Hurricane Harvey, which devastated the Houston, Texas area in August 2017, came ashore as a Category 4 storm under the classic hurricane threat scale, which is based solely on wind speeds. But it was not wind damage or storm surge that made Harvey the second-most damaging hurricane in U.S. history (behind 2005’s Katrina) — it was the 60 inches of rain that fell for days in and around Houston, causing catastrophic flooding. As with Florence, Hurricane Harvey caused far more death and destruction from inland precipitation than from coastal storm surge and erosion.
And then there’s Hurricane Sandy, whose winds weren’t even strong enough to warrant classifying the storm as a hurricane when it made landfall in the U.S. mid-Atlantic states in 2012. Still, this massive storm generated significant storm surge — around 14 feet — that had dramatic coastal impacts, ripping barrier islands in half, causing significant oceanfront property damage, and bringing severe flooding to New York City.
Global climate change will likely make the water-related impacts of tropical storms even more destructive.
All of these storms have one thing in common: The hazards they unleashed were not adequately described by the traditional hurricane classification system — the Saffir-Simpson Scale.
You may not know exactly what that is, but I bet you have some understanding that a Category 5 hurricane is supposed to be much more damaging than a Category 1 hurricane. These categories are taken from the Saffir-Simpson Scale, developed in the early 1970s by a meteorologist and a civil engineer as a simple way to characterize the strength of an approaching hurricane based solely on its sustained wind speed.
For almost 20 years, we and a few other coastal storm scientists have been urging the media to stop using the Saffir-Simpson Scale as the sole, or even the most important, marker for the magnitude of the threat from a hurricane. The real danger from all of these systems is water, not wind. And the Saffir-Simpson hurricane rating tells us absolutely nothing about the water. The most destructive storms the U.S. has experienced in the past decade (Florence, Harvey, Sandy, Katrina) have all wreaked havoc by generating a significant storm surge along with massive waves and/or by dumping huge amounts of inland rain, resulting in the inundation of places we didn’t even know were in the floodplain. This water can completely reconfigure a barrier island shoreline by opening new inlets, knocking down dunes, and pushing entire islands landward. The impact of wind can’t compare.
Global climate change will likely make the water-related impacts of tropical cyclones even more destructive. While there is no compelling evidence to suggest that a warming climate will increase the number of Atlantic hurricanes, it is likely that the storms that do form will be more intense and have higher rates of rainfall. These changes will increase the degree and likelihood of flooding and shoreline erosion.
The importance of water as a destructive force in tropical cyclones is not unique to the Atlantic Ocean basin and the U.S. coast. All of the hazards described above and the increased impacts resulting from global climate change will be felt in all of the world’s ocean basins. Yet, the approach utilized to classify tropical cyclones in the Pacific also relies on the speed of sustained winds.
We must find a better way to characterize and explain the potential severity of storm surge and precipitation-driven flooding for an approaching storm. The National Hurricane Center labels any hurricane at Category 3 or above as a “major hurricane.” I am confident that the people of the Carolinas would agree that Hurricane Florence (Category 1) was a major hurricane, regardless of where it fell on the Saffir-Simpson Scale.
There is no statistical relationship between the Saffir-Simpson rating of a tropical storm and the resulting storm surge.
It may seem intuitive that there should be a relationship between a tropical storm’s wind speed and the severity of its impact. Certainly, this is absolutely true for wind damage to structures associated with any hurricane. Many buildings begin to experience damage at wind speeds exceeding 70 mph and damage may be extensive over 130 mph. These impressive winds may be an eye-popping aspect of some major hurricanes, but they are typically a very small part of a hurricane’s actual impact. This is true whether one is talking about damage to infrastructure, loss of life, or impacts to natural systems.
Such was the case this week when Hurricane Michael — a Category 4 storm, with winds exceeding 155 MPH — slammed into the Florida Panhandle. Damages from such high winds were significant, especially in a narrow zone near the eyewall of the storm. But the vast majority of insured losses will come from Michael’s storm surge and flooding from heavy rains.
My colleagues and I have shown that there is no statistical relationship between the Saffir-Simpson rating of a tropical storm and the resulting storm surge. Yet the misperception still exists in the media that you can determine whether or not a storm is likely to be catastrophic based on the five-category ranking, and that this wind speed will also predict potential storm surge. In reality, storm surge is a complex interaction between storm meteorology, storm track, and the topography of the impacted shoreline.
In general, larger-diameter storms have an opportunity to push more water in front of them. Superstorm Sandy was a monster even though the wind speed at landfall — 80 miles per hour — was relatively low. In addition, a tropical storm that spends many days heading over open water directly toward a landfall target can push a significant amount of water in front of it. On the other hand, if the approach involves crossing islands, then the storm surge can be significantly limited. Hurricane Andrew was a Category 5 storm at landfall in 1992, but generated only around 8 feet of storm surge in Florida because the Bahamas were in the way.
A storm that hits the shore perpendicularly is of greatest concern. Such storms can push water in front of them for many days and then finish the job as they cross the shoreline head-on. Hurricanes Hugo (1989), Hazel (1954), and Katrina are good examples of this. Finally, the shape of the coast and the width of the continental shelf make a big difference. The same storm approaching a concave shoreline, such as the northern Gulf Coast, will generate a higher storm surge than a storm approaching a convex shoreline, such as Cape Hatteras. That’s because the water gets pushed toward the middle in the case of a concave coast.
In light of these complexities, the Saffir-Simpson Scale can tell you very little about the storm surge potential and potential coastal impacts of any tropical storm.
The same is true for a storm’s precipitation potential. Saffir-Simpson tells us nothing that can inform our ability to warn the public about the potential for precipitation and inland flooding. Like storm surge, precipitation potential is related to a complex interaction of storm meteorology with storm track and forward velocity. In the future, as warming oceans supply greater precipitation to tropical storms, the rainfall potential may need to be the headline grabber, not the winds.
Many meteorologists begged the public not to pay attention to the fact that Florence had dropped to a Category 1.
How do we replace the Saffir-Simpson Scale? Tropical storm impacts are simply too complex and dependent on too many variables to characterize hurricanes with a simplistic index. A good start would be a concerted effort on the part of meteorologists and emergency management personnel to downplay the Saffir-Simpson categorization of a hurricane and to emphasize potential surge and precipitation. When referring to the Saffir-Simpson Scale, weather professionals should always emphasize that it is a wind classification. Hurricane Florence, for example, would have best been classified as a “Wind Category 1” hurricane. The National Hurricane Center should also immediately drop the term “major hurricane” for storms solely because they are at Wind Category 3 or above. There is no useful basis for making that distinction.
In addition, new storm surge forecasts developed by the National Hurricane Center and the Coastal Emergency Risks Assessment should be relied upon more heavily. The National Hurricane Center also produces updated maps showing the potential for rainfall and flash flooding. Perhaps we could place the maximum surge potential into categories for easy communication, such as surge threats ranging from Category 1 — zero to 5 feet — to Category 5, exceeding 25 feet. A similar ranking could be created for rainfall potential.
The Saffir-Simpson wind classification only complicates the job of communicating risk. Many conscientious meteorologists begged the public not to pay attention to the fact that Florence had dropped from a Category 4 to a Category 1. They knew that the flood and surge potential were still major threats. Nevertheless, many North Carolinians admitted that they chose not to evacuate after Florence was downgraded to Category 1 and the National Hurricane Center’s official graphics no longer showed it as a “major” hurricane. It was then that an unprecedented deluge flooded eastern North and South Carolina, trapping and stranding tens of thousands of people.