuels, renewables, 5.8 percent of current nuclear generation, and 100 percent of nuclear presently under construction are used in this calculation. The plan then outlines a Best System of Emissions Reductions (BSER) used to calculate what carbon intensity a state can attain if they implement emissions reduction practices. This new carbon intensity is the goal that each state must meet by 2030.

Since current nuclear generation is only valued at 5.8 percent of its energy generation, the loss of one plant in a state has only a marginal effect on a state being able to achieve its goal under this standard. If this rule is intended to be a carbon regulation, then all energy sources should be valued based on their emissions and no technology should be given preference over another. Renewables, coal, natural gas, and others are given credit for 100 percent of their current capacity; nuclear energy should be no different.

UT students at the EPA Clean Power Rule hearing

(left to right): UT students Daniel Tenpenny, Gregory Meinweiser, and Remy Devoe at the EPA public hearings in Atlanta

Following these revelations, a triumvirate of three students from the University of Tennessee ANS student section attended the EPA hearing in Atlanta to share our comments on the new carbon regulations and draw attention to the subject. We were very fortunate that each of us was allowed to provide public comments, and the EPA responded by requesting detail on our analyses and an official comment on its plan. Afterwards, one of the panelists requested a conference call to clarify our points and asked for a personal copy of our analyses.

Below you can watch each of our public comments:

Daniel Tenpenny

Remy Devoe

Greg Meinweiser

To create a fully developed comment, we—with the aid of our advisor, Dr. Steven Skutnik—used the data provided by the EPA to perform our analyses. We simulated a hypothetical scenario in which all nuclear plants were shut down and their generation replaced by natural gas combined cycle units. The results of this analysis were astonishing. By crediting nuclear at only 5.8 percent of its generating capacity, 15 states were shown to have lower emissions rates under the rule as currently proposed when all nuclear generation was replaced by NGCC—a clear indication that the EPA’s method of emissions calculations is flawed. By valuing only a fraction of current nuclear generation, utilities are incentivized to shut down nuclear plants in favor of natural gas, the exact opposite of the EPA’s stated intent with this plan. Our analysis has exposed a perverse incentive for states to allow the retirement of carbon-free nuclear generation for replacement with carbon-emitting sources.

We must insist that the EPA considers the total generation from all energy sources in calculating carbon emissions intensity. If the EPA gives nuclear energy its full due, then every reactor in the nation must keep running or be replaced with other clean energy sources for a state to meet its goals, making currently operating nuclear units all the more valuable to states. We can make this happen, but only if the nuclear community rises to this urgent challenge, rallying together to push for a fairer, more effective rule that credits current nuclear generation at 100 percent of its current capacity in state-level emissions goals.

I urge each and every one of you to take a look at the Clean Power Plan and submit a comment. The comment period on the carbon rule is open until October 16, 2014, and the final rule will be implemented in June 2015.

coal and nuclear

Head ShotRemy Devoe is a graduate student in Nuclear Engineering at the University of Tennessee-Knoxville. He is currently working towards his Master’s degree in nuclear engineering and plans to pursue a PhD in the same. His research focus is in nuclear fuel cycles and used fuel management. He is currently the Vice-President of the University of Tennessee American Nuclear Society Student Chapter.

19 thoughts on “Unintended Anti-Nuclear Consequences Lurking in the EPA Clean Power Plan

  1. Ashley Finan

    Remy and Justin,

    Is there an explanation for why Kansas’ emissions don’t change when they replace nuclear with gas in your google docs spreadsheet?


  2. SKV

    Remy and Justin,
    Thanks for delving into this issue and ferreting out the enormous error. I agree with the others that an error of this magnitude hints at nefarious intentions. The current approach is significantly damaging to current (and future) nuclear capacity and does not make any sense. I just posted my comment on the proposed rule citing your work. Thanks again.

  3. Remy Devoe

    @Mac Cook

    You are correct in the majority of your assumptions except that hydro is not included at all in the CPP goal setting. You are correct that the analysis we have done is on the goal setting for 2030, the EPA is not mandating a certain energy mix, and the general language is favorable for nuclear. We do not believe that there was or is any ill intent on behalf of the EPA towards nuclear energy, but that does not mean that this error should be overlooked.

    You have highlighted a point that has been brought up to us by a few other colleagues that we do not know exactly how the EPA will determine compliance. There is a whole document on compliance that explains the use of meters and sensors on fossil-fuel plants, but does not explicitly say as far as I read that they will use the same equation used to set the state goals.

    The only reasonable way to measure compliance would be to use the same metrics used to assign the goals to a state. Because hydro and 94.8% of nuclear wasn’t included in the goal setting states would automatically have an advantage and most likely already have met their goals. The point of the CPP is not to spend all this time and money calculating costs and goals just to say your state has already met it’s goal. Every report shows that this plan will affect all states and the cleanest states the most.

    The point our analysis shows is that crediting 5.8% of nuclear generation will not “preserve” them and that in essence it’s like not crediting them at all. If the EPA doesn’t include nuclear in it’s carbon intensity calculations then a utility can shut a reactor down without consequence to a state’s carbon emission reduction goals.

    You also brought up a very good point that the states with reactors under construction now are faced with a liability if the reactors are not completed. The same would be true if current reactors are given 100% credit. This may be why we haven’t heard much from utilities because using 100% would mean each unit is a liability and that if it shuts down a new reactor must be built to replace the lost capacity. This may be why utilities have not been vocal on this issue besides Exelon.

    Thank you for your comment and to answer your question I would ask the EPA to do a straight forward calculation of carbon intensity using ALL sources. The rest of their plan would still work except there would be no gaming. Also nuclear is a cost effective means of producing electricity and should be used to replace coal and natural gas and not the other way around.

  4. Mac Cook

    Hey Remy & Co.,

    Thanks for the information and the helpful links! If you’re busy, don’t worry about reading or responding to this – I completely understand.

    I’ve been trying to interpret what the EPA is meaning by much of their policy and the proposed rule.

    The first clarification I’d like to make is that we are discussing the “Calculation of Carbon Emission Goals for States for the Year 2029 (or alternatively 2024), and the year 2030 and beyond (or alternatively 2025 and beyond).” These goals are set on a state-by-state basis based on each respective state’s current (2012) carbon emissions data. The purpose of the individual carbon emission goals is to create an “obtainable” target for each state in the near to mid future and beyond, based on the 2012 data. Please correct me if any of this appears to be incorrect.

    I am under the impression that these goals do not force a state to pursue a unique mix of energy production technology. These goals can be met by the state using any methods that the state chooses in accordance with the rules outlined in the documentation.

    Okay, so first I’ll state how I interpret the “5.8% rule” from Section VI.C.3

    From reading the accompanying documentation (available again here: http://www.gpo.gov/fdsys/pkg/FR-2014-06-18/pdf/2014-13726.pdf#page=42), the EPA seems like it is being friendly to the nuclear industry as it discusses this “5.8% rule” that it proposes based on data from EIA.

    The 5.8% reflects an industry-averaged “risk” associated with losing currently operating nuclear power plants. Instead of attempting to quantify the actual probability that a specific unit will close and affect a specific state’s carbon emissions, the EPA proposes to just average over all nuclear power (aside: At this point, I am all for simplifications, so, personally, no problem there).

    However, the terminology varies somewhat during the reading of the EPA docs. Sometimes this 5.8% is referred to as “at-risk” nuclear capacity, while at other times it is implied that the 5.8% is something referred to as “preserved” nuclear capacity. For example, please refer the section VI.C.3.b.2., titled: “Cost of CO2 Emission Reductions from Nuclear Generation,” available on page 43 of the PDF. This section makes the assertion that “at-risk” nuclear capacity should be “preserved” after an economic analysis of the benefits from carbon reduction. It overlooks the remaining 94.2%, and never seems to mention this part of what I believe they consider to be nuclear capacity “not at risk.”

    So, after reading this, I do not believe it is the intent of the EPA to undermine zero-carbon nuclear technology.

    In applying their suggested rules in section VII.C., Step 3 states, “We [The EPA] estimated the total quantities of generation from renewable generating capacity and from under-construction or preserved nuclear capacity using the approaches described in Section VI.C.3 above.” In this section the terminology “preserved” is again used.

    From the data you have presented both in the spreadsheets and in the TSD: “Goal Computation Technical Support Document,” the “preserved” and “at-risk” nuclear capacity is considered only the 5.8%, with the corresponding energy having a capacity factor of 90%. The remaining nuclear power contribution of 94.2% is not used at all in this data. Meanwhile, existing hydroelectric power is included in the “renewable energy” category for calculations.

    The effect of this oversight results in state goals that allow carbon emissions above what would result if zero-carbon nuclear was replaced by renewables, or in some cases as you pointed out, natural gas combined cycle plants.
    So far, though, we have only been discussing the carbon emission goals/targets for the future. These number are retroactive to 2012, thus from my understanding, these calculations cannot affect a state’s future choices because they are based on data from the past. This data is really just a benchmark to help “even the playing field” for states that may be heavily dependent on carbon-producing technologies. So, this rule hasn’t really damaged nuclear power up to this point.

    Now, what I don’t understand is how a state meets these goals. I would assume it would be a very simple calculation of [total carbon generated]/[total energy generated], in which case a state could use 100% of the nuclear credit, while not having to invest much more than before the rule came into effect.

    However, if only 5.8% of nuclear is considered in a state’s plan to meet the goal, than the overall effect on the nuclear industry is still pretty small because the baseline used to make the goal was only 5.8%, anyways. Essentially, the rule would incentivize plants to invest in existing plants to prevent shutting down. However, this rule was written using averaged values, so it doesn’t really do the costs and rewards justice, in my opinion.

    The last aspect is new nuclear power builds, which would require substantial investment, but get credited 100% in the calculation. Thus, the goals would be very hard to meet if a state (Georgia, South Carolina, or Tennessee in this case) had a utility that did not plan to bring the plants online within the timeframe discussed.

    So, I was thinking about commenting to the EPA, but I’m not for sure exactly what I should comment about. Should I simply ask for more clarity while asking that zero-emission nuclear is on a fair playing field with all other sources of power generation? Should I suggest that “not at risk” nuclear power be discussed more and used to evaluate goals in a similar way that hydro-electric power is used?

    Thanks for your input!

  5. Keith Pickering

    Davis Besse has a *gross* capacity of 925 MWe, but it’s net capacity is 894, according to this source. (The difference being the amount of electricity needed to run the plant itself.)

  6. mjd

    @Jim Hopf | August 25, 2014 at 19:49 |

    “Justin, Remy,

    Looking again (more carefully) at their Ohio example,…” I agree the whole methodology appears flawed, however in the Ohio example the spreadsheet has Davis Besse “current capacity” listed at 925.2 MW. This was the plant original nameplate capacity rating. The current nameplate capacity rating is listed at 889MW (Wikipedia). I doubt there is a power plant, of any type, in the country still rated at the original nameplate capacity. I don’t know how this affects all these calcs, but in the real world of CO2 production by power plants, the actual current plant ratings should be used for the baselines, not outdated original plant ratings. Most nukes have been uprated from original nameplate rating. Yes in forecasting future generation numbers average historic capacity factors have to be applied. However you should start with the current plant rating, which is the number the Utilities give to the various rate commissions to determine the “rate base.” If this is not an isolated example not only the methodology is bogus, but all the actual calcs may be bogus.

  7. Jim Hopf

    Justin, Remy,

    Looking again (more carefully) at their Ohio example, do I understand correctly that they are only including 5.8% of the state’s nuclear generation in the denominator, i.e., what they call the state’s overall power generation level (in MW-hrs)?! Do I also understand that this is only done to nuclear (i.e., that the entire amount of coal, gas and renewable generation is included in the computed total MW-hrs)?!

    In other words, when they refer to a requirement that the tons of CO2 per MW-hr generated be reduced by a certain percentage, they’re actually not referring to the actual, tons/MW-hr of the state’s (entire) power generation fleet? That is, what any reasonable person would assume they meant?

    Sorry, but this borders on being illegal, especially given the level of deception and the clear arbitrariness of it (where nuclear, specifically, is treated diferently than all other sources). This is clearly capricious and arbitrary, and completely against the spirit of emissions reduction (especially considering your replacing nuclear with gas example). Why would they go to the trouble of making something complicated and convoluted instead of simple, i.e., determining the emissions intensity by simply dividing overall CO2 emissions by overall power generation. There are no non-nefarious explanations.

    I mean it. Unless things are changed, the nuclear industry should sue EPA over this. Hell, the coal industry does it (sues) all the time, even over regulations that are completely justified and desperately needed.

  8. Remy Devoe


    The EPA sets state goals using MWh/year of generation as they should. What’s weird is that you get different answers for actual generation if you use the 5.8% value versus calculating from capacity factor. This may be due to the fact that 2012 is the base year and some units had outages. Also San Onofre wasn’t operating at all in 2012 though it’s listed as an operating plant. To be consistent, we scaled nuclear generation from the original spreadsheets except for the states that have under construction units. The EPA assumes a 90% capacity factor for these sites and we had to assume the same for the operating units to get the generation for those states.

  9. Rod Adams

    @Remy and Justin

    Thank you both very much for your work in exposing this buried treasure aimed directly at increasing revenues for the natural gas industry – aka “Big Oil.”

    Before I delve into your spreadsheets, can you answer one more question? Several times in your original post and in explanatory comments you used the word “generation” and then “capacity” interchangeably.

    For the average nuclear plant, there is only a 10% difference in value between its annual generation and its nameplate capacity.

    However, all other generation sources have more substantial differences between the amount of electricity that they generate over time and the amount of power they can produce when running at their physical design limit.

    Does the EPA formula use “capacity” or “generation” by applying some kind of average CF to each technology? If the former, there is an even larger problem than you describe.

  10. Remy Devoe


    Thank you for your comments and support.

    Here are the links to Google Docs that have our analysis and data.

    State Goal Data Calculations:

    Plant Level Data Calculations:

    These documents can be edited so you can view the equations. If you see any errors in the analysis please say so.

    The State Goal Data has our analysis of the plan. The first three sheets have the exact same format and formulas for calculating a state’s emissions rate goal. The first sheet is just a copy and paste of what the EPA had on there site using 5.8% of current generation and 100% of under construction nuclear generation. The second sheet adds 100% of nuclear generation to NGCC. The third sheet credits 100% of nuclear generation into a states emissions goal.The last sheet details the results.

    To note we did our analysis correctly, the total carbon emissions of the US for sheets 1 and 3 are the same. Also, the total electricity generation is the same for sheets 2 and 3. These values can be found in the results sheet.

    The amount of annual nuclear generation was calculated from the Plant Level Data document. This spreadsheet lists all generating units for each state and we simply summed the capacity of nuclear plants.

    Our idea was to use exactly what the EPA had in their state emissions goal calculation. To do this we had to paste values for the redispatched OG/Coal/Other columns because their value depended on what was in redispatched NGCC. Otherwise the data and formulas are exactly the same.

    Hopefully this data will convince anyone who is still skeptical. Thanks for reading.

  11. Robert Hargraves

    My reading of the EPA proposal is that states get credit for keeping existing nuclear plants in operation and completing plants under construction. I saw no credit at all for building new nuclear power plants. Accident? Deliberate? My mistake?

  12. Chris Bergan

    Good work gentlemen.
    Another route might be to require utilities to publish their CIPK. I would suggest a 2060 CIPK goal of 91 grams, which I believe works out to 200 Lbs per MWh. Have utilities publish this number monthly (or more often), and rate-payers will wonder why the Province of Ontario and the utility Energy Northwest (in Wash state) keep their rates in-line while producing even less CO2 than this marginal goal.

  13. Justin Knowles


    Thanks for your question. I thought that I would take a hack at it, and hopefully can provide you the answers you’re looking for.

    The reason for this is due to the way EPA calculates state-wide emissions “rates”. Your assumption stating that all existing nuclear generation is used in determining state emissions reduction goals is incorrect. To see this for yourself, I would recommend going to the technical documents (the “data provided by the EPA” link) supplied by the EPA for this plan. In the Goal Computation TSD and two supporting Excel files provided, you can find the total energy produced by nuclear in a state and compare it to the value that they use in calculating a state’s goal.

    In all states that do not have plants under construction, the amount of nuclear generation considered towards state emissions goals is 5.8%. For under construction plants, 100% of their expected production is considered. This makes a state like Georgia appear to have more nuclear energy than a state like Illinois; which of course is not true.

    With this in mind, you should see that a nuclear plant will have minimal effect on a state’s emissions “rate”. If the denominator (MWh) of an annual emissions rate changes by only 5.8% of the removed capacity, then it has minimal effect on the final goal.

    Mathematically speaking, this incentivizes natural gas because each installation has a larger effect on bringing a states emissions rate downward. This is because most states have an average emissions rate greater than that of a natural gas plant. Therefore, the shutdown of a nuclear plant will take the emissions rate upward only slightly, while replacement of generation with NGCC will bring emissions rates down considerably.

    Of course, all of these issues could be averted if the EPA only allows the option to consider emissions on a mass basis. With that said, if the EPA still wishes to calculate state emissions “rates” then they must consider all energy generation in the denominator.

    To answer your first question, Dr. Skutnik, Remy, and I are currently working on making our calculations publicly available.

  14. Jim Hopf


    You’re going to have to help me out here…..

    I would like to see your analysis showing that switching nuclear plants to NG would actually help a state meet the standard. If so, I’m clearly not understanding how EPA’s program works, and I’m a smart guy who read the EPA releases carefully (or so I thought). If what you say is true, they sure did a good job of burying it (i.e., the fact that it would help gas and hurt nuclear).

    My understanding was that, at the end of the day, the only tangible aspect of the EPA policies is a requirement that state utility sector CO2 emissions (from all generation sources) decrease by a certain amount. The only question was whether it was a required decrease in tons of CO2 or tons per MW-hr.

    EPA talked alot about things like how much nuclear they thought was at risk, and what they thought the best way of achieving reductions was, but the only relevence of all that discussion was EPA’s rationale for setting the reduction goals for each state. Once set, however, the goals effectively treat all (utility) sources of CO2 emissions equally, since the requirement is simply that CO2 emissions be reduced.

    My understanding is that, those state emissions reduction goals were set based on the assumption that ALL existing nuclear plants (including the 5.8 GW deemed to be at risk) continue to operate, and the five new plants are completed. I actually don’t understand the relevence, at all, of the 5.8 GW of “at risk” capacity, or how “only 5.8 GW is credited”.

    They set the reduction goals based on the assumption that all nuclear plants continue to operate. If ANY nuclear plant closes, the state’s emissions go up (relative to the expectations that were the basis of the goal), making them that much farther from the goal. The amount of low/non-emitting capacity that they will have to find would go up by the (closed) nuclear plant’s capacity.

    As for replacing nuclear generation with gas generation, that would simply make the state’s CO2 emissions go up, whereas they are required to go down.

    What am I missing here? What is the meaning of “credited” generation. My understanding is that it has no meaning, or relevence. The CO2 reduction goal is the reduction goal. If they replace a nuclear plant (other than the 5.8 GW of “at risk” capacity) and replace it with gas, to they get a new, higher baseline (that the reduction percentage would then apply to)? That was not my understanding.

    Anyway, is there a way that you could explain, in simple words, what I’m missing, and how the policy would allow an increase in emissions (from replacing nuclear w/ gas), while supposedly the state is required to reduce emissions, by a significant amount.

  15. James Greenidge

    With a green crowd at the helm, one must wonder just how “unintended” this issue was…seriously.

    James Greenidge
    Queens NY

  16. Engineer-Poet

    Mr. Devoe is too charitable to the EPA.  The rule’s incentive to kill existing nuclear plants is not an accident (nothing with this much money riding on it is an accident).  It is a gift to the gas industry, and the rule was certainly drafted with considerable input from them if not authored by them entirely.

  17. Frank Jablonski

    Can you post your database and calculations somewhere and provide a link?

    Also, thanks for reading through and sorting this out.

Leave a Reply

Your email address will not be published.

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>