Integrating stratigraphic process in 3D subsurface modeling is crucial because it helps to generate realistic subsurface representation, leading to accurate forecasting of reservoir production, carbon capture, utilization and storage capacity, or fluid migration for geothermal energy production. For this purpose, a physics based stratigraphic forward modeler (SFM) was developed, allowing for creating the deposition and erosion at each time step. These models are used as training data for an advanced fast 3D modeler that can be also conditioned to the observed well data. Our previous efforts proved that generative adversarial networks (GAN) can be used to rapidly generate 2D realistic stratigraphic models conditioned to the known measurements. In this abstract, we extend such effort to work in the 3D domain using an advanced GAN solution that treats 3D stratigraphic models as videos.

@inproceedings{zhu2022rapid,
                    title={A Rapid and Realistic 3D Stratigraphic Model Generator Conditioned on Reference Well Log Data},
                    author={Zhu, L and Tilke, P and Nagendra, S and Etchebes, M and LeFranc, M},
                    booktitle={Second EAGE Digitalization Conference and Exhibition},
                    volume={2022},
                    number={1},
                    pages={1--5},
                    year={2022},
                    organization={European Association of Geoscientists \& Engineers}
                  }