In the 1955, the chemist Charles Keeling performed a simple experiment. He measured the amount of carbon dioxide (CO2) at the air around Pasadena, California. This experiment will take him, in the next few years, in several places, in the west coast in the United States and Central America. He collected air samples every four hours, twenty-four hours a day. Measuring his samples back to his laboratory at Cal Tech, Keeling made a fascinating discovery.  No matter where he had collected his samples, forests, cities,  e.t.c, those gathered near midday yielded a nearly constant atmospheric CO2 concentration, about 315 parts per million (p.p.m.). As Keeling’s measurements have started to attract some attention, Harry Wexler, a visionary scientist of the Weather Bureau, the precursor agency of NOAA, invited him in Washington D.C. to discuss how stable the number 315 might be and whether it was as universal as Keeling believed.

Among the scientists who were present at this meeting, was Roger Revelle, director of the Scripps Institution of Oceanography in La Jolla, California. Ravelle together with Hans Suess, an oceanographer also from Scripps had been studying the increase of atmospheric CO2,  as well as the level of CO2 exchange between the atmosphere and the oceans. Suess had already established, by measuring the proportions of radio-active and non-radioactive isotopes of carbon (C-14), that some, but not all of the carbon released by burning fossil fuel would be absorbed by the oceans.  In their 1957 paper, Ravelle and Suess argued that “humans are carrying out a large geophysical experiment of kind that could not have happened in the past nor be reproduced in the future”. Contemplating “the probably large increase in CO2  production by fossil fuel combustion in the coming decades” Revelle and Suess concluded  “that a total increase of 20 to 40% in atmospheric CO2 can be anticipated”.[1]

They could not be more prophetic. The actual increase of atmospheric CO2 concentrations between 1958, when the measurements in Mauna Loa started, and 2000 was 17%; by December 2007 it has been increased by 22%. It was 315ppm in March 1958, 384 in December 2007. Today is 406ppm.

But, in 1957 there were many uncertainties and Ravelle’s and Suess study was inconclusive. Revelle wanted to establish a reliable baseline including consistent atmospheric measurements of CO2. He managed to secure some funds to set up a laboratory for continuous CO2 measurements and asked Keeling to run it.  Keeling built an infrared gas analyser which could measure the concentration of CO2 within an air sample and in March 1957, and he began continuous measurements of air.  Early in 1958, Keeling installed a second gas analyzer at Mauna Loa , a 13,680-foot volcano in the south central part of the island of Hawaii and he begun taking systematic measurements of atmospheric CO2 emissions. Located on the slope of a volcano 3,400 m above sea level and remote from external sources and sinks of carbon dioxide, the site was identified as suitable for long-term measurements which continue to the present day.

The first results of the measurements at the Mauna Loa Observatory showed that the concentration of CO2 in the air was  314p.p.m. pretty close to Keeling’s initial measurements.

In 1976, Charles Keeling and colleagues published in the Journal Tellus the classic paper “Atmospheric carbon dioxide variations at Mauna Loa observatory,” where they documented for the first time the rise of carbon dioxide in the atmosphere and revealed, for first time, “the effects of the combustion of coal, petroleum, and natural gas on the distribution of CO, in the atmosphere.” [2]

[1] Revelle, Roger, and Hans E. Suess (1957). “Carbon Dioxide Exchange between Atmosphere and Ocean and the Question of an Increase of Atmospheric CO2 During the Past Decades.” Tellus 9: 18-27.

[2] Charles Keeling, et al, Atmospheric carbon dioxide variations at Mauna Loa observatory DOI: 10.1111/j.2153-3490.1976.tb00701.x

Image Credit: https://scripps.ucsd.edu/programs/keelingcurve/tag/charles-david-keeling/