EMODNET Chemistry 7 th Steering Committee SYKE Helsinki

EMODNET Chemistry 7 th Steering Committee, SYKE, Helsinki, 7 June 2016 WP 2 Products in Mediterranean Sea Sissy Iona, HCMR-HNODC

DIVA maps • ec. oceanbrowser. net/emodnet • sextant. ifremer. fr/en/web/emodnet_chemistry/catalogue

Aggregated data collections Data Buffer Water body phosphate Water body ammonium Water body total phosphorus Water body total nitrogen

Aggregated data collections Data Buffer Water body nitrate plus nitrite Water body silicate

Aggregated data collections Water body dissolved oxygen concentration Chl-a: aggregation not complete yet

QC on aggregated data Methodology for data QA/QC and DIVA products (2015) http: //www. emodnet-chemistry. eu/documents. html Observations • Broad range control checks: – – High values Default not null values Negatives Zero values • Duplications checks • • No Spikes detection was applied No N: P ratio Interpolated data • Comparison with spatially averaged profiles

Further QC • For each region, mean vertical profiles on the vertical levels of the DIVA analysis arecomputed • Values outside the “mean ± 3 stdev” were rejected • Very extremes values (mainly coastal) are excluded in this way

Common format errors ERROR: Header Line. 'Depth[m]' is not in the semantic header. Cannot import this variable. Scripts to correct these are available

Format errors/possible cause • There are format errors in the harvested data – Make sure you use the “Import > Sea. Data. Net Formats” option to validate the SDN format and not the “Import > ODV Spreadsheet” – “Import > ODV Spreadsheet” will not detect wrong P 01 terms, mismatches between semantic and column headers, etc • Update the local odv txt files and inform MARIS for the corresponding CDIs updates • Scripts to correct these are available

QC on Oxygen Issues with wrong units/P 01

Glider Oxyg data with flag=2

Issues with the Instruments Oxygen-floats-west Med Although in P 01 there is no info for uncalibrated sensors

East Med Chl-a West Med Chl-a

QC feedback loop • Emails have been (will be) sent to Mediterranean cdi_partners and data originators for the results of the QC • Your response is needed to: – correct the sources and – Not exclude your data for the products

Influence of coastal data

Influence of coastal data Usage of variable Correlation Length depending on the depth in the next updates

Contaminants/Work so far • 330 P 01 terms in water profiles => 90 P 35 terms => 60 missing units conversions • 232 P 01 terms in sediment => 44 P 35 terms => 9 missing units conversions • 424 P 01 terms in time series Þ 54 P 35 terms List of missing units conversions were sent to Reiner but not the conversions factors New P 35 in SDN-BODC vocabs => aggregation will be repeated