ProPublica journalists Al Shaw and Lisa Song sifted through tons of data to deliver “Poison in the Air,” an award-winning reporting series in which they — alongside fellow reporters Lylla Younes, Ava Kofman, Maya Miller and photographer Kathleen Flynn — identify hundreds of industrial air pollution cancer-risk hot spots across the U.S. using publicly available EPA data. They even corrected reporting errors by the agency.
In a herculean effort to organize the data they gathered from their reporting and make it more accessible, the team created “The Most Detailed Map of Cancer-Causing Industrial Air Pollution in the U.S.”
This map allows readers to plug in an address to see what kind of industrial cancer risk is near it. Shaw and Song also used this tool to write numerous stories that highlighted the cancer risk data from various different angles. Their reporting spurred EPA reevaluations of regulations, increased local air testing and community engagement and activism.
In this “How I Did It,” Shaw and Song discuss their reporting approach for this series and offer tips to fellow journalists interested in covering industrial cancer risk.
The following conversation has been edited for brevity and clarity.
What sparked your interest in this story?
Shaw: “We had been working on a similar project a year before down in Louisiana looking at Cancer Alley. We were doing this project with Times-Picayune and the New Orleans Advocate looking at new chemical plants that were going in down there. There were a bunch of new chemical plants that had applied for permits in this stretch of the Mississippi river between Baton Rouge and New Orleans and that was already one of the most overburdened places in the country for chemical exposure. What we wanted to do is look at what the cancer risk in this already overburdened area would look like if all of these plants were built.
As we were trying to figure out how to do this, we came across this EPA database called Risk-Screening Environmental Indicators (RSEI), which is this crazily specific database that models air emission down to very small areas, about 800 square meters grid cells.
Essentially, they take data that facilities report to the Toxic Release Inventory (TRI) and combine that with air modeling and they show how that material would spread throughout neighborhoods around facilities.”
What was your approach for gathering data?
Shaw: “We took the existing RSEI data and hired a modeler to take the air permits that these facilities had applied for and put it through the same model to see how that would increase the concentrations of chemicals in that area. After we published this article, we asked ourselves, ‘Can we expand the scope of this nationwide?’ That ended up being a very difficult question to answer, mostly because the database is so large.
We started processing that data for the entire country, which took a while. We are looking at very small grid cells for every reported chemical in the entire country. But when we started playing around with it, all this stuff popped up that hadn’t been reported on. All these potential hotspots, cancer risk hotspots.
As we were looking through it, we realized, this is crazy [that] nobody has written about these places. That led us through a fact checking process to determine how dirty the data was, and it turns out there was a lot of misreported data in the RSEI database that we had to work through, but it did turn up a lot of places that hadn’t been reported on. Especially hotspots for ethylene oxide and other chemicals that we were able to write about.”
What was your reporting strategy?
Song: “Of course a lot of people know about Cancer Alley and that part of the Gulf Coast, and the huge numbers of plastics and chemical plants that are located there. That particular story of toxic air pollution in that area of the country has been reported on repeatedly, but there were also other places with much less media attention where similar things were happening. So we found various angles and communities to report out pieces of the story.
My focus, for example, ended up being on Pascagoula, Miss., and my story was really on a lack of air monitoring and a lack of enforcement from state and federal regulators. For years and years, this one community suffered and voiced complaints about the toxic air pollution, but received an underwhelming response from the government.
We started out by publishing a big overview story that included an interactive map, and had a national focus on the issue. It addressed some of the problems with our federal air pollution laws that we can’t get a handle on, and why the EPA hasn’t addressed them yet. And after that initial story, we followed up with individual pieces.”
What should health care reporters who are less familiar with the sacrifice zones and industrial air pollution know? What resources did you find the most helpful?
Shaw: “I think the first thing is to learn about the EPA databases that exist, because the EPA has so much data that is available on air quality around the country and in different categories, whether it be smog or PM2.5 or hazardous air pollutants.”
Song: “Chronic cancer risk from pollution is a complicated thing and I think the first thing you should do is talk to an expert, like an academic or advocacy group, because it’s a niche topic, even in the general field of environmental health.”
What was it like making the Map of Cancer-Causing Industrial Air Pollution for the public to use? What is your advice to health care reporters who want to do something similar?
Shaw: “The map was very complicated. This was probably one of the most sophisticated data analyses that we’ve done, mostly just because of how large it was. What we had to do was take this database that broke up every single chemical, by every single grid cell — and these grid cells are less than a kilometer wide — and apply the risk factor to every chemical, [and] sum them up over the entire country. So as you can see, it was a complicated undertaking to build it.
The second challenge was: How do we shove all of it into a web browser so that people can look at and use it? There were a lot of questions around how to present this and how to be legible and intelligible for people. The idea of risk can be a hard one to grasp. When someone has a ‘one in X chance’ of something happening, what does that mean? Risk drops off exponentially as you get away from facilities, so information design is all the more important.”
With so much data, how did you approach connecting the data to stories and people on the ground?
Shaw: “I think the first thing was looking for the outliers in the database, what were the places that jumped out at us as having verified risk. There are places around the country that have been well reported on, like Cancer Alley and certain areas of Houston, but there are certainly places that are less known.
In Laredo, Texas, there was a big sterilizing facility for medical equipment that uses a chemical called ethylene oxide to sterilize medical equipment. That was showing a big plume of cancer risk around the facility and that ended up being a story that hadn’t been written about.”
How did you find sources to interview?
Song: “In terms of finding the people to bring the stories to life, it was a combination of us finding the groups that were already focused on this, and sometimes we were the ones informing the local residents about the risk. We did a lot of outreach.”
What potential angles do you think need to be explored?
Song: “I think the idea of enforcement is important. What we often see with these types of stories is there’s an initial burst of interest, maybe some regulatory town hall where people talk about it, but then any kind of solution to the air toxics issue through regulation and enforcement usually takes many months if not years to actually get something. I think it’s really important for local reporters to stay on top of the story, so that there is continued pressure on the regulators. Otherwise, you are losing ground there and nothing will get done.”
