Radar image of Hurricane Matthew from 8 a.m. on Oct. 7. The eye was then just 35 miles east of Cape Canaveral. (NOAA and B. McNoldy)

Predictions for Hurricane Matthew were reasonably good. Meteorologists, emergency managers and elected leaders effectively made the case that it was a dangerous storm. Evacuation orders were issued. Yet, some were ignored. And, where they were followed, some may have backfired.

Meteorologists and public officials sometimes seem shocked when residents are reluctant to evacuate. Sometimes we criticize those who refuse to leave. But it would be more constructive to better understand them.

The reasons for staying home are complex. They are frequently not due to ignorance or complacency, but a result of the enormous disruption and even hardship that vacating one’s home presents. In other cases, they reflect a failure of meteorologists to characterize the range of dangerous hazards a hurricane poses, in addition to solely wind, which residents may not adequately appreciate.

It would serve meteorologists and other storm communicators well to try to see the situation through the public’s eyes, recognizing that the public cannot be lumped into one, single group. Residents of coastal communities impacted by Hurricane Matthew are composed of different age, ethnic and income groups. Each has its own, unique story; and each group makes evacuation decisions challenging for different reasons.

It would also help meteorologists to identify the misconceptions that different audiences have about storms to sharpen future communication efforts.

To demonstrate the complexity and diversity of evacuation decisions, I made the drive from Stillwater, Okla., to coastal areas in Florida, Georgia and South Carolina to speak with residents.

I spoke with local residents in two separate situations:  (1) locals who rode the storm out on Tybee Island, Ga., and ( 2) evacuees who were stuck in a traffic jam on U.S. Route 278 in South Carolina returning home after the storm.

In both situations, circumstances emerged that show why evacuation decisions are complicated.

Riding the storm out on a Georgia island

After reading a story that a number of residents of Tybee Island stayed behind, despite personalized calls from the mayor, I made the drive south onto the Island.

I spoke with five residents who rejected evacuation orders and stayed at their residences during the storm.

“We knew if we got off the island, we would have a hard time getting back on,” said Karen Kelly, a  Tybee Island resident.

Kelly said she was motivated to stay to help neighbors. She readied for the storm beforehand, bracing for power outages. She filled bathtubs with water and prepared food.

Kelly said she would probably make the same decision again. “If it’s a [Category] 4 or 5, I am most likely to leave. If it’s a 3 or less, I am most likely staying,” she said.

Two other residents I spoke to chose not to evacuate based on the predicted storm intensity. Calvin Ratterree, a bar owner on Tybee, said he would most likely not evacuate for a Category 1 or 2 storm. Barney Burke, another resident, said he would leave only if peak winds were forecast to exceed 120 mph.

Only one of the residents I spoke to declined to leave because of a lack of confidence in the forecast. “When they first told us about the storm, we didn’t think it was actually going to come up here,” said resident William Mackey.

Gridlock returning home after the storm in South Carolina

A number of residents who followed evacuation orders regretted it, in hindsight.

On Sunday, Oct.10, after Matthew had passed, hundreds of residents from Beaufort County, S.C., found themselves stuck on U.S. Route 278, unable to return to their homes. Frustration reached a climax. Those I spoke with admitted that they may not evacuate the next time.

Their reasoning was not due to a poor forecast of Matthew, but instead, the evacuation process, specifically the return home. Multiple residents realized that their neighbors who never left home rode out the storm unscathed.

The residents who evacuated were envious of their neighbors, who did not have to deal with the hassle of the evacuation process, including the congestion on the return home.

Some residents mentioned that evacuation was financially taxing. Evacuation requires, at a very minimum, housing for multiple nights, dining and transportation to and from their destination. The cost to evacuate is considerable, and some simply are challenged to afford it.


Meteorologists are often too eager to label those who do not evacuate as complacent. Yet, these decisions are often shaped by burdensome past evacuation processes. Those who choose not to evacuate often have understandable reasons for remaining behind.

This is not to say that reasons for not evacuating are the most prudent or safe, but we need to try to understand how the public makes these decisions and try to make the evacuation process less onerous.

Through my discussions with residents, a trend in the responses became evident: A majority of the residents who stayed behind made their decision based on the hurricane category. That is, they put misplaced trust in the Saffir-Simpson scale, which rates a hurricane based on wind speed.

By focusing on the Saffir-Simpson scale, residents discounted the risks posed by flooding rain and storm surge, which the scale does not address. Meteorologists and emergency managers should be aware of the public’s overreliance on hurricane category and consider deemphasizing it. Instead, they should stress the dangers from the full range of hurricane hazards in making the case for evacuation.

Most importantly, meteorologists and those who work with them must look at evacuation through the eyes and experiences of coastal residents rather than through their own.

The author, Jacob DeFitch, is a graduate student at Oklahoma State University focusing on social science issues in meteorology. He  is one of the founders of SunsetWx, which provides sunset forecasts, and previously worked at AccuWeather.