It was a clear but chilly Tuesday morning on Weather Service Road in Sterling. The temperature was in the low 40s, with light winds from the south. A nearly full moon had just dipped below the horizon, and the sun was pinkening the east.
But John Newkirk and Brian Guyer didn't care too much about what was happening on the surface of our planet. They were curious what it was like miles and miles above their heads, miles and miles above the cars starting to back up on the toll road, miles and miles above the jets taking off from Dulles International Airport.
They were getting ready to launch their weather balloon.
The two of them -- John, the veteran, and Brian, the new guy -- walked about 100 yards from the National Weather Service's local forecast office to a two-story cinder-block building. They opened a door, walked past dozens of tall canisters of helium, then passed into a room that resembled a large, high-ceilinged garage.
Brian was carrying a small cardboard box that he put on a large work table in the center of the room. "I never touch [the balloon] with my hand," he said as he opened the box.
"If you touch it," said John, "the oil on your hands will compromise the neoprene."
So Brian was careful not to touch the skin of the balloon. Instead, he grabbed the balloon's thick rubber neck and flopped the uninflated bladder like a dead snake a few times to unfold it and lay it out on the table. The shriveled balloon was about twice as long as his arm, colored a sickly greenish-beige.
Brian clamped the balloon's hoselike neck onto a vertical nozzle at the end of the table and turned a knob. The room filled with a hissing sound. The hissing continued for nearly 15 minutes, then there was a clank as the balloon reached its full, six-foot diameter and the helium shut off automatically.
John and Brian left the balloon bobbing at its mooring as they walked back to the forecast office to prepare the payload they would soon be sending skyward.
It's an instrument known as a radiosonde, and twice a day, every day, at exactly the same time all over the world, close to 1,000 radiosondes are launched to measure temperature, humidity, pressure and wind. Each little box costs about $100 and is roughly the size of a pair of disposable cameras stuck end to end. The guts look like the inside of a smoke detector -- metal discs, circuit boards -- encased in foam and wrapped in a thin cardboard the color of a cloudless sky. A tiny wire tail sticks out: the antenna.
"If we don't send these radiosondes up, these guys can't forecast," John said, motioning toward the meteorologists sitting behind banks of computer screens. "We've got all this technology. We've got radar and computers, but without that little box . . ."
Brian entered the radiosonde's serial number into a computer, along with the current weather conditions.
"Now I'm base-lining the instrument," he said, "initializing it with data on the ground."
There was a squawk as a rack of instruments known as the servo picked up the radiosonde's distinctive signal. It rose and fell like the atonal bleats of a snake charmer's flute.
John reached over to the servo and turned down the volume. Then he and Brian donned their jackets and walked back to the launch building. As they walked, they looked up at the scattering of wispy clouds.
"That might be a high alticu," said Brian, meaning alticumulus.
"Or a low cirrus," said John.
"Ever seen lenticular?" asked Brian, meaning the distinctive pancakelike clouds that form near mountains.
John said he had.
The balloon was where they'd left it. Brian picked up a phone mounted on an outside wall and called the Dulles tower. "Weather Service requests to release a balloon in one minute," he said.
Brian tied off the balloon's neck with a length of string, then John took the balloon from the work table. An orange plastic parachute was tied about 10 feet below the balloon, and the radiosonde tied about 60 feet below that. The parachute was a nice gesture, but John said no more than 30 percent of radiosondes are found and mailed back to be refurbished. Most of the ones launched from Sterling splash into the Atlantic.
John walked the balloon outside while Brian held the radiosonde. It was a little after 7 a.m. as first John, then Brian, let go, and the balloon rose steadily into the sky.
Brian walked to the controls that worked an antenna atop the launch building. He dialed the coordinates of the ever-shrinking balloon until the radiosonde's signal was captured. Then the pair went back to the office to wait for the data.
Ticking in the corner of their computer screen was the time elapsed since the balloon was launched. After 18.4 minutes, the radiosonde was about three miles up and the barometric pressure had dropped to 500 millibars. At 55.1 minutes, the pressure was 94.5 millibars and falling steadily.
John and Brian knew that as the pressure dropped, the balloon would expand.
Eventually the balloon would be so big -- almost 30 feet in diameter -- that it couldn't help but burst.
"If [the pressure] gets to 10, I'll be very happy," Brian said. "I don't think it will, though. It'll pop any second."
About 8:45 a.m., the balloon was 18 miles above the Earth, and the pressure was still ticking down: 11.5, 11.3, 11.1. Then there was a chirp from the computer, and Brian let out an involuntary "Oh!"
The balloon and the radiosonde were falling, falling, falling back to Earth.