The quality of the INSITU_GLO_BGC_DISCRETE_MY_013_046 product from the Copernicus Marine Service distribution is assessed through overall quality and spatial coverage of the following variables:
Chlorophyll
FLU2 (chlorophyll-a fluorescence; [mg m-3])
CPHL (chlorophyll-a concentration [mg m-3])
Oxygen
DOXY (dissolved oxygen [mmol m-3])
DOX1 (dissolved oxygen [ml l-1])
DOX2 (dissolved oxygen [μmol kg-1])
Nutrients
NTRA (nitrate [mmol m-3])
NTAW (nitrate [μmol kg-1])
SLCA (silicate [mmol m-3])
PHOS (phosphate [mmol m-3])
The particularity of any oceanographic in-situ measurements dataset is that the horizontal and vertical coverage of the dataset varies strongly along time following the technological developments and the international cooperation on observations networks. One should thus ensure that both the vertical coverage and the horizontal coverage fit his needs in terms of sampled ocean variability scales before using the data. Therefore, the horizontal coverage of the dataset for chlorophyll, oxygen, nitrate, silicate, and phosphate variables are shown.
As a measure of the overall quality of the data, the amount of data flagged as “good data” and “bad data”, respectively, during the quality control procedure is shown and discussed.
For additional information regarding the in-depth validation of this product, the calculation of the assessment metrics presented in this product other detailed information in quality and noticeable events please refer to the reference quid document CMEMC-INS-QUID-013_046.
Important notice:
The contents of this document are an assessment based on the best set of observations available for evaluation at the time the operational system was validated. The validation methodology was defined and agreed within Copernicus Marine Service, inheriting the long experience of MyOcean and MERSEA series of projects (Hernandez et al., 2018) The results presented in this report and derived estimated accuracy numbers (EAN) are representative of average error levels over large areas of the ocean. These numbers might be used as a mean error in one given point of the area, but in order to refine error estimates locally, the reader is invited to use complementary information from reference QUIDs (error maps for instance, when available).