Joint inversion based on variation of information-a crustal model of Wilkes Land, East Antarctica

Mareen Lösing, Max Moorkamp, Jörg Ebbing

Research output: Contribution to journalArticlepeer-review

12 Citations (Scopus)

Abstract

By combining gravity and magnetic data in a joint inversion approach, 3-D information on the crustal structure of Wilkes Land, East Antarctica, is obtained and possible geological features become evident. Both data sets are combined through a coupling method which decreases the variation of information (VI) so data misfit and model dissimilarity are minimized simultaneously. In this manner, statistically compatible inversion results are obtained. The suitability of the method is demonstrated through a synthetic example using magnetic data and pseudo-gravity. Subsequently, we apply the method to gravity residuals and magnetic data and identify matching features of high magnitude density and susceptibility. Prominent structures in NW-SE direction along the edge of the Mawson craton and at the presumed Australo-Antarctic and Indo-Antarctic terrane boundaries are enhanced. Given the structural similarity between inverted susceptibility and density, and a strong indication of a parameter relationship, we suggest a clustering approach in order to differentiate distinct groups with similar parameter properties. The spatial distribution of these clusters reveals possible geological structures that agree with previous 2-D studies and rock measurements from the Indian and Australian continents. This shows that the VI joint inversion is a convenient approach for remote regions like East Antarctica with sparse geological samples.
Original languageEnglish
Pages (from-to)162-175
Number of pages14
JournalGeophysical Journal International
Volume232
Issue number1
DOIs
Publication statusPublished - 1 Jan 2023
Externally publishedYes

Fingerprint

Dive into the research topics of 'Joint inversion based on variation of information-a crustal model of Wilkes Land, East Antarctica'. Together they form a unique fingerprint.

Cite this