Source: Robert Simmon/NASA

This Stunning Chart Helped Uncover Climate Change in 1959

'Oh my god': How a post-doc chemist stumbled on the pattern —and why the Carbon Clock started running backwards this week.

Spring is in the air, but what is it? The blossoms blossoming? The vernal equinox?

Yes to both, and one more: It's how a young American chemist named Charles David Keeling stumbled upon proof of a climate problem way back in 1959.

Here's how he made his astonishing discovery—and why the Bloomberg Carbon Clock just started running backwards. 

Keeling browsing through climate data in March 2004. Back in the '50s, his son recalled, he looked at his results "and realized: Oh my god."
Keeling browsing through climate data in March 2004. Back in the '50s, his son recalled, he looked at his results "and realized: Oh my god."
Photographer: Denis Poroy/AP Photo

Keeling earned a doctorate in chemistry from Northwestern in 1954 and headed west—the mountains seemed like a nice place—for a post-doc assignment at Caltech. Now, what to work on?

He posed for himself a hypothetical, possibly naïve question, a "sort of a what-if, just to do something that would give me a handle on studying the environment," he said in a 2005 lecture he gave upon receiving the Tyler Prize for Environmental Achievement.

It turned out, he realized, that nobody had a solid handle on what’s in the air. Specifically, the amount of carbon dioxide, a gas that was known since 1859 to absorb heat.

His ambition was modest. He wanted to do some basic experiments. It was 15 years before an environmental movement would emerge, and even then it had nothing to do with global warming. The phrase "global warming" itself wouldn't be coined until 1975.  Climate politics of the scalding kind seen today in the U.S. wouldn't emerge for nearly half a century. 

So Keeling made from scratch an instrument that measured the air’s CO2 content with great precision, and drew up a rigorous air-monitoring routine.

"Why did I devise such an elaborate sampling strategy when my experiment didn't really require it?" he asked in a 1998 essay. "The reason was simply that I was having fun."

He liked designing and putting together equipment. He liked camping out, even at the expense of waking to check everything several times a night. What he learned from those nights under the stars—away from the cities and forests that spit out and suck in CO2—is that the atmosphere has a steady concentration of the gas. CO2 clocked in just above 0.03 percent of the volume of well-mixed air. That was Keeling’s first discovery. 

By 1956, he had brought his experiment to the Scripps Institution of Oceanography, in La Jolla, California, where he would work until his death in 2005. The possibility that industrial CO2 emissions were warming the world was a viable but untested idea at the time. Several senior Scripps scientists had been wondering about the additional CO2 and what, if anything, it might mean. The U.S. had built a scientific outpost on Hawaii's Mauna Loa volcano. Keeling faced a phenomenal opportunity: to set up a monitoring station 11,135 feet up the side of a mountain, in the middle of the ocean, where his machine could sniff the clear air. 

"He got brought down to Scripps to run this enormous program at a very young age," Ralph Keeling, Dave Keeling's son and a prominent Scripps geochemist, said in an interview late last year. "In the generation after World War II, I guess you just threw people into things. You can kind of sense the postwar mentality."

Keeling started recording CO2 at the Mauna Loa facility in March 1958. Scientists think of atmospheric carbon not as a percentage but as “parts” of CO2 for every million parts of air, or parts per million (ppm). Keeling's first monthly average came in at 313.4 ppm. 

But by June, the observations had started to look like a noisy scatter of data. CO2 was up and down, barely distinguishable from poppy seeds remaining in the wake of a bagel.

Courtesy Scripps Institution of Oceanography

Now, it’s a cliché that scientists never say "Eureka!" when they make an important discovery. Instead, they usually suspect that something broke in the lab and the readings are wrong.

In fact, something had broken in the lab—a power generator. The relatively new outpost was fitted with a lowly five-kilowatt machine. It took more than a month to juice Keeling's equipment back up. Two more months of scattered data came in, with the readings bunched together several notches below the first batch.

Courtesy Scripps Institution of Oceanography

Then the generator blew again. Another two months in the dark. The power was back online before New Year's 1959. The new data seemed to leap up from their autumn doldrums, but the big picture still didn't make a whole lot of sense. 

Courtesy Scripps Institution of Oceanography

You know what happened next? The power blew. As disruptive as that must have been, at this point, the bigger problem was that the data just weren't telling a clear story. 

"He figured the system wasn't working, and he didn't know what was wrong," Ralph Keeling said. "And then, at some point, about a year into it, he realized: Oh my god, it's a seasonal cycle." 

Courtesy Scripps Institution of Oceanography

By the end of the winter of 1959, the Mauna Loa Observatory had secured a 10-kilowatt generator and a backup. Reliable power brought continuity to the CO2 measurements, and before long Keeling was able to make out the earliest trace of a pattern in the data. 

CO2 appeared to climb in the winter and drop beginning in the spring. This seasonal cycle is caused by plants and trees in the Northern Hemisphere, where most of them live, fusing atmospheric CO2 into their cells during growth. Since 1959, the season has been noteworthy not only for what's in the air—birds, pollen, love—but what we can practically see plants and trees snatching out of it: carbon dioxide.

Another year passed before Keeling delivered the data that basically kicked off the modern science of global warming. He wrote in the journal Tellus that the Mauna Loa measurement and another on Antarctica showed that the CO2 levels weren’t the same year-on-year. They were rising.

"Where data extend beyond one year,” he wrote, “averages for the second year are higher than for the first year."

The same phenomenon turned out to be true of the third and fourth years. And all the years since. As long as scientists have been measuring atmospheric CO2, the annual trend has been the same, despite the dip every spring.

Last year, at the end of its summertime decline, CO2 at Mauna Loa bottomed out at just under 398 ppm. From this year forward, it may not dip below 400 ppm in any of our lifetimes. Before industrialization it stood at about 280 ppm. The safe zone for CO2 may be below 350 ppm, a level last seen around 1988. The upper end of the safe zone—if you prefer, the beginning of the danger zone—is about 450 ppm. We're halfway there.

One lesson from Keeling's work concerns the human attention span. If you consider any chunk of time greater than a year or two, the Keeling Curve has a pretty clear upward trend. If you look at it only for a few weeks or months in the spring and summer, it's rolling down. Imagine it as a numerical counter, as we do with the Carbon Clock, and you'll see the CO2 number run backwards for a few months, before time marches on. 

Bloomberg Carbon Clock: Measuring Carbon Dioxide that Causes Global Warming


Before it's here, it's on the Bloomberg Terminal. LEARN MORE