A Tale of Two Hurricane Forecasts

Weather radar image of Hurricane Cleo over Miami, Florida, early on August 27, 1964

I’m a meteorologist originally from Birmingham, Alabama, so I am fairly well-qualified to discuss the forecasts of Hurricane Dorian and the non-threat it posed to my native state last week. To everyone from Donald Trump to his NOAA administrator-quislings I say, “YOU’RE FIRED!”

But the ego-driven distraction about Dorian has managed to shift our focus away from just how remarkable the non-sharpie-altered National Hurricane Center forecasts for Dorian were — and how far meteorology has progressed in my lifetime. To refocus us, I thought I’d compare and contrast the forecasts for Dorian and another hurricane that threatened south Florida: Hurricane Cleo in 1964.

I was in Miami for Hurricane Cleo — or, rather, proto-John was. My mom was about three months pregnant with me when our family was down in Miami, visiting my mom’s sister, Aunt Dardy. She lived on Key Biscayne for health reasons, for the ocean breezes, in the days before the Key became a Nixon-era destination overrun with high-rises. My dad’s pastor vacation came in August, so we went to see Dardy every August for many years (ah, the heat and humidity).

On the 26th of August, my family went to the set of a Miami Beach-based TV show my six-year-old brother liked, “Surfside 6.” On the way back to Key Biscayne, they encountered a huge, unexpected traffic jam (this was before southeast Florida became the most dangerous place in the country to drive). As the story was told to me, my family heard on another car’s radio that a hurricane was headed for Miami.

A hurricane? Nobody knew a hurricane was headed for Miami!

Hurricane Cleo had been an intense Category 4 storm near Hispaniola earlier that week, but had weakened somewhat while making its way westward through the Caribbean Sea. It was heading toward the Gulf of Mexico, with high pressure over Florida shunting the storm away from the Florida Peninsula. But the high weakened, and Cleo instead veered to the northwest and then north, an unusual path for that time of year. Cleo quickly cut across central Cuba and headed straight for Miami on short notice. The primitive forecast techniques of the era did not anticipate this northward path toward Miami until about mid-morning on the 26th, around the time Cleo was clearing the Cuba coastline and entering the Florida Straits.

The distance from Cuba’s north shoreline to Miami is about 125 miles. In-between the two lies the warm Gulf Stream ocean current.

Hurricane watches were hurriedly posted from Key Largo to West Palm Beach, FL at 2 pm on August 26, probably around the same time my family was at Surfside Six. They fought the traffic back to Key Biscayne and helped my aunt prepare her house for the storm. Meanwhile, Cleo intensified rapidly over the Gulf Stream, attaining 100-mph sustained winds. My mom always told me that our family was one of the last cars over the Rickenbacker Causeway out of Key Biscayne, with Biscayne Bay waters lapping up against the bridge.

Cleo made landfall at Key Biscayne with a punch around 2 am on August 27, 1964 as a strengthening storm, its central pressure dropping to 967.5 mb in North Miami. Hurricane Center meteorologists observed lightning in the eye wall as the storm went overhead. It was the first hurricane to hit Miami directly in 14 years; the next one to do it would be Andrew, 28 years after Cleo.

My family ended up evacuating inland to Winter Haven, FL, staying with my dad’s parents. Cleo went up the coastline from Miami to near Savannah, Georgia, preventing the Fort Lauderdale (FL) News from publishing for the first and only time in its history; delaying the grand opening of Florida Atlantic University for six days; and cutting power to over 600,000 Florida residents. There were no casualties in Miami, although locals at first feared the worst when they saw bodies floating in the water. However, the “bodies” were mannequins that had escaped from downtown department stores whose windows had been smashed by Cleo’s high winds!

It could have been worse. But the wave of panic that came over my family when they learned, in gridlock, that a hurricane was heading straight for us never completely subsided, not even when my mom retold the story to me decades later.

There are some similarities in the stories of Cleo and Dorian: an intensifying hurricane; lightning in the eyewall; a history of very high winds; and above all, close proximity to the southeast Florida coastline. Between 1964 and 2019, however, the stakes had been raised. The population of Florida increased by a factor of 3.75 over that 55-year period, from 5.8 million to 21.7 million, with much of it along the east coast. A repeat of Cleo’s surprise next-day visit to Miami would create much more than just gridlock — it would be a statewide, regional and perhaps even a national crisis. Last week we got a quasi-repeat: instead of an intensifying Category 2 hurricane barreling into Miami, as Cleo did, Dorian was a record-setting Cat 5 monster about the same distance from West Palm Beach as Cleo was from Miami when the hurricane warnings went up in 1964.

But widespread hurricane warnings for the Florida coast never were issued for Dorian. And they were never needed. Why not?

In another similarity between the two storms, the high pressure systems to the north of each broke down, permitting a northward path. In Cleo’s case, that allowed the storm to hit Miami instead of going west into the Gulf. But in Dorian’s case, the breakdown of high pressure allowed the storm to turn north before it hit Florida.

In 1964, this insight was found mostly after the fact, in a post-mortem study. There were no operational computer models for hurricanes back in 1964. There was only “forecast coordination.” There was no ability to simulate and anticipate, days in advance, the hurricane’s path based on solving the equations that describe the atmosphere.

In 2019, the National Weather Service and other organizations around the globe have the computing power to run models that produce forecasts of hurricanes many days in advance, and to do multiple “ensemble” forecasts to give a spread of possibilities for the hurricane’s path. This spread leads to the “spaghetti plots” of possible tracks that also inform the “cone of uncertainty” that has become familiar to users of NHC forecasts, from the public to the Oval Office. These computer forecasts get better each year, due to better computers, better and more data (especially satellite data) ingested into the computer models, better ways of ingesting the data into the models, and better science to understand what it all means.

The forecast models told a pretty consistent story by Sunday, September 1, 2019: that the storm would slow down, stall as an intense hurricane over the western Bahamas about 100 miles from the southeast Florida coast, and park there for a couple of days before heading northward and gradually weakening. A strange forecast, and one that reduced the chances of a Florida disaster.

And that’s exactly what happened.

That forecast took a lot of guts and confidence in numerical weather prediction. Twenty or thirty years ago, a Cat 5 hurricane 100 miles offshore would have led to a mass-panic evacuation of the east coast of Florida. In 1999, that’s just what happened with Hurricane Floyd. FEMA later said of the Floyd evacuation:

“Traffic engineers estimated 3 million people took to the highways in what was to become a frustrating effort. They sought only to flee the hurricane’s landfall. Instead, they created the largest, longest, and most incredibly snarled traffic jam ever known.”

Evacuations are dangerous business. When approximately 3 million people evacuated the Houston area ahead of Hurricane Rita in 2005, over 100 people died in the heat wave that awaited them once they left their homes and joined the gridlock. And Rita largely missed Houston. Even with Dorian, there has been at least one tragic evacuation-related death. Imagine what could have happened if Dorian had required the evacuation not of a relatively few coastal counties in Florida and up the coast, but instead all of the east coast of Florida!

Fortunately, the National Weather Service and its National Hurricane Center had the requisite expertise and confidence in its science to avoid telling millions that they had to flee. It was the right advice, and it saved our nation untold millions of dollars — and some lives, too. (Those who were told to evacuate because of Dorian and didn’t, such as hundreds of people on Ocracoke Island, NC, now regret it.)

While we are presented with the bizarre spectacle of a President doing after-the-fact editing of hurricane forecasts, let’s not forget who some of the heroes of Hurricane Dorian are: the meteorologists who stuck to their forecasts and gave us days of advance notice that Dorian would probably not devastate the Florida coast.

Thank a meteorologist today; they could use a morale-booster.

A geography professor and meteorologist at UGA in Athens, GA. I write about news, sports, weather, climate, education, journalism, religion, poetry, the South.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store