The data supplements the seabed susbtrate data produced by GTK with areas for which no actual survey data is available. The intention has been to determine the hard and soft seabed areas for Finland’s marine areas.

The data has been produced for supporting decision making, research and broad-scale marine spatial planning.

The data has been produced to a scale of 1:250 000, providing only general information about the occurrence of hard and soft seabed areas and therefore is not suitable for more precise scale inspection or use.

The data concerning areas restricted by the Finnish Defence Forces has been removed.

The organizations participating in the The Finnish Inventory Programme for the Underwater Marine Environment, VELMU.

The data has been produced using the ESRI ArcGIS environment.

The data has been modified from the marine geological seabed data produced by GTK to a scale of 1:20 000 and on the basis of the substrate observations made by SYKE and Metsähallitus. The GTK survey programme has gathered marine geological data since the 1970s. The quality and accuracy of the data has improved over the years thanks to improvements in methods and the introduction of global positioning technology.

This data on the soft and hard seabed areas was produced according to marine area, as Finland’s marine areas differ from one another in terms of geological features. The survey used the HELCOM 2013 division of marine areas (Bothnian Bay, Kvarken, Bothnian Sea, Archipelago Sea, Gulf of Finland) with the Gulf of Finland further divided into four sub-areas; according to GTK data, the Hanko Peninsula, Raseborg-Kirkkonummi, Greater Helsinki and Eastern Gulf of Finland differ from one another by their geological features.

The GTK marine geological data is classed into hard and soft seabed areas according to the classification used in the EMODnet Geology project. In addition, the original region-specific map data was modified into points with a frequency of either 500 metres or 200 metres depending on the area. The survey also utilised the seabed substrate observations of the Hertta Database. Furthermore, the video observations conducted by Metsähallitus on Kvarken and the Bothnian Bay were used to improve regional coverage. The seabed openness model and depth model, including its affiliated models used in the VELMU project were utilised as environmental variables.

The extents of hard and soft seabed areas for each environment variable combination were statistically checked per marine area. The area was primarily classified according to the dominant substrate type, i.e. if over 50% of environment variable combination observations were seen to be hard, all areas with similar combinations were classified as hard. Diverging from the aforementioned, statistical modelling, logical regression was utilised in the sand dominated areas of the Bothnian Bay.

The models for the hard and soft seabed areas produced for Finland’s different marine areas have been combined and the boundary areas for individual datasets have been inspected and manually edited wherever required. The data is generalized at a general scale of 1:250 000 and the areas with a size less than 0.3 square kilometres have been removed. The data has been supplemented with the substrate type data classed into hard and soft seabed areas produced in the EMODnet project for areas not surveyed by GTK. In addition, particularly with respect to open sea areas, for which very few substrate observations have been conducted, the data has been manually edited using the depth model and existing raw (1:1 000 000) data. The data has also been proportioned for the available substrate type data produced by Sweden, Estonia and Russia (EMODnet Geology Marine Geological Data). The data concerning areas restricted by the Finnish Defence Forces has been removed.

This data on the soft and hard seabed areas was produced according to marine area, as Finland’s marine areas differ from one another in terms of geological features. The survey used the HELCOM 2013 division of marine areas (Bothnian Bay, Kvarken, Bothnian Sea, Archipelago Sea, Gulf of Finland) with the Gulf of Finland further divided into four sub-areas; according to GTK data, the Hanko Peninsula, Raseborg-Kirkkonummi, Greater Helsinki and Eastern Gulf of Finland differ from one another by their geological features.

The GTK marine geological data is classed into hard and soft seabed areas according to the classification used in the EMODnet Geology project. In addition, the original region-specific map data was modified into points with a frequency of either 500 metres or 200 metres depending on the area. The survey also utilised the seabed substrate observations of the Hertta Database. Furthermore, the video observations conducted by Metsähallitus on Kvarken and the Bothnian Bay were used to improve regional coverage. The seabed openness model and depth model, including its affiliated models used in the VELMU project were utilised as environmental variables.

The extents of hard and soft seabed areas for each environment variable combination were statistically checked per marine area. The area was primarily classified according to the dominant substrate type, i.e. if over 50% of environment variable combination observations were seen to be hard, all areas with similar combinations were classified as hard. Diverging from the aforementioned, statistical modelling, logical regression was utilised in the sand dominated areas of the Bothnian Bay.

The models for the hard and soft seabed areas produced for Finland’s different marine areas have been combined and the boundary areas for individual datasets have been inspected and manually edited wherever required. The data is generalized at a general scale of 1:250 000 and the areas with a size less than 0.3 square kilometres have been removed. The data has been supplemented with the substrate type data classed into hard and soft seabed areas produced in the EMODnet project for areas not surveyed by GTK. In addition, particularly with respect to open sea areas, for which very few substrate observations have been conducted, the data has been manually edited using the depth model and existing raw (1:1 000 000) data. The data has also been proportioned for the available substrate type data produced by Sweden, Estonia and Russia (EMODnet Geology Marine Geological Data). The data concerning areas restricted by the Finnish Defence Forces has been removed.

Aineisto