IE 11 Not Supported

For optimal browsing, we recommend Chrome, Firefox or Safari browsers.

Hurricane Barry Showed Good and Bad of Storm Risk Assessment

Forecasters had correctly predicted that Barry, which briefly reached hurricane on July 13, would form in the Gulf a week earlier. But complex weather patterns made it difficult to explain its threats to the public.

(TNS) — Meteorologist Frank Marks, longtime director of the federal Hurricane Research Division, had a genuine bird’s-eye view of the effects of Hurricane Barry on New Orleans as the troublesome storm made landfall last month.

Sitting at a console of display screens next to a window aboard Kermit, one of his agency’s P-3 Orion Hurricane Hunter turboprop planes, as it pivoted over Louis Armstrong International Airport, Marks was keeping an eye on the information being collected by sensors on board the plane and by the cardboard tubes full of instruments called dropsondes that had been released on parachutes at various locations in and around the storm.

Forecasters had correctly predicted that Barry, which briefly reached hurricane strength that morning — July 13 — would form in the Gulf a week earlier. But complex weather patterns made it difficult to explain its threats to the public.

Barry presented a textbook example, Marks said, of the improvements made in hurricane forecasting over the past 20 years. Predicting where the eye of a storm will go and when it will arrive has improved dramatically. Forecasts of intensity have not improved nearly as much.

The improved accuracy, meanwhile, comes with a new problem: The public — and sometimes local emergency officials — often don’t understand where a storm’s worst effects will occur, how widespread the effects will be, or how long a storm will remain dangerous after its eye has moved inland, Marks said.

Ironically, some of the confusion may be a byproduct of the better accuracy, which has narrowed the size of the "cones" used to predict a storm's path.

The problem is driving a new round of revisions to both written and graphic forecasts issued by the National Hurricane Center and the National Weather Service, Marks said.

“I think Barry kind of showed the good and the bad,” Marks said in a recent telephone interview. “With that system, the models were actually picking it up almost a week ahead of time and saying there was going to be something dropping out of Georgia into the Gulf that would become pretty consequential, a hurricane. That was pretty amazing.”

The bad, Marks said, was the difficulty in getting people to understand how Barry’s slow intensification made it hard to say when and where the worst rainfall would take place. Other challenges: The storm’s potential to intensify and organize led to a scary storm surge forecast for an unprecedentedly high Mississippi River in New Orleans, and its failure to spin up then caused that threat to disappear.

Between 1990 and 2018, the track error for tropical storms dropped from just over 100 miles to just over 50 miles for 24-hour forecasts, from over 200 miles to 60 miles for 48-hour forecasts, and from more than 300 miles to 100 miles for 72-hour forecasts.

The result is that the forecast cone, that bubble on the map showing the area at greatest risk of having the storm’s eye pass over it in one, two and three days, has shrunk dramatically.

But the public — and some emergency managers — fail to recognize that a hurricane’s effects often stretch far beyond the cone’s area, even if the forecast track is accurate, said Michael Brennan, who oversees the team of hurricane specialists who write the National Hurricane Center’s forecasts. The track follows the storm's eye, but storms can be hundreds of miles wide, with storm surge and heavy rainfall affecting areas far from the eye.

“As the average error continues to get smaller, we almost always see impacts outside the cone,” Brennan said in a phone conversation. “And while the average track forecast errors are improving, we still have outliers, and it’s really hard to determine when that will happen, when we will have really big forecast errors.”

And, he added, “even a track shift of only 20 miles can make a huge shift in what happens on the ground, such as what happened during Hurricane Michael,” the Category 5 storm that destroyed Mexico Beach, Florida, in 2018.

Marks said Hurricane Harvey, which made landfall near Corpus Christi, Texas, on Aug. 25, 2017, as a Category 4 hurricane with top winds of 130 mph, is a good example of a tropical cyclone’s unpredictability. Harvey inflicted more damage in Houston, more than 200 miles away from that initial landfall, dropping more than 60 inches of rain on the huge city.

Continuing to improve

Meteorologists and scientists with the National Hurricane Center, the Hurricane Research Division and other organizations each year attempt to improve both the forecasts and the public dissemination of forecast information through the Hurricane Forecast Improvement Program. The program’s goals include:

Reducing by half, over the next five years, the rate of errors in predicting both where a hurricane or tropical storm will go and how strong it will be.

Extending the time for tropical forecasts to seven days, from the present five days.

Improving the information that forecasters give the public about “pre-formation disturbances,” like the low-pressure system in Georgia that turned into Barry this year.

Marks is most interested in a new effort to improve the way forecasters use social and behavioral science to communicate risk to the public.

The effort resembles a program that led to the introduction in 2017 of separate storm surge watch and warning messages and color-coded watch and warning maps, and of color-coded maps showing how high potential surge waters might be along areas of the coast.

A key part of the forecast improvement program is aimed at those now-ubiquitous, colorful animated computer models used to predict hurricanes.

NOAA uses its own Hurricane Weather Research and Forecasting model, known as HWRF, to track tropical disturbances in the Atlantic Ocean and Gulf of Mexico.

Weather models divide areas of the globe into trapezoid-shaped grids and then use complex numerical equations to predict what’s happening in the atmosphere above those grid spaces, stitching them together to create a broader animated forecast model.

The equations use data collected from ground-based weather stations, satellites and other sources. Air and water temperatures, wind speeds, humidity and evaporation rates, ocean currents and wave data — even measurements of the movements of foam created by winds atop the ocean — are plugged in to predict where clear skies, thunderstorms, tornadoes, hurricanes, snowstorms and heat waves will occur.

Over the years, the equations have evolved to allow models to “see” the atmosphere in three dimensions. But until a few years ago, those grid shapes were miles across, and represented only average conditions over broad spaces.

Newer models, like HWRF, use much smaller trapezoid “baskets” — only a mile across — within the larger grid pieces to capture what’s happening in the atmosphere around a tropical system.

The baskets capture much more information coming from satellites, from equipment mounted on Hurricane Hunter planes and from the instruments — the dropsondes — that are launched into the storm with a small parachute and fall a mile to the surface in about eight minutes.

A new project is taking those smaller baskets of data and plugging them into the Global Forecasting System, or GFS model, one of the broader, worldwide weather models. That allows them to capture the effects of a hurricane in one ocean on the formation or disruption of both tropical storms and other weather patterns half a world away.

The research program also is aimed at improving the data collection instruments aboard planes and elsewhere.

For instance, Marks’ plane has a new device attached to its side that uses beams of light to measure wind speeds by capturing the movement of aerosols — tiny particles — in the clear air between hurricane thunderstorms.

The plane already had more traditional Doppler radar, which measures wind speeds by bouncing radar waves off precipitation in the clouds.

Ocean-diving gliders

Researchers also are capturing more information about ocean temperatures and currents by using robotic gliders that dive into and then surface from the ocean about 15 times a day. First tested near Puerto Rico, this year the gliders are being used along the East Coast and in the Gulf of Mexico.

The gliders transmit information to shore when they surface. When a storm approaches, they can be programmed to stay ahead of the storm to determine if water temperatures are warm enough to trigger “rapid intensification,” in which a hurricane’s top winds can grow by 35 mph or more in just 12 hours.

That’s happened several times in the Gulf, such as when Hurricane Michael passed over the “loop current,” an eddy of much warmer water, sometimes as much as 250 feet deep, that moves around and was near the northern Gulf Coast at the time.

When a hurricane passes over warm surface waters, its own winds and the surge and waves it creates can quickly cool surface temperatures, helping limit its strength.

But when it passes over the deep, warm water of the loop current, which breaks off from the Gulf Stream most years and wanders around the Gulf, the hurricane is simply pulling more warm water to the surface, like turning up the flame under a frying pan.

Marks said researchers also are hoping to increase the time that the dropsondes can capture information within hurricanes through the development of new, inexpensive unmanned aircraft, or UAVs, that can be launched from the P-3 planes, NOAA’s larger G-IV Gonzo research jet, or the new G-550 research jet, whose outfitting will take another year or so to complete.

Contractors hired by NOAA will be testing cheap unmanned planes that can fly within a hurricane’s strong winds and thunderstorms for an hour or two before being expended. The hope is that they will cost only about $5,000 each.

Researchers also are working on improving stepped-frequency microwave radiometers that are mounted on several of the hurricane-chasing planes. They read radiation emitted from the ocean’s surface, including its spray and waves, to estimate surface wind speeds within a storm.


©2019 NOLA Media Group, New Orleans

Visit NOLA Media Group, New Orleans at

Distributed by Tribune Content Agency, LLC.

Special Projects
Sponsored Articles
  • How the State of Washington teamed with Deloitte to move to a Red Hat footprint within 100 days.
  • The State of Michigan’s Department of Technology, Management, and Budget (DTMB) reduced its application delivery times to get digital services to citizens faster.

  • Sponsored
    Like many governments worldwide, the City and County of Denver, Colorado, had to act quickly to respond to the COVID-19 pandemic. To support more than 15,000 employees working from home, the government sought to adapt its new collaboration tool, Microsoft Teams. By automating provisioning and scaling tasks with Red Hat Ansible Automation Platform, an agentless, human-readable automation tool, Denver supported 514% growth in Teams use and quickly launched a virtual emergency operations center (EOC) for government leaders to respond to the pandemic.
  • Sponsored
    Microsoft Teams quickly became the business application of choice as state and local governments raced to equip remote teams and maintain business continuity during the COVID-19 lockdown. But in the rush to deploy Teams, many organizations overlook, ignore or fail to anticipate some of the administrative hurdles to successful adoption. As more organizations have matured their use of Teams, a set of lessons learned has emerged to help agencies ensure a successful Teams rollout – or correct course on existing implementations.