 My name is Connor Goodwin and I'm ProPublica's Interim Director of Communications. Welcome to How Journalists Can Report on Toxic Cost Bots. For those new to us, ProPublica is a nonprofit newsroom dedicated to investigative journalism. Earlier this month, we published a groundbreaking map that showed the spread of cancer-causing air pollution from industrial facilities into nearby neighborhoods. Now, we'd like to empower journalists across the country to use this new tool to illuminate the risks their communities face and to hold government and industrial facilities accountable for pollution problems. Today's webinar is brought to you in partnership with the Society of Environmental Journalists and with support from McKinsey & Company. Now, allow me to introduce a few members of the reporting team behind this First of Its Kind project. Ava Kaufman reports on technology for ProPublica. She was previously a contributing reporter at the Intercept. Layla Eunis is a news app developer for ProPublica's local reporting network. She was previously a data reporter with the New York Public Radio and Gothamist. Maya Miller is also an engagement reporter with ProPublica's local reporting network. She works with journalists across the country on community-centered investigations. Our moderator today is Alex Zayas. Alex is an assistant managing editor at ProPublica, running a team of reporters and coordinating across the newsroom to enhance his special projects. Stories she edited have won two National Magazine Awards, two George Boca Awards, and the Pulitzer Prize for future writing. Before I hand it off, I want to know that the session is being recorded and a link to the video will be emailed tomorrow to everyone who registered. And once again, if you have a question, please click the Q&A icon at the bottom of your screen and type it there. Thanks again for joining us. I'll let Alex take it from here. Hey, everyone. Thank you so much for joining us. I see the numbers of attendees climbing. I'm super excited to have you here. As Connor said, we really want to empower reporters across the country to be able to use this first-of-its-time tool. The reporting that we've done recently has led us to believe that the EPA is taking this tool seriously, and we have seen reporters outside of ProPublica who have used this tool get some local impact. Just yesterday, we read a story out of the Ann Arbor News in which reporters identified some questionable emissions and state officials jumped on it. So we really want to make this tool something that you're able to use to bring authority to the work you're doing. We want to not just have people saying that they feel symptoms and complaining about the pollution around them, making them sick. Now we have black and white, some risk estimates that will actually show just how much risk people are being exposed to. So without further ado, I want to give a little bit more of an introduction to this all-star team here, Leila, who with her partner on the on the NewsApps team, Al, suffered for two years putting this tool together. She will describe for you guys how it works and what you can take from it. And she, along with Ava and Maya, have been talking to a lot of people on the ground, have started to do a lot of community reporting and also reporting with government sources. So they can answer your questions about that as well. We want to make this a dialogue. So please, if you have any questions that pop up, throw them in the chat and Conor and I will do our best to get to them. So without further ado, Leila, I'd love to have you give us a little walkthrough of the tool. Yes, absolutely. I'm going to share my screen. Can everyone see it? Okay, great. So yep, so this is what we call the most detailed map of cancer-causing industrial air pollution in the U.S. And at the top here, you see three different hot spots that our analysis identified. This one right here on the left is the most famous one. That would be Cancer Alley, Louisiana. In the center here, we have Houston, which we're going to explore together shortly. And then here on the right is an area in Charleston, West Virginia. So I'm going to scroll down to this zoomed in really crisp image of what our map looks like when it is zoomed in. What you see here are these individual boxes. We refer to them as grid cells. The model underlying our map and our analysis. It's an EPA model called RSEI-RISI, the risk screening environmental indicators model. If you'd like to read more about it, we wrote a very detailed methodology that sort of explains that model and how we created these risk estimations using it. What you should know here, however, is that each of the grid cells that the model outputs chemical concentrations into are 810 by 810 meters. That is less than a kilometer wide. So what you basically get is a super granular view of what's happening on the ground in these communities. So by using a model at this level of granularity, you can actually understand what the risk might be like at the fence line of a facility. Right up here, too, we show you the range of risk estimates. The way that risk is talked about in the EPA is an increments of one in X thousand. So this number right here, one in 100,000, means that if a community of 100,000 individuals are continuously exposed to a certain level of industrial pollution over a presumed lifetime of 70 years, one person would be expected to develop cancer from that exposure. It's important to note that that cancer risk is getting added on top of the cancer risk that we all already have just by virtue of being humans in the world, having family histories, being exposed to tons of stuff all the time, not just cancer causing air pollution. And so it's important to note that none of us start at zero and that this incremental lifetime cancer risk from toxic air pollution can be the difference between having a familial history of breast cancer and actually developing breast cancer, for example. The image that you see here is a zoomed in area of the Texas Gulf Coast. This right here is one of the most toxic plants that our analysis identified, Huntsman Petrochemical. And so let's now dive into the actual map. So it looks like a map, what it really is, is an interactive database. And basically, you can interact with it by clicking around, searching for addresses. We have multiple views on this map. What you see at the top here are a selection of hotspots. So we define a hotspot as a contiguous area of grid cells where the estimated cancer risk is above one in 100,000, that risk level that I mentioned just above. That level is right in the center of the EPA's range of risk benchmarks, of acceptable benchmarks. So at the high end of that range is one in 10,000. The EPA has said that anything worse than one in 10,000 is unacceptable. And at the low end of that range is one in one million, which is the risk level that many environmental advocates believe should be the actual risk level that the EPA uses. And it is the one that was included in the original Clean Air Act. So let's go ahead and take a look at some of the hotspots in the selection up here. I mentioned Houston, Texas, so we can just go there by clicking on it. And so this view that you're looking at right here is the hotspot view. So this is one of the two views in our map. The hotspot view gives you some sort of summary statistics of this hotspot, right, this contiguous area of grid cells with the estimated risk above one in 100,000. So you get the population of people within that hotspot, not within the city but within the area, right? And it's important to note that some hotspots can cover multiple cities or might only exist in a corner of a city. And so we give you the population within that hotspot, the number of people affected above that cancer risk benchmark. We also tell you what the average risk is within the hotspot. So if you average the estimated risk levels of all the grid squares within the hotspot, this is the average risk. And then we also tell you what the highest risk in the hotspot is. So that would be most likely the square that the plant exists in. The other thing that we give you in the sidebar in the hotspot view are the major facilities that are driving the risk in that hotspot. So I had mentioned Hudson Petrochemical, there it is, Seal & Knees and Equistar. All three of these plants are major emitters of the chemical ethylene oxide, which is one of the most toxic chemicals emitted by American industry. Our analysis identified it as the most toxic chemical, in fact, emitted by American industry. So if I want to exit the hotspot view and get into the location specific view, I can do a couple things. I can either search for an address or I can just click within the hotspot. So I'm going to click in the hotspot. And now you'll see that what happens, what exists within the sidebar has changed. So now, as you can see, I have my, this yellow locator icon shows up and it is clicked into the exact location that I had clicked on the map. And so it is basically going to give me information for this particular grid square. The information that it gives me is in the sidebar. It tells me the estimated cancer risk in that particular grid square. So in this case, it is 1 in 6,800, which is 1.5 times the EPA's level of acceptable risk, which was again, 1 in 10,000. So basically the EPA would consider the estimated cancer risk in this square to be unacceptable. Okay. And then, of course, we show you where that is on the range of risk on this legend right here. This is one of my favorite things that we have in the entire app, which is this little risk over time. I find it incredibly helpful because I think a question that many readers, myself included, have is, well, is it getting worse here? Is it getting better? Is it staying the same over the five-year period of our analysis? So this can help you to learn about that. And then similar to the hotspot view, you learn about the plants that are driving the risk in this individual grid square. Another really helpful thing I like here that we do is we give you the actual percent contribution on average of each of those plants over the five-year period. So equastar chemical, if you live in this particular grid square, is estimated to be contributing 61% of your cancer risk here. And then we list the chemicals that it is emitting that are driving the majority of that cancer risk. So again, ethylene oxide and acetyl aldehyde. I never pronounce that properly. So now let's go ahead and do an address search. So I'm going to look up, instead of an actual address, I'm just going to look up a city. So let's go to Memphis, Tennessee. The locator icon usually just gets plopped in the geographic center of whatever place I search for. And then since it's outside of a hotspot, the map prompts me to see the closest hotspot to me. So I click see hotspot and then I am in the hotspot view. Again, so this should look familiar. But again, if I decide that I want to exit the hotspot view and enter the location specific view, I just click within the hotspot. And the way to toggle between these two views is just to use this little button up here. So we always include in location specific view, this is part of a hotspot around Memphis, Tennessee. So I can click back here to kind of zoom back out into the hotspot view. I can also scroll over here and take a look at another big polluter in the area, SFI of Tennessee. And then, you know, I'll just call your attention to the fact that as you click around the hotspot, you'll see the risk change. And, you know, pretty logically the further away you get from the location of the plant, the risk decreases. So one thing that I do find really great about showing this granularity and really, you know, seeing the dispersion of the chemicals is that you can understand how the model is actually computing that dispersion, right? The model takes in all kinds of different inputs, including wind direction and topography, weather modeling data. And so, for example, and right here, we can see that because of the topography of the land because of all these different inputs, that pollution is kind of getting blown a little bit more in this direction, right? Rather than, you know, there isn't sort of like an equal amount of pollution on all sides of the epicenter of the plant. So I find that useful to learn and would call your attention towards that. Another interesting fact of this plant in particular is that we can see you have pretty low risk in these three years, and then you kind of have a jump in 2017. So that's something that I would, you know, that would catch my attention if I were reporting on this plant and, you know, would possibly, you know, make it or definitely make it into the list of questions that I ask this company and state regulators reporting on it. Another thing that we include here is that we kind of have links for each of these chemicals. So if I click on Chromium, for example, I am taken to the EPA's page on Chromium. Iris is the EPA sort of toxic chemical database. And this is pretty technical documentation, but we do find it helpful because it lists, you know, what the EPA considers, what parts of the body the EPA considers this chemical to affect. And it is important, of course, when you're sort of reporting on a plant and you're trying to find individuals who are potentially affected by that plant to make sure that the health effects they claim to have sort of align with the ones that are caused by the chemicals in the area. So that's why we sort of offer these chemical resources. And there are obviously other places that you can learn about these chemicals, but we think that Iris is a really good first step. And I will just, for fun, go to one other location that we are all familiar with. Let's go to Chicago. Chicago is an interesting city. It's sort of dotted with all sorts of tiny hotspots. And you can just click around the different ones of them and learn about them and see how their risks have changed over time. It's also useful to note that some of our hotspots are truly only one grid cell wide. That's because that is the only grid cell that none of the other grid cells around it are estimated to have a cancer risk greater than one in 100,000. And so, yeah, I want to ask now, my colleagues, if there's anything that I missed here that I should be explaining that I have forgotten to explain. I can chime in with some questions we're getting in the Q&A. I think it would be really helpful for folks to understand a little bit more about the risk levels. It looks like somebody is asking whether one in 100,000 is the base level of no cancer risk. And that's not quite right, right? I mean, there are plenty of people who live in one in a million or less risk. So can we talk about kind of the thresholds that we have for considering something is problematic and kind of, you know, how we think about these risks? Yes, sure, sure, sure. So there's definitely a long history with these risk thresholds. And I do encourage folks to read our methodology, which goes into this a little bit in our main story. But essentially, the EPA sort of included some, or Congress, sorry, included some pretty fuzzy language in the original Clean Air Act, which did not mandate the risk level of one in a million to be basically the hard line that chemical companies could not pass. The Congress said essentially in the first generic of 1970 that it wants to protect as many Americans as possible from a risk level greater than one in one million. What later happened was the 1989 benzene rule included language that introduced a different risk threshold, which was many, many times less protective. And that is the risk threshold of one in 10,000. So now environmental activists, environmentalists and scientists and the EPA kind of refers to this range of risk between one in 10,000 and one in a million as the fuzzy bright line. What we selected as the risk threshold that we wanted to use as a cutoff for our map was the level sort of in the exponential center of that range, which is one in 100,000, which is a risk level that again, environmental scientists that we spoke with believe is not even protective enough. So is that, I want to check with like Ava and Maya, does that seem like I go to explaining that. So we selected the one in 100,000. But of course, you know, you could imagine that our map would look substantially different if we had instead chosen to use the one in one million threshold. We were interested in being somewhat conservative in our risk estimates and sort of in the range of risk that we were showing. And the other factor is that the EPA only models out 50 kilometers around industrial facilities. When you're dealing with most chemicals, 50 kilometers out, you have very low or completely insignificant concentrations of those chemicals. However, with certain chemicals, such as ethylene oxide, it's a very high, highly potent chemical. 50 kilometers out from a facility, you can still have pretty substantial concentrations. And so even at that 50 kilometer mark, you can have sort of people experiencing one in a million cancer risk. So ultimately, the risk ranges are somewhat confusing. And that's why we feel that it's useful to discuss kind of the one in 10000 mark as sort of the EPA's, what the EPA considers unacceptable. However, that many individuals who we've interviewed believe that the one in a million mark is the appropriate risk benchmark. And for what it is worth, experts do say that one in 10,000 is too risky for people. Yes, exactly. That's it. And we have been interested in places that are doing whole big stories on places that are like one in 42,000, one in 20 something thousands. So one in 10,000 certainly should not be your cut off for being interested. Yeah, absolutely. There's a couple more questions involving the data. One is from Mark. Oh, that was just answered by Ava. I'll let Ava have just answered that question. Hi, Mark. So okay, we've got one from Jacqueline. Nobody's answered yet. Can you talk a bit more about those one cell hotspots? Are they worth investigating or reporting? Are they more like noise? I wouldn't consider them noise, because I do think that they're interesting. I think that they are less interesting than multi-cell hotspots. And I think that, again, it's always possible that our map is not encapsulating the full scope of the pollution that the industrial plants in our analysis are emitting. Plants are supposed to, for example, report their startup shutdown malfunctions releases, for example, any accidental leaks, but they don't always. And so it could be a useful place to just sort of do a little bit of reporting ground, see if people are bothered by the emissions of that plan. But I would say that if you are between looking into a one cell hotspot and a multi-cell hotspot, I would certainly go for the multi-cell hotspot. There's another question about the disconnect between risk levels and the permit requirements of EPA and the states for those chemicals. Do you guys want to talk a little bit about that? Sure. So, yeah, interestingly, many of the companies that we asked for comment when we published our first story and map told us, well, we are not violating any permit requirements. And based off of their reported emissions alone, that is, in many cases, correct. However, even still, even with their permitted, even staying within their permitted emissions, if you actually do the modeling, they are, in some cases, for example, huntsman, elevating cancer risk above levels that the EPA, the agency considers unacceptable. So, that is a disconnect that is a result of a regulatory framework that does not sort of, you know, connect, right, the state and federal sort of what the state governments are permitting and what the federal government has said is an acceptable range of risk. And that is, you're right to call it a disconnect because it is one. Looks like the reporters are answering some other questions on here. But I'd love to move along to a little bit more of our shoe leather efforts. We'll certainly get back to the data. But, you know, Ava, I'd love to have you talk a little bit about storytelling choices and taking, you know, this tool, which is really just a starting point and trying to, you know, figure out how it is impacting people and start to hold government and industry accountable. So, do you want to talk through that process? Yeah, definitely. So, we, you know, the first step for kind of telling a story, I think, with this app is kind of understanding the model, which is why these questions are really helpful. And I think Layla's overview gives, you know, a good sense of kind of what it can provide. But the next step we found is actually just verifying these emissions data. And I see that there's a question about that in the chat that's relevant. So, you know, in some stories, you know, you might approach kind of who you're writing about later in the process. In this case, we really recommend if you're writing about a hotspot or a group of facilities, and we did this for, you know, the stories we have that are about to come out and the main story that we published alongside the map, contacting these facilities, you know, early and often, and asking them to confirm that the data they submitted to the federal government on which this map is based is accurate. And that's because all of this data is self-reported. And it's supposed to be looked over by the federal government. There's supposed to be a quality assurance process. You know, obviously, one could argue that it's in the company's own interest not to report data that would make them look, you know, really bad. There's a transparency incentive built into the database. However, we found, you know, all kinds of errors, which we sussed out and corrected, especially for the largest facilities. So, as a first step, we recommend going to the company, asking them to confirm the data, even if you don't know whether or not you're writing a story. They can update the data any time of the EPA, which we would then, you know, update if notified in our own app. And also letting them know that it appears that based on the data they reported, there's an elevated cancer risk in the area so that they have a chance to kind of, you know, respond and understand that kind of, you know, initial claim as you're working through the story. You know, the next step we recommend is also kind of digging into state and local data. So the way this whole system works is, you know, we're making this map based on federal data, but the EPA delegates much of its authority, and certainly all of the permitting process to states, local, and tribal government partners. So often they have their own error emissions data. And in some cases, you can use that data to supplement, maybe to even tell a sharper story. There might be gaps in the data, which is, you know, a story on its own and something that EPA sources have confirmed, you know, both with us and in the many reports they've produced is just a major problem. You know, the data is itself like kind of part of the story we found. And so it's really worth just like digging and then figuring out what is and isn't available and how you're going to frame the story around that kind of right away. You know, once you kind of have those requests out and have that going as you're localizing, you know, a story around, you know, a particular area or hotspot, it's also great, of course, to kind of blanket that area with shoe leather reporting. You know, we kind of get into this in our reporting recipe, but everything from calling local community groups, you know, church leaders, you know, officials in the area, political, you know, everyone from lobbyists who work with industry to state legislators, to teachers, to concerned parents, there's often environmental advocacy groups already present, active and thriving in a lot of the hotspots that we've identified. And they're often doing their own research, their own air monitoring and can be incredible resources to lean on and that we've, you know, found, you know, to be, you know, beyond valuable sources. And, you know, as, you know, finally, I guess the last step I would say if there is a recipe for doing this is just putting in a lot of records requests as well. You know, TRI and EPA data is just one source of data. Someone mentioned in the questions, you know, what about OSHA data? That's something that we've pulled for some of our stories. You can also reach out to like local fire departments. There's emergency planning, like data that facilities are supposed to keep, you know, reports of the amount of chemicals they have on hands in case of emergencies or spills or leak. That could be a really good way to kind of ground truthing what you have available. Permits in terms of construction and zoning can also give a sense of waste and also relationships with local officials. And lastly, I would say like digging into the health effects is obviously key. You know, this map and this pollution is meaningful because it's causing real harm. And real health effects in people's lives. And while obviously we're only showing estimates, you know, between tumor and just use, which are sometimes useful, although measured in a different way, we got we did get a question of that. And talking to city officials who may have conducted their own health studies and research, and also looking at lower level health effects like asthma, not just cancer, which our map shows, but, you know, other indicators that there may be a pollution problem like asthma or difficulty breathing, COPD, autoimmune diseases, all of those are really key to dig into. So I would say those are the kind of really big picture things that we would recommend, you know, getting started on as you're rounding out this reporting. Maya, you're up. We'd love to hear a little bit more about the way that you have conducted outreach around this project. Survey for those who live and work near these facilities, the mailers and opportunities for local tip collaboration. For sure. I'm really excited to be with all of you and talk about this. We've been steeped in it for a while, so it's fun to be able to share and brainstorm ideas with you all in the chat. Yeah, so one of the things that I was working on as we were kind of launching this was talking to people in hotspots across the country to just really do some deep listening and understand the questions that they had. And we started hearing repeated questions and so in conjunction with the map, we also published a service post that has kind of frequently asked questions. So as you're doing this reporting, if you hear questions that come up a lot, feel free to point people to that. We hope it can be a helpful resource. And then after we publish, we've been hearing from a lot of people across the country, someone asked about the small grid cells. And we actually heard from someone in one of those small grid cells who like lives in a place where the estimated cancer risk is three times average. And upon talking to them in an interview, they mentioned a lot of other types of nuisances or pollution. So I know we mentioned this in the map and Ava and Leila mentioned this, but we really see this as a starting point. So I definitely think it's worth reaching out to people even if it's small cell, because you might learn about other types of pollution or nuisances that the community has been living with for a long time, that maybe state agencies or local officials haven't adequately responded to. So yeah, that's a note on that. And then local tip collection efforts. So we put out this call out like we do often with our stories to hear from residents and really understand how this is how the regulatory failures are affecting people and have been for the last couple of decades. And so we've heard from over 315 people at this point, and I've gotten dozens of emails, we're working on some follow up stories, but we're probably not going to be able to individually interview everyone who's written in. And so if you all are interested in localizing stories, please feel free to email TalksMap at ProPublica.org. And we can keep you in mind in January once we kind of wrap up the next couple of stories we have, and we could see if you all are interested in getting some of the tips that we're not going to be able to have long conversations with folks about we'd love to pass those alongs. So yeah, that's kind of what we're doing on that front and hope to be able to collaborate with you all on that tip and off in the future. So we will go shortly to the Q&A. Do we want to briefly touch on the chromium kind of complicated chromium issue with the map because you guys might encounter that? And it's worth understanding a little bit about it from our reporters. Yeah, sure. I'll jump in there. So one of the most kind of toxic chemicals and carcinogens that often really drives the risk is known as hexavalent chromium. It was made famous in the movie Erin Brockovich. It was found in the water in Hinkley, California. And it has a cousin called trivalent chromium, which is not associated with any cancer effects. One of the really frustrating things about the EPA's reporting system is that it lumps together hexavalent chromium, highly toxic carcinogen with trivalent chromium, which might have some issues, but it's certainly not a major driver of cancer risk as chromium or chromium compounds. And facilities, even if they wanted to, have no way of kind of distinguishing, telling us, telling the EPA how much of which part of that they're emitting. So the EPA's workaround to this is assigning what it assumes based on the industry category and how much a certain type of industry like aerospace or chrome plating might emit a kind of toxicity weight. So it'll say, okay, this chrome plating facility we know from the industrial processes that it performs, we expect about 30% of the total chromium compounds to be hexavalent chromium. And that toxicity weight is based in science, it's a best effort, but it's ultimately a guess. And so it's really tricky as you end up with facilities that might be massively overweighted, where it looks like they're really seeing far more of this highly toxic chemicals than they are, and facilities that might be underweighted. And so we have notes with all of those facilities that release chromium on our map, letting people know about how the EPA's reporting system works and encouraging them to just go to the facility very early in the reporting process and see if they can provide you with additional documentation that shows or refute to provide some more context into actually how much of that chromium is toxic. All right, I'm going to delve into some of the questions that we got prior to this, because they're good and I think they're useful to a lot of people. The first one is from Bruce who asks, the data range is an intensity on the map. As we report on this, is there guidance as to which spots, if any, are not intense enough to include in the story, trying to cover an entire state? And I'm afraid if I list all the locations, people will just tune out due to info overload. We talked a little about this, but I'd love to hear from you guys what your kind of judgment is on what is not intense enough to include. To me, it depends on what people on the ground are saying, right? No, Leila. Yeah, yeah. No, I totally, I would agree with that. I would say if folks nearby are complaining a lot about a particular source, even if it doesn't look terribly bad on our map, there could be more happening that our map doesn't capture. And so it's probably worth examining. And, you know, in certain cases, however, there could be chemicals that residents are not aware of because they are colorless and odorless. All industrial chemicals are different. And some of them are much easier to perceive than others. Benzene, for example, smells like gasoline, so you can't really miss it. But ethylene oxide is a lot harder to know whether or not you're being exposed to it. So I would say that in those cases, it might be useful to take a look at what the trend lines are for a particular company. Is it getting a lot worse? Did its emissions crash to zero in the past three years? You know, those are the types of, you know, is it sort of stagnant and continuously bad? Are they adding new chemicals to the toxic soup? And one thing that I also really like to do is a lot of times if you do a quick clip search, you can tell whether or not a facility is going to be expanded in the near future. And so that can also kind of be an interesting news peg sometimes to say, well, we expect this facility or this area to sort of experience even more pollution in the near future. And so then you sort of have an avenue for impact. Also recommend like looking at the facilities through the lens of the EPA's enforcement and compliance database, because you might see bad actors, you know, they are only releasing so much, but they have a huge history of violations. And that's obviously, you know, really interesting to dig into as well. One other thing I'd recommend from the community standpoint is to potentially go to local Facebook groups or Reddit, subreddits, et cetera, and do searches by the name of a facility or by pollution. If people are complaining a lot and have been for a while, it could be interesting or even requesting complaint records from the state environmental agency to see if people have been calling in or writing in a lot complaining about it. So yeah. These are really good suggestions. Thank you guys. We've got a question. Maya, you might be able to handle this one. It's from Christian who says, what organizations exist to assist rural typically underrepresented low-income areas in reporting and maintains accurate measurements in the areas surrounding these hot spots? Where would a concerned citizen start to gain some real actionable traction? Yeah, that's a great question. I think a story is going to be coming about this soon, but air monitoring is one of the best ways to kind of understand what's really in the air. And historically, the EPA has not had a robust budget for that to be done on the local level. But we have seen instances in the past where when residents are faced with this, they've kind of advocated to their local environmental agency or federal agency of like, you know, let's get some air monitoring and see what's going on. And I think that will give you your best measurement. And I think even if in rural or urban areas, we've seen communities that have pushed for air monitoring have a better sense for what's really in the air that they're breathing. And just to explain a little bit about kind of who to hold accountable here, state agencies are the ones that permit polluters. And they're the ones that would come crack down on on facilities if they're emitting past their limits. But the EPA's regional offices have just as much power to to do some community air monitoring. So it's both, you know, the state agency and the EPA regional offices that if air monitoring is not being done in a neighborhood, you know, the buck stops at both of them and they should, you know, both be asked about it. Okay, we have another question here about going to the polluting facilities. How would you approach the polluting company in a way that would likely end up with them responding to the story? This is James in Louisville, where there is a welding facility that puts a lot of nickel in the air around a busy apartment complex. Ideally, James would like to have that facility explain how it happened. Yeah, that's that's a really great question. I'll grab that one. So I think I mentioned, you know, you can obviously always send a kind of boilerplate email, there's contacts to like who the environmental health and safety manager is actually within the EPA's own TRI database. So those might be depending on the size and scope of the facility, you know, a good person to reach out to. But it's also totally possible that you send that email, and you don't get any response, or they feel like you're just trying to write a thought you story and not just trying to actually understand what's going on. So another way in as a first step that I'd recommend is also trying to find, you know, whether or not the workers are unionized, kind of talking to the local union, saying if there's people there who might be able to connect you with, you know, employees or former employees that can help you just understand, like, what's the actual industrial process going on here? Why do they need this chemical? Why are they using it? You know, is it inextricable from, you know, what they have to make? Or is this, you know, a pollutant that's been phased out years ago that they're kind of, you know, emitting an excess despite the fact that, you know, most other companies in that, you know, industrial category wouldn't be using it anymore? Like, all of those are just so key to like, know and understand. Sometimes companies' emissions might spike because they were paving the road, and they have to include that in terms of like the chemicals they were using. So I think that's a really great question, and going to them as soon as possible, and with as, you know, open in mind as possible to really just get a sense of kind of what's going on in here to learn, whether that's through current employees, former employees at the union, reading company literature, you know, sometimes they'll have PDFs if it's a big company online that, you know, help you get a sense of what they do, you know, internal magazines and promotional materials, that kind of thing, really helpful. We have reported our way out of stories after having gone to the companies and them, you know, noting that their emissions were wrong and then, you know, then saying, thank you very much for filing corrections. So, you know, do it early, because that could save you a little bit of time. Alex, can I add one thing on that, too, is just that I would recommend asking the facility, if they come back and say, these emissions are wrong, to submit an updated R form in the TRI program, because then they're attesting to the federal government that actually what they submitted was incorrect, but and here are updated emissions, otherwise it's just kind of like their word and it's always nice to have that extra layer of someone saying attesting to the federal government that we're doing this right, because, you know, if that's a lie, then it's fraud. So, just pushing facilities to do that, if they come back and say, these are wrong numbers. Yeah, thank you for reminding. It's worth mentioning that we have not revised our map unless a company does do this revised form, so that, you know, we keep it, we keep it honest. So, this is, there's another good question. Here, Belen asks, does the project contemplate if and which minority groups could be impacted more, more impacted by the pollution? Layla has done a very interesting data analysis on this, would you like to talk about that? Sure, yes. So, REC data is not only aggregated at in these little grid squares, but you can also aggregated at the level of census tracks. And so, we did an analysis using the census track level REC data. And we sort of connected that REC data with census demographic information to try to understand whether minority communities or minority, majority minority census tracks are experiencing on average more toxic air pollution than majority white tracks. And we found that that is, in fact, true that census tracks where the majority of residents are non-white are experiencing approximately 40% more cancer causing cancer risk from toxic air pollution than majority white tracks. And that predominantly black tracks are experiencing more than double the level of cancer risk from toxic cancer causing air pollution. And so, that analysis is not sort of baked into our app per se, but I would say that if you are interested in those questions, then it would be what I usually do actually is I take, I pull up census reporter and I look up an area and I put that on one side of my screen and then I put the map on the other side of my screen and I try to understand if the places on our map where the the pollution is the worst, if those areas are what the sort of demographic breakdown is of those areas. That's great. Thank you, Leila. So, let's see, our next question. What are some of the best ways you've found to connect and build trust with the communities most impacted by this sort of pollution? Would you like to take this one, Maya? Yeah, absolutely. So, yeah, that's a great question and something that we've been wrestling with, you know, for the last six, seven months. I think that some there's kind of two camps from which we've been learning from is that there's some people, as we mentioned earlier, who kind of have known about this pollution experiencing it, have seen cancer in their families, have seen other types of pollution emitted by these industrial facilities affecting them through asthma and other health issues. And then there's a group of people who we found, like Leila said earlier, who have no idea that this is in their communities. And so there's kind of different approaches we take for each of them when doing this work. If there is kind of a movement and organizing around this, it's worth just, you know, speaking to the community leaders who've been doing this kind of work for a long time, just listening to them doing community research. So, I often check the American Community Survey, which is census data to understand the demographics of the community, whether it's a majority English-speaking or other speaking community, whether people are online and offline, and then look at how people are organizing there to really meet them where they are, and to ensure that whatever questions you're asking is accessible to the people being impacted. For the people who haven't heard about this information before or are unaware of the pollution, it can often be a shock when you're telling them. So, kind of just, I would recommend just trying to bring together people in a group conversation with somebody who's well trusted, and just, yeah, having one person's buy-in, someone to vouch for you, and then kind of explaining here's what we've learned. I've similarly done something that Leila did earlier, which is sharing my screen and kind of walking people through the map and answering questions I have. And that tends to kind of engender trust. But, yeah, I'd say that obviously the people impacted are why we did this in the first place. And they have a lot to teach us. So, they are experts in their own right about the conditions on the ground. They have receipts to hold companies accountable in some instances and government agencies. So, just really treating people like the experts they are has helped us build trust. Absolutely. Paul is asking what toxic dumps Brownfields or abandoned industrial sites are in our area, Bucks County, Pennsylvania, like steel plants, chemical, plastic, sugar, or water and sewage treatment facilities are in my area. Yeah, that's a great question. And if I didn't already emphasize it already, we definitely recommend we're just looking at air pollution from industrial sources. And there are so many other types of pollution, so many other types of cancer causing pollution that are also being, are supposed to be regulated by the EPA. And so, looking at their other databases, there's places in the EPA, in the EPA's kind of whole data set where you can search Brownfield sites, Superfund sites, you can look at information collected from RICRA and TRI, which we just looked at the air kind of data, also collects amazing water information, which is allegedly easier to measure than air. So, you might even have an easier time looking at water and information about onsite and offsite releases. And then on top of that, there's also a database called the National Emissions Inventory, which states submit to the federal government and often also will include some of their monitoring data and data about kind of shipyards and ports and diesel trucks that's all represented in the National Air Toxic Assessment, which can be another really good resource to kind of check against our map since, you know, automotive emissions are such a big source of pollution as well and something to consider. If you're looking at a hotspot, you see five highways going through it. It's not just the cumulative risk of the four facilities that were big enough to report to this database that we visualized. It's also the risk from all of these highways, from small metal foundries that might not even have to report to the federal government from a train that might be going through the neighborhood. And so, really thinking about the cumulative impact and the way that that affects an entire neighborhood, I think is really key to localizing these stories and also showing the harm. There's a great question here from Danielle. What can we do if the EPA doesn't have the full picture, such as in El Paso with the air pollution from Ciudad Juarez? Yeah, I can take this one because we've been working on a story that will also be coming soon. That's also on the border of the US and Mexico. Mexico does have its own database about toxic air pollution, but they don't include all of the same chemicals that the US does. For example, ethylene oxide, which, as we know in our national story, is responsible for most of the industrial toxic air pollution, cancer risk in our country. And so, you can look at that database, but yeah, I think there's some kind of blind spots, and it's an interesting question to raise. If this air pollution doesn't see borders that we've drawn up, and so what's happening to people on both sides of the border as where the US is housing these industrial facilities that are putting toxic air pollution in the air? So not a fully satisfying answer. It's something we're grappling with as well, but I'd be happy to chat more and send that database that Mexico has over after as well. Someone is asking how they can get photos and videos to us of dirty flares coming from a facility just north of Corpus Christi. I would say probably email us right at ToxMap at ProPublica.org. That reaches the entire reporting team. By the way, if you have any additional questions or you run into any other roadblocks, you can reach out to us there. And most importantly, if you do a story and you actually get some impact out of it, please let us know by emailing us there. We're also going to look into other ways that we can keep in touch with some of you guys as you continue to do your reporting. So we still have some more questions here. Let's see. I live in the city of Whittier, California, and I believe we have a highly toxic site on the corner of Whittier Boulevard in Washington Boulevard that affects a four mile radius. I believe the site also has contaminated our water tables at the site. My concerns are that we drink water provided by our own wells. I want to know how I can find information that would indicate the site is in fact contaminating our drinking water. I can take a stab at this one and then I'll kick it to you, Eva, because this is a question we've been getting a lot of people who've been writing into us with tips. There are other databases that you can check. TRI actually also has, if you go to there in the reporting recipe, we have a link to TRI and you can kind of separate out air releases and water releases. So you can see if the facility is releasing water or chemicals through the water. You might also want to check with local health departments would be good for this one or local environmental agencies. They often put out alerts when there's some drinking water issues. But Eva, I want to give you a chance to answer two in case I there's anything I didn't cover. Okay. No, it's great. So we've got nine minutes left, everyone. Ask your questions now. Okay, good. We just got one from Rebecca. How do we know that the EPA takes the time to verify changes to emissions and form ours from industrial companies? Eva? So the EPA does have a quality assurance process, but obviously there's, you know, 21,000 companies that report to the toxics release inventory. I mean, it's the largest, I think inventory of toxic releases in the world. So it's not, you know, humanly possible and they haven't developed some sort of machine learning algorithm to actually check that what the companies are saying is true. What they try to do is use kind of the incentive that a company has to, you know, legally certify. They have to actually sign an official has to sign that they attest that the emissions are accurate. But enforcement of the database and like enforcement, if your emissions turn out to be inaccurate later, we found is pretty spotty. So that's what we think the best thing is to just triple check with the companies, even though it's supposed to have been, you know, verified and checked with the EPA, as you're working on the story. And, you know, all of the reporting we've done to be clear is done after those checks with the companies after they say, yeah, we are releasing this many pollutants. And while we did hear back from some companies that said, oh, you know, we made an error here or there and areas were various scales. Some of the errors were, you know, quite small. And, you know, others did, you know, affect how a company looked in the map. We did hear back from lots of companies that said, yeah, this is this is how much we are releasing to the best of our knowledge using reasonable estimates. And, you know, that's why that that should definitely be the first step. Because the short answer to your question is we don't we don't know. One last I have a question for the audience, which is, are you running into any obstacles within your newsroom to being able to report based on our map? And is there anything we can do to help mitigate these obstacles? In the meantime, we've got a question. Somebody is asking what kinds of questions would you ask elected representatives and state legislatures in Congress representing these areas? Yeah, that's a great one. I can I can jump in there too. So we actually made a list of some of the questions we recommend that people ask the things that were kind of guiding us in our reporting. And they include just kind of getting a sense of how the air toxics, the local air toxics program works. You know, you also want to kind of ask it legislators, you know, what kinds of, you know, conversations are they having with industry? How much, you know, is industry's voice given a seat at the table, as opposed to community members? How does that shape the state's legislation? One thing we found, and of course, this is something that anyone who reports on the environment already knows is, you know, regulations differ, you know, quite substantially from state to state. So, you know, that's the kind of importance of really holding state officials and these local regulators to account. It's also great to ask about the funding that they've provided to the state agency, how much money they want to put aside for things like monitoring the air, which obviously provides much better data than self-reported emissions. And whether or not they're considering the disproportionate impact on communities of color that, you know, across the nation, these facilities continue to have when considering new permits, when and other industrial processes in the area. So just to break that down, I think the first part of Eva's answer is really like what you would ask the state agency, who I would say is probably actually the most important body that you're going to want to actually sit down and talk to when you're doing this reporting. And then the sort of more broad questions about how much are you funding your environmental agency, that would be to your to your congressperson and to your representative and so on. But the state agency, I think like, which wasn't exactly, you know, I think that they really weren't a real sit down. And that conversation can be somewhat technical, like, do you model, you know, the emissions from plants? What are you looking for when you do your modeling? Do you require environmental impact assessments? But those, I found it very helpful to ask the state that I'm investigating those questions and then to compare their answers to that of a different state. Some of the sort of more golden examples include Massachusetts and Maine, California. And so, you know, sometimes it's useful to be able to say things like, well, in this state, they do it this way. Why aren't you doing it this other way? That also reminds me of cumulative risk. So I think, and is it true that New Jersey might be one of the fewer only states that looks at the cumulative? Yeah, Layla, do you want to talk about that actually a little bit? Yeah. So New Jersey passed sort of a landmark law last year that requires permit the permitting authority, the state permitting authority to, to consider the preexisting conditions in an area before permitting new plants. So basically, they must assess whether or not an area is already overburdened by toxic pollution before permitting a new plant. It's sort of the first law that gives the permitting authority the power to deny a permit on the grounds that a community is an environmental justice community is an overburden community. And so that's going to be an interesting law to watch because it's possible that other states sort of follow suit and begin adopting these laws and there certainly will be, I think, pressure and there already is at the federal level for the federal EPA to adopt a similar law.