NOTICE (March 2018): This website provides access to the CDIAC archive data temporarily. It will be gradually transitioned into data packages in the new ESS-DIVE archive. This site will continue to operate in parallel during and after the transition, and will be retired at a future date. If you have any questions regarding the data or the transition, please contact

image image image image

Atmospheric CO2 Record from In Situ Measurements at Amsterdam Island

graphics Graphics   data Data


A. Gaudry, V. Kazan, and P. Monfray
Centre des Faibles Radioactivités,
Laboratoire de Modélisation du Climat et de l'Environnement,
Centre d'Etudes de Saclay, Bâtiment 709,
Orme des Merisiers, 91191-Gif-sur-Yvette Cedex, France

Period of Record



Until 1993 air samples were collected continuously through an air intake located at the top of a tower, 9 m above ground and 65 m above mean sea level. Since 1994, the intake has been situated 20 m above ground and 76 m above mean sea level. The tower is located at the north-northwest end of the island on the edge of a 55 m cliff. The air is dried by means of a cryogenic water trap at -60°C. Until 1990, determinations of CO2 were made by using successively two Hartmann-Braun URAS 2T nondispersive infrared (NDIR) analyzers. Since 1991, CO2 determinations have been made using a Siemens ULTRAMAT 5F NDIR. Standard gases in use from October 1980 to December of 1984 were CO2-in-N2 mixtures certified by Scripps Institution of Oceanography (SIO). The N2 scale was corrected for the carrier gas effect to obtain the air scale (WMO mole fraction scale). In 1985, CO2-in-air mixtures expressed in the 1985 WMO mole fraction in air scale were introduced. In 1990, a new series of 12 primary standard gases were gravimetrically prepared, then linearly adjusted at the laboratory and checked several times (i.e., 1990, 1992, and 1993) through intercalibrations with DSIR in New Zealand and NOAA/CMDL, which both used the 1985 mole fraction scale. The agreement was always better than 0.1 ppm (Monfray et al. 1992). Since 1993, the 1993 mole fraction scale has been used thanks to a new series of 10 cylinders provided by SIO.

Hourly mean CO2 concentrations are compared with hourly mean meteorological conditions (i.e., wind direction and velocity). For the samples to be considered indicative of background air conditions, the following criteria must be met: (1) wind direction must be from 260-300° and wind velocity must be >8 m/s, or wind direction must be from 300-360° or 0-50° and wind velocity must be >4 m/s; (2) the individual hourly means must be part of a sequence of four consecutive hours where the above conditions are met; and (3) the standard deviation of the hourly means must be less than 0.2 m/s. The wind criteria are indicative of maritime air masses, and cancel southerly winds, which are frequent. Data are reported in the Centre des Faibles Radioactivités' 1990 scale. For further details on the sampling methods, reference gases, and data selection criteria, see Gaudry et al. (1983, 1991).

Map showing location of Amsterdam Island, Indian Ocean

Amsterdam Island, Indian Ocean
Island seashore cliff
37°47' S, 77°31' E
70 m above MSL


The Centre des Faibles Radioactivités has been monitoring atmospheric CO2 concentrations since May 1980 at the station of Amsterdam Island, Territoire des Terres Australes et Antarctiques Francaises. The station, 5000 km off South Africa, is part of the Global Atmospheric Watch (GAW) network of the World Meteorological Organization (WMO). The sampling point is located at the north-northwest end of the island, on the edge of a 55-m high cliff. For further details on the site and sampling methods at Amsterdam Island see Gaudry et al. (1983, 1991).

The yearly mean CO2concentration increased from 338.6 parts per million by volume (ppmv) in 1981 to 358.4 ppmv in 1995; however, the annual rate of increase varied significantly during this period. Lambert et al. (1995) reported the atmospheric CO2 growth rate has decreased from ~1-2 ppmv per year, typical of the years 1980 to 1990, to less than 1 ppmv per year in 1992-1993. For further discussion on results, influences, and trends in the atmospheric CO2 record from Amsterdam Island see Ascensio-Parvy et al. (1984), Gaudry et al. (1987, 1990, 1991), Lambert et al. (1995), and Monfray et al. (1987).


  • Ascensio-Parvy, J.M., A. Gaudry, and G. Lambert. 1984. Year-to-year CO2 variations at Amsterdam Island. Geophysical Research Letters 11(12):1215-17.
  • Gaudry, A., J.M. Ascensio, and G. Lambert. 1983. Preliminary study of CO2 variations at Amsterdam Island (Territoire des Terres Australes et Antarctiques Francaises). Journal of Geophysical Research 88:1323-29.
  • Gaudry, A., P. Monfray, G. Polian, and G. Lambert. 1987. The 1982-83 El Niño: A 6 billion ton CO2 release. Tellus 39(B):209-13.
  • Gaudry, A., G. Polian, B. Ardouin, and G. Lambert. 1990. Radon-calibrated emissions of CO2 from South Africa. Tellus 42(B):9-19.
  • Gaudry, A., P. Monfray, G. Polian, G. Bonsang, B. Ardouin, A. Jegou, and G. Lambert. 1991. on-seasonal variations of atmospheric CO2 concentrations at Amsterdam Island. Tellus 43(B):136-43.
  • Lambert, G., P. Monfray, B. Ardouin, G. Bonsang, A. Gaudry, V. Kazan, and G. Polian. 1995. Year-to-year chnages in atmospheric CO2. Tellus 47(B):53-5.
  • Monfray, P., A. Gaudry, G. Polian, and G. Lambert. 1987. Seasonal variations of atmospheric CO2 in the Southern Indian Ocean. Tellus 39(B):67-71.
  • Monfray, P., M. Ramonet, A. Gaudry, G. Pearman, D. Beardsmore, M. Manning, and P. Pohl. 1992. An intercalibration of CO2 measurements between France, Australia, and New Zealand. WMO/Global Atmosphere Watch Report No. 77. World Meteorological Organization, Geneva.

CITE AS: Gaudry, A., V. Kazan, and P. Monfray. 1996. Atmospheric CO2 record from in situ measurements at Amsterdam Island. In Trends: A Compendium of Data on Global Change. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, U.S. Department of Energy, Oak Ridge, Tenn., U.S.A.