Sub3DNet1.0: a deep-learning model for regional-scale 3D subsurface structure mapping

Jiang, Zhenjiao; Mallants, Dirk; Gao, Lei; Munday, Tim; Mariethoz, Gregoire; Peeters, Luk

This study introduces an efficient deep-learning model based on convolutional neural networks with joint autoencoder and adversarial structures for 3D subsurface mapping from 2D surface observations. The method was applied to delineate paleovalleys in an Australian desert landscape. The neural network was trained on a 6400 inline-formulakm2 domain by using a land surface topography as 2D input and an airborne electromagnetic (AEM)-derived probability map of paleovalley presence as 3D output. The trained neural network has a squared error inline-formula<0.10 across 99 inline-formula% of the training domain and produces a squared error inline-formula<0.10 across 93 inline-formula% of the validation domain, demonstrating that it is reliable in reconstructing 3D paleovalley patterns beyond the training area. Due to its generic structure, the neural network structure designed in this study and the training algorithm have broad application potential to construct 3D geological features (e.g., ore bodies, aquifer) from 2D land surface observations.

Zitieren

Zitierform:

Jiang, Zhenjiao / Mallants, Dirk / Gao, Lei / et al: Sub3DNet1.0: a deep-learning model for regional-scale 3D subsurface structure mapping. 2021. Copernicus Publications.

Zugriffsstatistik

Gesamt:
Volltextzugriffe:
Metadatenansicht:
12 Monate:
Volltextzugriffe:
Metadatenansicht:

Grafik öffnen

Rechte

Rechteinhaber: Zhenjiao Jiang et al.

Nutzung und Vervielfältigung:

Export