Oral argument — National Parks Conservation Association v. US Environmental Protection Agency

And we have Mr. McFedrin, Ms. Bowers, Mr. Riley, and Mr. Thomson. Mr. McFedrin, Mr. Riley, Ms. Bowers, Mr. Thomson, Mr. Riley, Ms. Bowers, Mr. Riley, Ms. Bowers, and Ms. Bowers, Ms. Bowers, and Ms. Bowers, Ms. Bowers. Whenever you're ready. Good Morning. May it please the Court, – Charlie McFedrin – for Conservation Groups. I would like to reserve 14 minutes of my time for rebuttal. That's fine. Thank you. Thank you. EPA approved a plan that provided for no emission reductions at any source. EPA admits that this approval is based on cursory information. This plan does not solve this problem. It does not address this problem, and it does not require any reductions. Emissions from Pennsylvania sources – As I understand, there's probably at least 10 things that you claim were a problem. Let me start with the first one only, the challenge to the reliance on the transport rule. It looks like in June of 2012, there was a FIP, and that would appear to be national, if that's the case, and it would then appear that that could only be challenged under statute in the D.C. Circuit. What am I missing? Well, it is being challenged there, but there's also a requirement that the Pennsylvania Haze Plan address best available retrofit technology here in Pennsylvania, and that substitution was based on a prior version of the cross-state rule. That substitution was based on a version of that rule that provided for implementation of emission reductions in 2012 and 2014, did not provide for implementation of reductions in later years. The plan has now been amended so that those two phases of reductions didn't start until 2015 and 2017. It seems like the reasoning relies on the nationwide rule that the transport rule is better than BART, and if that's the case, I don't want to tread on the turf of what has explicitly been given to the D.C. Circuit, and you're saying, I'm not quite understanding why it is that this particular rule, we'll get to the other nine later, why is this particular rule, especially now that you are challenging it, you say before the D.C. Circuit? It's been steady pending the resolution of the cross-state rule. The better than the cross-state rule case has been steady pending the resolution of the cross-state rule, and in the cross-state rule case, the implementation of the cross-state rule itself has been delayed. A delayed rule is not the same as a timely rule. For example, a dollar today is not the same as a dollar tomorrow. I think the point is that if the D.C. Circuit is already considering the so-called transport rule and whether it's better than BART and that should apply, that is something that is nationally applicable and therefore not really for a regional circuit to address, and so isn't it best that we just not rely on that, dismiss that part of this petition, allow the D.C. Circuit to rule, and let us focus on those areas? Yeah, especially when it's already teed up in some way. We don't think it's wise to dismiss that part of the petition. Well, you asked for an alternative to hold it in abeyance or sever, maybe transfer it, but why would we hold it in abeyance if there is other grounds that we clearly have the authority to evaluate and we won't have the authority to evaluate that rule? Sure. The problem is there's been no finding of equivalence. There's no finding that the Pennsylvania plan, relying on that part of that national rule that's been approved by the D.C. Circuit, is a big issue in Pennsylvania. But the point is it's before the D.C. Circuit. I guess at least where my questions are coming from on this subject is why would we even evaluate this? This is for the D.C. Circuit to address and shouldn't we be focusing on your other grounds for challenging the EPA approval of the Pennsylvania plan? We've petitioned for review under a portion of Section 307 that provides for resolution of issues of regional significance in circuit courts. That's why we're in the circuit court. And this plan needs to account for this piece of this issue. We've provided the option of a stay so that after that ruling in D.C. EPA can make a record regarding power plants and EPA can come back and put that record before our members. But then it would be a stay because we wouldn't have the record. When we say something it's because there's a record in front of us to decide and we're withholding for some other reason. Here we'd have to go through the exercise of remand to allow them to develop a record on the cross state rule. We think the gap in EPA's decision, there is no record to support at the moment, there's no record to support EPA's shift from the old cross state rule to the new cross state rule. Don't you have other grounds that you're also attacking the approval of this plan other than the cross state rule? Sure, we do. Okay, why don't we talk about those? Sure. Okay. You talk about the fact that there needed to be some kind of concrete cost effectiveness threshold. Right. Where is the requirement that a concrete cost effectiveness threshold be set? May I put that in context to the broader issues? Mm-hmm. So the statute requires that EPA consider best available retrofit technology. It's a statutory requirement, not something invented by EPA in a regulation. It's something that This is a detailed evaluation of the best technology for reducing pollution of haze causing pollutants, which are, in this case, are nitrogen oxide, sulfur dioxide, and particulate matter at each plant. It's a detailed evaluation. It requires looking at all available control technology. It requires sifting this technology by cost and tons reduced. It requires a decision about how much that's going to benefit air quality, including from one plant to multiple park receptors, multiple parks that receive air pollution from that one plant. And it requires EPA to make a documented decision that people like us can review. The question of cost effectiveness goes to how many dollars per ton it costs to reduce pollution. Picking up on that question, where do the BART guidelines require the use of a cost effectiveness threshold? We think a cost effectiveness threshold helps EPA develop a record that's defensible. No, I understand. And that's a good policy argument. But where do the BART guidelines require the use of a cost effectiveness threshold? We think that as a matter of non-arbitrary decision making, EPA needs to develop cost effectiveness data and needs to present it in a way that ranks options. But it looks like when they did the Wyoming SIP that they said that BART does not require any threshold level of control. So they're just being consistent when they're doing the same thing, applying the same thing to Pennsylvania. I don't have a cite to the BART guidelines in front of me, Your Honor. I believe we discussed that in our brief. But I think the principle on cost effectiveness is the same as the principle on some of the other BART issues that we've identified, which is EPA needs to show the data that it relies on, it needs to evaluate the options, and it needs to look out in terms of cost effectiveness at where benefits will be felt. If that's at multiple parks and wilderness areas, then EPA should consider multiple parks and wilderness areas. Let me turn to the failure, you say, challenge to the emission limits associated with controls. You say there's a failure to consider emission limits developed as best available control technology, just B-A-C-T or BAST, how do you pronounce it? You say BACT. BACT. And they also developed as part of lowest achievable emission rate. That's right. You say LARE. LARE, okay. Where do the guidelines require BART emission limits to be based on either BACT or LARE? They require the BART emission guidelines, and our argument is that those limits under BART need to be based on, by reference to the two standards you've identified, they also need to be considered maximum achievable control technology, which is mentioned in the guidelines as a streamlining process. And they need to be based on the sort of source-by-source analysis that we contend wasn't done. Is there something, I think, going back to what Jambo was asking, is there something in the guidelines that say you have to use BACT or LARE as your benchmark? The guidelines refer to BACT and LARE. But not require consideration. I think that's correct. All the guidelines say is that, quote, technologies required as BACT or LARE are available for BART purposes and must be included as control alternatives. That's correct. I mean, that's exactly our approach, because our approach is not to say… But where does it say anything as to whether the state must also consider BACT or LARE emission limits? Well, emission limits are under the statute. BART is an emission limitation under the statute. So we don't think there's any issue between, if you find the issue between emission limits and emission control devices that the other side raised. We answered that briefly in our reply brief, I think, right near the beginning. If I could just finish the range of technology I was mentioning before. You know, there are these benchmarks. There's BACT and LARE and MACT. I'll use the acronyms since you asked about them. And there's the case-by-case determination. And then there are all these other permits that we describe in detail in our comments, which are in the Joint Appendix, and also mention some of that material in our briefs, where other sources, similar sources for similar pollutants have imposed much tighter limits than that 0.1 limit that DEP and EPA have relied on. We not only provided, to go to your question, Judge Ambrose, we not only provided the limits in the text of our comments, but we provided the actual permits and attachments. So EPA, DEP, anybody could take a look at not just the text, not just the limits we mentioned, but also the underlying permits and have complete information. So across that suite, I think I mentioned five different sources of authority. EPA could have looked at any of those, and the state could have looked at any of those. And when you look at the totality of those, that wide range of options, we don't say that anything in particular had to be considered, but there's no record to support EPA's finding that 0.1 is defensible. Okay, thank you. We'll get you back on rebuttal. Ms. Bowers. Good morning, and may it please the Court. Kate Bowers here on behalf of the United States EPA. With me at counsel's table today are Donna Mastro, the Office of Regional Counsel, EPA Region 3, and Matthew Marks with the Office of General Counsel, EPA. I'm going to take 11 minutes today, and then Pennsylvania DEP will argue for three minutes, counsel for Homer City for one minute. That's fine. Whatever you decide, and then we may have additional questions in any event. Why don't you start with that first issue as to the transport rule, whether it belongs here or in the D.C. Circuit, and if it belongs in the D.C. Circuit, can we transfer only part of the case to the D.C. Circuit? Well, there would be no need to transfer it to the D.C. Circuit, because the D.C. Circuit is already considering the entirety of the national rule in the Fed. So we should just withhold our hand on that one? That's correct. That's correct. I'd like to just make an additional point on that, Your Honor. I think this is fairly well set out in the briefing, but the petitioner's argument that Casper's better than BART determinations technical basis has changed because the implementation timeline has changed, that argument was raised for the first time in the petitioner's reply brief. That's been waived. And furthermore, those implementation deadlines were changed well after EPA received Pennsylvania's SIP for review, and after it acted to issue a limited approval on limited disapproval of that SIP. So were petitioners to challenge that technical basis underlying the better than BART determination, based on the new timeframe, the proper course of action would be to file a petition for reconsideration with the agency. But again, I think that's fairly well set out in our brief, and what I'd like to focus on today, if the Court will indulge me, is EPA's exercise of its discretion in reviewing Pennsylvania's source-by-source BART determinations and its decision that while parts of Pennsylvania's analysis were not as robust as they might have been, whether EPA still acted reasonably in determining that any deficiencies in the analytic process that Pennsylvania went through did not materially affect the reasonableness of Pennsylvania's conditions. Was there any analysis of cumulative visibility impact? Pennsylvania itself did not analyze cumulative impact. Were they required? What I want to make sure that we're clear on the distinction here is we're looking at analysis of cumulative visibility benefits, and this is in Step 5 of the BART determination, where you're analyzing the visibility benefits that would be associated with the installation of additional controls at a given source. Now, EPA conceded in its 2014 revised approval that Pennsylvania should have considered the benefits of additional controls at multiple sources. But EPA has made clear in other SIP rulings, this is particularly clear in the Arizona SIP approval, that consideration of cumulative benefits does not require a strict summing of the deci-view values associated with improvement at all affected Class I areas. And what Pennsylvania did do, although it didn't actually do a cumulative analysis itself, was for many sources it did provide those delta deci-view benefits at more than one affected area. Hasn't the EPA disapproved SIPs in other places that didn't have this cumulative analysis as part of their SIP? And if that's the case, how can we find the agency acted not arbitrarily here when it has a track record of requiring that elsewhere? It's done both, Your Honor, and it's context dependent. EPA has disapproved SIPs that, among other things, failed to do a cumulative benefits analysis, where EPA's own review showed that those cumulative benefits would be significant. And by significant, here I'm talking 20 deci-views. This is double-digit deci-views. On the other hand, EPA has approved other SIPs that failed to include this cumulative analysis, where EPA was able to determine that a cumulative look wouldn't have resulted in significant improvement levels. If you decide that it's close enough for horseshoes in certain cases, don't you have to give a reason why it's close enough for horseshoes? Well, that was what EPA did here. And that doesn't require setting a particular threshold, be it for cost effectiveness or be it for what kind of visibility benefit would be considered. Obviously, somebody's saying, okay, this one's close enough, that one's not close enough. But if you're a reviewing court on a petition for review, don't we need some reasons why you believe it is close enough? Otherwise, how can we do our job? Well, what EPA did here in the revised final rule was indicate that those cumulative benefits, for most sources, would have been at the less impacted areas, would have been in the hundreds or the thousands of the deci-view. And EPA pointed out that where benefits are that small, no state has required additional controls. So EPA was clearly relying on a body of technical expertise that it has acquired in part through its review of SIPs all across the country. And it was able to supplement the state's analysis with its own expertise. I agree with everything you say is correct. But don't you have to state why one passes muster in your view and the other does not, even though they didn't take into account the cumulative effect when they should have? Well, I don't believe that it's required that the agency expressly say, we did X in this other SIP, we're doing Y in this other SIP, here's why it's different. But it's clear from the decision-making record that EPA was not taking action that was inconsistent. Well, put yourself in our position then. How are we to decide, absent a remit? Well, what you can do, Your Honor, is take a look at the rule, take a look at the other SIPs that are cited in the party's briefing, and it should be pretty clear from both the rule here and the agency's analysis in these other cited SIPs that it's disapproving SIPs where failure to perform a certain part of the BART analysis would have, was actually significant. You're taking me now back to law school because in law school when I had ad law, one of the key cases was Chenery. And if the reasons are not given by the administrative agency, then the court is not to supply the reasons. Right. Well, what's required here under the standard of review that we're looking at is that the agency articulate a satisfactory explanation for its action, including a rational connection between the facts found and the choice made. Right. So tell us where we would know that from the EPA ruling here. We'll focus on cumulativeness. Right, right. And the print is small and there are many columns on the page. It's so general that how would we know why it was okay for them to have Pennsylvania not to have done it here, but not okay in Wyoming, Arizona, Nevada, the Navajo Nation, the various other SIPs we were directed to look at? How would we know that? I think that's the difficulty, at least I'm having, and I don't know why the EPA decided it was nonetheless okay that they didn't do it here, but not okay elsewhere. Right. If I remember correctly from my read of the Federal Register notice, EPA acknowledged that Pennsylvania should have done this analysis, but pointed out that from its review of the BART memos, that these cumulative benefits would have been quite, quite small. And I believe that EPA did point out that they would have been in the hundreds or the thousands of a deciview, and that no other source elsewhere in the country had required controls when the benefits were in that range. I think there was something about the National Parks Service had some observations that had one considered using the Glatt Felter Facility's contributions. And if you looked at it from the Brigham Teen area, from Dolly Sods, from Otter Creek, and you put them all together, there would be a consequence on the cumulative visibility. And if we use that as just an example, isn't it a big failing not to have considered this, and that's just one facility where there was an impact, at least from National Parks Service's comment. Right. And EPA acknowledged that Glatt Felter was an exception, but with the exception of Glatt Felter, the cumulative benefits looked at at the less impacted sources, less impacted Class I areas for each source, wouldn't have amounted to any significant value. But at Glatt Felter, EPA did acknowledge in the 2014 Federal Register notice that the benefits of additional controls would have been more significant. So we can put Glatt Felter into a slightly different category. But what EPA did with Glatt Felter was refer back to the fact that that source's overall contribution to visibility impairment was so low, so its visibility impacts were so low, that under typical analysis, it wouldn't have even been considered subject to BART. And I think this is probably a good point for me to just identify the biggest piece of factual context here that sets what Pennsylvania did apart from what other states preparing regional haze plans have done, which is Pennsylvania took all 34 of its BART-eligible sources, and where the BART guidelines kind of provide these threshold off-ramps to say, okay, you have your BART-eligible sources, but do they emit pollutants that could reasonably be anticipated to cause or contribute to visibility impairment at a Class I area? And if they do, then you deem those sources subject to BART, and you go on and you do this five-factor BART determination that results in an emission limit. And if not, they're just out of the pool. But because Pennsylvania was part of this regional agreement, it agreed to do full-on five-factor BART determinations for all 34 of these sources, even though only eight of them had visibility impacts of over half a deciview, which is considered the threshold for contribution to visibility impairment. And it's actually only four sources that were demonstrated to have a visibility impact of over half a deciview, an additional four. Once they chose that avenue, and from your point of view, I guess you're saying they did more than they may have been required to do because not all their facilities qualify. But once you go down that avenue and they present their plan, EPA's got to review the plan under those guidelines. And even in the commentary that appears in the rule, there are statements by the EPA that say cursory information doesn't allow us to make an assessment of certain things. I mean, as a court reading that, if the EPA had cursory information that didn't allow for an assessment of certain things, how can this court determine whether it was a reasonable exercise of authority to approve it? Well, given that Pennsylvania had undertaken to do BART, the full five-factor BART determinations for these sources, EPA was reviewing those determinations and was reviewing them to ensure that they were reasonable. What EPA was able to do, and at this point I do also just want to point out that the BART memos do have much more substantive analysis than what was suggested in the petitioner's brief. They do vary, but by and large, a lot of what you would expect to see in the five-factor analysis is in there. EPA was able to evaluate the information that Pennsylvania provided, as well as the information that EPA has accumulated in its experience of reviewing other SIPs, and determined that even where Pennsylvania hadn't provided the full panoply of information and analysis that it should have, that its cost estimates, that its control effectiveness estimates, and that its visibility improvement estimates were within the range of reasonableness that EPA was able to review, based on EPA's understanding of different sources. You mentioned cost, or referred to what I understood to be cost effectiveness. At 79 Fed Register 24343, I have what looks like four instances where in the final rule, the EPA says that it cannot assess Pennsylvania's cost effectiveness calculations and visibility modeling, because supporting documentation was never made available to the agency. And again, it's at 79 Fed Register 24343. I'll read just one or two of them. For almost all of these sources, the Pennsylvania Regional HAYS SIP contains very limited information describing Pennsylvania's analyses and considerations of the BART factors. It is difficult to assess the estimates of the improvements in visibility associated with various controls given the limited information in the SIP as to the assumptions relied on in the modeling and the summary nature of the results provided. EPA agrees with the commentators, however, that many of the controls and their considerations were likely cost effective measures. Unfortunately, where controls were estimated to be more cost effective, EPA cannot assess the extent to which Pennsylvania's analyses are reasonable instruments for purposes of making a BART determination. As a matter of logic, if you cannot assess, that is, if you cannot make a determination, how can you, as you say, make a determination that it's okay? Well, what EPA did was a couple of things. First, they acknowledged that Pennsylvania itself hadn't always provided the information on which EPA would make an assessment, but EPA has the discretion to supplement a state's analysis with its own analysis and its own data, and that was what EPA did here. Is that contained in the final rule with regard to what we just talked about? EPA makes reference to it, and we believe that it's sufficient reference that the analytical path of the agency follows there. These are highly qualified staff people, and it would seem that when you say that I, as somebody who delves in this area and know my way around, if I cannot make an assessment, how can I make an assessment, even with my own information? What EPA was unable to do at times was make a full assessment based on the information that Pennsylvania had provided. But here it just said that Pennsylvania hasn't provided, at best, limited information, and therefore you can't make a determination of whether they're compliant. If you can't make a determination of whether they're compliant, how can you make a determination that they are compliant? Well, I believe that the particular point you've pulled out is with respect to cost effectiveness. That's just one. If you continue to read the rule, what EPA goes on to find is that the key part factor here that was driving Pennsylvania's conclusions and what drove EPA's finding that those conclusions were reasonable is looking at the visibility improvement. And based on the very, very low visibility improvement that was associated or would have been associated with installation of controls in almost all of these sources, EPA was able to conclude that ultimately Pennsylvania's determinations not to require additional controls under BART were reasonable. That almost sounds like a Gestalt analysis. That doesn't help us, I don't think. They're supposed to give you, they have a SIP, and they're saying we're compliant. And you're to review and to see if, in fact, what they're suggesting they're doing, are they doing it? And if so, is it compliant? And when you say I don't have enough information to make that determination or perhaps those determinations, then how are we to go and say, well, let's just let it roll? Well, it is within the agency's discretion, and I think that this is rooted in the standard of arbitrary and capricious review to approve a plan where the ultimate conclusions are determined by the agency to be reasonable, where the agency determines that there are some analytic deficiencies but they don't materially affect the ultimate conclusions. And that was what EPA did here. It conceded that there were parts of the analysis in the SIP that should have been there but that were not in there. And it's absolutely appropriate as part of EPA's review there to consider the factual context in which Pennsylvania was making these determinations. As part of that context, it's also important to keep in mind that states are given a fair amount of discretion in terms of how they weigh the different BART factors and how they go through that analysis. To do what you just said, you would say we have not been given enough information by this particular state. We have gone off on our own, and we have the following additional information. Based on that following additional information, we think that the state is compliant with the Clean Air Act. Boom. But I don't find, and this is one example that I said to you with regard to cost effectiveness, where Pennsylvania has, or EPA has gone and said, okay, we've done or have obtained our own independent information, and we believe that it is compliant for X, Y, and Z reasons. So I do want to point out one bit. This is on 79 Federal Register 24342 in the second column. Just to give you one counterexample, this is about halfway down the page. EPA says that the information Pennsylvania did provide led EPA to conclude that Pennsylvania's ultimate BART determinations were nevertheless reasonable, and then goes on to say, based on the cost estimates for other BART sources in other states, EPA has concluded that Pennsylvania's cost numbers appear to be generally consistent for such controls. EPA, you're correct that EPA did not go through and identify what those numbers were from other states. That's not in the text of the Federal Register notice. We would maintain that that level of detail wouldn't be required, and what EPA did here, which was clearly make reference to its understanding of what controls were required for other states, what emission limits those controls were associated with, and what visibility impacts those controls were associated with, led EPA to come to the conclusion that it did here. And our position is that that would certainly surpass the arbitrary and capricious threshold. Well, we'll hear from Mr. Riley then. Thank you very much. Good morning, Your Honors. My name is Robert Riley, and I represent the Pennsylvania Department of Environmental Protection in this matter. What I'd like to do is tell you that under the guidelines, the states are free to determine the weight and significance of each factor. And what we did with our BART analysis is we looked at all five of the factors, and we weighed cost and visibility improvement more heavily. Let me see if I understand correctly, because I may not. Pennsylvania, I guess, as part of the way of trying to be efficient, was part of a regional planning organization, which I guess I call Mainview. Yes, that's correct, Your Honor. And specifically, Mainview made a decision, I guess about 10, 11 years ago, to treat all BART-eligible sources as subject to BART. Yes, that's correct. So did they do that? So, yes. So what we did is we did an analysis, and we came up with 34 BART-eligible sources. And then what you have to do is you have to determine whether or not those sources should be subject to BART. And you do that to determine whether or not that source causes or contributes to visibility impairment in a Class 1 area. And so for 26 of those sources, we didn't even have to do an analysis. But because of our commitment to Mainview, we said that we would do an analysis of all 34 of the BART sources. And I think you have to put this in context, because really what we're talking about, even with all 34 of the sources, is that the sources are in Pennsylvania, but the Class 1 areas that we're talking about, the Mainview Class 1 areas, are in New England. So they're hundreds of miles away, except for the Brigantine. Brigantine is in New Jersey. So for the most part, the visibility impact of these sources is going to be small to begin with. So any visibility improvement is going to be small. So the point of BART is not to put controls on sources. The point of BART is to do an analysis to determine if additional controls are going to be cost-effective and are going to improve visibility. And what Pennsylvania determined was, no, in our case, number one, these sources, it's not going to be cost-effective. It's not going to be cost-effective from a dollar-per-ton basis or a dollar-per-decibute basis, number one. And number two, the visibility improvement that you're going to get, as DOJ said, is in the hundreds and sometimes the thousands of a decibute. And so it's very difficult for us to go to these sources and to say, well, you know, you're really not having any impact on these Class 1 areas. And the technologies that we want you to put on really aren't going to improve visibility at all, but we want you to do it anyway. How would you know that, though, if there's no analysis of the cumulative impact of all the sources together? Well, Your Honor, it is Pennsylvania's contention that we are not required under the law to do a cumulative impact analysis. Nowhere in the guidelines does it say that we are required to do that. Now, we did ask EPA because the federal land managers asked us if we would do a cumulative impact analysis. And so we said, well, it's not required. We did ask EPA. EPA did not provide us any guidance at all. Now, had EPA provided us with some guidance, certainly then we would have done a cumulative impact analysis. But they never did. And really this is where Pennsylvania farts company with EPA, because the cumulative impact analysis is not required under the guidelines. And there was clearly a distinction between what EPA did out west with New Mexico and Arizona and what they are doing here in the east. And one of the things that you had brought up was glass filter. And if you sum up glass filter, well, if you sum up glass filter, it comes out to be about a 0.6. EPA did a, in Colorado, Colorado had a similar issue. And they said, well, you know, Colorado, you should have done a cumulative impact analysis. He didn't do one. But in this case, if we sum up all of the impacts, it comes out to be a 0.67. And so that's not significant enough. Well, that's really what it comes out to be for glass filter. And quite frankly, for glass filter, two of those Class I areas are not even in Mainview. So they're in West Virginia and Virginia, which are not part of Mainview. And so, and I believe the other one is Brigantine, which comes out to a 0.2. So for glass filter, we didn't even have to do the analysis, number one. And number two, it's only impacting just one Class I area in Mainview. All right. Thank you very much. Thank you, Your Honors. Mr. Thompson. Good morning, Your Honors. Chet Thompson on behalf of Intervenor, Homer City Generation. I want to pick up on the issue of cumulative impacts. And like Pennsylvania, we, too, do not believe cumulative impacts is required under the BART analysis. If you look at JA-632, where we actually cite the BART guidelines, it says two states that when they're talking about where to put the receptors, it says you should locate them in the nearest Class I area. For other Class I areas, you may model a few strategic receptors. So the guidelines are clear, again, at JA-632, that it's not a mandatory requirement. How does that say it's not cumulative effect consideration? It's not a mandatory consideration? Well, it talks, Your Honor, as to where the receptors are supposed to be located. And it says, again, only that you should locate a receptor at the nearest Class I area. It does not say you should locate receptors at any impacted part. But isn't all that a separate issue? The EPA said in its approval that they agreed with the commenters that in considering visibility improvements expected from the use of controls, Pennsylvania should have taken into account the visibility impacts at all impacted Class I areas. So what are we doing with the EPA approval here? Your Honor, with all due respect to our friends at EPA, I think they actually misquote what they say. If you look at JA-site 31, which is the original 2012 approval, what they say is that they recommend that Pennsylvania do a cumulative impact, but say, and I quote, that BART rule does not require a cumulative impact analysis as part of BART. That's at JA-section 32. In the 2014 rule, they say that all those comments are incorporated by reference in the new document. So, again, with all due respect to the agency, they're not a collective corrective member. How are we to determine then whether their action is reasonable or not? Well, with respect to this one, I think what you should do is look at the plain language of the guidelines, which controls. When petitioners throughout their brief, they ask this court to look at the plain language of the guidelines elsewhere, they should be held to that standard with respect to cumulative analysis. That, again, is clear on its face. One other point that's worth mentioning, Council 3K refers to SIPs. You know, out west, they are not SIPs. They're actually called FIPs, Federal Implementation Plans. And why is that important? Because we are not saying cumulative impacts analysis are prohibited, only that they're not required. So when EPA does their own plan, they're free to do a cumulative impact. We take no exception to that. What's happening here was whether the Commonwealth was required to do that, and they're not required to do that. So if you go down the list of all the sites that petitioners go to, they're all Federal Implementation Plans. The two that aren't, Colorado and Florida, are the very states that actually EPA said you didn't do one, but we're fine with that given the low impacts. Thank you very much. Thank you. Mr. McRibbon? Just one or two items. On the cumulative impacts issue, we documented that for 25 of the 34 plants we're talking about, EPA has no record on cumulative impacts. So they compound their failures on technology that we talked about before with all these different sources of information. What about Mr. Thompson's point that cumulative impact or cumulative effect is not required? Well, EPA has to make a decision on the record for this decision. EPA has said in this case that Pennsylvania should have taken into account the visibility impacts at all Class I areas. We'll agree that EPA should make a decision for Pennsylvania about Pennsylvania impacts, but when you look at where Pennsylvania impacts downwind sources, the diversity of sources in Pennsylvania, it goes right into EPA's wheelhouse that haze is caused by the cumulative impacts of many sources. That's in EPA's own regional haze rule. And the cumulative impacts analysis is absent here for most of these plants, and it's based on a technology determination that's also absent, that cursory information. Okay. Thank you. Thank you very much. Thank you to all counsel for well-presented arguments and well-done briefs. We'll take the matter under advisement and recess for the day.