3D Reservoir Modelling: Concepts, Workflow, and Applications

1. Introduction

3D reservoir modelling is a fundamental tool in modern petroleum exploration and development. It integrates geological, geophysical, petrophysical, and engineering data into a unified, three-dimensional representation of the subsurface. The goal is to reduce uncertainty, quantify hydrocarbons in place, assess field performance, and optimize development plans. As reservoirs become more complex and fields mature, 3D models provide a realistic framework for predicting fluid flow, guiding drilling decisions, and evaluating development scenarios.

---

2. Objectives of 3D Reservoir Modelling

• Characterize reservoir architecture (stratigraphy, facies, and structural features) in 3D.

• Quantify petrophysical properties such as porosity, permeability, water saturation.

• Estimate STOIIP and GIIP with uncertainty ranges.

• Guide well placement (producers, injectors, horizontal wells).

• Support dynamic simulation for forecasting reservoir performance.

• Assist in EOR planning by simulating fluid flow behaviour under various recovery methods.

• Integrate multidisciplinary data to ensure consistency and reduce uncertainty.

---

3. Key Data Sources

A robust reservoir model relies on multidisciplinary input, including:

3.1 Seismic Data

• Defines structural framework (faults, horizons).

• Provides seismic attributes for facies prediction and porosity trends.

• Time-to-depth conversion using velocity models.

3.2 Well Data

• Well logs (GR, density, neutron, sonic, resistivity).

• Core measurements (porosity, permeability, capillary pressure).

• Well tests and production history.

• Image logs for fracture identification.

3.3 Geological Data

• Stratigraphic correlations.

• Depositional environment interpretation.

• Facies types and reservoir geometry.

• Structural maps and tectonic history.

---

4. 3D Reservoir Modelling Workflow

4.1 Structural Modelling

The first step is constructing a 3D structural model, including:

• Faults (geometry, throw, sealing behavior).

• Reservoir horizons and surfaces.

• Grids (corner-point grid, pillar grids, or unstructured grids).

A realistic structural framework is essential because errors propagate into volumetrics and simulation.

---

4.2 Stratigraphic and Facies Modelling

Facies modelling captures reservoir heterogeneity and depositional architecture.
Common techniques:

• Object-based modelling (channels, lobes, bars).

• Pixel-based modelling (Sequential Indicator Simulation).

• Rule-based or process-based modelling for complex depositional systems.

Facies mapping is guided by:

• Core descriptions.

• Well log facies.

• Seismic attributes.

• Depositional models.

---

4.3 Petrophysical Modelling

Petrophysical properties (porosity, permeability, clay volume, water saturation) are distributed within the facies framework.

Techniques include:

• Sequential Gaussian Simulation (SGS).

• Co-kriging with seismic attributes.

• Machine-learning guided property distribution.

Trends are applied vertically or laterally to mimic geological processes.

---

4.4 Rock Physics Integration

Important for seismic-reservoir property linkage:

• P-wave and S-wave velocity predictions.

• Elastic property modelling.

• Fluids and pressure effects.

Rock physics constraints help validate seismic-driven property models.

---

4.5 Upscaling

The detailed geological model often contains millions of cells—too many for flow simulation. Upscaling reduces the grid size while retaining key reservoir behaviour.

Two main types:

• Static upscaling: porosity, permeability.

• Dynamic upscaling: transmissibility, relative permeability.

---

4.6 Dynamic Simulation

The upscaled model becomes the basis for reservoir simulation:

• History matching (aligning model with actual production).

• Forecasting field performance.

• Evaluating well placement and drilling scenarios.

• Running sensitivities for water/gas injection.

• Assessing EOR methods (polymer, miscible gas, ASP, WAG).

Simulation results help optimize field development plans.

---

5. Uncertainty and Sensitivity Analysis

Reservoir models inherently contain uncertainties due to data limitations. These are managed by:

• Generating multiple realizations.

• Varying structural scenarios (fault throw, depth uncertainty).

• Petrophysical property ranges.

• Different facies modeling assumptions.

Probabilistic estimates (P10/P50/P90) ensure robust decision-making.

---

6. Applications of 3D Reservoir Models

• Exploration and Appraisal

  • Volumetric estimation.

  • Risk assessment and ranking prospects.

• Field Development Planning

  • Optimal well spacing and trajectory design.

  • Injection strategy and facility sizing.

• Production Optimization

  • Identifying bypassed hydrocarbons.

  • Reservoir surveillance planning.

• Enhanced Oil Recovery (EOR)

  • Screening and designing EOR pilots.

  • Forecasting incremental recovery.

---

7. Challenges in 3D Reservoir Modelling

• Limited or sparse well data.

• Seismic resolution limitations.

• Complex fault networks or fracture systems.

• Heterogeneous carbonate reservoirs.

• Uncertainty in fluid contacts and pressure regimes.

• Time-consuming history match process.

---

8. Conclusion

3D reservoir modelling is a powerful tool that integrates geology, geophysics, petrophysics, and engineering into a coherent framework for understanding and managing hydrocarbon reservoirs. It reduces uncertainty, enhances prediction accuracy, and plays a critical role in decision-making during exploration, development, and production. Continuous advances in machine learning, seismic imaging, cloud computing, and geostatistics are pushing reservoir modelling toward higher fidelity and faster workflows.