NOTICE (September 2017): Data previously stored at the CDIAC archive are being transitioned to the new, ESS-DIVE archive. This website provides access to the CDAIC data during the transition. If you have any questions regarding the new archive or the data transition, please contact

image image image image

Modern Records of Atmospheric Oxygen (O2) from Scripps Institution of Oceanography


This page provides an introduction and links to records of atmospheric oxygen (O2) concentrations at nine currently active stations. Records since 1989 are available from Scripps Pier and Alert, Alaska, although these are not continuous. Continuous records from seven stations extend back to 1993, and data for the other two stations (Cold Bay, Alaska and Palmer Station, Antarctica) are available back to the mid 1990s. These data are from remote locations or other locations situated so that they represent averages over large portions of the globe rather than local background sources.

These data have graciously been made freely available for access and distribution by researchers at the Scripps Institution of Oceanography; the original investigators made the effort to obtain the data and assure their quality. To assure proper credit is given, please follow the instructions at the bottom of this page when using any of this material. If data accessed from this site are to be used in a publication we strongly recommend some contact with the data contributor at an early stage of the work to be sure the data are being interpreted and used correctly. Neither the principal investigators nor CDIAC is responsible for misuse of these data.


Ralph Keeling, Scripps Institution of Oceanography (Scripps O2 Program)

Period of Record

1989 - current

Station Locations


Station Information

Station Latitude Longitude Elevation Beginning
(meters above msl) of Record
Alert 82.3° N 62.3° W 210 11/89
Cold Bay 55.21° N 167.72° W 21 8/95
Scripps Pier 32.87° N 117.26° W 10 5/89
Mauna Loa 19.54° N 155.58° W 3397 1/91
Cape Kumukahi 19.52° N 154.82° W 3 6/93
American Samoa 14.25° S 170.56° W 30 6/93
Cape Grim 40.69° S 144.69° E 94 1/91
Palmer Station 64.92° S 64.00° W 10 9/96
South Pole 90° S   2810 11/91

graphics Graphics

Graphics Plots of the monthly data for each site:

image Data

Daily and monthly averages for each station:


Measuring changes in the amount of oxygen in the atmosphere is difficult because there is so much of it. If you see a crowd of 400 and you go away and come back again, you would notice if 100 of them had gone somewhere else. If the crowd numbered 200,000 people a lot more would have to have gone away before you would notice any difference in the size of the crowd. This analogy holds for measuring oxygen, which is about 21 percent of the earth's atmosphere on a molecule-by-molecule basis.

Several techniques have become available for measuring atmospheric oxygen. The one used at Scripps Institution of Oceanography is interferometry, which exploits the refractive properties of different precisely known wavelengths to measure the oxygen/nitrogen ratio (O2/N2) (Keeling et al. 1998). Other methods have since been developed, including: paramagnetic analysis (Manning and Keeling, 1999), gas chromatography (Tohjima, 2000), vacuum ultraviolet absorption (Stephens and Keeling, 2003; mass spectrometry (Bender, 2005) and differential fuel-cell analysis (Stephens et al. 2007).

δ (O2/N2) is defined as:

The units of δ (O2/N2) are per meg; one per meg is one molecule of oxygen out of a million molecules of oxygen. Currently, this is roughly one molecule out of 4.8 million molecules of all gases in the atmosphere, not including water vapor molecules. Thus, 4.8 per meg is roughly one in a million molecules of dry air, or one part per million (ppm).

For more information on the stability of the reference gases, see Keeling et al. (2007).

Atmospheric Potential Oxygen (APO)

Atmospheric Potential Oxygen (APO) is defined so as to minimize the influences of the terrestrial biosphere, thereby providing information useful in quantifying the oceanic components of O2 and CO2 uptake in the context of the CO2 provided to the atmosphere by fossil fuel combustion. A simplified definition of APO (in ppm) is:

APO = δ(O2/N2) + 1.1 × CO2.

The factor of 1.1 arises from additional O2 (probably associated with oxidation/reduction reactions involving nitrogen) that is emitted by the terrestrial biosphere in addition to that typically represented in the equation for net photosynthesis:

6H2O + 6CO2 → C6H12O6 + 6O2.

Sometimes a constant value (e.g. 350 ppm) is subtracted from the CO2 term, to represent the oxygen concentration the air would have if photosynthesis/respiration could adjust the atmospheric CO2 concentration to that constant value.

To obtain an approximate expression for APO in units of per meg, simply multiply the first expression for APO, above, by the approximate conversion factor from ppm to per meg.

4.8 × APO (ppm) = 4.8 × δ(O2/N2) (ppm) + 4.8 x 1.1 × CO2 (ppm)
= APO per meg = δ(O2/N2) (per meg) + 4.8 × 1.1 × CO2 ppm,

which is the form given by Battle et al. (2006).

For a more information about per meg units or APO, see:

For more information about APO, see Battle et al. (2006); for a more technical discussion see Stephens et al. (1998).


Oxygen concentrations are currently declining at roughly 19 per meg per year, or about 4 ppm per year. One "per meg" indicates one molecule out of 1,000,000 oxygen molecules, or roughly one molecule in 4.8 million molecules of air.

Oxygen Depletion

We are occasionally reminded that fossil fuel burning is depleting atmospheric oxygen at a rate of almost 1000 tons per second. There are about 32 million seconds in a year, so that somewhere around 30 billion tons of O2 are being converted to CO2 annually. There are about 1,200,000 billion metric tons of O2 in the atmosphere, so we can keep burning fossil fuels at the present rate for 40,000 years before we run out of oxygen. By then, all of the world's fossil fuel supply will have long since been exhausted. For a more complete, but less detailed, discussion of this topic see Et tu 02 by Wallace Broecker.

If we take the worlds supply of fossil fuel to be 10,000 billion metric tons of carbon, as per and we oxidize all of it we would get about 37,000 billion metric tons of CO2, and about 27,000 billion metric tons of O2 would have been consumed. Some additional O2 would have also been consumed by oxidation of hydrogen in the (hydrocarbon) fuel, so that roughly 38,000 billion metric tons of oxygen would have been consumed. This is about 3.3 percent of the atmosphere's oxygen. Such a loss would be equivalent to increasing your elevation from sea level to about 330 meters, or about 1100 feet.


  • Battle, M., S, Mikaloff-Fletcher, M Bender, et al., 2006. Atmospheric potential oxygen: New observations and their implications for some atmospheric and oceanic models. Global Geochemical Cycles, 20 (4).
  • Bender, M.L,, D.T. Ho, M.B. Hendricks, et al. 2005. Atmospheric O2//N2 changes, 1993–2002: Implications for the partitioning of fossil fuel CO2 sequestration. Global Biogeochemical Cycles, 19 (4).
  • Keeling, R.F. 1988. Measuring correlations between atmospheric oxygen and carbon dioxide mole fractions: a preliminary study in urban air. Journal of Atmospheric Chemistry 7: 153-176.
  • Keeling, R.F., A.C. Manning, E.M. McEvoy and S.R. Shertz. 1998. Methods for measuring changes in atmospheric O2 concentration and their application in Southern Hemisphere air. J. Geophys. Res. 103: 3381-3397.
  • Keeling, R.F., A.C. Manning, W.J. Paplawsky and A.C. Cox. 2007. On the long-term stability of reference gases for atmospheric O2/N2 and CO2 measurements. Tellus 59B, 3-14.
  • Manning, A.C., R.F. Keeling and J.P. Severinghaus. 1999. Precise atmospheric oxygen measurements in a paramagnetic oxygen analyzer. Global Biogeochemical Cycles, 14: 1107-1115.
  • Manning, A.C. and R.F. Keeling. 2006. Global oceanic and land biotic carbon sinks from the Scripps atmospheric oxygen flask sampling network. Tellus 58B, 95–116.
  • Stephens, B.B., R.F. Keeling, M. Heimann, et al. 1998. Testing global ocean carbon cycle models using measurements of atmospheric O2 and CO2 concentration. Global Biogeochemical Cycles 12: 213-230.
  • Stephens, B.B., R.F.Keeling and W. J.Paplawsky. 2003. Shipboard measurements of atmospheric oxygen using a vacuum –absorption ultraviolet technique. Tellus 55B: 857-878.
  • Stephens, B. B., P.S. Bawkin, P.P. Tans, et al. 2007. Application of a differential fuel-cell analyzer for measuring atmospheric oxygen variations. J. Atmospheric and Oceanic Tech. 24: 82-94.
  • Tohjima, Y. 2000. Method for measuring changes in the atmospheric O2/N2 ratio by a gas chromatograph equipped with a thermal conductivity detector. J. Geophys. Res. 105: 14575-14584.

Citing These Data

Please cite as: Keeling. R.F. Atmospheric oxygen (and/or APO) data for (site name or names) (for oxygen data), (for APO data) and/or (for graphics), and the date the file was accessed.

For example:

Keeling, R.F. 2013, Atmospheric oxygen and APO data for Alert, Canada and the South Pole, (for oxygen data), and (for APO data), accessed June 11, 2013.

We also recommend citing Keeling 1988, above.

If accessing the data from this site: please also cite: Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, U.S. Department of Energy.