Articles | Volume 15, issue 17
Development and technical paper
08 Sep 2022
Development and technical paper |  | 08 Sep 2022

DeepISMNet: three-dimensional implicit structural modeling with convolutional neural network

Zhengfa Bi, Xinming Wu, Zhaoliang Li, Dekuan Chang, and Xueshan Yong

Related authors

ClinoformNet-1.0: stratigraphic forward modeling and deep learning for seismic clinoform delineation
Hui Gao, Xinming Wu, Jinyu Zhang, Xiaoming Sun, and Zhengfa Bi
Geosci. Model Dev., 16, 2495–2513,,, 2023
Short summary

Related subject area

Solid Earth
Simulation of a fully coupled 3D glacial isostatic adjustment – ice sheet model for the Antarctic ice sheet over a glacial cycle
Caroline J. van Calcar, Roderik S. W. van de Wal, Bas Blank, Bas de Boer, and Wouter van der Wal
Geosci. Model Dev., 16, 5473–5492,,, 2023
Short summary
AdaHRBF v1.0: gradient-adaptive Hermite–Birkhoff radial basis function interpolants for three-dimensional stratigraphic implicit modeling
Baoyi Zhang, Linze Du, Umair Khan, Yongqiang Tong, Lifang Wang, and Hao Deng
Geosci. Model Dev., 16, 3651–3674,,, 2023
Short summary
PySubdiv 1.0: open-source geological modeling and reconstruction by non-manifold subdivision surfaces
Mohammad Moulaeifard, Simon Bernard, and Florian Wellmann
Geosci. Model Dev., 16, 3565–3579,,, 2023
Short summary
Reconstructing tephra fall deposits via ensemble-based data assimilation techniques
Leonardo Mingari, Antonio Costa, Giovanni Macedonio, and Arnau Folch
Geosci. Model Dev., 16, 3459–3478,,, 2023
Short summary
IMEX_SfloW2D v2: a depth-averaged numerical flow model for volcanic gas-particle flows over complex topographies and water
Mattia de' Michieli Vitturi, Tomaso Esposti Ongaro, and Samantha Engwell
Geosci. Model Dev. Discuss.,,, 2023
Revised manuscript accepted for GMD
Short summary

Cited articles

Alon, U. and Yahav, E.: On the bottleneck of graph neural networks and its practical implications, arXiv [preprint],, 9 June 2020. a
Bi, Z., Wu, X., Geng, Z., and Li, H.: Deep relative geologic time: a deep learning method for simultaneously interpreting 3-D seismic horizons and faults, J. Geophys. Res.-Sol. Ea., 126, e2021JB021882,, 2021. a
Bi, Z., Wu, X., Li, Z., Chang, D., and Yong, X.: Training and validation datasets for “Three-Dimensional Implicit Structural Modeling Using Convolutional Neural Network”, Zenodo [data set],, 2022a. a
Bi, Z., Wu, X., Li, Z., Chang, D., and Yong, X.: zfbi/DeepISMNet: DeepISMNet: Three-Dimensional Implicit Structural Modeling with Convolutional Neural Network, Zenodo [code],, 2022b. a
Calcagno, P., Chilès, J.-P., Courrioux, G., and Guillen, A.: Geological modelling from field data and geological knowledge: Part I. Modelling method coupling 3D potential-field interpolation and geological rules, Phys. Earth Planet. In., 171, 147–157, 2008. a, b, c
Short summary
We present an implicit modeling method based on deep learning to produce a geologically valid and structurally compatible model from unevenly sampled structural data. Trained with automatically generated synthetic data with realistic features, our network can efficiently model geological structures without the need to solve large systems of mathematical equations, opening new opportunities for further leveraging deep learning to improve modeling capacity in many Earth science applications.