The primary goal of the WMO/IAEA Round Robin Comparison Experiment is to assess the level to which participating laboratories maintain their link to the WMO mole fraction scales using normal operating procedures. Maintaining a direct link to the WMO scales and successfully propagating the scales to working laboratory scales are fundamental to the measurement process. To meet the recommended WMO network compatibility goal of 0.1 and 0.05 ppm CO2 in background air in the northern and southern hemispheres respectively [WMO, 2012], results from these laboratory comparisons of reference gases must differ by much less than target compatibility levels of actual air measurements in the field which include additional uncertainties.

The primary focus of the Round Robin experiment is comparison of CO2 scales. Many participating labs are also able to make measurements of other greenhouse gas and related tracers including CH4, CO, H2, N2O, SF6, O2/N2, and δ13C and δ18O of CO2. Participating labs are encouraged to make these additional measurements provided the effort does not extend beyond the allotted time.

Ongoing and direct atmospheric air comparison experiments between laboratories is still our best strategy for assessing atmospheric measurement compatibility between independent programs. However, results from the periodic WMO/IAEA Round Robin experiments have proven useful to understanding the cause(s) when measurement differences between laboratories are observed.

If you are making ongoing, high-precision atmospheric measurements of CO2, maintain a laboratory CO2 scale, and would like to participate in or learn more about the WMO/IAEA Round Robin Comparison Experiments, please contact us at


The first Round Robin comparison in 1984-1985 was terminated because two of the circulating high-pressure cylinders were leaking [WMO, 1986]. Participants of the 3rd WMO Expert meeting on atmospheric carbon dioxide measurement techniques held at Lake Arrowhead, California in November 1985 emphasized the importance of experiments designed to compare national laboratory calibration scales. The group outlined specific requirements for future WMO comparisons, which were refined at the next WMO Expert meeting held in Gaithersburg, Maryland in 1987. As historical background information, we provide the recommendations for future WMO Round Robin comparisons from the 1987 meeting report [WMO, 1987].

  • The NOAA/GMCC group be requested by WMO to coordinate further intercomparison between national programmes for atmospheric CO2 measurement.
  • That this intercomparison takes the same general form as that carried out in 1986. That is that 3 cylinders of CO2-in-air gas mixtures be sent to a number of laboratories for measurement and that the results of all such measurements be published as soon as practicable.
  • That the CO2 concentration of intercomparison gases be within 15 ppm of the annual average for global background air.
  • That laboratories that did not participate in 1986 be given priority in arrangements for the next circuit.
  • That participating laboratories treat the intercomparison gases as if they were tertiary (working) standards and assign a concentration for them as if they were for use at a field station.
  • That each laboratory provides a short but detailed report of their measurements together with supporting information including at least the following:
    1. the CO2 concentration scale used (SIO, NBS, etc.)
    2. the number of (secondary) standards used directly and indirectly in determining the concentrations
    3. cylinder identification numbers for the standards
    4. the carrier gas for the standards (air or nitrogen) and, in the case of the air carrier gas whether this is a synthetically mixed air, or derived from natural air
    5. the assigned concentrations and date of calibration for each standard
    6. the type of calibration curve used for the analyzer (linear, second, third order, etc.)
    7. the method used to determine the calibration curve (e.g., least squares fitting)
    8. analyzer make and model
    9. sample and reference (comparison, zero) gas flow rates, and flushing and measuring times (time of day and frequencies)
    10. water vapour conditioning procedures
    11. whether flowing or static measurements are made
    12. for each intercomparison gas and for each day it is measured, the number of measurements and their average value
  • That the timetable be based approximately on two weeks shipping time from one laboratory to the next and two weeks for measurements at each laboratory. As a guideline each laboratory is expected to make separate measurements on three different days, although more measurements may be made provided these are consistent with a maximum amount of gas set by the coordinating group.


  • 16th WMO/IAEA Meeting on Carbon Dioxide, Other Greenhouse Gases, and Related Measurement Techniques (GGMT-2011), Wellington, New Zealand, 25-28 October 2011, 67 pp. October 2012 (Report No. 206)
  • Report of the WMO Meeting of Experts on Carbon Dioxide Concentration and Isotopic Measurement Techniques, Lake Arrowhead, California, 14-19 October 1990 (Report No. 77)
  • Report of the NBS/WMO Expert Meeting on Atmospheric CO2 Measurement Techniques, Gaithersburg, USA, 15-17 June 1987. December 1987 (Report No. 51)
  • Report of the Third WMO Expert Meeting on Atmospheric Carbon Dioxide Measurement Techniques, Lake Arrowhead, California, USA, 4-8 November 1985. October 1986 (Report No. 39)