Storm surge from Hurricane Dennis in 2005 (USGS)

The National Hurricane Center (NHC) is running an improved version of its old, reliable storm surge model on faster computers to help emergency managers update hurricane evacuation plans.

When a storm threatens any place in the U.S. NHC forecasters will use the improved, faster-running model to predict its surge.

Hurricane forecasters know those who might be in a hurricane’s path tend to focus on the storm’s wind speed, which define its one-to-five category.  “The categories put a focus on wind; people are absorbed by wind,” says Jamie Rhome, the NHC’s storm surge lead, but “water is what kills the vast majority of people. Evacuations are done for surge.”

Surge is the water a storm pushes ashore as it arrives. Its height, extent, and areas flooded depend on many factors, including the speeds of the storm’s winds, the direction the storm is coming from, how fast it’s moving, the storm’s size, the depth and shape of the ocean bottom offshore, and what the surge encounters on land such as dunes, buildings, embankments, or rising land.

Only a powerful computer can make sense of all of this.

The NHC’s storm surge model is called SLOSH*, for Sea, Lake, and Overland Surges from Hurricanes.  (There’s also a version of SLOSH for extratropical cyclones, such as our winter Nor’easters.)

The Hurricane Center uses the model to make advance calculations of potential surge for 37 “basins,” which include all of the U.S. East and Gulf coasts, the Bahamas, Puerto Rico, the Virgin Islands, and parts of Hawaii and Guam.

To prepare the “hazard analysis” for each basin that emergency management officials use for evacuation planning, NHC runs hundreds of simulated hurricanes. These are done for all five storm categories changing various factors such as forward speeds, directions of approach, and landfall locations. Each model run for a category creates a separate Maximum Envelope of Water (a MEOW). These are combined into a Maximum of Maximums (a MOM) for each category.

Faster computer speeds are enabling the NHC to run more scenarios than in the past, which gives a better picture of potential surges. “In the past we did 10,000 to 15,000 scenarios for the hazard analysis” Rhome says. “We’re working on one now for Louisiana and are more than doubling the number of scenarios to 50,000. “

Emergency managers use these hazard analyses to create evacuation zones long before storms hit.

Two Tampa Bay maps show the NHC Model results for the Tampa Bay area and the surge zones derived from it and other data. (Top) The model shows higher water levels over the Bay and the Gulf of Mexico near the coast as well as over land. (Bottom) The zones map shows areas flooded by the five categories of hurricane. Each category, of course, would flood the area for all lower categories in addition to the area it only would flood (National Hurricane Center)

The NHC normally updates surge data for five or six basins a year. These are done after a storm changes the shape of a coastline, the high-resolution land elevation for the area has been updated, or local emergency managers need updated surge data for planning, Rhome says.

Surge data for the Lower Chesapeake Bay, Norfolk area, was updated last year and the Upper Chesapeake Bay is now being updated. Rhome stresses that “surge is not just a coastal event, the tidal Potomac River up to Washington, D.C. is vulnerable.”

When a hurricane threatens a particular basin the NHC runs a few hundred scenarios with slight changes in the hurricanes’ strength, exactly where and when it could hit, the storm’s size, and other factors that could affect surge. The results help emergency managers decide which zones to evacuate.

Another big change this year is potential surge is being described in terms of depth of water above ground level at each location, instead of ”feet above normal tide levels” as in the past. If you’re in a potential surge area you won’t have to know how far above “normal tide levels” your house is to know how deep the water around it could be.

Related: Weather Service Sandy assessment emphasizes importance of effective risk communication

An early version of a storm surge hazard map being developed by the National Hurricane Center shows potential flooding in the Fort Myers, Fla., area for a storm similar to Hurricane Charley, which struck the area in 2004. Center officials already have decided to change the label on the map for 3 feet or less above ground, which is blue, to “substantial,” to show any flood waters could be dangerous. The NHC expects to begin issuing such maps to the public in the 2014 season. (National Hurricane Center)

As with all forecasting, surge forecasting is done in terms of probabilities. Rhome comments that it’s always a fine line between crying wolf and giving adequate warnings. “Under evacuation is a terrible thing to think about. Imagine if I told an emergency manager five feet and it turned out to be 15 feet.”

Still, “its an exciting time. This year’s changes are the biggest ever.” he says.

Additional reading

Washington-area surge if Sandy had come up Chesapeake Bay

NHC Historical surge maps, including Isabel’s Chesapeake Bay surge in 2003

* What’s in a name?

If the acronym SLOSH makes you wince, you’d love its parent, an even more tortured acronym, SPLASH (Special Program to List Amplitudes of Surges from Hurricanes.)

Despite the name, SPLASH is an honored ancestor to today’s model.

During the 1960s Chester Jelesnianski and his colleagues at the Weather Bureau’s Technique Development Lab in Silver Spring, Md., had been working on ways to predict surge and had SPLASH ready for prime time before Category 5 Hurricane Camille hit Mississippi on August 17, 1969. As Camille headed for the Gulf Coast Jelesnianski ran the model and told Bob Simpson, then Hurricane Center director, that Camille could bring ashore 20 feet of surge. Some others at the Hurricane Center said such a surge was ridiculous, but Simpson―hedging his bet a little―issued a statement predicting 15 or more feet of surge.

In doing this Simpson violated the Weather Bureau’s policy at the time of not using predicted wind speeds or surge heights in warnings.  Forecasters were supposed to use terms such as “strong winds and dangerously high water are expected.”

As it turned out, Camille’s surge reached at least 22.6 feet, and surge modeling has been one of the Hurricane Center’s tools since them. The old Technique Development Lab is now the National Weather Service’s Meteorological Development Lab, still in Silver Spring.