When Hurricane Florence made landfall in the United States last month, few could anticipate the strange journey the hurricane would take. While many major hurricanes have hit the Carolinas, it turns out, none have taken the same path as Florence.
Harry Stevens, a data and visual journalist at Axios, analyzed the data collected by the National Oceanic and Atmospheric Administration (NOAA) to create a stunning visual comparing the hurricane to past ones. Storybench spoke with Stevens about his process for creating the visual.
What was the main objective of this visual?
I wanted to show that Florence’s trajectory was unusual for a major storm in that location. When I started working on the map, Florence was projected to make landfall as a Category 4, so it would have been the eighth storm to hit the Carolina-Virginia coast as a Category 3 or higher. Fortunately for the people in its path, Florence ended up making landfall as a Category 1. That partially undermined the purpose of the graphic, but it was still a worthy project.
What was the inspiration for the project?
I wanted to make something similar to what my colleague Lazaro Gamio made last year for Hurricane Harvey, but when I put the previous hurricane paths on a map near the Carolinas, it looked like an incomprehensible spaghetti of lines. So then I thought, ‘How is Hurricane Florence different from other major hurricanes that have struck the Carolina/Virginia coast in the past?’ And it was obvious that its trajectory was really unusual. The major storms that have struck that area since 1850 all came up from the south, while Florence came west across the Atlantic. Last Monday, when I started working on this, Florence was still projected to make landfall as a Category 4, so it made sense to compare it to other “major hurricanes,” which are defined as Category 3 or higher. There have been just seven to make landfall in South Carolina, North Carolina, or Virginia as Category 3 or higher since 1850, and filtering the data down like that made the map legible.
How did you acquire the data from past hurricanes?
What were the other tools used to visualize these data?
What was the process for making this visualization?
What were some of the challenges you faced while trying to make this project?
Animating the paths of the previous storms was challenging. D3 has functions that make animating transitions easy, but the data for each storm included different numbers of path segments that would appear on the map. I wanted each storm, rather than each line segment, to take the same amount of time animating, so I had to do some math so that each path segment delayed for the right amount of time and animated at the right speed. It was also somewhat challenging get the labels to fit on a small screen (the smallest screen we design for is 280px wide, so it’s pretty tight). If I had more time, I would have improved the labelling. I probably also would have simplified the color scheme to improve the discriminability of the different hurricane paths. In journalism, you only have so much time to make stuff.