PhD Position in ML-Based Interpretation of Geophysical Legacy Data

Delft University of Technology is offering a four-year PhD position focused on transforming legacy petroleum exploration data into resources for deep mineral exploration. The candidate will combine quantitative geophysics, basin analysis, and machine learning to characterize the Kupferschiefer sediment-hosted copper system. Responsibilities include analyzing well logs, correlating seismic data, and developing robust multimodal machine learning models incorporating physical constraints. The researcher will join an interdisciplinary team within the Faculty of Civil Engineering and Geosciences as part of the EU-funded INTELLIGENT project. Successful applicants will collaborate with international academic and industry partners to build regional-scale mineral system models.

• Analyze and extrapolate well-log data to identify target horizons across multiple wells.
• Utilize seismic reflection data for cross-well correlation and spatial characterization.
• Investigate the roles of stratigraphy, faulting, and basin evolution using basin analysis techniques.
• Develop interpretable baseline machine learning models for well-log analysis.
• Build sequential and multimodal deep learning architectures incorporating physical and geological constraints.
• Integrate geological, geochemical, and mineralogical information to identify mineralization fingerprints.
• Contribute to the creation of a regional-scale mineral-system model and transferable exploration workflows.
• Collaborate with international academic partners, geological surveys, and industry stakeholders.