The work entitled ‘In a Notoriously Polluted Area of the Country, Massive New Chemical Plants Are Still Moving In’ has been awarded the prize of Climate Change and Enviromental Commitment Best Graphic Award. The work was published in ProPublica (USA) and the team behind the piece, Lylla Younes, Al Shaw and Claire Perlman, shares with us some of what went in to creating the piece and the challenges they faced.
Could you briefly explain the idea of the story?
The story is about a set of towns along the lower Mississippi River in Louisiana known as “Cancer Alley” because of the concentration of chemical plants there. With 12 plants slated to be either newly built or expanded in the area, we wanted to see just how much worse the air would get in an area that already has some of the most toxic air in the country and some of the loosest air toxics regulation. To do that, we used a government database that models emissions down to a grid of less than a square kilometer to isolate the hotspots and hired our own modeler to show how newly permitted facilities would make toxic air worse. The result is some of the most detailed maps to date of precisely how far toxic plumes spread from their sources and how they combine in deadly ways.
What was the process of working on the graphic?
The thing that fascinated us most and inspired us to make this a graphic was the level of detail in the chemical data. Being able to visualize concentrations in small grid cells let us show how individual chemicals overlapped in communities in a way we had never seen before. We started by building a proof of concept in Observable that let us see how different chemicals interacted along the river with a histogram of how they combined by grid cell. This also let us see exactly where aggregated cancer-weighted concentrations were being pushed over certain risk thresholds.
We went through many versions of maps and histograms of combined chemicals by color before ultimately removing the histogram and moving to a palette of one color at a time (red for aggregated toxicity and yellow for isolated chemicals) to guide readers through the hardest hit neighborhoods.
We overlaid parcel footprints for planned facilities over the grid of cancer-weighted toxicity from our modeler to show how emissions would spread from facilities that didn’t exist yet.
What has been the challenge of this story?
One of the principle challenges of this story was understanding the data underpinning the map, and translating it for our readers. The EPA’s model for environmental risk screening, RSEI, uses industry-reported emissions data to estimate the concentrations of toxic chemicals in the air around industrial plants in 810 by 810 meter squares across the country. We were interested in visualizing the cumulative burden of all the cancer-causing chemicals in Louisiana’s river corridor. To do that correctly, we spent many hours on the phone with the EPA contractors who work on RSEI. They helped us understand how to accurately combine and filter the data, and how to translate the results into simple and correct language. Another challenge was the sheer size of our data. In its compressed form, the RSEI microdata was over 120GB, and some of the queries we used took over a day to run. We had to write a set of scripts to synthesize the data and create images that are renderable on the front end.
Lylla Younes, Al Shaw and Claire Perlman, ProPublica