Jeff Berardelli is WFLA’s Chief Meteorologist and Climate Specialist

TAMPA (WFLA) – In the last seven years, we’ve had eight major hurricanes landfall on the Gulf Coast. But each time the Tampa Bay area has dodged a direct hit. In fact, the last time the Tampa Bay was hit directly by a major hurricane (Category 3 or higher) was the 1921 Tarpon Springs Hurricane which produced 115 mph winds and up to 12 feet of storm surge.

If you have lived here long enough, you’ve come to expect hurricanes to somehow avoid slamming into Tampa Bay. Some say we are protected by the Native American Burial Grounds, some say it is luck, while others say that’s just the way it is around here.

As a long-time Florida meteorologist, I can tell you we have a distinct geographical advantage, tucked away in the eastern Gulf of Mexico it’s hard for hurricanes to turn in our direction. Still, only one major hurricane in the last 100 years – what’s up with that? Is it luck or is that just a part of the normal Tampa Bay climate?

That’s a question that’s almost impossible to answer given that we have limited reliable data which only goes back a couple of centuries. But to answer this question accurately we need thousands of years of data.

So to overcome this limitation, the Max Defender 8 weather team decided to do something unprecedented. We partnered with my Alma mater, Columbia University’s Climate School in New York City, to conduct a cutting-edge, first-of-its-kind Tampa Bay area study to find out once and for all if Tampa Bay has just been lucky for the past century or if big hurricanes missing the Tampa Area is just a normal part of our climate.

The study lasted the better part of the summer and generated results that surprised even the most experienced hurricane forecasters, including me!

Study Background

Columbia University is known to have the biggest and best climate school in the country. And just as importantly, they are one of the only institutions in the world that can do what we needed for this study: Use a climate model to simulate millions of synthetic storm tracks to give us a much larger sample size of data. That’s where the CHAZ model comes in.

Depiction of the CHAZ model simulated synthetic hurricane tracks. Image: WFLA

The Columbia Hazard Model (CHAZ) – a statistical dynamical downscaling system – was created by Dr. Chia Ying-Lee and colleagues at Columbia’s Lamont Earth Observatory to generate projections of hurricane activity in a changing climate.

The model uses real monthly climate data from 1950-2019, which effectively replicates the Global climate of each month in the computer model environment. The CHAZ model then estimates where tropical systems may form, intensify and track.

The team working on our project was comprised of two professors – Dr. Lee and Dr. Suzana Camargo – and four graduate students – Olivia Cosio, Arjana Ejupi, Jillian Nash and Katie Parker. The study was part of the students’ final work in the Columbia Climate School’s MA in Climate and Society, known as a Capstone Project.

The WFLA-Columbia Climate School Hurricane Study Research Team

The project was separated into two parts. First, they analyzed the history of hurricanes in the Gulf of Mexico and Florida and then they analyzed the CHAZ model output to see how it compared with the historical record.

For those interested in getting into the weeds, the monthly climate data is from the fifth generation of the European Centre for Medium-Range Weather Forecasts (ECMWF) reanalysis (ERA-5) dataset. For the historical analysis, the team used the Atlantic hurricane best-track dataset from the National Hurricane Center (NHC) called HURDAT2 which dates back to 1851 and ends in 2022.

The concept for the study was to ascertain just how vulnerable the Tampa area is to hurricanes – especially major hurricanes (Cat. 3 and up) – and compare that to New Orleans and Miami. Each study area had a radius of 150 km from the city center. In addition, they also analyzed Florida’s risk overall.

WFLA-Columbia University Hurricane Study Areas

The 150 km Tampa Bay study area spanned from around Cedar Key to North Captiva Island. The team also performed an analysis on a smaller buffer zone of a 50 km radius from Tampa, spanning from New Port Richey to Bradenton.

In total, the CHAZ model simulated 1.8 million storms across the Atlantic Basin. That’s more than 100,000 years’ worth of storms when using the 20th-century average of 12 named storms a year. Then they analyzed how many of those storms crossed into the study area buffer zones.

Study Results

Their historical analysis revealed what is well known to the Max Defender 8 team – since 1851 only six major hurricanes have entered the 150km Tampa Bay Study area, including hurricanes Ian, Charley, Easy and storms in 1935, 1926 and 1921.

Historically speaking, while the number of low-end Category 1 hurricanes passing within the study areas are only slightly more for Miami and New Orleans, than for the Tampa Area, the number of major hurricanes are around three times less in the Tampa study area. Historically, six, 22, and 16 major hurricanes made landfall in the Tampa Bay, Miami and New Orleans study areas, respectively, in the period of 1851 to 2022.

This part is not surprising given we are aware of Florida’s hurricane history and areas like Miami and New Orleans are more exposed – not as geographically protected – as the Tampa Bay area.

(It’s worth mentioning that right before the official record begins there was a hurricane called the Great Gale of 1848 – estimated to have made landfall as a category 3 or 4 storm – which made landfall in Clearwater, Florida, bringing 15 feet of storm surge.)

Now, let’s examine what the CHAZ model found. For the Tampa 150 km study area, the CHAZ model finds low-end hurricanes should make a direct hit every 6 years, which matches well with the historical record of about once every 4.6 years.

For major hurricanes (Category 3+) the CHAZ model finds the same 150 km Tampa study area should experience a direct hit once every 33 years. Again, this fits well with the historical data of a major hurricane return period of once every 29 years.

For the smaller 50 km Tampa study area (spanning from New Port Richey to Bradenton), the CHAZ model finds the return period of a low-end hurricane should be once every 27 years. History shows it is once every 29 years. The model matches our lived experience.

For major hurricanes (Category 3 and higher), the CHAZ model finds a major storm should make a direct hit on the immediate Tampa Bay area once every 170 years. The limited historical record shows once every 173 years. Another accurate match, at least going back to 1851.

What does this all mean? The CHAZ model says that what the Tampa Bay area has experienced over the past couple of centuries is pretty much what we should expect to see in the climate of the 1900s and early 2000s. In other words, it’s not luck, this was just normal for the Tampa Bay area during the previous seven decades.

So why is the Tampa area more protected from major hurricanes? We are not only tucked away in the eastern Gulf of Mexico, but the winds that steer hurricanes also assist us. The prevailing steering winds during the peak of hurricane season blow from southeast to northwest.

The result is that the majority of storms hit Florida’s East Coast, move up the Eastern Seaboard or emerge out of the Caribbean and hit the central or western Gulf Coast. For a storm to hit the Tampa Bay area, it would have to turn at the right angle and at the right time – like threading a needle. It can and does happen, but it is more rare.

When hurricanes hit the Tampa Bay area, it is more often during fall when steering winds shift and storms in the Gulf are more likely to turn eastward toward Florida’s West Coast.

Going a step further, the CHAZ analysis reveals that in the 150 km Tampa Study region, from Cedar Key to North Captiva Island, a strong Category 4 or 5 storm should only happen once every couple to few hundred years.

And this is where the model diverges from the historical reality. Since 2004, there have been two strong Category 4 hurricanes which made landfall in the study area – Charley in 2004 and Ian in 2022. Is this just chance, is the CHAZ model risk too low for strong hurricanes, or has climate change loaded the dice for stronger hurricanes in the Gulf in recent years?

That brings us to the next phase of this project.

Next Phase of WFLA-Columbia University Project

In the past seven years, there have been eight major hurricanes in the Gulf of Mexico. Although there have been both quiet and busy periods in the past, this degree of Gulf major hurricane landfalls in such a short time is unprecedented in the historical record.

There’s no way of knowing for certain if this is just a random occurrence or if this is the beginning of a permanent increase in stronger Gulf hurricane landfalls. But it certainly begs the question of whether a warmer climate is behind the amplification of recent storms.

With this in mind, the Max Defender 8 Weather Team plans to team back up with the Columbia Climate School again next year to figure out if Tampa’s risk will change – for the better or worse -in a warmer climate. This will enable us to learn if current and future warming changes our risk of hurricanes and major hurricanes.

8 Major Hurricane Landfalls in the Gulf in the last 7 years. WFLA

Here’s some background. Over the past century, the Gulf of Mexico has warmed around 2 degrees Fahrenheit. NOAA says it is warming about twice as fast as the global ocean average. Simple physics tells us that hotter water can enable hurricanes to get stronger. But when the climate changes, it is not just the water temperatures that change – other factors adjust as well – which can either enhance or lessen hurricane activity.

With that in mind, a paper published this April, by leading climate and hurricane experts – focusing on US coastal hurricane risk due to climate change – found that a changing climate will continue to enhance hurricane frequency for the Gulf of Mexico and lower East Coast regions. The study attributes the increase in coastal hurricane frequency to changes in steering flow and reduced wind shear.

If this paper is correct, the Tampa Bay Area, Florida and the Gulf in general may be exposed to greater hurricane risk in the future. The next phase of our study should help to uncover that.

Lastly, I want to thank the team at Columbia University for their tireless work this summer in helping Tampa Bay viewers better understand their risk from hurricanes. Their work and expertise is greatly appreciated. We look forward to the next phase of this project.