Physically grounded model improves seismic and well data matching
Physical Cross-Modal Masked Autoencoding for Seismic-to-Well Representation Learning
Machine Learning
Summary
Seismic surveys give broad, low-detail pictures of underground formations, while wells provide detailed but very localized data. The authors developed a new computer model that learns to connect these very different types of data by looking at both together, improving predictions about one from the other. Their approach uses a special way to represent location and focuses on reconstructing missing data from one type using clues from the other. They tested this on extensive U.S. seismic and well data, showing the model can better predict well measurements from seismic data and capture geological structures like salt formations, even in new regions.
What this means in practice
- •For oil and gas exploration teams: Predict high-resolution well measurements from seismic data to improve subsurface characterization and reduce drilling uncertainty.
- •For geophysical data processing groups: Enhance seismic data reconstruction and interpretation by integrating sparse well log information to better delineate geological features.
Authors
Meher Gajula, Keyla Gonzalez, Ben Lasscock, Alejandro Valenciano
Abstract
Learning from scientific measurements often requires aligning modalities with different spatial support and resolution. Subsurface characterization is an extreme dense-sparse problem in which 3D seismic provides volumetric but indirect measurements and well logs provide high-resolution 1D measurements at sparse borehole locations. We introduce a physically grounded cross-modal masked autoencoder (CM-MAE) for seismic-to-well representation learning. The model jointly tokenizes seismic volumes and well-log depth patches, embeds both modalities in continuous physical coordinates using four-axis rotary position embeddings, and reconstructs masked targets with a cross-modal decoder. Sparse Mixture-of-Experts layers provide modality-specific capacity while dense attention allows information exchange between modalities. Pretraining uses 23 seismic volumes covering approximately 178,000 square km across U.S. offshore and onshore basins, together with approximately 92,000 wells. Matched-mask ablations show strongly asymmetric information flow. Seismic context improves held-out well-log reconstruction by 9.73%, while well-log context improves seismic reconstruction by 0.65%. We evaluate seismic-only pseudo-log predictions against independent interpreter-drawn salt-geobody masks to determine whether they contain geologic signal. Across offshore U.S. surveys, compressional-slowness-derived salt scores reach an AUROC of 0.910. In onshore basins, predicted compressional slowness preserves formation-scale structure and partial relative organization in an unseen survey despite calibration drift. CM-MAE learns useful seismic-to-well representations under extreme modality asymmetry, although absolute pseudo-log calibration remains survey-dependent.