Assessment of seismic reflectivity and lithology discrimination using 3D seismic data in a mature sedimentary basin.

Project Abstract

This study investigates the efficacy of seismic reflectivity attributes and advanced waveform analysis for discriminating lithologies within a mature sedimentary basin using 3D seismic data integrated with borehole information and petrophysical constraints. The primary aim is to develop a robust workflow that enhances lithology classification, improves reservoir delineation, and reduces uncertainty in characterizing siliciclastic and carbonate intervals commonly encountered in mature basins. The methodology combines pre-stack and post-stack seismic attribute analysis, spectral decomposition, and machine learning-based discriminant models trained on calibrated well logs and core data. We begin with meticulous data conditioning, including noise attenuation, de-ghosting, amplitude preservation, and time-to-depth conversion using a well tie-derived velocity model to ensure accurate seismic response in depth domain. Key seismic attributes explored include instantaneous phase and amplitude, Li(e)ghtness, cosine of instantaneous phase, kurtosis, GLCM-based texture metrics, spectral amplitudes across multiple frequency bands, AVO/AVA response, and guided wave indicators to capture porosity and mineralogical variations at multiple scales. A hierarchical lithology discrimination framework is developed, leveraging well-correlation panels to link seismic facies with lithology classes such as sandstones, shales, carbonates, and diagenetically altered units. The framework integrates rock physics modeling to relate seismic velocity, density, and elastic moduli to lithology, porosity, and fluid content, enabling more physically meaningful classifications. Machine learning classifiers (random forest, gradient boosting, and deep learning architectures) are trained on a multi-modal feature set that fuses low-frequency reservoir-scale information with high-frequency detail attributes. Cross-validation is performed with out-of-sample wells to assess generalization, and feature importance analysis identifies the most discriminative attributes for lithology separation. Uncertainty quantification is embedded through Bayesian posterior updates and ensemble methods to quantify confidence in lithology labels, particularly in regions with sparse well control. The workflow also incorporates semantic interpretation of seismic facies to connect reflectivity patterns with depositional environments and diagenetic overprints, thereby improving reservoir prediction and risk assessment. Results demonstrate improved lithology discrimination in both clastic and carbonate sequences, with notable gains in distinguishing lithic sandstones from tight carbonates and differentiating laminated shales from clay-rich siltstones. Case studies within the mature basin show enhanced reservoir connectivity assessment, more reliable pore throat estimation, and better prediction of overpressure zones and diagenetic alteration fronts. The study contributes a reproducible, scalable methodology for integrating 3D seismic data with lithology-oriented analytics, providing actionable insights for exploration strategy, field development planning, and reservoir management in mature basins. It also offers a framework adaptable to other basins with varying sedimentary architectures, enabling transferable best practices for seismic-based lithology discrimination and lithofacies mapping.

Project Overview

What This Project Is About

---

A straightforward, beginner-friendly look at how geologists study rocks beneath the earth using 3D seismic data. The project explores how reflections of sound waves help identify different rock types and their properties in a mature sedimentary basin.

---

---

The Problem It Addresses

---

In mature basins, rocks can look similar on the surface but differ in control of fluid, porosity, and rock type. Traditional methods may miss subtle differences. This project investigates how 3D seismic data can reveal these differences more clearly, aiding accurate lithology discrimination and basin understanding.

---

---

Objectives of the Project

---

1. Learn the basics of seismic waves and how 3D seismic surveys are collected.
2. Explain what lithology means and why it matters for reservoirs and groundwater.
3. Apply simple methods to distinguish rock types using seismic reflections.
4. Identify limitations and uncertainties in seismic interpretation.
5. Present a clear, practical workflow that a junior geologist could follow.

---

---

What You Will Do Step by Step

---

1) Read introductory materials on seismic data and lithology. 2)Review a small, publicly available 3D seismic dataset. 3)Learn basic interpretation tools and create a simple stratigraphic sketch. 4)Compare seismic signals with known rock types from nearby wells or literature. 5)Document assumptions and uncertainties. 6)Summarize findings in a short report with visuals.

---

---

Expected Outcome

A clear, approachable demonstration of how 3D seismic reflections can help distinguish rock types in a mature basin, with a simple workflow and common-sense caveats. The project should result in practical guidance for entry-level geologists and a foundational understanding suitable for further study.