Degree
Doctor of Philosophy (PhD)
Department
Earth and Energy Sciences
Document Type
Dissertation
Abstract
Unconventional shale resources have become an important source of oil and gas in the United States, but their characterization remains challenging because of strong heterogeneity and complex elastic and geomechanical behavior. Important properties, such as total organic carbon (TOC) and brittleness, play a key role in evaluating source quality, reservoir behavior, development potential, and subsurface storage suitability. This dissertation develops artificial intelligence-based geophysical techniques to improve unconventional resource exploration and characterization by integrating seismic attributes, well-log data, deep learning, multicomponent seismic registration, and joint PP-PS inversion. A TOC estimation technique is developed and applied to both the Tuscaloosa Marine Shale (TMS) and the Bakken Shale to examine how seismic attributes relate to organic richness in two different unconventional systems, showing that TOC prediction is formation dependent and requires basin-specific relationships for reliable interpretation. In addition, a deep-learning-based workflow is developed to improve PP-PS registration, which is a major challenge in multicomponent seismic analysis because PP and PS data record the same geology at different times and with different wave behavior. The improved registration is then used in joint PP-PS inversion to estimate elastic properties and derive geomechanical attributes, including brittleness-related parameters, for Bakken shale characterization. Overall, this dissertation demonstrates that artificial intelligence-based geophysical workflows can improve TOC estimation, multicomponent seismic registration, and geomechanical characterization, and can provide a useful framework for unconventional shale exploration and subsurface storage assessment.
Date
2-6-2026
Recommended Citation
Hoque, S M Shamsul, "Development of Artificial Intelligence Based Geophysical Technique for Unconventional Resource Exploration" (2026). Doctoral Dissertations. 52.
https://scholarshub.louisiana.edu/dissertations/52
DOI
https://proquest.com/docview/3347815148
First Committee Chair
Rui Zhang
First Committee Member
Gabriele Morra
Second Committee Member
Jorge A. Villa
Third Committee Member
P. Io Ioannidi