Abstract. The measurement of greenhouse gases in the atmosphere is one of the activities to monitor changes in the global environment, as some of these gases are presumed to contribute to climate change. The CCQM Gas Analysis Working Group has organised a pilot comparison for carbon dioxide and methane at ambient levels to compare primarily the capabilities of national metrology institutes (NMIs) for measuring these species in air. The participation in this comparison was not limited to signatories of the MRA, as is usually the case. Laboratories from the World Meteorological Organisation (WMO) were invited as well, and two participated.

The project has been split in two parts, a comparison of the (analytical) measurement capability, and a high-accuracy comparison of primary standard gas mixtures for greenhouse gases. In this first part, the measurement capability was compared between NMIs and WMO laboratories. The nominal amount-of-substance fraction levels are 365 µmol/mol for carbon dioxide and 1.8 µmol/mol for methane. The matrix was synthetic air, simulated by a mixture of nitrogen, oxygen (209 mmol/mol) and argon (9.3 mmol/mol). The measurements took place in the spring of 2003.

The protocol used for this comparison was the same as used so far in the gas analysis area for key comparisons and studies of this type. The reference values were assigned to each gas mixture on the basis of the gravimetric preparation, taking into account the results from the purity verification. In particular for the value assignment of methane, traces of methane in the matrix gases (nitrogen, oxygen and argon) can influence the value assigned to the gas mixture appreciably.

The results for methane agree within 4% relative, and for most participants even within 2% relative at an amount of substance fraction level of 1.8 µmol/mol. For carbon dioxide all results agree within 1% relative, and for most the agreement is even better: within 0.5% relative at an amount of substance fraction level of 365 µmol/mol. The methane data show a discrepancy of approximately 0.025 µmol/mol between the averages of the WMO laboratories and the NMIs, which confirms results of earlier comparisons. Measurement traceability is quite differently established in NMIs and WMO laboratories, and may account for this difference. Further work is needed to find out the causes of this difference.

Main text. To reach the main text of this paper, click on Final Report.

The final report has been peer-reviewed and approved for publication by the CCQM.