Estimating subsurface CO2 leakage pathways using 3D seismic attributes and probabilistic joint inversion under uncertain geostatistical priors
Table Of Contents
Chapter ONE
INTRODUCTION
- 1.1Introduction
- 1.2Background of Study
- 1.3Problem Statement
- 1.4Objective of Study
- 1.5Limitation of Study
- 1.6Scope of Study
- 1.7Significance of Study
- 1.8Structure of the Research
- 1.9Definition of Terms
Chapter TWO
LITERATURE REVIEW
- 2.1Review of Geophysical Methods for Subsurface Characterization
- 2.2Seismic Attributes and Their Applications in Reservoir Monitoring
- 2.3Concepts in CO2 Sequestration and Leakage Scenarios
- 2.4Inversion Techniques in Geophysics: Deterministic and Probabilistic Approaches
- 2.5Geostatistical Modeling and Uncertainty Quantification
- 2.63D Seismic Data Acquisition and Processing Advances
- 2.7Multidimensional Joint Inversion Frameworks
- 2.8Machine Learning and Pattern Recognition in Geophysics
- 2.9Case Studies of CO2 Geological Storage Sites
- 2.10Gaps and Future Directions in Subsurface Leakage Research
Chapter THREE
RESEARCH METHODOLOGY
- 3.1Research Design and Philosophy
- 3.2Data Requirements and Sources
- 3.3Data Preprocessing and Quality Control
- 3.4Seismic Attribute Extraction and Calibration
- 3.53D Seismic Inversion Methodology
- 3.6Probabilistic Joint Inversion Framework
- 3.7Geostatistical Priors and Uncertainty Modeling
- 3.8Model Validation and Uncertainty Quantification
- 3.9Simulation of CO2 Leakage Pathways
- 3.10Software, Tools, and Computational Resources
- 3.11Ethical Considerations and Reproducibility
Chapter FOUR
DATA PRESENTATION AND ANALYSIS
- 4.1Study Area and Geological Setting
- 4.2Data Presentation and Processing Workflow
- 4.3Seismic Attribute Analysis Results
- 4.4Inversion Results and Parameter Estimation
- 4.5Uncertainty Assessment and Sensitivity Analysis
- 4.6Visualization of Leakage Pathways in 3D
- 4.7Comparison with Traditional Methods
- 4.8Implications for CO2 Storage Security
Chapter FIVE
SUMMARY, CONCLUSION AND RECOMMENDATIONS
- 5.1Summary of Findings
- 5.2Theoretical and Practical Implications
- 5.3Limitations and Recommendations for Future Research
- 5.4Conclusions
- 5.5Potential Applications and Policy Relevance
Project Abstract
Accurate characterization of subsurface CO2 leakage pathways is critical for assessing the integrity of geological CO2 storage sites and mitigating environmental risks. This study presents an integrative workflow that combines 3D seismic attribute analysis with probabilistic joint inversion to detect, delineate, and quantify potential leakage conduits in heterogeneous reservoir-caprock systems under uncertain geostatistical priors. We begin by deriving a suite of seismic attributes sensitive to fluid substitution effects, microfracturing, and caprock integrity, including amplitude variation with offset (AVO), spectral attributes, impedance contrasts, and P- and S-wave velocity proxies. These attributes are mapped to petrophysical properties using data-driven calibration anchored by available well logs, dynamic baseline models, and rock-physics relationships that account for CO2 saturation, salinity changes, and shale content. A Bayesian joint inversion framework is then developed to fuse multi-attribute seismic information with dynamic monitoring data, such as time-lapse (4D) seismic, microseismic events, and wellbore pressure observations, while explicitly incorporating geostatistical priors that capture spatial continuity, anisotropy, and uncertainty in reservoir properties. The forward model employs a coupled flow-geomechanics simulator to generate CO2 migration scenarios and their seismic responses, enabling likelihood computations that reflect measurement noise, model inadequacy, and prior uncertainty. We implement hierarchical priors to quantify uncertainty at multiple scales, from facies-level lithologies to fracture networks, allowing the inversion to prefer physically plausible leakage pathways without overfitting sparse data. The methodology is tested on synthetic benchmarks with realistic heterogeneity and on a field-scale dataset from a CO2 storage pilot, where we demonstrate the ability to recover subtle leakage channels that are corroborated by independent geophysical and geochemical indicators. Key contributions include (1) a robust, uncertainty-aware attribute-based screening workflow for leakage indicators, (2) a probabilistic joint inversion scheme that seamlessly integrates seismic, time-lapse, and wellbore data under uncertain priors, (3) a scalable, parallelizable computational strategy leveraging surrogate modeling and progressive unfolding to reduce runtime while preserving accuracy, and (4) practical guidelines for monitoring program design, including data acquisition strategies, priors selection, and decision thresholds for risk-informed management. Results indicate that joint inversion substantially improves leakage pathway localization and saturation estimates compared to single-dataset interpretations, particularly in complex faulted and layered sequences where CO2 migration is channeled through high-permeability conduits. The framework also provides quantitative uncertainty bounds essential for risk assessment, operational decision-making, and regulatory compliance in CO2 storage projects. This work advances geophysical monitoring by delivering an integrated, uncertainty-conscious toolkit for proactively identifying and characterizing subsurface CO2 leakage pathways.
Project Overview
What This Project Is About
This project looks at how scientists can map hidden channels where carbon dioxide might escape from storage sites underground. It uses three-dimensional seismic data, which are like a 3D map of the underground, and combines different clues to spot possible leakage paths. The goal is to create a clear, practical method to identify and understand how CO2 could move through rocks over time, even when data is imperfect or uncertain.
The Problem It Addresses
Leakage from CO2 storage sites is a major safety and climate concern. Data that show underground conditions are often incomplete or noisy, making it hard to be sure where leakage could occur. This project tackles how to better infer leakage paths when information is uncertain, by using advanced data integration and probabilistic methods.
Objectives of the Project
- Explain the concept of CO2 storage and why monitoring leakage matters.
- Introduce 3D seismic data and how it helps visualize underground features.
- Describe probabilistic joint inversion in simple terms and its role in combining clues.
- Develop a step-by-step workflow to identify leakage pathways from data.
- Apply the workflow to synthetic and real data examples for demonstration.
What You Will Do Step by Step
Step 1: Learn the basic ideas behind seismic data and why uncertainty matters. Step 2: Gather and preprocess 3D seismic data from a test site. Step 3: Define possible leakage paths using simple geological ideas. Step 4: Combine different data clues with probability to rank likely pathways. Step 5: Test the method on sample data and compare results with known outcomes. Step 6: Interpret results and discuss practical implications for monitoring. Step 7: Write a clear summary of methods and findings.
Expected Outcome
The project should deliver a straightforward method to identify potential CO2 leakage paths using 3D seismic data and probabilistic thinking. The outcomes include a user-friendly workflow, simple visuals showing leakage likelihoods, and guidance on how to apply the approach in real monitoring programs.