ENVIRONMENTAL IMPACTS OF CLOSURE PRACTICES IN MINING AND OIL EXPLORATION: STRATEGIES FOR REHABILITATION AND SUSTAINABLE DEVELOPMENT

Abstract

Geophysical monitoring of geological carbon dioxide storage sites (CCS) represents an indispensable component for validating long-term safety and permanence. However, mathematical modeling of the seismic response to CO₂ injection faces challenges that fundamentally exceed the conventional frameworks of exploration geophysics. This paper demonstrates that accurate modeling requires abandoning simple acoustic or elastic models in favor of advanced poro-viscoelastic frameworks integrated within fully coupled Thermo-Hydro-Mechanical-Chemical (THMC) simulations. While the canonical Gassmann equation serves as an industry standard, it systematically fails to predict the seismic response to 〖"CO" 〗_2 due to its omission of rock-frame weakening, chemical dissolution, and wave-induced fluid flow (WIFF). To ensure rigorous operational execution, this revised study explicitly details the numerical solver architectures, grid resolutions, boundary conditions, and advanced uncertainty quantification (UQ) workflows—utilizing physics-informed neural networks (PINNs) and surrogate-assisted Monte Carlo simulations—to manage high degrees of freedom, reduce prediction uncertainty bounds, and calibrate multiphysics parameters from limited field data