Geophysics

New submissions

• New submissions

New submissions for Fri, 31 May 24

[1]  arXiv:2405.19767 [pdf, ps, other]

Title: MAE-GAN: A Novel Strategy for Simultaneous Super-resolution Reconstruction and Denoising of Post-stack Seismic Profile

Authors: Wenshuo Yu, Shiqi Dong, Shaoping Lu, Xintong Dong

Subjects: Geophysics (physics.geo-ph)

Post-stack seismic profiles are images reflecting containing geological structures which provides a critical foundation for understanding the distribution of oil and gas resources. However, due to the limitations of seismic acquisition equipment and data collecting geometry, the post-stack profiles suffer from low resolution and strong noise issues, which severely affects subsequent seismic interpretation. To better enhance the spatial resolution and signal-to-noise ratio of post-seismic profiles, a multi-scale attention encoder-decoder network based on generative adversarial network (MAE-GAN) is proposed. This method improves the resolution of post-stack profiles, and effectively suppresses noises and recovers weak signals as well. A multi-scale residual module is proposed to extract geological features under different receptive fields. At the same time, an attention module is designed to further guide the network to focus on important feature information. Additionally, to better recover the global and local information of post-stack profiles, an adversarial network based on a Markov discriminator is proposed. Finally, by introducing an edge information preservation loss function, the conventional loss function of the Generative Adversarial Network is improved, which enables better recovery of the edge information of the original post-stack profiles. Experimental results on simulated and field post-stack profiles demonstrate that the proposed MAE-GAN method outperforms two advanced convolutional neural network-based methods in noise suppression and weak signal recovery. Furthermore, the profiles reconstructed by the MAE-GAN method preserve more geological structures.

[2]  arXiv:2405.20144 [pdf, other]

Title: Flowy: High performance probabilistic lava emplacement prediction

Authors: Moritz Sallermann, Amrita Goswami, Alejandro Peña-Torres, Rohit Goswami

Subjects: Geophysics (physics.geo-ph); Computational Physics (physics.comp-ph)

Lava emplacement is a complex physical phenomenon, affected by several factors. These include, but are not limited to features of the terrain, the lava settling process, the effusion rate or total erupted volume, and the probability of effusion from different locations. One method, which has been successfully employed to predict lava flow emplacement and forecast the inundated area and final lava thickness, is the MrLavaLoba method from Vitturi et al. The MrLavaLoba method is implemented in their code of the same name. Here, we introduce Flowy, a new computational tool that implements the MrLavaLoba method of Vitturi et al. in a more efficient manner. New fast algorithms have been incorporated for all performance critical code paths, resulting in a complete overhaul of the implementation. When compared to the MrLavaLoba code, Flowy exhibits a significant reduction in runtime - between 100 to 400 times faster - depending on the specific input parameters. The accuracy and the probabilistic convergence of the model outputs are not compromised, maintaining high fidelity in generating possible lava flow paths and deposition characteristics. We have validated Flowy's performance and reliability through comprehensive unit-testing and a real-world eruption scenario. The source code is freely available on GitHub, facilitating transparency, reproducibility and collaboration within the geoscientific community.

• New submissions