Behind the Scenes Q&As

“Useful, beautiful and innovative.” How the Washington Post’s Harry Stevens reaches high engagement on climate and environment reporting

There’s a difference between reading about climate change and vividly seeing it in action. Last year, the Washington Post launched Climate Lab with Harry Stevens as the climate change, data analysis and graphics columnist. To make journalism more accessible and transparent, the column provides a visual and data-driven column about climate, environment and extreme weather.

Harry Stevens's headshot
Harry Stevens, courtesy of the Washington Post.

Stevens has found engaging ways to change how climate news is reported. Before Climate Lab, he was part of the team working on the series “2C: Beyond the Limit,” which won the Pulitzer Prize for explanatory journalism. Stevens previously worked for the Hindustan Times and Axios (Storybench profiled his work there visualizing hurricanes in the Carolinas).

The following interview has been edited for length and clarity.

How have you made the data and information in your articles digestible for your audience and those who may not have a scientific background?

I think you just have to put yourself in your reader’s shoes. It’s pretty easy for me to relate to somebody who doesn’t have a science background because I’m one of them. Now, I’m more familiar with scientific terminology than I was before, but I can still understand what would make somebody feel intimidated or what might make their eyes glaze over so I try to just strip out all the boring or confusing stuff and just keep the stuff that’s essential. I still want to capture as much of the nuance of the science so that people actually learn something, but also keep in mind that readers aren’t in the laboratory themselves, so they don’t need to be familiar with all the technical jargon. They just need to know the essence of what’s interesting.

In Watch the Earth Breathe for One Year, viewers interact with a real-time view of how carbon dioxide builds up in the atmosphere. What was your thought process when creating the story?

NASA has this thing called the Scientific Visualization Studio. They make videos of various atmospheric phenomena. They got data for 72 layers of the Earth’s atmosphere. It’s interesting, but it’s just a video, so you can’t interact with it. I thought it would be cool for users to be able to spin [the planet] around. I called the people at NASA who had been the scientific advisors for that project and got them to explain the basic science to me. I also got them to send me a data set where the atmosphere was compressed into just one layer instead of all 72 and make an image of the earth for each. If each of those images is 1,000 pixels wide and 500 pixels tall, they would be pretty compressed, but it would still be 500,000 pixels and you need to compute the position of each one of those pixels 60 times per second to make it look smooth. Now you’re talking about 30 million calculations per second.

WebGL (Web Graphics Library), lets you use the GPU (graphics processing unit) instead of the CPU (central processing unit). Because of that, you’re reprojecting all of these pixels 60 times per second. And then it’s also a video because that is what allows you to show the animation. You have to keep updating the pixels that are getting reprojected, so there’s a lot of computation taking place. Fortunately, we have the WebGL technology that lets us do that.

DON’T MISS  How one data-driven art and technology company is understanding cities with AI

On average, how long does it usually take to publish interactive articles?

The column comes out every two weeks. That includes data analysis graphics, reporting, writing and design. I write the Climate Lab column but [the rest of the team] works on lots of other stuff. They do these really deep investigations into scientific phenomena and how it’s impacting people around the world. Sometimes those take months because they’re doing newer data analysis and making more involved graphics, which take more reporting.

We did a series last year called Unearthing, for example, about how human beings have created a new era in geological time called the Anthropocene. You can see that if you dig into the earth, how things have changed because of human beings’ behavior. If an alien were to come to earth in a million years and dig down to the surface and do scientific analysis. They would see the telltale signs of humans because we change the planet. One story in the series takes you to a lake in Canada and it dives down below the surface of the lake and explains the buildup of sediment there and using 3D graphics of the lake. This is not the sort of thing that you could make in two weeks.

What inspired you to make interactive visualizations rather than a typical bar chart? Can you describe the feedback you have received?

Before I knew how to make interactive graphics, I saw them on the internet. I want to know how to make it myself and so that still is why I do it. There’s a little bit of an interesting debate about what deserves to be interactive because they take longer to make. For things like a bar chart, we make bar charts too and those don’t need to be interactive. You can show all of the information without any interactivity and you should never hide information behind layers of that interactivity. We shouldn’t make somebody hover on something to get the data when you can just put a label on it and so I think that’s an important debate that you shouldn’t be making interactives when they’re unnecessary.

At the same time, they can differentiate a newsroom from any of the stuff that you can get for free. One of the things that sets the Washington Post apart from Substack or The New York Post is that we have this offering. We’re trying to give readers something that’s worth paying for a subscription. I think it’s a good way to set us apart and say that we have something that you can’t get anywhere else, which is these really incredible experiences that aren’t only fun and cool to look at, but also illuminate the world in a way that you can’t without a more elegant and interactive display.

DON’T MISS  How The New York Times predicted your politics using a quiz

What are your top three prime goals for data visualization?

The first thing is just to show the thing that’s interesting by picking the data visualization that brings out the pattern in the data.

The second thing is beautiful and fun to look at. “Beautiful to look at” is really important, because it becomes more memorable. If information is presented in a beautiful way, you’re more likely to remember it.

If it presents the data in a new way that can be fun to look at. We’ve all seen a million billion bar charts and a million billion line charts and the reason that people use those chart styles is because they’re good. If you want to compare relative values, a bar chart is what you want. The reason they get used a lot is because they’re very effective; but I like to see innovation in my field. So if there’s some sort of more complex data and somebody comes up with a clever way to show it, that’s always exciting. I would [aim for] useful, beautiful, innovative.

What do you think the future of data visualization will look like? Where do you think it’s headed?

Right now, you need some specialized technical skills to be able to make data visualizations. Even simple things like to make a bar chart or a line chart. That’s gotten better because now you have tools, for example, DataWrapper where you can just paste your data in and get a beautiful chart back. I would like to see that expanded into all types of visual information and all types of data sets. Potentially chatbots or artificial intelligence have a lot of potential to make it easier for anybody to work with data.

Imagine if you could just put in your data set and say, “Find an interesting pattern in this data and make me an appropriate graphic to show that pattern,” and then it just did it for you and explained what the interesting pattern was. You can say “make a map of this county data set” and it just makes a map. It picks a good color scale, you don’t need to know how to code or anything. I think that would be really cool, because the technical part of creating the graphics is just sort of an onerous thing that you need to learn in order to do it, but that’s not the creative part. That’s not the part about finding interesting insights. Anybody should be able to find interesting insights without having to do all that onerous learning how to code stuff — and I say that as somebody who did all the onerous learning how to code stuff.

Leave a Reply

Your email address will not be published. Required fields are marked *

Get the latest from Storybench

Keep up with tutorials, behind-the-scenes interviews and more.