A Framework for Modeling Subsurface Heterogeneity in Seismic Wave Propagation

Thesis Abstract

In the exploration of subsurface geological formations, the heterogeneity inherent in geological materials significantly influences seismic wave propagation, affecting the accuracy of seismic imaging and resource exploration. Despite advances in seismic modeling, existing frameworks often assume homogeneous or simplified subsurface conditions, limiting the precision of seismic interpretation in complex geological settings. This study aims to develop a comprehensive modeling framework that captures the variability of subsurface heterogeneity and improves the predictive capability of seismic wave behavior within variable geological media. The specific objectives include (1) analyzing the spatial statistical properties of subsurface heterogeneity through high-resolution borehole and seismic reflection data; (2) integrating these heterogeneity characterizations into a probabilistic seismic wave propagation model; and (3) validating the model against empirical data from controlled seismic surveys. The research adopts a mixed-methods approach, combining quantitative analysis of geophysical data and theoretical model development. The population comprises 20 subsurface profiles from seismic survey areas within sedimentary basin settings, with a sample size of 15 profiles selected through stratified random sampling to ensure representativeness of heterogeneous geological conditions. Data collection instruments include 3D seismic reflection datasets, borehole logs, and laboratory-measured physical properties of core samples. Quantitative techniques employed encompass spatial statistical analysis, variogram analysis, and geostatistical modeling to quantify heterogeneity patterns. The core methodological framework integrates stochastic simulations, such as geostatistical kriging and Monte Carlo methods, with physical modeling based on elastic wave equations, guided by the theoretical underpinnings of Biot’s theory of poroelasticity and the Effective Medium Theory. Data analysis involves multivariate regression to identify key heterogeneity parameters influencing seismic velocities and attenuation. The results inform the development of a layered probabilistic model that characterizes subsurface heterogeneity as a function of spatially variable parameters, thus enabling more realistic seismic wave simulations. The validation process utilizes seismic attribute analysis and comparison of modeled waveforms against observed data, employing root-mean-square error (RMSE) and correlation coefficients to measure model fidelity. Anticipated findings include detailed quantification of heterogeneity parameters that significantly impact seismic wave behavior, along with a validated modeling framework capable of integrating heterogeneity characteristics into seismic interpretation workflows. This study contributes novel insights into the spatial variability of subsurface geological properties and introduces an integrated probabilistic modeling approach to seismic wave propagation that accounts for heterogeneity at various scales. The framework offers a significant advancement over traditional homogeneous assumptions, providing geophysicists with a robust tool for more accurate seismic imaging and resource evaluation in complex geological environments. The findings are expected to influence seismic data processing protocols and geological interpretations, particularly in sedimentary basins with complex stratigraphy. Concluding, the research underscores the importance of detailed heterogeneity modeling in seismic analysis and advocates for the operational integration of the developed framework in standard geophysical workflows. Recommendations include further refinement of the probabilistic model through larger datasets and the incorporation of anisotropic heterogeneity features. Future research directions suggest extending the framework to include temporal variations in subsurface conditions and adapting it for use in marine and unconventional resource contexts. Ultimately, this study offers a significant step toward advancing seismic modeling methodologies, enabling more accurate subsurface characterization and informed decision-making in geological and resource exploration endeavors.