Quality control procedures and methods of the CARINA database

• Published in: Earth System Science Data
• Main Authors: Tanhua, T., van Heuven, S., Key, R. M., Velo, A., Olsen, A., Schirnick, C.
• Format: Article in Journal/Newspaper
• Language: English
• Published: Copernicus Publications 2010
• Subjects: article; Verlagsveröffentlichung; Arctic; Southern Ocean
• Online Access: https://doi.org/10.5194/essd-2-35-2010; https://noa.gwlb.de/receive/cop_mods_00029305; https://noa.gwlb.de/servlets/MCRFileNodeServlet/cop_derivate_00029260/essd-2-35-2010.pdf; https://essd.copernicus.org/articles/2/35/2010/essd-2-35-2010.pdf

Data on the carbon and carbon relevant hydrographic and hydrochemical parameters from previously not publicly available cruises in the Arctic, Atlantic and Southern Ocean have been retrieved and merged to a new data base: CARINA (CARbon IN the Atlantic). These data have gone through rigorous quality control (QC) procedures to assure the highest possible quality and consistency. All CARINA data were subject to primary QC; a process in which data are studied in order to identify outliers and obvious errors. Additionally, secondary QC was performed for several of the measured parameters in the CARINA data base. Secondary QC is a process in which the data are objectively studied in order to quantify systematic differences in the reported values. This process involved crossover analysis, and as a second step the offsets derived from the crossover analysis were used to calculate corrections of the parameters measured on individual cruises using least square models. Significant biases found in the data have been corrected in the data products, i.e. three merged data files containing measured, calculated and interpolated data for each of the three regions (i.e. Arctic Mediterranean Seas, Atlantic, and Southern Ocean). Here we report on the technical details of the quality control and on tools that have been developed and used during the project, including procedures for crossover analysis and least square models. Furthermore, an interactive website for uploading of results, plots, comments etc. was developed and was of critical importance for the success of the project, this is also described here.