On this page
Most students meet the carbon cycle as a diagram with arrows. Charles David Keeling turned it into a measured line on a graph, one air sample at a time. This post tells how he did it, what the curve shows, and how to read it yourself with real numbers.
Who was Charles David Keeling?
According to the Scripps CO2 Program biography, Keeling was born in Scranton, Pennsylvania, on 20 April 1928, earned a PhD in chemistry from Northwestern University in 1954, and worked at Scripps Institution of Oceanography from 1956 until his death in 2005. He received the US National Medal of Science in 2002 and the Tyler Prize for Environmental Achievement in 2005.
As a postdoctoral researcher at Caltech, he built a precise instrument for measuring carbon dioxide in air. The Scripps account of the early Keeling curve says he took it to Big Sur and sampled air through the day and night. Air held more CO2 at night, when plants and soil respire without photosynthesis, but in the afternoon he kept getting about the same value, 310 ppm. He found the same afternoon value in the rain forests of the Olympic peninsula and in Arizona's mountain forests.
Why did a steady number matter so much?
Earlier measurements of carbon dioxide had varied widely, so nobody could say whether it was rising. Keeling's afternoon readings showed that, away from cities and the local influence of plants, there was a stable "background" value. If that background could be measured precisely in one remote place, year after year, any change would show.
In 1956 he proposed a programme of remote stations for the International Geophysical Year. The Scripps account says one of his infrared gas analysers was installed at Mauna Loa in March 1958, and on its first day it recorded 313 ppm.
What did the first two years show?
The concentration rose to a maximum in May 1958, fell to a minimum in October, then rose again and repeated the pattern in 1959. Keeling described it, in words quoted on the Scripps page, as "nature's withdrawing CO2 from the air for plant growth during summer and returning it each succeeding winter". The yearly average was also higher in 1959 and higher again in 1960. He reported both findings in his 1960 paper in Tellus, whose abstract describes "a systematic variation with season and latitude" in the northern hemisphere and "a small but persistent increase" in Antarctica.
Later, the Scripps biography notes, Keeling used the carbon-13 to carbon-12 ratio to show that the seasonal cycle carries the signature of land-plant photosynthesis and respiration. That links the curve straight to your biology syllabus. Our guide to nutrient cycles covers the carbon cycle that sits behind it.
How do you read the Keeling curve with real numbers?
NOAA now runs the Mauna Loa record and publishes it as plain text. Its annual mean file notes that data from March 1958 to April 1974 came from Keeling at Scripps. Here is a worked example using those annual means.
Step 1: the total rise. 1959: 315.98 ppm. 2025: 427.35 ppm. Rise = 427.35 − 315.98 = 111.37 ppm. As a percentage: 111.37 ÷ 315.98 × 100 ≈ 35.2%.
Step 2: the average rate. 111.37 ppm over 66 years = 111.37 ÷ 66 ≈ 1.69 ppm per year.
Step 3: is the rate changing? Compare two decades:
1959 to 1969: 324.62 − 315.98 = 8.64 ppm, or 0.86 ppm per year.
2015 to 2025: 427.35 − 401.01 = 26.34 ppm, or 2.63 ppm per year.
2.63 ÷ 0.86 ≈ 3.1, so the yearly rise is now about three times faster. The curve is not a straight line; it is getting steeper.
Step 4: the seasonal swing. From NOAA's monthly file, May 2025 averaged 430.51 ppm and September 2025 averaged 424.37 ppm, a swing of 6.14 ppm in four months. That is the northern hemisphere's plants drawing CO2 down over the growing season, exactly the pattern Keeling first saw in 1958.
What mistakes do students make with this graph?
Mixing up the wiggle and the trend. The yearly zigzag is plants; the upward slope is the extra carbon from fossil fuels and land-use change. Compare the same month across years, or use annual means, to see the trend.
Using a single year's change as "the rate". One year can be pushed up or down by events such as El Nino, which the Scripps biography links to year-to-year swings. Average over a decade.
Forgetting units. ppm means parts per million by number of molecules in dry air. Read the file header before you compare data sets.
What can a student take from Keeling's way of working?
Keeling's own essay, Rewards and Penalties of Monitoring the Earth (1998), describes four decades spent keeping one measurement going. His lesson for students is simple: precision and patience can turn a small, careful number into a discovery. He also did not explain away the night-time readings; he worked out why they were high.
If the link between photosynthesis, respiration and the seasonal swing still feels abstract, the Fermi Concept Library has every concept in physics, chemistry, biology, mathematics and computing with its own lesson, so you can revisit the carbon cycle and come back to the data.
Open NOAA's monthly file today, pick any five years, and calculate the May-to-September drop for each one.
