Data harmonisation and data validation Issues identified in

Data harmonisation and data validation Issues identified in the Arctic region Ann Kristin Østrem Institute of Marine Research, Norway The European Marine Observation and Data Network (EMODnet) is financed by the European Union under Regulation (EU) No 508/2014 of the European Parliament and of the Council of 15 May 2014 on the European Maritime and Fisheries Fund.

Definition of the Arctic region 62 - 88°N, 42°W – 70°E 10/28/2021 Regional meeting Baltic/Arctic, 17 June 2020 2

Issues identified - Eutrophication data ODV file format: No errors found this time, the issues from the harvesting in 2018 have been corrected Metadata: Positions on land Bottom depth Default values: 0, -9999, -9, 9999, … Larger than max depth of the region Less than max sample depth 10/28/2021 Regional meeting Baltic/Arctic, 17 June 2020 3

Issues identified - Eutrophication data Data: Default values: -10, -9. 98 -> should be empty, with QF = 9 Negative values: not possible (except for temperature) -> QF = 4 Zero values: Default value or below LOD? ? Default value -> should be empty, with QF = 9 Below Limit of detection -> QF = 6 Decimal point errors: Oxygen values 1/10 of normal (30 umol/l instead of 300), silicate values 10 times higher than normal Very low nitrate values – probably nitrite Data with QF = 0 QF = 2 used for bad data Profiles with same value for all depths 10/28/2021 Regional meeting Baltic/Arctic, 17 June 2020 4

Data examples - Oxygen Too low values for open sea Profiles with values 1/10 of normal Profiles with only zero values, and spikes/default values 10/28/2021 Regional meeting Baltic/Arctic, 17 June 2020 5

Data examples - Nitrate Negative values in CTD data Very low values – probably nitrite -9. 98 and 0 used as missing value 10/28/2021 Regional meeting Baltic/Arctic, 17 June 2020 6

Data examples - Silicate -9. 98 and 0 used as missing value Values 10 times higher than the rest of the data 10/28/2021 Regional meeting Baltic/Arctic, 17 June 2020 7

Issues identified - Contaminants data Many profiles with wrong primary variable – ADEPZZ 01 instead of COREDIST for sediment data Some files with errors in variable names Probably wrong depth unit in some sediment data – m instead of cm? Very little data for many parameters – difficult to QC 10/28/2021 Regional meeting Baltic/Arctic, 17 June 2020 8

@EMODnet www. emodnet. eu Your gateway to marine data in Europe