# Oral argument — National Parks Conservation v. USEPA > Briefs, arguments, decisions, and more. URL: https://www.frixlaw.com/law-library/documents/argument%3A39471 ## Record - **Collection:** Oral argument - **Document type:** Oral argument - **Published:** May 16, 2014 ## Text Good morning. May it please the Court, Jenny Harbine for Conservation Petitioners. My intention is to reserve two minutes of time for rebuttal this morning. Congress directed EPA to eliminate human-caused haze pollution over national parks and wilderness areas. Here, EPA promulgated a Montana Haze Plan that leaves visibility impaired for more than 400 years. Well, wait a minute. All they've done is do something for the next five years, right? Five years from now they can go and they can change it. This is the first planning period and the planning period in which EPA is required to make BART determinations. And it's important to remember the function of BART in the statutory scheme. Because of their age and size, BART sources make an outsized contribution to regional haze and offer the greatest potential for cost-effective emission reductions. This is the low-hanging fruit. And if EPA fails to maximize emission reductions from these dirty sources, then restoring natural visibility becomes very difficult, if not impossible. And here EPA rejected SCR, the most effective control for NOx pollution, on Montana's dirtiest sources, these BART sources. And I'd like to use my time today to explain why EPA's decision to reject SCR is contrary to the agency's own analysis and contrary to the law. First, EPA's analysis strongly favored SCR. The analysis showed SCR was reasonable at each stage of the five-factor BART analysis, including cost and visibility benefits. These are the two factors that EPA says it gave the most weight. I thought there was just a very slight differential in the deciview improvement between SCR and the next combination, and that that was the factor that it just didn't justify the huge additional cost for what little gain in visibility there would be. Did I misread the record? No, that was EPA's explanation. But, in fact, SCR has one-and-a-half to two times the visibility, greater visibility benefits than SNCR, the inferior technology tool. Let me ask the question in a different way, Ms. Harmon. As I understand it, we're trying to get the biggest bang for the buck here, and EPA decided that for the incremental expense that would be required, we just weren't going to get much bang for it, that there wouldn't be that much of an increase in improved visibility. Why is it arbitrary and capricious for the agency to make that determination mandating reversal? With respect, the statutory standard is not getting the most bang for your buck. The statutory standard is achieving natural visibility conditions and doing so at a reasonable rate of progress. But I thought cost and improved visibility are two of the five steps, are they not? They are two of the five steps. How is that irrelevant to the agency's analysis? I don't understand your argument. EPA appears to have used a bang for the buck test. Let me tell you why. First, it's not appropriate, and second, why EPA misapplied it here. Can I just go back for a minute to the question of what incremental change there was? I gather, I'm just looking at some pages from the proposed order of Federal Register, the only one I could find where you actually had the visibility differential on the same page or metric as the cost. And I gather that the differential is, you said it was two and a half times as much. It doesn't seem to show that. But it seems to show a differential that's of the same magnitude approximately as the difference between SOFA plus SNCR and SOFA alone. That's right. And it's generally one and a half to two times as much for each affected Class I area. And the number I see is .404 versus .264. Is that wrong? Well, if we're looking at a cold-strip unit... That's the general order of magnitude. But it's also the same order of magnitude on the next level, which would be .182 versus .264. Right. And you're comparing the most affected Class I areas against one another, and that is about one and a half times, I believe. That's right. The increments are about the same. And one of the arguments, one of the points we have made, is that there's no standard attached to EPA's not worth it determination or bang for the buck determination. The same rationale could be used to reject every single BART control. What you're really saying is they don't have a metric. But the question is, would they have a metric? That argument goes to both sides. You're both saying they don't have a metric. They don't have a metric. And it's impossible to discern one here. If you look at a bang for the buck analysis for SCR versus SNCR, if SCR achieves one and a half to two times the visibility improvement and is two times the cost, the bang for the buck analysis would appear to be about equal. But we don't know what EPA did here. I'm looking at cold-strip unit one, the BART analysis for nitrogen oxide, and I'm looking at a very small incremental improvement in visibility from 0.249 to 0.378 using the most expensive technology and a cost differential of $13 million versus $83 million, or an average cost effectiveness per ton of $1,564 versus $3,195 a ton. I still understand why it's irrational for the agency to look at that and say, that's just not a sufficiently incremental improvement in visibility to justify the enormous cost differential. And you keep telling me that that's not part of the standard, but it's two of the five factors in the BART statute. So help me with your argument here, Counselor. I'm really struggling to understand that. You appear to be looking at the visibility improvement for UL Bend. And UL Bend is actually a good class one area for us to look at to illustrate this problem. The cold-strip and correct power plants are the largest stationary source contributors to haze at UL Bend. And EPA's analysis showed that SCR on the cold-strip units and on correct collectively would improve visibility at UL Bend by one and a half decivues. That's clearly visible, clearly noticeable visibility improvement. Counselor, I'm looking at this chart. I think it's ER1, but I'm not sure. So let's at least agree on what page we're looking at. Is that right? I agree with your numbers for cold-strip units. So I don't understand. It's not visible. We're dealing in increments of decivues, right? Right. And what I was referencing was the statistic. If you have SCR on all four cold-strip units and correct, you achieve one and a half decivues of visibility improvement. And this level of visibility improvement actually puts UL Bend on pace to achieve natural visibility by 2064, which is EPA's benchmark. But what EPA did is without SCR, they achieved about 0.5 decivue improvement collectively at UL Bend. So let's go back to the conceptual question then. Counselor, you're down to about two minutes, just to remind you. How should they go about making this decision of what the comparative is between the visibility increment and the cost? I mean, that's a legitimate concern, right? Those are our two factors in the statute. So at some point they can say this costs too much. You're not doubting that, right? That's right. How do they – and they kind of did an eyeball, right? And you're saying – and the company is saying they can't do that. They have to do something else. What's the something else? Well, let's remember, first, that EPA found the cost of SCR to be well within the range that EPA finds reasonable. So this isn't a scenario where we have extraordinary costs. Well, that's for a ton, but not necessarily per – I understand they're not doing it per deciview. So that's what's making this odd, because they're saying we're not really doing it per deciview. We're doing it per ton, but what we ultimately have to know is how it affects the views, right? Or the haze. That's right. What should they do? Just some possibilities. Well, the standard set by the EPA is they have to adopt measures that are calculated to achieve natural visibility conditions. If they don't maximize emissions from bark sources, we know they can't get there. That's the design of the Clean Air Act's regional haze program. So EPA had to, at a minimum, explain how its decision achieves that statutory standard. Here EPA didn't even offer that explanation. That doesn't deal with the cost and the – how you integrate the two factors. Well, EPA has used cost, essentially, to override the statutory mandate, and that's not an appropriate way to integrate the two factors. So your argument is that you don't – the cost can't override it, essentially, until they've actually achieved the statutory standard. Is that your argument? My argument is that, yes, cost can override the statutory standard, and regional haze, by its nature, is caused by the cumulative impact of individually relatively small contributions. Congress tasked EPA with eliminating those contributions, and if it fails to do so because it finds it's not worth it, then it can't meet the statutory mandate. Thank you, Counsel. Your total time has expired. We'll hear next from Counsel, who will identify herself in just a moment. Thank you, and may it please the Court. My name is Lisa Blatt, and I represent Petitioner PPL Montana. And I'd like to focus on what we see are the two most glaring defects with EPA's selection of SNCR and the fourth scrubber at Coal Strip. Preliminary question, to what extent are you and the conservation parties in agreement or disagreement? I think we're all in agreement that we have no clue what EPA was doing, but that's where we are. So do you want to take a stab at my question? So what should they have been doing? Well, I know what they – With some specificity. Sure, sure. I mean, what they just did with that good neighbor provision that the Supreme Court just upheld, they actually had thresholds. They had a $500 per ton threshold, and the highest one was 2,400, and they had factors. Here, there's nothing. It's whatever EPA said, which is what is so astonishing and dramatic and inexplicable. I thought they had something. Didn't they have something that said that if the plant wasn't of a certain size, then it wasn't supposed to? No, they have nothing. I mean, what they did was impose an SNCR and scrubber, and they looked at the global, the composite, of that second technology over a first technology and just said this is all reasonable. What they didn't, and again, let me just get to the two points why I think there has to be vacature. The first is they didn't justify that these technologies had very high incremental cost and very low incremental visibility benefits. And second, they quite arbitrarily, inexplicably, and incoherently rejected other controls with dramatically better cost-benefit ratios. And that's where I can see why NPCA is going, what happened? They actually rejected controls that look so much better by their own math. And we can get straight to the inconsistency, but I'll start with the incremental. I think we start with common ground that EPA has got to look at incremental cost and benefits when it adds a technology under another, and that's because if you just look at the sum, it hides the value of the component parts, and we offered that Hank and Tommy Aaron analogy. And here, and so what the guidelines do is they say regulator, do not impose the more expensive option just because its visibility benefit appears to be worth the cost. If there's a cheaper option that gets you almost as much visibility benefit, but for less cost. So you have to justify the high incremental cost to achieve a low benefit. But for here, and what is really quite dramatic, is the SNCR had two times the cost of SOFA at half the cost of the visibility benefit, and EPA does not mention this, much less justify it. Now, this is not in the brief, so I do want to disclose this to the panel, but at ER30, EPA does look at incremental cost for the force scrubber. And this is not in the briefs by any party, so you should look at it. But the industry is complaining, hey, you didn't look at incremental cost. And so what the agency says is, okay, fine, this trivial .055 visibility benefit is worth it for the $2,400 cost. That's just reasonable, and that's all it says. You started by saying that there were other methods, ones that were just dumped out at an early stage. Is that what you're saying? I'm sorry, say that again? You began by saying that there were other methods that would have been more efficient, which I don't remember really seeing any much discussion of in the No, they start with what they should have done, which is the lime injection and SOFA, which the industry doesn't contest, which gets you almost all the way. It's basically most of the work. But focusing just on this $2,400 figure, when they finally say, hey, we'll look at incremental benefits, they say nothing. They give ipsy-dixit reasoning that doesn't tell you why they thought this was worth it. But even if this I just don't understand where you started. You started out by saying there were other things they could have done that would have achieved a lot, but they didn't do it. Oh, no, they could have just stopped at SOFA, which gives you most of your bang, and they could have and should have stopped at lime injection. And what they did was they just blindly added the middle and did not look at the incremental cost. So with EPA's defense, I love how we've all assumed that they had a bang for the buck. That's something made up in their briefs. They try to say they're Goldilocks here, that SCR was too hot, SOFA too cold, so let's just look at SNCR, baby bear, middle ground. But here baby bear got help from his mother. The option they picked was SNCR plus SOFA. So the brief does the very same mistake that the guidelines forbid. SOFA is the mother in your hypothesis? Yes. Mothers do all the work. I just want to make sure I understand who mother is. Yes. And so they in their brief still do not grasp that SNCR is a $3,000, $3,000 incremental benefit at a .085 visibility benefit. And they have no defense to that other than, well, we want to look at it globally, enough bang for the buck. Now what makes this case so easy for this panel is the controls they rejected, which, again, have dramatically better cost-benefit ratios. Being just the SOFA by itself and the lime injection by itself. No, those they imposed. What they rejected. That's what I've been asking you. What's what they rejected? Lime injection. Okay, so they at Caret, which is a whole other plant, they rejected two controls, SCR, which is what the other side was just complaining about, and they rejected SOFA at Caret. SOFA at Caret has a .087 visibility benefit and only costs $1,487. And they said that's not good enough. They said to the environmental people and the public, that's too much money to spend for this visibility benefit. We can't impose on industry $1,400 for .087. I think they said this, but it seems fairly self-evident, that the difference is that Caret is a much smaller plant. Okay, well, they made the same insistence. So, therefore, there's also an absolute number that's relevant here, i.e., not just what you get per ton or per dollar. Yeah. Well, there's a reason why that plant size is not in the guidelines or in the final rule, because it's irrelevant. I mean, this is something, again, they made up in their brief. From the camper's perspective, the camper doesn't care how big the plant is. The camper cares about visibility. From the industry's perspective, they care about cost. That's why the NPCA is going to put a visibility number. I mean, if you have whatever it may be per ton, it's still only contributing, you know, one iota. Judge Berzon, the plant is much closer at Caret, where they rejected it. These are very far away at Coal Strip. It doesn't matter the plant size. And, again, if EPA wants to stand by that, they better say that on remand. It's not in their guidelines. So I ask you again, because this is where I get stuck in this case. What is it you want them to do? You want them to come up with thresholds, come up with ratios, come up with something that translates over. They have two choices. Their guidelines say, do whatever you want. They say, just cite the statutory factors regulator, and hopefully that will fly. Well, they're essentially applying a factor test instead of a. . . Right. So they have two choices. They can either go that and hope it works in court and that courts won't say you need a little bit more description, or they better be consistent. And what you do when you tell a regulator, do whatever you want, is they have the very same rulemaking is all over the place. They have vastly different. They're rejecting things as too expensive and not enough visibility benefits, yet they're imposing controls that cost more, got smaller benefits, and were way less cost effective. What is most embarrassing and devastating for the EPA is what they did at Coal Strip, obviously where plant size doesn't matter. Because remember, they rejected Papa Bear, which is SCR. That has an incremental benefit. And this is the only thing they say on the record, which is what you just pointed out, Judge Berzon. They said .4 was too small of a benefit at a godforsaken $3,200 cost, and yet SNCR was $3,300. I gather it's not .4. It's .4 as compared to .26. They said what mattered to EPA, which is what matters to the public and courts and industry, is they thought this cost-benefit ratio of $3,200 was too much money to achieve a .4 visibility benefit. Counselor, you're below two minutes. I'd like to reserve the right. You may do so. Thank you very much. We'll hear from the EPA. Good morning. My name is Daniel Pinkston. I'm with the Environmental Defense Section of the U.S. Department of Justice. With me at the council table is Elizabeth Dawson, my co-counsel, and we also have Sarah Loughman, who's an attorney with EPA Region 8 as well. We have 20 minutes, as has been mentioned. We have divided up our time between Ms. Dawson and I. I'm basically prepared to talk primarily about why, if all of our calculations are correct, the result is the correct result in this case. Let's get to Ms. Blatt's point. Why did you reject Mama Bear at Courette and do something completely different at Colstrip? Well, first of all, BART is a case-by-case analysis. As we know, it's based on a number of factors. And foolish consistency is the hobgoblin of administrative agencies? I'm not sure where the argument goes. I mean, you've got an inconsistency here. And what's the explanation for the difference? Well, one of the explanations is that they're not identical. Because of their size? There is a difference in the size and also the effect of the different technologies that we analyzed on visibility. But Courette is much closer to Class I areas than any of the units at Colstrip. But that also has to do with the amount of emissions as well. What we looked at and we described in our brief is that there are differences. If you look at SOFA plus SNCR at both Courette and Colstrip, for example, the visibility effect at Courette is much less if you use those technologies. So the technologies don't apply across the board. You're basing this on the basis of CalPUF? It seems to me you're pushing the model to its limits and suggesting that it can effectively measure changes in visibility in excess of 200 kilometers. Are you not? And that's the only available technology we have right now to do that measurement. CalPUF is the approved method for dispersion modeling. And we have taken note that there are concerns about how accurate CalPUF is at a variety of differences if it's over 200 kilometers. And one of the ways, as we described in the brief, we tried to deal with that, was to use some statistical measures so that you're not using the outliers. There are definitely – I'm sorry. Judge Berzon has a question. I was going to say, and I think Judge Tomlin was essentially going in the same direction, there's a tendency in administrative agencies to say a bunch of code words over and over again, what you've been saying, which is discretion and individual determinations and so on. And when we read your brief, it happens about every page and a half. But in the end, you need to engage in rational decision-making, and the question is how – not only how – what are you doing, but how do we know what you're doing if essentially you're saying, well, we say it and so it's okay? Well, I think the answer to that is the word discretion to some extent. But discretion to engage in rational decision-making. So how then, if you have – if you look at these numbers, for example, and you just chose the middle one, why? Well, rational decision-making is evident when we have followed the statutory scheme and the regulatory scheme. And that means there are five factors. We are going to tell you what each of the five factors is, and then we're going to tell you which one we – we're going to – we'll make little charts, and we'll tell you which one it is, and then we're just going to choose one. Well, I hearken back to the question you asked some of the prior attorneys here at the podium, and that is what is the test otherwise? If EPA had tried to – One option is – I mean, there are various ways you could cabinet, either with floors, ceilings, ratios, or you could at least provide more of a narrative about why it is, for example, that in this instance the SOFA plus SCR, which is too much, costs too much, even though the increments are about the – am I right about the fact that the increments between each level is about the same? Well, the incremental cost between SOFA – I think I meant the cost, the incremental visibility differences. I think they're roughly the same. All right. So essentially what you're saying is, well, seven – I guess you must be saying that $2,200 for that increment is worth it, but $7,000 – well, that's not the difference. It's the difference between – it's about double and then double again in cost per ton, right? And you're saying the second doubling is not worth it. Well, one of the elements of the test that Congress set out is what's the cost. Well, I understand that, but then what do you do with the cost? I know the cost. I know the visibility, but I don't know how they fit together. So is it sufficient to just say what the cost is, say what the visibility is, and say we think one is enough, that this one's enough and that one's too much? Well, there is no – the BART guidelines, for example, say there is no bright line standard. I understand that. You keep saying that, but I'm asking you what is rational about saying we're just going to choose a number? Well, what's rational here is that – and you're probably not going to – I think what we're looking for is a reasoned explanation to defend the ultimate decision, and we're all having some difficulty with that. I understand. And in this case, what we have done is to lay out all of the facts, which include the cost, the visibility benefits, and so forth. We've compared the various – I mean, I suppose another thing you could do is you could demonstrate that the visibility increment between the SNCR and the SCR is not going to make enough of a difference in the long run in terms – it's not going to matter in the long run because you're going to be able to make it up later and you're going to be able to get to the ultimate goal even without that. There are various things you could do in terms of either narrative or calculations, but you've done none of them. Well, what we have done, we think, is to follow the scheme which Congress set up and was fleshed out by EPA. And it should be recalled as well that – What is the scheme? The scheme is you have a set of factors to be considered and you look at a variety of – And you put them together however you want. Well, you couldn't say BART is nothing. You couldn't say BART should be a million dollars a ton. Somewhere in between, but there is no bright-line standard. It's a matter of judgment and discretion, which is entrusted to the agency when it's doing fairly a fit, as in this case. I assume – Mr. Pinkson, one of the problems I have, going back to my concern with the cowpuff model, is that the margin of error of the model is so high, I'm not even sure how the agency can measure the change in visibility improvement because it's so small. What's your response to that question? Your Honor, if I could defer to – Should I ask Ms. Dawson? Okay. Be loaded for bear on that one, since we're hunting bears. I forgot which one was Papa Bear. But I think it does come back to the question, if we had tried to set out a bright-line standard of some sort – I'm not suggesting – One option would be a sliding scale or a matrix or a narrative or something that tells you how this is going to end up 50 years from now and why it doesn't make a difference and why it's not worth it. But you've done just none of it. You've just said, we'll take that number, please. Well, the other side of this that we haven't really talked about is the reasonable progress side. And that is not BART determinations, but that is where, since there are planning periods that go out for 2064, there is some opportunity to make up some ground. That's what I'm saying. And if that point had been made, i.e., it's okay to miss this increment now because we're going to make it up later, but you haven't really shown that either. And going back to my CalPUF concern, I don't understand how the agency is going to be able to know whether there's been an improvement because it's beyond the capability of the model to measure, unless I'm reading the technical data wrong. Well, to some degree, CalPUF provides a measure, a comparative measure. And there is, admittedly, no model is perfect. And the other problem is that it's not discernible to the human eye unless it's an order of magnitude of one deciview or greater. And we're talking about less than one deciview in incremental improvement, are we not? Yes, although... So how does the agency know whether or not this technology is actually going to work, even if it's employed and the companies have to spend all this money putting in this very expensive technology? Well, CalPUF has been used in many contexts for many years. And as you know, CalPUF is approved in the BART guidelines, which doesn't make it perfect. Well, also it's cumulative, right? I mean, in other words, this is .4, but something else is .3 and something else is .2. The nature of regional hay is that it's many sources creating a problem together, and the BART part of it seeks to attack that by getting to particular impacts of particular sources as opposed to regional projects. Are these the only two sources in Montana that are subject to the BART standard? No, there are some others. Kelsey, you are impinging on your co-counsel's time. Thank you, Your Honor. Thank you. Good morning, Your Honors, and may it please the Court. My name is Elizabeth Dawson, and I am here, as my co-counsel said, with the Department of Justice on behalf of EPA. Can I begin by asking you whether you accept my understanding of this representation about CalPUF not being within the margin of error? That's not your numbers. That's the company's numbers. Correct. Are you contesting the notion of what they're claiming the margin of error is? Well, we certainly don't concede that the particular number that PPL put in their brief is the margin of error, and in that respect, their reply brief does. Do you have a different margin of error? No, Your Honor. We don't have a specific margin of error calculated for the model because we believe that while there are inherent uncertainties in the model, the use of the 98th percentile, which would be the eighth highest day of emissions, adequately accounts for visibility impacts and other uncertainties such as distance. And I think it was important to note what my co-counsel said, that CalPUF is, to some extent, a relative determination. Even if the model itself is not perfect, it is sufficient enough as part of the five-factor analysis to look at the results for one technology versus another. Their number, their margin of error number, came from studying one place for one year, right? I believe that's correct, Your Honor. And it just hasn't been tested. It's just an assertion. Right. And we do not concede that that would be correct. And to go, Judge Tallman, to your point about the small amount of benefits that are shown by the model, I think it's important to remember that while human perceptibility might be at one deciview, that is not the threshold for what might be an appreciable benefit as far as setting a BART determination. Indeed, the statute requires that sources that cause or contribute to visibility impairments have to be subject to BART. And the EPA and the BART guidelines determine that one deciview is sufficient to cause visibility impairment, but .5. How do you account for natural contributions such as wildland fires? I'm thinking of a fire on the order of magnitude of the Yellowstone fire that burned hundreds of thousands of acres and burned for weeks. Yes, Your Honor. And I believe that that is considered as part of the natural baseline that is considered. So it's my understanding that EPA does not base its calculations on pristine conditions, but on what is a natural baseline. So it does include, I think, to a certain extent, fires. Is that evident from the record in this case? I believe it discusses that in the BART guidelines. And in the proposed rule, I can't be certain of the page, but where it discusses the baseline for determining what sources are subject to BART. But going back to the point about the thresholds, I think it's important to note that some sources are going to be considered to contribute when they have .5 deciviews of contribution, which means that when BART is set and their visibility improves, it's reasonable to assume that the improvement will be less than .5. And in that instance, that is... Well, here we're talking .085, right? In some instances, yes, Your Honor. Which is significantly below .5. It is significantly below .5, I would agree, but I think... But we're not talking about an absolute number. That's what's missing from the charts. I mean, all of these plants are all contributing more than .5. It's the increments that we're talking about. Yes, I understand that, Your Honor. I'm just saying as far as why it might be reasonable that the benefits themselves are lower than .5. It's because they don't contribute that much. Well, here you have two sets of unhappy people who are unhappy in different directions but are for essentially the same reason, which is they say that they can't figure out what you're doing. And frankly, I can't much figure it out either. So what do we do about that? Well, Your Honor, I think I would go back to the fact that the BART guidelines do set out a set of factors. Well, I understand that. But if you're going to tell me again that you have discretion and that everything is different and that seems to me is just a fancy way of saying we can do what we want. Well, Your Honor... We'll come back to reasoned explanation. That's your obligation. And I think we're looking for the reasons in the explanation. I understand that, Your Honor. And I do think that we do have a significant discussion of the facts in the record. I think to the extent that there is a question with this Court and it seems that there might be... Well, could you paraphrase for me the explanation of why you chose in, for example, with regard to Coal Strip Unit 1 to impose SOFA plus SNCR, but not SOFA plus SCR, given the fact that SCR would provide somewhat an increment in visibility improvement similar to the one between SOFA and SOFA plus SNCR. What is the reason? Well, it's... So how much more, Matt? It costs more, but again, the increment is about, or the ratio is about the same, i.e., double and then double again. Well, if you look at the specific numbers in the proposed rule, it does appear that the increment... I want to know what the agency has said about why they chose the middle number. Well, Your Honor, the agency talked about the two highest, the two most affected Class 1 areas, Theodore Roosevelt National Park and UL Bend Wilderness Area, and said that particularly looking at the fact that those were not projected to meet the uniform rate of progress, that the improvements using SOFA and SNCR were substantial, and that is on page 24027 of the proposed rule. So it must be that the improvement the next level up would also be substantial because it's about the same. Well, I think it would be, I think that would be a substantial improvement as well, Okay, so that can't be the reason. I think that the improvement in that scenario was not justifiable. And I think there is... Again, the increment's about the same. It's about double from SOFA to SOFA plus SNCR, and then it's about double from SOFA plus SNCR to SCR. So it's a double-double. Yes, Your Honor, and I think that in this instance where the Court seems to be concerned about the explanation that EPA gave, I think it is important to note the case law that suggests that and that requires that even if an agency explains its decision with what a court determines to be less than ideal clarity, that decision should not be upset if the agency's path may reasonably be discerned from the record. And I think here, I think certainly there is... More explanation, I think, can always be done. I think there's a question of what explanation is enough. But here, where there are all the facts in the record and where it is noticeable that there is a difference between the $13 million initial cost for SCR versus 80-some for SCR, that it would be rational for EPA to choose... But you also said that the cost effectiveness for the SCR was within the range that's been accepted. That's correct, Your Honor, but SCR has also been rejected at other places, too, and that's where we do go back to the case-by-case determination, that each plant is different, it will be at a different distance from any number of Class I areas and will have differing effects on visibility. Ms. Dawson, if we were to remand this to the agency for further explanation, does that reopen the entire public comment process? Your Honor, I believe that... I think if it's just for further explanation, I do not believe that additional public comment would necessarily be required. However, the EPA could decide to have a public comment period for that. But if vacater were the decision, then I believe EPA would have to start over. What is the interim effect of... Are these currently in effect, these standards? Well, there's a compliance period that is set. So PPL has a certain number of years not to necessarily install this particular technology because BART is just an emissions limitation. It's not a particular technology. But it has a certain number of years to meet the BART emissions limitation. How many years? Five, I believe, in this case. So essentially we could send it back for further... There wouldn't be much of a reason to vacate it because it's not even operative right now. I mean, it's not... It's not due yet, I think, is... So they haven't actually started to implement any of this technology? To the best of my understanding, no. Again, this would be outside of the record, but it's my understanding that CORET might in fact close at some point soon. And at Coal Strip there might be the installation of a different technology to account for NOx. But I'm not... I mean, that was one of the things I was wondering about. I mean, that's interesting because I had this feeling that there was a subtext to the CORET thing, which was it probably wasn't going to be around for very long. So why did you say that? Well, Your Honor, that's not in the record. Well, I know, but it has to be what was going on. Well, to the best of my understanding, that was not before the agency at the time that it made its decision. We found out about that at a different time. But I think as far as the compliance period, that might not make a difference. But I think it is important to remember that the BART is an emissions limitation. So understanding that PPL has taken issue with the technologies that the DPA assessed, PPL can choose to achieve the emissions limitation in another way. Without any further questions, I seem just about out of time. Thank you very much, Counsel. Thank you. Ms. Black, you have some reserve time. So, Judge Berzon, on the math, SOFA alone, this is all giving EPA the benefit of the doubt, so I'm putting everything in the light helpful to EPA. SOFA alone is a visibility improvement of 1.82. The incremental is half for SNCR at .085, and you get a whopping doubling at .222 incremental with SCR, and all that's on page 98 of the record. I think that's Cold Script Unit 1 and for Theodore Roosevelt. So those are the numbers in the light most favorable to EPA. And the scrubber numbers look terrible for EPA because lime injection got them 2.84, and the scrubber vessel only got them .066. And, again, that's the most helpful. The number they put in the record was .055. But it could be, we just don't know because they haven't told us that there's some reason why that increment matters because in the long run you need, you know, you're not going to be able to. Who knows? Right, but Cold Script is where they didn't even address the incremental, and I'm sorry, NOx is where SCR, they just missed the boat, and they at least addressed incremental on page ER 30, but they did it in numbers that sort of killed them on correct. So the record, what you, all you need to look at is ER 98, and then the correct numbers are on 114 of the ER. And on the number, the analysis is also on those pages of the record, but on page 62 of EPA's brief where they're responding to MPCA's argument, they said, hey, we can't have $3,300, imposing $3,200 on industry for a .4 visibility. I assume that all of these changes are going to result in an increase to the rate payers. They're just going to pass the costs on. Is that part of the? I don't know. I mean, I assume that's what Congress was thinking because BART is basically retrofit, so it's only things that were involved in 1977 and through, I think, 65 through 77. No, my question is actually more broad. My question is, is there any assessment of what the impact on the rate payers are going to be? I don't recall PPL putting that in their comments, but their full comments are in the record. I just don't know. If I could talk about CalPUF briefly, I think, Judge Barzahn, you raised a fair point on the margin of error. We say it's .39, they say 98th percentile, so it is a he said, she said, and there might be room for deference. But where EPA can't defend itself is on the distances beyond 200 meters, kilometers. Actually, that's there, they said, more than they said about anything else. As I understood, they said it's no good beyond 300. 200 and 300, you just have to be careful, but it's not useful. Likely problems is what they said on the final record ER6. And then they said, well, we don't care about those likely problems because we'll use the 98th percentile. But what is so incomprehensible is that 98th percentile of the year applies at 1 kilometer, 2 kilometers, 8 kilometers, 199 kilometers. In other words, they just didn't account for the second error that they conceded exists. Well, or they were overly careful at the beginning. No, they said, look, we recognize there are two problems with the model, one at all distances, so we have to go 98th percentile. We recognize there's another inherent problem in the model, anything between 200 and 300, but we at least applied the margin of error that we applied for 1 kilometer. Meaning there's just simply nothing in the record that explains what you're supposed to do or how they gave the industry a cushion or another margin of error for that second problem. Now, they could address it. Well, is it possible that they were giving you more of a margin of error at the front end, so it could still be an adequate margin of error? No, because all they did, we deserve them. No, they didn't change it, but that could be because they were overcompensating at the beginning. Well, they didn't say that. They just said we did the 98th percentile. And the brief doesn't say that either, Judge Berzon. I mean, you're sort of kind of inferring what they might have been thinking, but the record doesn't say that. Counsel, our questions have taken you two minutes over time, but I do have one follow-up question going back to the question you heard from Judge Tallman to Ms. Dawson. What is the significant difference between a remand for explanation and a vacator here? And what is it that your side is asking for? If all we had was lack of explanation, it would be a remand. But because you have vastly incoherent, inconsistent things, I think you do have to vacate because you can't uphold it. Vacature would open up the comment period. And in terms of why it matters is the compliance period, although it's in the future, obviously to get it done the industry had to start already. So it's not like they say in two years you have to have it done and it takes one minute to install all these technologies. Very well. Thank you, Counsel. Your time has expired. The case just argued will be submitted for decision. --- Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/argument%3A39471. Public record. Not legal advice.