This Simple but Ingenious Instrument Helped the World Measure Carbon Dioxide Levels in the Atmosphere
Scientist Charles Keeling’s invention had a profound effect on scientists’ understanding of the severity of the climate change crisis
To unlock the secrets of the atmosphere, first, travel away from cities, forests, cars and people, to somewhere the air blows clean. Say, the top of a Hawaiian volcano. Bring two round glass flasks, volleyball-sized but with the air vacuumed out, wrapped in surgical tape for safety. Point the glass nozzle jutting from each toward that fresh air. Now hold your breath and walk at least ten feet into the wind. Open the valve, let the air rush in and, well after the hissing has stopped, close it back up.
“Then you can breathe once you’ve done all that,” says Tim Lueker, a scientist with the Scripps Institution of Oceanography. “Now you have an air sample, and I’m still amazed that it works.”
Charles David Keeling established this method in the mid-1950s at the California Institute of Technology. Back then, scientists didn’t know how much carbon dioxide naturally occurred in the atmosphere—measurements varied widely. Keeling cracked how to measure it precisely. Then he took this endeavor to new heights—11,000 feet above sea level.
At a 1956 meeting with a research director at the U.S. Weather Bureau (a precursor to the National Weather Service), Keeling made an ambitious pitch: continuously monitoring CO2 using a gas analyzer, a high-tech but as-yet-unproven tool for the job. The Bureau’s new research station on Hawaii’s Mauna Loa volcano, with its high altitude and barren landscape, would prove to be the ideal collection spot. Along with that continuous monitoring, scientists would also gather five-liter air flask snapshots, sampled in pairs, at remote locations around the world, analyzing the results under controlled laboratory conditions. Nearly on the spot, Keeling was offered funding to build a CO2 program, taking it to Scripps in La Jolla, California. He began measurements in 1958.
As months and years of data arrived, Keeling’s rigorous work revealed two global phenomena: Earth’s plants inhaling CO2 each summer and exhaling it each winter, and CO2 levels’ unrelenting annual increase. The undulating graph, marking the atmospheric CO2 concentration, would become known as the “Keeling Curve.”
On January 29, 1982, at the Mauna Loa Observatory, a NOAA employee first filled glass flask I-457 (shown above), when atmospheric CO2 levels measured 341.28 parts per million (ppm). Analyzed at the Scripps lab, this sample also could now reveal CO2’s fingerprints from sources like living things, fossil fuels and volcanoes.
/https://tf-cmsv2-smithsonianmag-media.s3.amazonaws.com/filer_public/d4/fa/d4faf2d9-2c95-4442-a521-f9c8b1b585aa/caljsioa_2006-20_02-37-01.jpg)
In the early 1980s, Lueker joined Keeling’s lab in La Jolla. The researchers were grappling with major questions of global carbon math, from the ocean to the sky. Activities like burning fossil fuels and making cement release CO2. But only half of the carbon dioxide stays in the atmosphere. Land-based plants and the ocean (to a limit) absorb the rest. Determining those fractions and limits—the mathematical destination of fossil fuel CO2—was a contentious scientific question, complicated by the burning of the Amazon rainforest to clear land for farming.
Essential to much of this work was understanding how and where the planet cycles one of its most widely distributed elements through the atmosphere, ocean and land. Those answers hissed out of glass flasks like I-457, gathered at stations from near the North Pole to Mauna Loa, and from the tiny Christmas Island to the South Pole.
Subscribe to Smithsonian magazine now for just $19.99
This article is a selection from the June 2025 issue of Smithsonian magazine
/https://tf-cmsv2-smithsonianmag-media.s3.amazonaws.com/filer_public/75/ba/75ba8acc-e09c-44f7-aab1-27a80d94c1ee/bb3209394w_2.jpg)
In his later years, Charles Keeling reflected on the slow-footed public response to growing evidence of climate change. People, he wrote in 1998, should “heed the rise in atmospheric CO2 concentration as serious unless proven to be benign.” He died in 2005.
On June 7, 2022, flask I-457 took its own last breath on Mauna Loa before embarking on a different journey—one to the Smithsonian’s National Museum of American History. Curator Kristen Frederick-Frost remembers the flask arriving at the museum, perfectly cradled in its custom padding. She marveled at the box’s design, made to ship fragile flasks all over the world. “One of my big regrets is not collecting the box. We had to send it back,” she says.
When the flask took in its last sample in 2022, atmospheric CO2 levels registered 421.36 ppm, a concentration that Lueker calls “unbelievable.” When the instrument first began collecting samples in 1982, CO2 levels had been 80 ppm lower. “At that point, the fires, the droughts, the floods, the powerful hurricanes, that was all hypothetical,” he says, “and now we’re living through it.” With global CO2 continuing to rise faster than expected, the Scripps team is part of a monitoring effort trying to understand why. The Los Angeles Megacity Carbon Project, a collaboration between Scripps and other organizations, seeks to zero in on the localized sources and forms of pollution in Southern California cities to better address it. Flask samples still deduce sources’ CO2 fingerprints.
“It’s almost like a crystal ball, the glass flask,” Lueker adds. “It’s just stunning that such a simple invention could have such a profound effect.”