The hurricane forecast cone that for two decades has accompanied newspaper headlines and broadcast news, to become one of the most widely-viewed warning graphics for a looming disaster, may be getting a makeover.
After finding widespread misinterpretation of the cone, researchers led by the University of Miami Rosenstiel School of Marine and Atmospheric Science say it’s now time for an update.
“Graphical communication strategies need to be revised to better support the different ways in which people understand forecast products,” the team wrote in a study published in the Bulletin of the American Meteorological Society. “And these strategies should be tested for validity in real world settings.”
The team is now working with the National Oceanic and Atmospheric Administration, the National Weather Service, the National Center for Atmospheric Research and others to determine improvements.
When asked about the changes to the cone, the National Hurricane Center said in an email it is gathering recommendations and would take up the matter “in earnest” once the hurricane season ends.
People view it as an impact cone, meaning that if you're in it, you're in trouble. If you're out of it, you're fine. That is not at all what it's intended to beBrian McNoldy, UM hurricane researcher, on the perceptions of the forecast cone
The study was published in August, before Hurricane Ian barreled ashore near Fort Myers a month later. But confusion about the forecast cone before and after the storm helped confirm its conclusion.
“People view it as an impact cone, meaning that if you're in it, you're in trouble. If you're out of it, you're fine. That is not at all what it's intended to be,” said Brian McNoldy, a co-author and UM hurricane researcher. “It’s not meant to show where impacts are going to be experienced. It’s only for track uncertainty.”
Yet in the days before landfall, with the center of the cone headed north toward Tampa Bay, some assumed that meant the most severe impacts would be felt there, not farther south where it made landfall at Cayo Costa.
“It was always in the two-thirds likely zone to get hit by the storm,” McNoldy said of the area. That also still left a third of a chance the storm lands outside the cone.
Read more: As Ian's death toll rises, questions swirl on why more Floridians didn't evacuate
Immediately after the storm, debate over the cone spread. Acting hurricane center director Jamie Rhome saidin a web post by the National Weather Service on Monday that he held off commenting given the magnitude of the disaster. But he said it’s clear there is confusion over predicted hazards.
“It’s tempting to want to engineer a quick fix to the cone, but we need to be scientifically disciplined and wait for the body of evidence to come forward and then determine how to best apply it,” he said, adding that he doubted the cone would be abandoned altogether.
“It still has merit for conveying high-level information about where the center of the tropical cyclone is likely to move, especially for storms where the entire track is over water,” he said.
The cone is just one of many graphics
And while it gets the most attention, the cone is just one of many graphics and maps produced by forecasters.
“They actually have nudged the cone off to, like, fourth [or] fifth ranked now,” McNoldy said, instead focusing on a ‘key messages’ graphic usually posted on social media that includes both texts and maps. “Something that is more of an impact-based statement," he added.
Among the other graphics are interactive storm surge maps, maps with wind, forecasted speed locations and arrival times, and rainfall graphics. It’s a long way from the all-caps bulletins issued by the center when Hurricane Andrew roared ashore. The forecast cone was introduced in 2002, to help depict more likely areas of landfall based on concentric circles using past forecast accuracy.
For the study, a team of hurricane researchers along with a psychologist and experts in graphics and media, surveyed more than 2,800 people across the state, asking questions about the meaning of the cone.
Participants were also asked where they got information and which they trusted. Participants were also broken down by age, income, race, education, location and other things that might influence the findings.
Forty-four percent incorrectly said the cone showed the forecasted size of the storm and 40% said it identified areas where damage would occur. Just 18% correctly answered all five questions about the cone.
The survey also found that while participants trusted television reports the most, they looked at social media more. But those who relied on social media were most often wrong. They also found that experience with hurricanes, rather than level of education, was more likely to lead to correct answers.
“That one finding kind of made us take a pause and think, well, we need to understand that more, because more education may not be the solution to misinterpretation,” said Scot Evans, the study’s lead author and a UM community psychologist who met with focus groups to design the survey.
One of the bigger challenges in changing the cone is how familiar it has become, he said.
“What do we replace it with, because people have gotten so familiar with that as the go-to information,” Evans said. “The flood of information from lots of different sources can overwhelm folks and that may be one of the reasons why they go, 'Give me the cone. I need some simplicity.'”
The study also revealed a surprising misunderstanding of storm surge, he said.
“They understood flooding. They understood wind threat. Tornado threat. But storm surge, even for folks here in Miami at the time when we were doing these conversations, they didn't really understand what it meant,” he said of the powerful hurricane-driven waves that can carry far more force than winds.
Since completing the study, researchers have been holding workshops to ask about possible design improvements. Many have asked for more interactive maps to show specific impacts, Evans said. But because forecasts are based on complicated model runs that take time, how interactive they can be made may be limited, he said.
Even if the cone survives, McNoldy suspects warnings will shift even more to impacts.
“I think that's where they're going and where we're going with our research is impact-based, to put less emphasis on where it's going and and instead focus on what it's going to do," he said.
