In recent years, deep learning (DL) has emerged as a promising alternative approach for various seismic processing tasks, including primary estimation (or multiple elimination), a crucial step for accurate subsurface imaging. In geophysics, DL methods are commonly based on supervised learning from large amounts of high-quality labelled data. Instead of relying on traditional supervised learning, in the context of free-surface multiple elimination, we propose a method in which the DL model learns to effectively parameterize the free-surface multiple-free wavefield from the full wavefield by incorporating the underlying physics into the loss computation. This, in turn, yields high-quality estimates without ever being shown any ‘ground truth’ data. Currently, the network reparameterization is performed independently for each dataset. We demonstrate its effectiveness through tests on both synthetic and field data. We employ industry-standard Surface-Related Multiple Elimination (SRME) using, respectively, global least-squares adaptive subtraction and local least-squares adaptive subtraction as benchmarks. The comparison shows that the proposed method outperforms the benchmarks in estimation accuracy, achieving the most complete primary estimation and the least multiple energy leakage, but at the cost of a higher computational burden.

Seismic angle gathers and spectral seismic attributes offer complementary insights to improve understanding of complex subsurface characteristics. However, the labor-intensive process of subsurface characterization, data annotation, limited labeled data, and subsurface complexity make it difficult to leverage these insights via supervised learning approaches.

To overcome such challenges and benefit from the strength of spectral seismic attributes, this study introduces a novel hierarchical Self-Organizing Map (SOM) framework to integrate spectral seismic attributes like scalograms and spectrograms (joint time-frequency analyses) extracted from angle gathers.

In our current research, firstly, we trained individual SOMs, as an unsupervised pattern recognition algorithm on reflectivity images, angle-gathers, and the spectral seismic attributes extracted from angle-dependent data. Secondly, we deploy a hierarchical SOM network to combine and analyze all these datasets. Thirdly, we evaluate the hierarchical approach and standalone analyses of clustering quality and information content using the binary boundary maps and the performance metrics. Our findings indicated that, the scalogram-based hierarchical SOM, containing information of different angles, achieves the lowest Quantization Error and Davis-Bouldin Index, indicating optimal feature representation and well-separated clusters. The findings stress the potential of hierarchical networks and joint time-frequency analyses from angle gathers for robust seismic interpretation workflows.

Downward continuation is a critical task in potential field processing, including gravity and magnetic fields, which aims to transfer data from one observation surface to another that is closer to the source of the field. Its effectiveness directly impacts the success of detecting and highlighting subsurface anomalous sources. We treat downward continuation as an inverse problem that relies on solving a forward problem defined by the formula for upward continuation, and we propose a new physics-trained deep neural network (DNN)-based solution for this task. We hard-code the upward continuation process into the DNN’s learning framework, where the DNN itself learns to act as the inverse problem solver and can perform downward continuation without ever being shown any ground truth data. We test the proposed method on both synthetic magnetic data and real-world magnetic data from West Antarctica. The preliminary results demonstrate its effectiveness through comparison with selected benchmarks, opening future avenues for the combined use of DNNs and established geophysical theories to address broader potential field inverse problems, such as density and geometry modelling.