Department of Energy Argonne National Laboratory Office of Science NEWTON's Homepage NEWTON's Homepage
NEWTON, Ask A Scientist!
NEWTON Home Page NEWTON Teachers Visit Our Archives Ask A Question How To Ask A Question Question of the Week Our Expert Scientists Volunteer at NEWTON! Frequently Asked Questions Referencing NEWTON About NEWTON About Ask A Scientist Education At Argonne Determining Past CO2 Levels
Name: Skyler
Status: Student
Grade: 9-12
Location: Other
Country: Canada
Date: Summer 2009

How did scientists manage to measure past levels of carbon dioxide (CO2) within the atmosphere without actual air samples?

Hi Skyler,

There are a number of possibilities that have been used in the past few years to measure historical levels of CO2.

One of the most reliable methods is polar and mountaintop ice. As snow falls, it traps small pockets of air. As more snow falls on top, the layers underneath can be compressed by the weight to become ice, still with tiny bubbles of trapped air. In conditions where that ice has not had a chance to melt, the trapped air can be recovered and examined by drilling cores from the ice and carefully melting them so that the released air is captured. In effect, we do have samples of air from thousands of years ago.

Another less direct method is to examine the growth rings of trees. Trees grow more rapidly when the climate is warm and moist, but they also grow more rapidly when CO2 levels are higher. It is possible to make estimations of CO2 levels in the past by examining the growth rings of trees, either as trees or as found in timbers of old buildings or ships and so on.

Some forms of plankton form shells of calcium carbonate, by removing the dissolved carbonates from sea water. When CO2 levels are increased, the level of dissolved carbonate will also be increased. This can be reflected in growth patterns in microscopic fossil shells found in ancient limestone for instance.

All of these methods require the ability to accurately date the material you are examining. Scientists from around the world have developed a growth chart for trees which allows dates to be determined by comparing significant events which have left visible records in tree growths. In 1816 for instance, a volcanic eruption caused so much shading of the northern hemisphere that there was no summer that year. Trees around the whole of the northern hemisphere can be dated accurately if scientists can find the tree ring pattern with the missing summer. By comparing trees and timbers from different sources, scientists have developed a time line of tree ring patterns which stretches back hundreds of years.

Snow layers can be dated by counting the layers back from the surface, and again significant events may also leave tell-tale signs in the layers.

Nigel Skelton
Tennant Creek High School


Ice cores from Antarctica, Greenland, and glaciers all over the world have yielded trapped air within them, the CO2 content of which can be measured. The ice cores can yield ice that was produced from precipitation hundreds of thousands of years ago. There are a number of interesting sites on the internet that show CO2 concentrations found in ice cores: and

Note that air temperatures have been slightly higher at times in the past, but more importantly, we are in a natural high cycle of CO2 concentration and air temperature even aside from our man-made contributions of CO2 (see the second web site above). Note that CO2 concentration in the air now is higher than it appears to have been for hundreds of thousands of years.

Something that the ice cores show is that global warming precedes an increase in CO2 concentration (see show-cosub2sub-increases-lag-behind-temperature-rises-disproving-the- link-to-global-warming.html). There are feedback mechanisms that cause this, but it's not clear how they have contributed to the rise in CO2 and temperature over the last 150 years. Over the past 6 years, global temperature has not risen significantly and may have actually decreased the past two years. We won't know for some time whether we've reached the 100,000+ year peak of temperature or whether our man-made contribution of CO2 will cause further increases in air temperatures.

One last note. As air temperatures near Earth's surface have been rising recently, the temperature in the Stratosphere has been decreasing, an apparent response to the warming of the air near the surface.

David R. Cook
Climate Research Section
Environmental Science Division
Argonne National Laboratory

Hi Skyler.

Thanks for your question. I discovered a lot myself while researching this question.

The topic of human contribution to global warming is still being argued among scientists. Sometimes the arguments are quite heated. But to answer your question,

"How did scientists manage to measure past levels of carbon dioxide (CO2) within the atmosphere without actual air samples?"

Here are some papers that contradict each other in their conclusions, but provide a view of how scientists measure past levels of CO2.

1) This paper refers to data from air trapped in air bubbles in glaciers, 19th Century measurements from Scientists in the 1800s, and evidence from Stromatalytes.

Stomatas are pores in plant leaves...please see

This paper concludes that: The pre-industrial CO2 level was not significantly lower than current levels. Neither they nor the present readings are high relatively to the geologic record. The entire output of computer climate models begins with the assumption that preindustrial levels were measurably lower. Elimination of this assumption further undermines the claim that the warming in the industrial era period was due to human addition of CO2 to the atmosphere. Combined with their assumption that CO2 causes temperature increase when all records show the opposite then it is not surprising that IPCC predictions of temperature increase are consistently wrong.

2) This paper refers to evidence from air bubbles in glaciers.

3) This paper talks about data collected from "Proxy Data." "Proxy data are indirect measurements of CO2 - they are a measure of the effects of CO2," said Dr. Park, professor of geology and geophysics at Yale who created the computer simulations for the project. "While we cannot actually measure the CO2 that was in the atmosphere millions of years ago, we can measure the geologic record of its presence. For example, measurement of carbon isotopes in ancient ocean-plankton material reflects atmospheric CO2 concentrations." The full article can be found at:

Sincere regards,
Mike Stewart


Carbon dioxide is one of the many molecules in the air we breathe. It is the fourth most common component in air (after nitrogen at 78.08%, oxygen at 20.95%, and argon at 0.93%) and it accounts for only 0.038% (or 380 ppm) of all molecules in air!

If enough snow falls in one region it can turn into glacial ice. The glacial ice in Antarctica has been accumulating for several hundred thousand years. Geologists drill into the ice and analyze the composition of the material that makes up the ice – carbon dioxide is one of the components. The Vostok ice core of Antarctica has a provided a very reliable and long record of carbon dioxide concentrations for hundreds of thousands of years. The samples of air that we use are preserved in ice.

Leslie Kanat, Ph.D.
Professor of Geology
Department of Environmental Sciences
Johnson State College


One way is to take ice core samples. As the ice builds up, it traps gases of that era/age. So the deeper you go into the ice, the older the air sample trapped in. Taking the ratio of CO2 in the air sample gives an indication of the amount of CO2 in the air of that era/age. This can then be correlated to other sources.

Greg (Roberto Gregorius)
Canisius College

Click here to return to the Environmental and Earth Science Archives

NEWTON is an electronic community for Science, Math, and Computer Science K-12 Educators, sponsored and operated by Argonne National Laboratory's Educational Programs, Andrew Skipor, Ph.D., Head of Educational Programs.

For assistance with NEWTON contact a System Operator (, or at Argonne's Educational Programs

Educational Programs
Building 360
9700 S. Cass Ave.
Argonne, Illinois
60439-4845, USA
Update: June 2012
Weclome To Newton

Argonne National Laboratory