Mark J. Dekkers, Author at Eos

Example of the convolutional neural network (CNN) approach from an area in the Black Hill norite

Machine Learning for Magnetics

by Mark J. Dekkers 21 October 202020 December 2021

Classic interpretation of aeromagnetic anomaly maps involves several steps with limiting boundary conditions; a recent study develops convolutional networks largely bypassing these issues.

Plot showing two known endmembers fitted with several magnetic components

Editors' Highlights

Unmixing Magnetic Components – An Experimental Twist

by Mark J. Dekkers 19 October 202021 October 2021

Various unmixing approaches are used in environmental magnetism, each starting from a different premise; now they are put to the test by scrutinizing experimental mixtures of known endmembers.

Graph, based on a dataset of paleomagnetic data from a Swedish lake binned in 150-year intervals, showing the classical uncertainty measure plotted against the new proposed uncertainty measure that takes propagated measurement uncertainty into account.

Editors' Highlights

Uncertainty Propagation in Paleomagnetic Data Quantified

by Mark J. Dekkers 29 July 202030 September 2022

In classic paleomagnetic data processing uncertainties are calculated at a single level only, but there is now a more lucid way to include error propagation.

Editors' Highlights

Extraordinary Polar Wander During the Late Jurassic?

by Mark J. Dekkers 22 April 201927 January 2023

Small amounts of polar wander have occurred during geological history, but whether larger amounts occurred is still controversial. Did a truly large polar wander event really happen?

Features from AGU Journals