Data for Synthetic Shelf Sediment Maps for the Greenland Sea and Barents Sea

This dataset represents the sediment properties and physical environment of the seabed for the Greenland Sea and Barents Sea shelf area. The data were produced at a spatial resolution of 0.01 by 0.01 degrees. Available variables include: whole sediment mean grain size, mud, sand and gravel percentag...

Full description

Bibliographic Details
Main Author: Laverick, Jack H.
Format: Dataset
Language:English
Published: NERC EDS UK Polar Data Centre 2022
Subjects:
"EARTH SCIENCE","OCEANS","MARINE SEDIMENTS","GEOTECHNICAL PROPERTIES"
"EARTH SCIENCE","OCEANS","MARINE ENVIRONMENT MONITORING"
Arctic
Barents Sea
Greenland Sea
Sediment maps
Sediment properties
Arctic Ocean
Greenland
Online Access:https://dx.doi.org/10.5285/fd971fc7-a730-4c68-9a02-76022e56ddab
https://data.bas.ac.uk/full-record.php?id=GB/NERC/BAS/PDC/01629
Description
Summary:This dataset represents the sediment properties and physical environment of the seabed for the Greenland Sea and Barents Sea shelf area. The data were produced at a spatial resolution of 0.01 by 0.01 degrees. Available variables include: whole sediment mean grain size, mud, sand and gravel percentages, rock cover, porosity and permeability, carbon and nitrogen content of sediments, depth, slope, roughness, terrain ruggedness index, topographic position index. The dataset also includes a seasonal cycle of monthly natural disturbance and bed shear stress. This dataset was produced by the MiMeMo project (NE/R012572/1), part of the Changing Arctic Ocean programme, jointly funded by the UKRI Natural Environment Research Council (NERC) and the German Federal Ministry of Education and Research (BMBF). : This dataset was produced by a random forest model trained on bathymetric properties and bed shear stress to predict sediment classes as defined by the Norwegian geological survey (NGU). Sediment classes were predicted for a larger area in the Barents Sea than covered by NGU, and also for the Greenland Sea. The discrete sediment classes were then decomposed into continuous variables before exploiting relationships with sediment grain size to produce fields of additional sediment properties. : The model was implemented in the R programming environment (v.4.0.2) using H2O (v3.32.0.1). : Average model accuracy was always > 92% in recreating Sediment classes. For more detailed assessment of model accuracy see the associated manuscript.

Similar Items