Confidence in the classification of substrate type in the 2023 EUSeaMap broad-scale predictive habitat map. Values are on a range from 1 (Low confidence), 2 (Moderate confidence), 3 (High confidence). Substrate type is one of the layers of information used to categorise physical habitat types in EUSeaMap; these layers of information are collectively known as 'habitat descriptors'. The substrate layer confidence was obtained from reclassification and standardisation of the confidence scores associated with each primary layer used to create the Substrate types layer. A report on the methods used in the 2023 version of EUSeaMap and reports on previous versions (v2019 and V2021) are linked in Online Resources.

Predictive Biological Zone layer produced by EMODnet Seabed Habitats as an input layer for the 2023 EUSeaMap broad-scale habitat model. The extent of the mapped area includes the Mediterranean Sea, Black Sea, Baltic Sea, and areas of the North Eastern Atlantic extending from the Canary Islands in the south to the Barents Sea in the North. The map of biological zone was produced using underlying physical data and thresholds derived from statistical analyses or expert judgement on known conditions. The model is produced in R and Arc Model Builder (10.1). The model was created using raster input layers with a cell size of 0.00104dd (roughly 100 meters). The model includes the sublittoral zone only (Infralittoral to Abyssal zone). An accompanying confidence layer is available for viewing and download from EMODnet Seabed Habitats. Reports detailing the methods used in the previous versions of EUSeaMap (v2019 and V2021) are linked in online resources a new report is in progress. Vasquez M., Manca E., Inghilesi R., Martin, S., Agnesi S., Al Hamdani Z., Annunziatellis A., Bekkby T., Pesch R., Askew A., Bentes L., Castle L., Doncheva V., Drakopoulou V., Gon├ºalves J., Laamanen L., Lillis H., Loukaidi V., McGrath F., Mo G., Monteiro P., Muresan M., O'Keeffe E., Populus J., Pinder J., Ridgeway A., Sakellariou D., Simboura M., Teaca A., Tempera, F., Todorova V., Tunesi L. and Virtanen E. EUSeaMap 2019, A European broad-scale seabed habitat map, Technical Report, 2019. Available from: https://archimer.ifremer.fr/doc/00636/74782/ Vasquez Mickael, Allen Harriet, Manca Eleonora, Castle Lewis, Lillis Helen, Agnesi Sabrina, Al Hamdani Zyad, Annunziatellis Aldo, Askew Natalie, Bekkby Trine, Bentes Luis, Doncheva Valentina, Drakopoulou Vivi, Duncan Graeme, Gonçalves Jorge, Inghilesi Roberto, Laamanen Leena, Loukaidi Valia, Martin Simon, McGrath Fergal, Mo Giulia, Monteiro Pedro, Muresan Mihaela, Nikilova Christina, O'Keeffe Eimear, Pesch Roland, Pinder Jordan, Populus Jacques, Ridgeway Amy, Sakellariou Dimitris, Teaca Adrian, Tempera Fernando, Todorova Valentina, Tunesi Leonardo, Virtanen Elina (2021). EUSeaMap 2021. A European broad-scale seabed habitat map. D1.13 EASME/EMFF/2018/1.3.1.8/Lot2/SI2.810241– EMODnet Thematic Lot n° 2 – Seabed Habitats EUSeaMap 2021 - Technical Report. https://doi.org/10.13155/83528 Credit: Licensed under CC-BY 4.0 from the European Marine Observation and Data Network (EMODnet) Seabed Habitats initiative (www.emodnet-seabedhabitats.eu), funded by the European Commission.

Confidence in the classification of biological zones in the EUSeaMap (2023) broad-scale predictive habitat map. Values are on a range from 1 (Low confidence), 2(Moderate confidence) and 3 (High confidence). Biological Zone is one of the layers of information used to categorise physical habitat types in EUSeaMap; these layers of information are collectively known as 'habitat descriptors'. Confidence in the classification of a Biological Zone at any location is driven by both the confidence in the values of the input variables, and the confidence in the classification based on proximity to, and uncertainty in, the boundary between classes (i.e. areas closer to a boundary between two classes will have lower confidence). Layers are also available showing confidence in the values of the input variables used to model Biological Zones. Detailed information on the modelling process is found in the EMODnet Seabed Habitats technical reports and its appendices (links in Resources). Created by the EMODnet Seabed Habitats project consortium.

Classified seabed substrate types for European seas. Produced by EMODnet Seabed Habitats as an input layer for the 2023 EUSeaMap broad-scale habitat model, based on a combination of EMODnet Geology seabed substrate products and biological substrates extracted form individual habitat maps from surveys around European seas. The extent of the mapped area includes the Mediterranean Sea, Black Sea, Baltic Sea, and areas of the North Eastern Atlantic extending from the Canary Islands in the south to the Barents Sea in the north. The layer of seabed substrate was produced using data from EMODnet geology at the following scales: - 1:5k, 1:10k, 1:15k, 1:20k, 1:25k, 1:50k, 1:100k, 1:200k, 1:1M (these were all update in 2023) Biological substrates were included in the 2023 version of EUSeaMap to assist in the classification of biogenic habitats for the 2019 version of EUNIS. The Folk 5 classification of substrate is adopted because it is compatible with both the 2007-11 and 2019 versions of EUNIS, both of which have been applied in EUSeaMap 2023. A report on the methods used in the 2023 version of EUSeaMap and reports on previous versions (v2019 and V2021) are linked in Online Resources.

Confidence in the classification of: - the Marine Strategy Framework Directive (MSFD) Benthic Broad Habitat Types (V. 2017) - the EUNIS 2019 habitat types - the Helcom Hub regional classification system habitat types - the Barcelona Convention classification system habitat types in the EUSeaMap (2023) broad-scale predictive habitat map. Values are 1 (Low confidence), 2 (Moderate confidence) or 3 (High confidence). The final habitat type is classified by overlaying several layers of information; these layers of information are collectively known as 'habitat descriptors'. Habitat descriptors differ per region but include: Biological zone Energy class Oxygen regime Salinity regime Seabed substrate Riverine input The confidence in the classification of the habitat type is taken as the minimum of the confidence in all of the relevant habitat descriptors at that location. Confidence values are also available for each habitat descriptor and input data layer. Detailed information on the modelling process is found in the EMODnet Seabed Habitats technical reports and appendices (links in Resources). Created by the EMODnet Seabed Habitats project consortium. Credit: Licensed under CC-BY 4.0 from the European Marine Observation and Data Network (EMODnet) Seabed Habitats initiative (www.emodnet-seabedhabitats.eu), funded by the European Commission. It is important to note that a habitat type confidence score is only relevant to that particular level of the classification system.

Output of the 2023 EUSeaMap broad-scale predictive model, produced by EMODnet Seabed Habitats. The extent of the mapped area includes the Mediterranean Sea, Black Sea, Baltic Sea, and areas of the North Eastern Atlantic extending from the Canary Islands in the south to the Barents Sea in the north. The map was produced using a "top-down" modelling approach using classified habitat descriptors to determine a final output habitat. Habitat descriptors differ per region but include: Biological zone Energy class Oxygen regime Salinity regime Seabed substrate Riverine input Habitat descriptors (excepting Substrate) are calculated using underlying physical data and thresholds derived from statistical analyses or expert judgement on known conditions. The model is produced using R and Arc Model Builder (10.1). The model was created using raster input layers with a cell size of 0.00104dd (roughly 100 metres). The model includes the sublittoral zone only; due to the high variability of the littoral zone, a lack of detailed substrate data and the resolution of the model, it is difficult to predict littoral habitats at this scale. EUSeaMap is classified into EUNIS 2019 level 3 (or more detailed levels where appropriate), EUNIS 2019 level 2 , EUNIS 2007-2011, the MSFD benthic broad habitat types, the HELCOM HUB classification in the Baltic, and the recently revised habitat classification in the Mediterranean. In the Black Sea, EUSeaMap is not classified into EUNIS 2007-2011 (due to inapplicability), but is classified according to a classification that was developed by EMODnet Seabed Habitats (Populus et a, 2017, and for a revised version Vasquez et al, 2020, See Online resources). Reports that provide methods used for the classification of the predicted habitats into the new 2019 EUNIS classification, regional classifications, and MSFD BBHT (v.2017) are linked in Online Resources. A report on the methods used in the 2023 version of EUSeaMap and reports on previous versions (v2019 and V2021) are linked in Online Resources. Credit: Licensed under CC-BY 4.0 from the European Marine Observation and Data Network (EMODnet) Seabed Habitats initiative (www.emodnet-seabedhabitats.eu), funded by the European Commission.

Confidence in the classification of the EUNIS 2007 habitat types in the EUSeaMap (2023) broad-scale predictive habitat map. Values are 1 (Low confidence), 2 (Moderate confidence) or 3 (High confidence). The final habitat type is classified by overlaying several layers of information; these layers of information are collectively known as 'habitat descriptors'. Habitat descriptors differ per region but include: Biological zone Energy class Oxygen regime Salinity regime Seabed substrate Riverine input The confidence in the classification of the habitat type is taken as the minimum of the confidence in all of the relevant habitat descriptors at that location. Confidence values are also available for each habitat descriptor and input data layer. Detailed information on the modelling process is found in the EMODnet Seabed Habitats technical reports and its appendices (links in Resources). Created by the EMODnet Seabed Habitats project consortium. Credit: Licensed under CC-BY 4.0 from the European Marine Observation and Data Network (EMODnet) Seabed Habitats initiative (www.emodnet-seabedhabitats.eu), funded by the European Commission. It is important to note that a habitat type confidence score is only relevant to that particular level of the classification system. For example, a cell of A3.1 high energy infralittoral rock with ÔÇÿlowÔÇÖ energy class confidence, ÔÇÿmoderateÔÇÖ biozone confidence and ÔÇÿhighÔÇÖ substrate type confidence would have an overall ÔÇÿlowÔÇÖ confidence. However, moving up the hierarchy to EUNIS level two (A3 infralittoral rock) removes the energy class; therefore, the confidence of the EUNIS level two habitat type would only consider the ÔÇÿmoderateÔÇÖ biozone confidence and ÔÇÿhighÔÇÖ substrate type confidence, resulting in an overall ÔÇÿmoderateÔÇÖ confidence.

Energy class layer produced by EMODnet Seabed Habitats as an input layer for the 2023 EUSeaMap broad-scale habitat model. The extent of the mapped area includes the Baltic Sea, and areas of the North Eastern Atlantic and Arctic extending from the Canary Islands in the south to Norway in the North. The map of energy classes was produced using underlying wave and current data and thresholds derived from statistical analyses or expert judgement on known conditions. This layer was updated in EUSeaMap 2023 using a new wave Kinetic energy at the seabed layer for the European Shelf area. A report on the methods used in the 2023 version of EUSeaMap and reports on previous versions (v2019 and V2021) are linked in Online Resources.

Confidence in the classification of energy level in the EUSeaMap (2023) broad-scale predictive habitat map. Values are on a range from 1 (Low confidence), 2 (Moderate confidence), 3 (High confidence). Energy level is one of the layers of information used to categorise physical habitat types in EUSeaMap; these layers of information are collectively known as 'habitat descriptors'. Confidence in the classification of an energy level at any location is driven by both the confidence in the values of the input variables, and the confidence in the classification based on proximity to, and uncertainty in, the boundary between classes (i.e. areas closer to a boundary between two classes will have lower confidence). Layers are also available showing confidence in the values of the input variables used to model energy levels (kinetic energy at the seabed and wave exposure). A report on the methods used in the 2023 version of EUSeaMap and reports on previous versions (v2019 and V2021) are linked in Online Resources. Created by the EMODnet Seabed Habitats project consortium.

Confidence in the classification of biological zones in the EUSeaMap (2023) broad-scale predictive habitat map. Values are on a range from 1 (Low confidence) to 3 (High confidence). Biological Zone is one of the layers of information used to categorise physical habitat types in EUSeaMap; these layers of information are collectively known as 'habitat descriptors'. Confidence in the classification of a Biological Zone at any location is driven by both the confidence in the values of the input variables, and the confidence in the classification based on proximity to, and uncertainty in, the boundary between classes (i.e. areas closer to a boundary between two classes will have lower confidence). Layers are also available showing confidence in the values of the input variables used to model Biological Zones. Detailed information on the modelling process is found in the EMODnet Seabed Habitats technical reports and its appendices (links in Resources). Created by the EMODnet Seabed Habitats project consortium.