Analysis of total and organic carbon and total nitrogen in settling oceanic particles and a marine sediment: an interlaboratory comparison

10 laboratories, using their routine methods, collaborated in a comparison of analyses of total carbon, nitrogen and organic carbon in a single sample of marine sediment and a pooled sample of settling particulate material collected by sediment trap. The standard error of the means from all individual laboratories was +/-3% of the mean for all total carbon results, and for total nitrogen in the marine sediment, but was +/-7% of the mean for total nitrogen in the sediment trap material. Larger errors (+/-8% of the mean) were found for organic carbon results from both the marine sediment and the sediment trap material, with a range of results (expressed as a percentage of the untreated dry weight) for the sediment trap sample analyzed here of 5.60-8.03%. When this range is compared with that of particulate organic carbon concentration in sediment trap material obtained by different workers at various deep ocean sites (2.4-10.2%) [Wefer, G., 1989. Particle flux in the ocean: Effects of episodic production, in: W.H. Berger, V.S. Smetacek, G. Wefer (Eds.), Productivity of the Ocean: Present and Past, John Wiley & Sons, pp, 139-154.] it is apparent that a large proportion of the observed oceanic variability could be accounted for by difference in analytical technique. We suggest that the problem stems from the difficulty of accurate separation of the organic and inorganic phases and discuss the inaccuracies involved in the separation. We present evidence that the range of results is partly due to either volatilisation of organic carbon during acid treatment, or incomplete removal of organic matter during thermal treatment. Lf results from different laboratories are to be compared, as is presently required for basin scale and global studies of ocean carbon cycling and budgets, there is clearly a need for suitable reference materials, rigorous intercalibration and technique development. (C) 1998 Elsevier Science B.V.