In this week’s episode, host Kristin Hayes talks with Marissa Childs, a postdoctoral fellow at the Harvard University Center for the Environment, about a recently published study that explores the prevalence and dangers of wildfire smoke in the United States. Childs discusses changes in the location and frequency of wildfire smoke, the degree to which increased prevalence of air pollution can be attributed to wildfire smoke, and the disproportionate effect of wildfire smoke on certain groups of people.
Listen to the Podcast
- Wildfire smoke varies in intensity as a pollutant: “There’s a lot we’re still learning about what else is in wildfire smoke, which can include things like lead and heavy metals, and is a function of what is burning, whether that’s homes or chemicals or industrial facilities; how hot the fire is burning; how far the smoke is traveling; and how long it ages … There’s a lot more to unpack to start understanding the effects of wildfire smoke.” (3:39)
- Wildfire smoke is increasing as a pollutant: “Unsurprisingly to most folks, one of our big findings is that smoke is increasing fastest in the American West. In some places, wildfire smoke is contributing up to 5 micrograms per cubic meter more to average annual PM2.5 levels than it was a decade ago. To put those 5 micrograms into context, the annual standard from the Environmental Protection Agency is 12 micrograms per cubic meter. This is a large increase over just a decade.” (10:31)
- Wildfire smoke and related air pollution are difficult to mitigate: “Wildfire smoke PM2.5 … is currently considered an ‘exceptional event’ under the Clean Air Act. Because the Clean Air Act has been so successful … wildfire smoke now is making up a very large portion of the pollution that people are exposed to, but [the Clean Air Act] does not regulate the pollution [that wildfire smoke produces].” (16:08)
Top of the Stack
- “Daily Local-Level Estimates of Ambient Wildfire Smoke PM2.5” for the Contiguous US by Marissa L. Childs, Jessica Li, Jeffrey Wen, Sam Heft-Neal, Anne Driscoll, Sherrie Wang, Carlos F. Gould, Minghao Qiu, Jennifer Burney, and Marshall Burke
- All We Can Save: Truth, Courage, and Solutions for the Climate Crisis edited by Ayana Elizabeth Johnson and Katharine K. Wilkinson
The Full Transcript
Kristin Hayes: Hello, and welcome to Resources Radio, a weekly podcast from Resources for the Future. I’m your host, Kristin Hayes. My guest today is Marissa Childs, who currently is a postdoctoral fellow at the Harvard University Center for the Environment. Dr. Childs recently earned her PhD from Stanford University’s Emmett Interdisciplinary Program in Environment and Resources, and her research broadly focuses on the intersections of environment and human health.
Today, we’re going to talk about research that was recently published by Marissa and a number of coauthors about the prevalence and dangers of wildfire smoke in the United States. Honestly, without even realizing I was doing it, I’ve scheduled myself two Resources Radio conversations in a row on air pollution and, this time, we’re covering a topic that has engendered growing concern in many spots in the United States. Stay with us as we dive into this important topic.
Marissa, thanks so much for coming on Resources Radio, and it’s really nice to talk with you today.
Marissa Childs: Thank you, Kristin. I’m happy to be here.
Kristin Hayes: While reading about your background, I saw that you study a range of topics that lie at the intersection of human health and the environment. Before we turn to the paper itself, tell us a bit about the breadth of your areas of interests and how you came to be interested in those.
Marissa Childs: I don’t think I need to tell anyone in your audience this, but human health and environmental change are intimately connected. Often, when we talk about environmental change, we focus on images of polar bears on melting sea ice, but equally devastating and sometimes less photogenic is the toll of environmental changes on humans—in particular, the human health effects. I originally started research wanting to better understand these connections and focus rather narrowly on vector-borne disease transmission, but over time the range of projects I work on has somewhat organically expanded to include everything from COVID 19–transmission modeling to wildfire smoke.
Kristin Hayes: You work at an interdisciplinary program in environmental resources—what are the disciplines that are merged together in the program? What are the areas of study that you actually brought to your focus on environment and human health?
Marissa Childs: My research borrows on disease ecology, which focuses on infectious diseases, and climate econometrics, which tries to understand how environmental change affects all sorts of societal outcomes.
Kristin Hayes: Let’s turn to the topic at hand and, maybe at some point, we could talk further about the disease work, because that also sounds very interesting. But today’s topic certainly is relevant for the United States and also something that Resources for the Future has also looked at a lot. Let’s dive into the conversation on air quality and wildfire smoke.
As our listeners may know or remember from the aforementioned recent episode on air pollution, a key type of pollution of concern for human health is fine particulate matter, also referred to as PM2.5. I wanted to confirm with you first that that is the primary pollutant of concern that we’re talking about when we think about wildfire smoke?
Marissa Childs: PM2.5 is a major concern for health, and it’s often the primary pollutant that we talk about when we talk about air pollution. That’s true in our research on wildfire smoke, as well. But I do want to say that there’s a lot we’re still learning about what else is in wildfire smoke, which can include things like lead and heavy metals and is a function of both what is burning whether that’s homes or chemicals or industrial facilities; how hot the fire is burning; how far the smoke is traveling; and how long it ages. Much of the research that has occurred so far—including our work—has focused on PM2.5, but there’s a lot more to unpack to start understanding the effects of wildfire smoke.
Kristin Hayes: I remember reading in the paper that not only is PM2.5 a particular area of focus in this research, but the particulates in wildfire smoke that we’re talking about, even when compared to other compositions of PM2.5, may be particularly concerning. Did I interpret that correctly? Is that something you can say a bit more about?
Marissa Childs: We don’t quite know on this one yet. PM2.5 is really about the diameter of the particle—how big the particle is—which determines how far it can get into a human body, into people’s lungs, sometimes all the way into their bloodstreams. But that doesn’t say anything about the chemical composition of the particulates.
We’re still learning more all the time about how smoke PM2.5 might differ from other sources of PM2.5. There’s been one study in Southern California that suggested that smoke PM2.5 might have a 10-times larger health effect. They were particularly looking at respiratory hospitalizations, but there’s still a lot more that we need to figure out here, and one of the things that I hope that these estimates can help us do—now that we have better estimates of smoke PM2.5—is to try and understand the relative harms from wildfire smoke and other sources.
Kristin Hayes: Let’s turn to the paper in particular. I wanted to note that you and your coauthors worked to significantly advance the methodology of how we understand the distribution of PM2.5 for wildfires in the United States. In many ways, the paper is a methodological paper. What were the challenges in previous methodologies that you were working to overcome? What were some of the innovations that you all brought to bear and how you think about understanding the prevalence of PM2.5?
Marissa Childs: I think this paper’s main offering is that it takes a first step towards trying to quantify smoke PM2.5 over the entire United States for as long of a time period as we could, which, in this case, was 15 years. Previously, we just didn’t have these types of estimates, which made all sorts of other studies that looked at both drivers and impacts of wildfire smoke much more difficult to do. In doing this research, there were definitely some challenges, and there are two big ones that I want to highlight. The first of those challenges in measuring wildfire smoke is, How do we attribute PM2.5 to wildfire smoke?
What we observe on the ground is total PM2.5. But, given that total daily concentration, it’s not obvious which of the PM2.5 should be attributed to wildfire smoke. On extreme-pollution days, there’s not a question of whether wildfire smoke is affecting air quality, but on the subtler-but-still-important days, with less elevated PM2.5, it’s more difficult.
Our solution here was to combine data from these ground-based air-pollution monitors with satellite-based estimates of where wildfire smoke is. We used the two data sources together to define what background PM2.5 levels would be when wildfire smoke wasn’t affecting air quality and then to attribute everything over that background level to wildfire smoke—when we know that smoke is overhead.
This lets us define smoke PM2.5 in the locations where we have monitors. But the second challenge is that these ground monitors only exist at a couple thousand locations in the contiguous United States. Large expanses of the United States, especially in the West, lack monitors.
To come up with spatially continuous estimates of smoke PM2.5, we trained a machine-learning model on satellite data to predict smoke PM2.5 in locations with the monitors and then to predict smoke PM2.5 in new locations to come up with these continuous estimates over the contiguous United States. I want to note that while these were advances in coming up with smoke PM2.5 estimates, they build on our rich, existing literature that has isolated smoke PM2.5 in other contexts or predicted total PM2.5 in other locations.
Kristin Hayes: This reminds me of the recent conversation about air pollution, where some of the very same challenges were called out: the lack of widespread monitors (particularly for certain types of pollutants) and the need to combine data sources in order to get robust pictures. I’m definitely getting a stronger sense of how folks are innovating when it comes to air-quality research these days.
So, where did that satellite data come from? Is that widely available or is that a novel data source that you brought to bear here?
Marissa Childs: The estimates of smoke plumes that we used are from the National Ocean and Atmospheric Administration’s hazard mapping system; they have analysts that draw plumes over the United States every day using geostationary satellites. That’s a publicly available data source that’s been used for many years to estimate where wildfire smoke is occurring. But one of the challenges with that is that we just are seeing from the sky, so we don’t know, given a plume, where in the atmosphere the smoke really is and whether it’s affecting people on the ground.
Kristin Hayes: Let’s turn to some findings, and there are many in this paper. You were able to look at exposure to wildfire smoke across a range of dimensions, and maybe I can start with the geographic dimension. You were able to extrapolate to at least the lower 48 United States. Could you also share a bit about the temporal dimensions? You did also note that there are a number of years of data here. I’d love to hear more about how exposure has changed, how it differs based on geography, and how it has changed over time. Can we start there?
Marissa Childs: Unsurprisingly to most folks, one of our big findings is that smoke is increasing fastest in the American West. In some places, wildfire smoke is contributing up to 5 micrograms per cubic meter more to average annual PM2.5 levels than it was a decade ago. To put those 5 micrograms into context, the annual standard from the Environmental Protection Agency is 12 micrograms per cubic meter. This is a large increase over just a decade.
A second temporal finding is that we’re seeing increases not just in average smoke PM2.5, but also in the number and extent of extreme smoke days. We considered a whole bunch of definitions of “extreme,” but I’ll focus on a central one, which we defined as days with smoke PM2.5 concentrations above 100 micrograms per cubic meter, which would be unhealthy in terms of the air quality index. Over the last decade, we estimate a 27-fold increase in the average number of people exposed to at least one extreme smoke day per year. In 2020 alone, we estimate almost 25 million people were exposed to at least one day with extreme smoke.
Kristin Hayes: You did also touch on some demographic variables such as race and income in terms of looking at exposure, as well. I know there’s probably a lot more that you could do there, but can you say a little about what you have been able to look at so far and what you might be able to do moving forward?
Marissa Childs: We compared smoke exposure a decade ago to now for different census tracts and looked at the racial and ethnic breakup of those census tracts. We found that census tracts with a higher percent Hispanic population are seeing smoke exposure growing fastest. This is somewhat unsurprising, given the geographic distribution of those places and the geographic distribution of smoke; we’re seeing a lot more smoke in the western half of the country and especially in the Southwest, which is also where we see many census tracts with higher percent Hispanic populations.
We also see a slightly larger increase for higher-income census tracts, which is counter to the usual pattern of total PM exposure, where we often see really high PM2.5 in lower-income places. I do want to caveat all of that by saying we’re measuring smoke PM2.5, which is occurring on top of background levels. So, while these exposures might be different from the patterns of total PM2.5, it’s important to think about the fact that these operate on top of existing PM2.5 levels and on top of existing vulnerabilities.
Kristin Hayes: You shared a bit of the intuition of why Hispanic populations might have been disproportionately exposed, but I’m curious about the high-income piece. Is that because of where locations of expensive properties are in relation to where wildfires are happening, or can you say a little bit about the intuition there?
Marissa Childs: Part of that might be driven by the relatively wealthy census tracts in many parts of California. We didn’t break that down in particular, and that’s something we could do more with, but I think that’s an explanation for the patterns we see.
Kristin Hayes: I’m going to home in on a statistic that brought home for me what we’re dealing with. You wrote in the paper that some of these western regions saw decadal increases in annual smoke, PM2.5, of five micrograms per cubic meter or greater. The paper notes that that is an amount comparable in absolute magnitude to the reduction in PM2.5 brought about by the Clean Air Act in the United States.
I want to restate that, because it really struck me. We’re talking about decadal increases that are comparable in magnitude to the reductions brought about by the most comprehensive clean air regulation in the United States. This is illustrative of the magnitude of the challenge that we’re talking about here. At the risk of making you responsible for solving a very significant problem, I did want to ask: What else do we need to do? What other research advancements, in your discipline or in others, do we need to make to be able to help the United States get its arms around this problem?
Marissa Childs: There are lots of big questions here, and maybe I’ll separate them into questions about the impacts of wildfire smoke and questions about the drivers and determinants of wildfire smoke.
On the impact side, which is where my research lies, one big question is to understand the long-term health effects of smoke exposure and the health effects of chronic repeated exposure, as it’s becoming a problem for a lot of people in the West.
On the drivers and determinants side, a question that I am wildly ill-equipped to answer is whether and how to think about regulating wildfire smoke PM2.5, which is currently considered an “exceptional event” under the Clean Air Act. Because the Clean Air Act has been so successful, as you mentioned, wildfire smoke now is making up a very large portion of the pollution that people are exposed to, but it does not regulate the pollution under there.
We also need to figure out how to prevent these enormous fires that produce so much smoke. That’s going to involve both more prescribed burns, thinning—this is not something within my discipline, but it is an important thing for us to figure out, because wildfire smoke is just growing so rapidly as a problem.
Kristin Hayes: Are there any kind of areas of interest for you personally that you’re looking to do next on this general topic?
Marissa Childs: I didn’t set out on this research to do this study at all. I started out wanting to study the health effects of wildfire smoke. I was wondering, What are the mortality effects? Do they differ from total PM2.5? I realized that that study was challenging to do, because we didn’t have estimates of smoke PM2.5; we didn’t know how much people were being exposed to.
It led us down this long rabbit hole of the entire study. I’m excited now to get to use these estimates. It’s a terrible pollution problem that we’re facing, but I think this data set is the beginning of how we can start to study it for some things. After all the effort that we put into making these predictions, I hope we can now put them to good use.
Kristin Hayes: We’ll welcome you back on the podcast anytime to talk about the further use that you’re able to get out of that dataset, because it sounds like it was a lot of work,has led to some important thinking, and can build a good foundation for future research.
Marissa, thank you so much for talking through this work with me. I did want to close by asking you what’s on the top of your stack. If you have more good content of any media variety—it could be on this topic or not—I would welcome your recommendations for what our listeners might want to take a look at. So, Marissa, what’s on the top of your stack?
Marissa Childs: Sitting on my bookshelf right now is a book called All We Can Save, which is a collection edited by Ayana Johnson and Katharine Wilkinson. I’ve been looking forward to it for a while but haven’t had a chance to sit down with it. I hope that this winter when it gets cold I’ll sit down and read it.
Kristin Hayes: Well, it’s rapidly gotten cold here in Washington, DC, and it’s about 51 degrees right now. Our listeners have at least a few more days of cold to enjoy a good book, and I certainly welcome that recommendation.
Thank you so much, and I look forward to talking with you soon.
Marissa Childs: Thank you so much, Kristin.
Kristin Hayes: You’ve been listening to Resources Radio, a podcast from Resources for the Future. If you have a minute, we’d really appreciate you leaving us a rating or comment on your podcast platform of choice. Also, feel free to send us your suggestions for future episodes.
This podcast is made possible with the generous financial support of our listeners. You can help us continue producing these kinds of discussions on the topics that you care about by making a donation to Resources for the Future online at rff.org/donate.
RFF is an independent, nonprofit research institution in Washington, DC. Our mission is to improve our environmental, energy, and natural resource decisions through impartial economic research and policy engagement. The views expressed on this podcast are solely those of the podcast guests and may differ from those of RFF experts, its officers, or its directors. RFF does not take positions on specific legislative proposals.
Resources Radio is produced by Elizabeth Wason, with music by Daniel Raimi. Join us next week for another episode.