Use of reference materials for nutrients in seawater and comparability of nutrients in the world's oceans

Following issuance of the IPCC report in 2007, there has been renewed effort to improve the global comparability of nutrient measurements from seawater samples. The most recent inter-laboratory comparison studies were led by scientists at the Meteorological Research Institute (MRI), Japan, and have included many key laboratories around the world including the one at Scripps Institution of Oceanography/Shipboard Technical Support (SIO/STS). These comparison studies were made possible by the production of Reference Materials for Nutrients in Seawater (RMNS) by the General Environmental Technos Co., Japan (Aoyama et al 2006, Ota et al., 2010). Some of the participating laboratories have developed protocols for the use of RMNS to improve comparability between laboratories and cruises. SIO/STS participated in all of the recent comparison studies and helped develop protocols for use of the reference materials. The nutrient analysis section of the WOCE manual from the 1990s has also been updated and rewritten for GO-SHIP as part of these efforts. The original manual called for 1% accuracy or reproducibility for nitrate, 1-2% for phosphate, 1-3% for silicate (Joyce et al., 1991). The new GO-SHIP manual states, “These materials [RMNS] along with the use of best practice in using analysis equipment and improved internal standardization should make it commonly possible to achieve comparability of nutrient analyses to a level better than 1%.” (Hydes et al.,2010). The results from the comparison studies show that there is very good comparability and traceability between the participating core labs for nitrate, nitrite, and phosphate (Aoyama et al., 2006, 2007, 2008, and 2010); however, there are differences between some of the core labs in silicate. These inter-lab differences are on the order of 2% and are consistently present over the course of the comparison studies. The results suggest that the differences lie in standardization and lack of a primary silicate standard, thus indicating that the 2% silicate difference seen between some of the CLIVAR repeat cruises could be accounted for with the use of the reference materials. The chemistry lab at SIO/STS has been using the reference materials on cruises since 2009, specifically CLIVAR cruises P06, S04P, A20, A22, and P02, and the U.S. GEOTRACES cruises in 2011 and 2013. Comparison of the data from the recent CLIVAR cruises with the archived WHP data show very good agreement when SIO/STS was the analyzing lab for both expeditions.