As Hurricane Dorian homed in on the South Carolina coast last week, forecasters put out what seemed like a dire warning: Storm surge pushed onto land could reach as high as 4 to 7 feet.
A map, published on the website of the National Hurricane Center, showed multiple feet of inundation on the fringes of the region's islands and peninsulas. The image, which modeled what could have happened with a 7-foot surge, showed much of Sullivan's Island underwater.
So why didn't the water come?
"We got lucky. That’s really all it was," said Ron Morales, a meteorologist for the National Weather Service in Charleston.
Dorian's track was just about 20 miles away from a far different outcome, as the storm's strongest winds were clustered tightly around the eye.
"I would do the same thing (with the surge forecast) again given the information I had," Morales said. "Unfortunately, when people say, 'Oh it didn't happen,' they don't know what they're talking about. They don't know how close it was."
The peak surge — or the wall of additional water that a hurricane produces on one side as it whirls counterclockwise — arrived as the tide was falling in Charleston Harbor. The timing was similar in Savannah, Morales said, with the primary wall of water coming near the low point of a tidal cycle.
But forecasters have to warn the public and emergency planners about surge often days before it arrives. Surge warnings and the graphic were published on Sept. 3, but the eye of Dorian reached its closest point offshore of Charleston two days later.
This map, published on Sept. 3, showed widespread flooding that did not reach the Charleston area. Blue areas show more than a foot of water, …
An array of subtle factors, including the storm's angle, small shifts in its track, its intensity and its speed can all change how much water ends up covering dry land. When a cyclone is moving parallel to land, as Dorian was, tiny movements east or west can make a big difference.
It's for that reason that forecasters write their predictions largely assuming the surge will arrive at high tide, Morales said. The surge amounts and map were created based on what's called a "reasonable worst-case scenario," meaning there was only a 10 percent chance that water levels could exceed the predictions.
Norm Levine, the director of the Lowcountry Hazards Center at the College of Charleston, said the forecasters' warning was an appropriate one in an evolving situation.
The Charleston area had seen elevated tides for days before the storm approached because of the phase of the moon, and when forecasters were warning of the surges, on Aug. 27, the storm had restrengthened to a Category 3.
Dorian weakened again to a Category 2 as it got closer and also slowed down, changing the timing of when the worst impacts would arrive.
"Just because the prediction didn't seem to match that height (on the ground) doesn't mean the predictions were incorrect," said Levine, who has developed his own flood modeling for the Lowcountry. "When they're producing these products, they are showing, 'here is the potential for what it can do.'"
Levine said another factor was the many water pumps the city of Charleston deployed to the peninsula, which may have obscured just how much water had arrived as they sucked it off the ground.
Ultimately, tidal data showed that Dorian did push water levels higher as it passed, anywhere from 3 to 5 feet above the normal levels expected.
Morales said he understands that what's now become an annual warning every time a cyclone draws near is tiring for residents of the region, and can prove expensive when evacuations are called.
But there are two months left in this year's hurricane season, and the Lowcountry's coastal zone remains at risk as far inland as Moncks Corner.
"Everywhere you look, everywhere, unless you’re (from) Summerville westward, is vulnerable to storm surge," Morales said.