The idea that carbon dioxide levels in the atmosphere might be changing was not new when Keeling arrived on Mauna Loa. In 1938, the British engineer and amateur meteorologist Guy Stewart Callendar compared CO2 measurements taken at Kew between 1898 and 1901, which averaged 274 parts per million, with readings from the eastern United States in the late 1930s, which averaged 310 ppm. Callendar concluded that concentrations were rising due to the burning of fossil fuels. But his evidence was patchy, drawn from disparate locations with inconsistent methods, and the scientific community remained skeptical. What was missing was not the hypothesis but the data: a continuous, carefully calibrated record from a single site that could show a trend beyond seasonal noise and local contamination.

Keeling, a researcher at the Scripps Institution of Oceanography at UC San Diego, had spent years perfecting his measurement techniques at sites ranging from Big Sur near Monterey to the rain forests of Washington's Olympic Peninsula and high mountain forests in Arizona. He understood that meaningful atmospheric measurements required isolation from local sources and sinks of CO2, which led him to Mauna Loa, where clean air from the free troposphere reaches instruments uncontaminated by continental activity. His measurements used a nondispersive infrared sensor calibrated to World Meteorological Organization standards, a descendant of the capnograph first invented by John Tyndall in 1864. Keeling's 1960 paper in the journal Tellus presented the first monthly records from both Mauna Loa and Antarctica, identifying a "distinct seasonal cycle" and, more cautiously, "possibly a worldwide rise in CO2 from year to year."

The Keeling Curve reveals two stories at once. The first is seasonal: CO2 concentrations drop by about six parts per million from May to September as Northern Hemisphere vegetation absorbs carbon through photosynthesis during spring and summer, then rise again through fall and winter as plants decay and release it back. Because most of Earth's landmass and forests are in the Northern Hemisphere, this seasonal pulse dominates the global signal. The second story is the relentless upward slope. In March 1958, atmospheric CO2 stood at 313 ppm. By November 2018, it had reached 406 ppm, rising at a rate of approximately 2.5 ppm per year and accelerating. Measurements of ancient air trapped in polar ice cores show that CO2 concentration held between 275 and 285 ppm throughout the Holocene epoch, the 11,000 years of stable climate during which human civilization developed. The Keeling Curve showed that this stability was over.

On May 9, 2013, the daily mean CO2 concentration at Mauna Loa crossed 400 parts per million for the first time in the history of the measurement program. Paleoclimate estimates suggest that CO2 had not reached this level since the mid-Pliocene epoch, two to four million years ago, when sea levels were significantly higher and the Arctic was forested. The milestone drew worldwide attention. Harvard historian of science Naomi Oreskes called the Keeling Curve one of the most important scientific works of the twentieth century. In 2015, the American Chemical Society designated it a National Historic Chemical Landmark, installing commemorative plaques at both Mauna Loa Observatory and the Scripps Institution of Oceanography. That same year, Pope Francis cited the observatory's 400 ppm reading in Laudate Deum, his apostolic exhortation on the climate crisis.

Keeling supervised the Mauna Loa measurements until his death in 2005, fighting throughout his career for the funding to maintain what he knew was an irreplaceable record. Gaps in the data would have been permanent, and the curve's power lies precisely in its continuity: nearly seven decades of unbroken observation from the same location, using instruments calibrated against the same standards. After Keeling's death, oversight passed to his son, Ralph Keeling, who wrote in Science magazine on the fiftieth anniversary of the project about his father's tenacity and the program's evolution. Today, about 100 stations around the world monitor atmospheric CO2 through the Global Greenhouse Gas Reference Network, and Europe maintains its own version, the ICOS Curve, drawn from 38 monitoring sites. But no station on Earth has as long or as consequential a record as the instruments on the north flank of Mauna Loa, measuring what one man decided to measure because he believed precision mattered.