Oral argument — Natural Resources Defense Coun v. EPA

Petitioners v. Environmental Protection Agency et al. Ms. Shea for the petitioners and Ms. Shea for the petitioners and Ms. Shea for the petitioners and Ms. Shea for the petitioners and Ms. Good morning. Good morning. Good morning your honors and may it please the court. My name is Margaret Shea and I'm appearing on behalf of petitioners. If EPA's definition of natural event ended after the first sentence, petitioners would not be here in court today. But the second sentence of EPA's definition allows an unlimited amount and an unlimited proportion of man-made industrial pollution to qualify as part of a natural event. Is that right? Unlimited? Have you maybe overstated their position? Well, in the preamble, the agency says virtually unlimited. So anything short of 100%, so 99.99% would be enough. And although the agency's brief attempts to gloss over this significant loophole, the interpretation is a post-hoc interpretation and it cannot be squared with the actual text of the regulation. The crux of the problem here lies in the second sentence of the definition. When read on its own, the first sentence appears to require nature to be the predominant cause of both the event and the emissions that together comprise a natural event. But the second sentence eviscerates this limitation because it allows virtually any amount. So if we just have the first sentence, you think it's permissible that even though we're on the natural prong, the agency can permit human activity that plays a little role? That does not go too far. Yes, that's correct, Your Honor. We agree with that interpretation. And the second sentence allows a natural event to incorporate virtually any amount and proportion of industrial manmade pollution. In the words of EPA's own words at the time of rulemaking, and this is at JA-74, this is the case regardless of the magnitude of the emissions that are generated by these reasonably controlled anthropogenic sources and regardless of the relative contribution of these sources. As written, the second sentence thereby creates a significant loophole in the Clean Air Act and it threatens to engulf the narrow exception that Congress intended to create through Section 319. Why doesn't the phrase reasonably controlled do the limiting work that you think is necessary? Yes, Your Honor. It does not do so because even when sources are reasonably controlled, they can still emit vast amounts of pollution. To give an example, a single coal-fired power plant can generate 16,000 tons of sulfur dioxide in a single year. And this is a power plant that is subject to reasonable controls. And just to put things into perspective, that is 20 times the amount of sulfur dioxide that is produced by all of the sources within the entire District of Columbia within a certain year. And so EPA's brief... Wait, what does that have to do with how much is emitted during an event? Your Honor... That source may be a problematic source. Otherwise, we're talking about emission during an event. That's correct, Your Honor. And EPA's brief represents that the second sentence is essentially about high wind events. So that is the context that we're focusing on here. So EPA is representing that during a high wind event, if you have power plants that are reasonably controlled, that are putting a large amount of industrial pollution into the air, and those emissions are swept up by the high winds, then they can qualify as part of a natural event. So you do need that piece of nature that's part of the event, but the emissions component can be coming from anthropogenic sources. Suppose the regulation were limited to what I think EPA regards as the paradigmatic event, which is, right, windstorm sweeps through the desert and it's kicking up dust all over the place and happens to kick up some dust coming off of Interstate 10 as well as the desert. And they say, we're going to just treat that as an act of nature. Would you quarrel with that? Your Honor, if the dust that's coming off of the interstate just composes a little amount of the overall emissions, we don't quarrel with that. But if you have the situation where the dust from the desert is 98%, 50%, but 50% of the dust is coming off the interstate, then we do think that's not consistent. But if that's true, then aren't you saying that the regulation might be overbroad in some of its applications, even though the core of it, or at least what EPA thinks is the core of it, is probably okay? Your Honor, the problem with that is that there's nothing in EPA's regulation that constrains the agency to the interpretation that it's advancing its briefs. And we would think that the majority or many of the applications that are impermissible would actually be allowed by the actual regulatory text. And I'd like to draw the Court's attention to footnote 8 on page 18 of EPA's brief. And that's where we think the agency draws a very critical distinction. The agency distinguishes between, and we're talking about just high wind events, the agency distinguishes between dust that is lifted off the ground by the wind, on the one hand, and pollutants that are already airborne at the time that they are entrained by high winds. So in the latter category, we have not only particulate matter that is put into the air by smokestacks and other anthropogenic sources, but we also have the large amounts of gaseous pollution that come from reasonably controlled sources, but that never settle to the ground. You've actually gone one step beyond where I was in the analysis, and that's fine. Don't we start with what does the phrase natural event mean? I mean, Congress has said a natural event, and what's the meaning of that? Yes, Your Honor, and Congress has made it very plain by use of that term. The statute does not provide a definition of the word natural, and so there doesn't seem to be any distinction. That's critical, right? We start with that. Congress hasn't defined the phrase natural event. So what are we to do then to decide whether it's clear or whether it's ambiguous? That's the next step in our reasoning. Yes, Your Honor, and in that situation, the courts have said if Congress has not defined the term, then we go with the ordinary meaning of the term. And natural means of, arising from, or forming part of nature. And it defies any ordinary understanding of the word natural to think that a cloud of power plant pollution can be natural simply because it's being blown around by wind at 25 miles per hour. Is that right? When a hurricane ravages an area and kicks up pollutants, that's not a natural event? Your Honor, I think that goes back to the idea of reasonable controls and the distinction that EPA draws. So if the hurricane... No, I think I'm focused right on what does it mean to say something's a natural event. A natural event is... A hurricane that kicks up pollution is not a natural event. It is. It certainly is, Your Honor. And the reason for that is in EPA's limiting construction, which is when it's nature that is putting pollution into the air. And air pollution is defined as emissions that are in the air, then that is natural. But when you have smokestacks that are putting the pollution into the air, then nature is not the source of those pollutants. And Congress made this very clear in the congressional report. It said that when you have an ecological process that itself generates pollutants, that can be a natural event. And the examples that Congress provided were volcano eruptions and forest fires. And in both of those situations, you have nature putting the pollutants into the air. In contrast, Congress made it very clear that when you have climatological occurrences that merely influence pollutant behavior, that that aren't themselves generating the pollution, that's not a natural event. Where did Congress make that clear? In the congressional committee report. And what should that have to do with our understanding of their use of the phrase natural event? Your Honor, this court has said that the legislative history is appropriate to consider during the first step of the Chevron inquiry. And so that informs what Congress plainly meant by the term natural event. And it goes to the understanding that the emissions that are imputed to a natural event must not only be naturally dispersed, but also naturally generated or naturally put into the air. How do you understand that we count pollution that is already there one minute before the bad occurrence? It's emitting, it's there. And then we have the high winds. Sure, Your Honor. This is the piece I'm not really getting. Are you saying that the government means to discount that, count that? In other words, it's already there. And then a minute later, there is this event. That same pollution in the same volume as there, just swirling more because there are winds. What is your understanding about what we do with that? Your Honor, we're not talking about that body of pollution. So let's say a smokestack is putting pollution in the air, and that would normally be counted by the monitors. That is not part of the natural event. That's just the background emissions. So what we're talking about is that the emissions that normally wouldn't be picked up by those monitors because they're being blown in from further away. So on a normal wind day, let's say that the pollution would only be dispersed within a 50-mile radius. But on a day when the winds are at 25 miles per hour or above, we have pollution coming in from beyond the 25-mile radius. So it's that body of pollution that we're talking about here. Now, does that body of pollution count twice for where it originates and where it's being blown to? Yes, Your Honor, it does if it's registering on those monitors where it's coming from. And for those monitors, it would normally be part of the background pollution anyway. But so what really matters here is that it's also registering for these monitors where it normally wouldn't reach. And the Clean Air Act has provisions that deal with this kind of interstate transportation of pollution. There's Section 110, which makes it very clear that a jurisdiction is responsible for making sure that its pollutants aren't contributing to exceedances in other jurisdictions. And so this is part of the scheme that Congress intended. And EPA's enlarged definition of natural event threatens to undermine that scheme. And regardless of how reasonable the limiting interpretation that EPA has put forth in its brief is, the problem is that there's no support for that narrowed interpretation in the actual text of the regulation. So someone who is just reading EPA's regulation would have no basis to think that the regulation can be applied only in the very narrow way that EPA is now suggesting. Well, suppose there was an opinion that confirmed that. Yes, Your Honor, if there was an opinion that confirmed that, then that would perhaps provide notice. But both this Court and the Supreme Court have made it very clear that an agency's interpretation of its own regulation is not entitled to any deference where it is contradictory and plainly inconsistent with the actual text of the regulation. And that's the Drake v. FAA case, the Christopher v. Smith claim. Well, that assumes that it's plainly inconsistent. I thought your argument was that their reasonable interpretation is there. We understand it, but we can't count on it being applied later. And I asked you, well, suppose we say they gave a reasonable interpretation of what they're proposing to do, and that's the basis upon which we deny the petition. Let's see how that can be a worry for you then. Your Honor, that is an interpretation that has been offered by the first time by appellate counsel. It's not the interpretation that has been offered by the agency itself contemporaneous with the rulemaking. There's nothing in the text of the rule that says that. There's nothing in the preamble or other documents that say that. But most importantly, it's not in the text of the regulation itself, Your Honor. And this Court and the Supreme Court have made it very clear that unless the understanding is articulated in the plain text of the regulation, that cannot be credited to the agency at this point. There's also an interpretive presumption that we construe. We try to construe text to make it lawful rather than unlawful. So if that limiting construction of the regulation were necessary to make it a reasonable interpretation of the statute in a Chevron sense, why wouldn't we impose it rather than just throw the whole thing out? Your Honor, the Court has addressed situations like this before. So, for example, in the Cuomo v. Clearinghouse case, you have a regulation that's very broad, and then you have an agency interpretation that is narrower and perhaps reasonable, and that narrower interpretation could fit within the broader interpretation. However, the Court found there that when you're just looking at the broad regulation, it's inconsistent with the narrow one being advanced by the agency because someone reading the broader actual language would not have any idea that you could actually – that what the agency meant was limiting construction. See, I mean, you have a problem there because you're talking about a limited population of people for whom we would worry about giving notice, and they would all have notice. If there was an opinion that said that's all this means, there's no problem. Sure. Notice has been given, and the people who are fighting about this would understand that the Court said that's all this regulation means. Yes, Your Honor. I mean, I can understand in other cases where your argument about notice would be compelling. It wouldn't be compelling here. Mm-hmm. If inside of the Drake and the Christopher cases, if this Court does decide that the agency's – You are right that you can find cases that say what you're saying, but we can also find cases that do it the other way as well. Sure. Which say there's clearly a limiting construction. We don't see how you can read it otherwise because otherwise it would make the regulation nonsensical to go further. I mean, there are cases that say that as well. Okay. Yes, then, in your honor, that's correct. And an opinion that memorializes or gives legal effect to the agency's limiting construction would probably put the public and others on notice of that. But we would ask the Court to make that clear in its opinion so that the agency is not free to change its interpretation at some future point. And – Great. Unless there are further questions from my colleagues, we'll give you back two minutes. Thank you. We'll hear from the government now. Good morning, and may it please the Court. My name is Sue Chen, and I represent the United States. With me today is Emily Seidman from EPA's Office of General Counsel. There's been a lot of confusion this morning about the meaning of emissions, so I'd like to make that clear before turning to explain how the definition works and then briefly addressing Petitioner's argument that EPA's interpretation is somehow post hoc rationalization. As EPA explained at JA38 and 101, emissions means aerodynamic entrainment. So in plain English, to emit means to put into the air, to make something airborne. So, for example, wind sweeping up dust from the ground is generating emissions, and those emissions are said to result from the wind. A factory that releases pollutants into the air through its smokestack, perhaps, is generating emissions, and those emissions are said to result from the factory. And that's true even if later on wind comes by and transports those emissions elsewhere, because that does not change the fact that it was the factory, not the wind, that generated the emissions, that it was the factory, not the wind, that released the pollutants into the air. Whether it's within the region or elsewhere? Whether it's within the region from which the smoke originally emitted or transported to another region? That's true regardless where it ends up. But you're saying your rule doesn't change that? That still counts? That in that case, it's still the factory that's considered to have generated emissions. Yes, Tom. So everything coming out of the smokestack gets counted? As the factories. As the emissions generated by the factory, it's not the emissions generated by the wind. I want to be clear about that. All right. So with that in mind... But the point is it gets counted, right? Those emissions are counted. They're not excluded from the report. They're not excluded from... That's right. If the emissions generated by the factory ends up near a monitor, that's going to get recorded. Right. It's going to record as part of routine emissions. So with that in mind, let's turn to the definition of natural event. Now, there are two sentences in the definition, and you have to read them both, and you have to read them in the order in which they appear. So start with the first sentence, which says, Natural event means an event and its resulting emissions, which may recur at the same location in which human activity plays little or no direct causal role. There are two basic elements required for a natural event. You need an event, and you need its resulting emissions. Think of these two elements as a door to the universe of potential natural events, so that if you don't have an event, you can't get in the door. If the emissions that you're talking about did not result from or not generated by the event, you can't get in the door either. And this is where petitioners trip up. They completely ignore that requirement. The routine industrial emissions they're worried about cannot get past the door. And why is that? Because routine industrial emissions are not an event. An event is a deviation from normal conditions, and routine industrial emissions are routine. They are the emissions generated by industrial sources operating as a matter of course. So there's no event here. And for the same reason, routine industrial emissions did not result from an event. And so routine industrial emissions get shut out of the definition. You don't even proceed to the second sentence. Petitioners skip straight ahead to the second sentence, and they read it to say that routine industrial emissions are a natural event or can be part of a natural event, and they're wrong for two reasons. The first is that their interpretation ignores the first sentence's special requirement that you need an event and its resulting emissions. Nothing in the second sentence exempts you from that requirement, and as I've explained, routine industrial emissions do not satisfy that requirement. So for this court to accept petitioners' interpretation, you would have to strike out the words, an event and its resulting emissions from the first sentence. The other reason that they're wrong in their interpretation is that the second sentence does not actually talk about routine industrial emissions. The second sentence says, for purposes of the definition of a natural event, anthropogenic sources that are reasonably controlled shall be considered to not play a direct role in causing emissions. What are we talking about there? I'm sorry? What are we talking about there? That's what I'm getting to, Your Honor. I'm getting excited waiting. I'll push you out of your suspense. These emissions refers back to the events resulting emissions in the first sentence because that is the only other place in the entire definition that talks about emissions. Now, under the first sentence, the emissions have to result from not just any event because if you read on, you'll see that the event has to be one in which human activity plays little or no direct causal role. In other words, an event that's predominantly natural, what we call an act of nature. So the emissions have to result from, have to be generated by an act of nature. And you put all this together, and what the second sentence says, and all that the second sentence says, is that reasonably controlled anthropogenic sources do not play a direct causal role in the emissions generated by an act of nature. It does not say anything about routine industrial emissions, and it certainly does not say anything about routine industrial emissions or any other human activity being a natural event. As we explain in our brief, the purpose of the second sentence is to clarify this unique situation where you have emissions generated by the high-speed winds that include particles that originated from an anthropogenic source. And it tells you in that situation whether the anthropogenic source played a direct causal role in the emissions generated by the wind. Suppose you have a different example. You have a natural event causing emissions, very substantial emissions whose source was human-made. So, I don't know, a bolt of lightning hits a reasonably controlled container of toxic gas. And as a result, the only emissions resulting from that event are the toxic gas stored in the container. That seems a little different from the wind in the desert case. So, in that situation, I am not sure if it would be considered a natural event or whether it would be considered an event caused by a human activity unlikely to recur. This is a very fact and sense analysis. But your regulation would classify that one as natural, even though it seems like it's a much closer question. That would, I think, depends on the facts. But I think to get to your question, maybe another hypothetical that's less problematic would better illustrate the point that you're concerned about, Jess Casas. So, in our dirt road example, let's say all the dust swept by the wind came from the dirt road. In that case, provided the dirt road was reasonably controlled, that would be a natural event. Yeah. It just, that hypothetical seems unlikely to me. It is. And the more realistic hypothetical with the dirt road, most of the dust will come from the desert, and that's part of the intuitive appeal of the regulation. But the regulation isn't limited to that kind of case. Your paradigmatic case is natural event when emissions are natural, right? The dust from the desert as opposed to the dust from the road. Right. But in your example, it's not, it could be the case that because the human source, well, in your case with the explosion, I think it is possible that it would be considered an event caused by human activity unlikely to recur. But even if it is evaluated under the natural event framework, it would still be the case that without the natural event, you wouldn't have all that pollutants in the air in the first place. And remember the context that we're talking about here is the Clean Air Act. You wouldn't have it, and therefore what? I want to make sure I understand your conclusion. You started a sentence and then moved on to another part. You said, what do you mean to say? So in the example of a lightning strike, you would have an act of nature, which is the lightning strike, and you would have, and that event then caused pollutants to get in the air. So you would have its resulting emissions. And under the second sentence, if those emissions contain particles that came from an anthropogenic source, and that source was reasonably controlled at the time the event happened, then you would have a natural event here. And that makes sense because it was the act of nature. You would still say the emissions were caused by the lightning strike. Because if the lightning strike was what put the pollutants in the air, yes. Yes, that is the analysis. And remember, we're talking about air pollutants. Just to be clear, but if that lightning strike had hit a smokestack, were pollutants already going in the air, then? I'm sorry, if the pollutants were already in the air? Yeah. No, then that would be counted as the emissions generated from the factory and would not be part of the natural event analysis. So the question really is, what put the pollutants in the air? I'd like to very briefly address Petitioner's argument that EPA's interpretation is post hoc rationalization. There is nothing post hoc about any of this. EPA's reading of its natural event definition reflects the agency's well-established position that although natural events can include some minor human component, human activities are not natural events. And you can trace this position back over 20 years. The 1996 guidance at JA202-203, EPA recognized that natural event can include human contribution only under very limited and narrow circumstances. In its 2007 definition, EPA similarly explained that a natural event is an event in which human activity plays little or no direct causal role. And in its response to comments in this rulemaking at JA147, EPA addressed the natural event definition, and it said, under this definition, repeated and long-term human activity would preclude the event from being natural. So EPA's definition and its interpretation of that definition articulate the agency's longstanding and considered judgment about the permissible relationship between natural events and human activities. I'd like to leave the court with one final thought, which is what you have heard here today are two starkly different readings of the natural event definition. Petitioner's reading ignores key parts of the regulatory text and produces this absurd result that human activities are natural events. EPA's reading, in contrast, addresses every word of the definition and results in a sensible treatment of natural events that's consistent with its longstanding position. To me, you've jumped right over the Chevron 1 analysis. You're just assuming that there's ambiguity here, and your argument strikes me as being all in Chevron 2 land. Is that right? Do you want to say anything about Chevron 1, why you think this might be ambiguous, or is it a gap filler? Well, the term natural event is not defined in the statute, and the term natural is— But it is in the dictionary, right? But that doesn't tell you what natural really means. One of the first things they said today was that it's okay for a natural event to include some minor human component. That right there tells you that it's not all that clear what natural means. Unless the court has any further questions? Thank you. Thank you. Your Honors, there are just three points I'd like to make. The first is that if, as Judge Edwards suggests, the court believes that EPA's interpretation is reasonable, petitioners think that that would address the problem, but we would ask the court to memorialize that in a written opinion because we don't think that the actual text of the regulation conveys that limited interpretation, and we would wish that the agency is held to that narrow interpretation. The second is that, and related to the first, is the actual text of the regulation does not say what EPA is now saying that it says. EPA's counsel had mentioned JA088 and the quote talking about routine emissions generated by and transported from anthropogenic sources are not exceptional events. But again, that's just talking about normal emissions on days without a natural event. It's talking about the smokestack emissions on days when you don't have high winds. And the universe of the events that we're concerned about are the high wind days when you have the pollution that's being blown over and that's being considered part of the natural event. And the agency says that this new interpretation is well established. It talks about the 2007 regulation. But it's critical to note that the second sentence, that's in EPA's 2016 definition, the current definition, that second sentence was not in the 2007 regulation. And all that the 2007 regulation said was something very similar to the first sentence of EPA's current definition, which is that nature must play the predominant role in the natural event. I'm sorry, how is your broader reading consistent with the word resulting in the first sentence? Natural event means an event and its resulting emissions. So if you have the smokestack that is emitting pollution regardless, and then you have the very high wind day, I would think those emissions will not be resulting from the event. Your Honor, that is certainly one plausible reading of that word. And, again, if the court were to accept that, we would just ask that it memorialize it. But we believe that as written, when you're reading the phrase resulting from, that the pollutions that are already there, that are picked up. Would that be sufficient to address your concerns in this case, if we said that much and no more? Yes, Your Honor, it would be. Thank you. Thank you very much. The case is submitted.