Flying into the churning belly of the beast

The hurricane hunters had arrived at the heart of Tropical Storm Barry, and the center looked unfinished, a semicircle that convulsed and heaved.

“It’s a mess,” declared Paul Reasor, one of the meteorological researchers who had crowded into the cockpit to take a look.

None of those aboard the plane expected to be flying now — not in July, so early in the six-month Atlantic hurricane season. But Barry had formed in unusual fashion, as a disturbance moving down from the South into the Gulf of Mexico, where warm summer waters had fueled its development.

Like doctors taking a CT scan, the hurricane hunters from the National Oceanic and Atmospheric Administration, based in Lakeland, Florida, and their partners went to get a snapshot of the storm’s insides Thursday.

Although Barry was forecast to make landfall along the Louisiana coast as a Category 1 storm Saturday morning, it was not yet a hurricane. This only made it more critical for researchers to study it. Their models had predicted that the tropical storm would intensify more quickly, but it had not. The team wanted to know why.

The pilots pointed the nose of the aircraft directly at the messy center. An alarm rang: Fasten your seatbelts.

“Folks,” warned Mike Holmes, the flight director, “we’re going to punch through some tough air here.”

Hurricane hunters do much more than confirm a storm’s location and strength. (Air Force Reserve reconnaissance flights do that, too.) The planes that NOAA uses, two Lockheed WP-3D Orions nicknamed Kermit and Miss Piggy, are sophisticated flying research labs that launch probes and collect real-time data that is crucial to understanding hurricanes across the globe. The planes logged their first flights in the mid-1970s.

It is especially important to gather data from weather systems like Barry that defy predictions: The weirdest storms can sometimes produce the best science.

“To me, these are more challenging than a traditional storm,” Holmes said from his seat behind the cockpit of Kermit on Thursday. “When a storm is getting its act together or weakening, that’s when they get more sporty.”

On the side of each plane are dozens of stickers, one for each state and country where it has flown missions — and for each of the storms it has crossed. Miss Piggy — which has not flown a mission since late 2016, when it began major renovations that are almost complete — has 93 storm stickers. Kermit, which was previously upgraded, has 110.

Flying these planes is not for the faint of heart. The preflight safety briefing includes warnings about snug oxygen hoods that can suffocate the wearer if misused, and about “immediate ditching” — when the four-engine turboprop plane must try to land on the water. Everyone aboard is encouraged to keep a blue plastic sick bag at the ready. Violent turbulence can strike at any time.

It would take eight hours for the 14-member crew of NOAA Flight 42, all with earplugs handy, to reach Barry. They would then carefully crisscross it in what is known as a butterfly pattern, communicating over the roaring engines through headsets and radio.

On board, radar under Kermit’s belly and tail turned on. The crew prepared to launch probes known as dropsondes at points along the storm’s edge, midpoints and center. Countdown: Three, two, one. A switch was flipped. The cylindrical probe whooshed out of a chute. In front of the technician, data — humidity, temperature, pressure — started to chart in red, green and purple.

A single flight would not solve the mystery of why a storm did not develop as expected. But the real-time data went to the National Hurricane Center before its next forecast advisory. And, once on land, the scientists could plug the new data into their models to make them more accurate.

For now, the conclusion was this: On the north side of Barry, the air was dry, and the storm struggled to get organized. On the south side, there was the bottom half of a full-fledged — and very wet — system.

“It’s a tale of two storms,” said Robert Rogers, a meteorologist with the Hurricane Research Division in Miami who was the lead scientist on the flight.

The three pilots and two flight engineers worked on a rotation, taking breaks at a galley in the back of the plane. They opened lunch bags full of homemade salads, leftover tortellini and ham-and-Swiss sandwiches. Other times, they had frozen burritos from Trader Joe’s.

For a while, the radio picked up air traffic from commercial flights. But as Flight 42 approached the storm’s core, the traffic stopped. In just moments, the azure sky and gleaming sun in the windows turned an opaque white. Rain was everywhere.

Then came the wind. It buffeted the plane to the left near the cockpit, and to the right near the tail. It pushed up, making the passengers feel heavy. It pressed down, making them feel light. Kermit trembled and rattled.

On the radio, Holmes apologized as he scrutinized the radar. “I wish I could see the end, but it just keeps building!” he said. “This never ends, guys. It just keeps building in, the whole way.”

Finally, about 90 harrowing minutes later, a break appeared in the massive storm blob. The seatbelt sign went dark. Passengers exhaled.

“When people ask me what it’s like, I tell them, ‘Think about putting a GoPro in the dishwasher — and then running it,’” said Cmdr. Nathan Kahn, one of the pilots. “That’s what you see.”

The scientists stood up and peered at one another’s computer screens. For a few hours, they worked in relative calm, squinting at the tiny numbers and streaks of color that would soon help cities below them prepare for the onslaught. Then it was time to turn around; to head back south, into the nasty stuff. In the cockpit, the pilots felt a bump.

“Come on, Barry, don’t be like that,” one of them said.

And they went in, again.

This article originally appeared in The New York Times.