2015's Hurricane Patricia, one of the strongest tropical cyclones ever recorded, shows where the "category" system breaks down.NOAA

On Wednesday, Hurricane Florence weakened to a Category 3 storm, and then again to a Category 2. Simultaneously, meteorologists warned that Florence looked even more dangerous than it did before.

How can a storm weaken and also look more life-threatening?

The answer lies in the “hurricane category” system, and how it can distort the most important information about a tropical cyclone.

As a storm bears down on the coast, just about everyone—except, perhaps, veteran meteorologists—tends to fixate on its category. Most people know that the bigger the category, the scarier and more notable a storm.

That rule of thumb has the benefit of being true: It was legitimately worrying when, earlier this week, Hurricane Florence seemed like it might become the first Category 5 storm to strike the East Coast north of Florida. Only 33 Category 5 storms have ever been observed in the Atlantic Ocean, and as President Donald Trump exclaimed last year: “I never even knew a Category 5 existed.”

But this rule can also guide families to ruin, especially if they make a survival decision on the basis of category. A family might decide to ignore an evacuation order since it’s survived a Category 4 storm before. But a storm can be scary and notable without having a high category. That’s because only one trait determines a storm’s categorial intensity: its maximum sustained wind speed.

The “category” system—officially known as the Saffir-Simpson Hurricane Wind Scale—is a very simple scheme: The faster a hurricane’s maximum sustained winds, the higher its category. If scientists record winds anywhere in the hurricane between 111 and 129 miles an hour, the storm is Category 3. As soon as they detect 130-mile-an-hour winds, it becomes Category 4.

The scale’s inventors—Herbert Saffir, an engineer, and Robert Simpson, a meteorologist—hoped their scale would measure a storm’s total destruction potential when they developed it in 1971. But their system omits a tremendous amount of information about a storm’s ferocity.

“They took only into account the maximum sustained wind speed,” which is concentrated in a small part of the storm, says Rosimar Ríos-Berríos, a research meteorologist at the National Center for Atmospheric Research. “This doesn’t tell us anything about the storm’s size, or how far from the center the strong winds are located. And it doesn’t tell us anything about rainfall and other hazards.”

She continued: “The problem with [the Saffir-Simpson scale] is that not all hurricanes are the same. We can have two hurricanes with the same category, but one may be smaller and one may be large. Even though they are the same intensity, the impacts may be greater for a large system because the strong winds extend over a large area.”

The category also only loosely describes the most severe hazard of a hurricane: storm surge, which is responsible for half of all hurricane-related deaths in the United States. “The leading cause of fatalities due to tropical cyclones is not the wind, but the water,” Ríos-Berríos said.

Nevertheless, a hurricane’s sustained winds inform the entire way that meteorologists talk about a storm. A hurricane whose peak winds blow a little faster has “strengthened.” If winds lag a bit, then the hurricane has “weakened” or “been downgraded.” This is in fact exactly what happened to Hurricane Florence on Wednesday afternoon. The storm weakened slightly through the day: Its most ferocious winds blew at only 120 miles an hour, not the 130 miles an hour they had achieved earlier this week. This 10-mile-an-hour change was enough to downgrade it from Category 4 to Category 3.

But the size of the storm had actually expanded, wrote Stacy Stewart, a senior hurricane specialist at the National Hurricane Center:

While the hurricane hasn’t strengthened in terms of peak winds, the inner-core and outer wind fields have continued to expand, resulting in an increase the cyclone’s total energy, which will create a significant storm surge event.

In other words, the hurricane looks like it will have worse storm surge now, even though it is officially “weaker” than it was on Tuesday. Hurricane Florence was also predicted to take a meandering, dangerous track that will strafe against the coast of North and South Carolina on Wednesday—another new and scary aspect of its forecast.

There’s one more problem with the Saffir-Simpson scale: You literally can’t have a hurricane stronger than Category 5. Every storm with peak winds above 157 miles an hour is ruled as such.

Sometimes it seems like there should be a Category 6. In 2015, Hurricane Patricia had wind speeds of 215 miles an hour, but it is categorized identically to 2016’s Hurricane Matthew, which had wind speeds of 160 miles an hour. “With climate change, the oceans are going to warm up and hurricanes are likely to intensify,” says Lakshmi Kantha, a professor of earth science at the University of Colorado at Boulder. “Calling them Category 5 is no longer acceptable, then. You have to bump it up so you can go beyond 5.”

There are candidates to replace the Saffir-Simpson scale. Researchers at the National Center for Atmospheric Research have proposed a “Cyclone Damage Potential Index,” which takes into account maximum surface winds, the radius of hurricane-force winds, and how quickly the hurricane moves along its track.

Kantha has suggested a similar system in which hurricanes are categorized through a combination of their peak winds and the radius of their hurricane-force winds. Florence’s peak sustained winds are 120 miles an hour, and its hurricane-force winds extend to 70 miles, he told me. This would make the hurricane Category 2.6 in his reading, he said.

“Florence, however, will go over the warm Gulf Stream before hitting [the] Carolinas, and she is moving slow enough to draw upon the enormous heat content of the Gulf Stream just before hitting the coast,” he added. “So she might intensify very rapidly, and so these indices may go up just before landfall.”

At least these problems aren’t limited to meteorology. Seismologists use the moment-magnitude scale, or Richter scale, to rank the intensity of earthquakes. But a famous 2000 article by a U.S. Geological Survey scientist starts with the following declaration: “I hate the Richter scale.”

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.