Predictive modelling of grain-size distributions from marine electromagnetic profiling data using end-member analysis and a radial basis function network

View/ Open
Date
2018-04-17Author
Baasch, Benjamin
Concept link
Müller, Hendrik
Concept link
von Dobeneck, Tilo
Concept link
Metadata
Show full item recordCitable URI
https://hdl.handle.net/1912/10701As published
https://doi.org/10.1093/gji/ggy152DOI
10.1093/gji/ggy152Keyword
Neural networks; Fuzzy logic; Statistical methods; Electrical properties; Magnetic properties; Marine electromagnetics; Controlled source electromagnetics (CSEM)Abstract
In this work, we present a new methodology to predict grain-size distributions from geophysical data. Specifically, electric conductivity and magnetic susceptibility of seafloor sediments recovered from electromagnetic profiling data are used to predict grain-size distributions along shelf-wide survey lines. Field data from the NW Iberian shelf are investigated and reveal a strong relation between the electromagnetic properties and grain-size distribution. The here presented workflow combines unsupervised and supervised machine-learning techniques. Non-negative matrix factorization is used to determine grain-size end-members from sediment surface samples. Four end-members were found, which well represent the variety of sediments in the study area. A radial basis function network modified for prediction of compositional data is then used to estimate the abundances of these end-members from the electromagnetic properties. The end-members together with their predicted abundances are finally back transformed to grain-size distributions. A minimum spatial variation constraint is implemented in the training of the network to avoid overfitting and to respect the spatial distribution of sediment patterns. The predicted models are tested via leave-one-out cross-validation revealing high prediction accuracy with coefficients of determination (R2) between 0.76 and 0.89. The predicted grain-size distributions represent the well-known sediment facies and patterns on the NW Iberian shelf and provide new insights into their distribution, transition and dynamics. This study suggests that electromagnetic benthic profiling in combination with machine learning techniques is a powerful tool to estimate grain-size distribution of marine sediments.
Description
Author Posting. © The Authors, 2018. This article is posted here by permission of The Royal Astronomical Society for personal use, not for redistribution. The definitive version was published in Geophysical Journal International 215 (2018): 460–473, doi:10.1093/gji/ggy152.
Collections
Suggested Citation
Geophysical Journal International 215 (2018): 460–473Related items
Showing items related by title, author, creator and subject.
-
Parametric study of the physical properties of hydrate-bearing sand, silt, and clay sediments : 1. Electromagnetic properties
Lee, J. Y.; Santamarina, J. Carlos; Ruppel, Carolyn D. (American Geophysical Union, 2010-11-09)The marked decrease in bulk electrical conductivity of sediments in the presence of gas hydrates has been used to interpret borehole electrical resistivity logs and, to a lesser extent, the results of controlled source ... -
Regeneration in the era of functional genomics and gene network analysis
Smith, Joel; Morgan, Jennifer R.; Zottoli, Steven J.; Smith, Peter J. S.; Buxbaum, Joseph D.; Bloom, Ona E. (Marine Biological Laboratory, 2011-08)What gives an organism the ability to regrow tissues and to recover function where another organism fails is the central problem of regenerative biology. The challenge is to describe the mechanisms of regeneration at the ... -
Electromagnetic flow sensors
Lawson, Kenneth; Kanwisher, John W. (Woods Hole Oceanographic Institution, 1974-03)Flow sensors based on the principle of electromagnetic induction were investigated as alternatives to commonly used mechanical devices utilizing rotors and propellers. Prototype sensors were constructed showing ...