Full Waveform Inversion (FWI) — A Complete Overview
Full Waveform Inversion (FWI) — A Complete Overview

Introduction
Full Waveform Inversion (FWI) is one of the most advanced technologies in seismic imaging. It uses the entire seismic wavefield — not just reflections — to estimate high‑resolution velocity models. FWI has transformed subsurface imaging in complex environments such as salt bodies, carbonates, and deepwater basins.
This article explains what FWI is, how it works, and why it’s becoming essential in modern seismic processing.
1. What Is Full Waveform Inversion?
FWI is a data‑driven inversion technique that matches recorded seismic data with synthetic data generated from a velocity model. The model is updated iteratively until the synthetic and recorded wavefields align.
FWI uses:
Reflections
Refractions
Diving waves
Surface waves
This makes it far more powerful than traditional tomography.
2. Why FWI Matters
✔ Ultra‑high‑resolution velocity models
FWI can resolve features as small as tens of meters.
✔ Better imaging beneath complex structures
Salt, basalt, and carbonates become clearer.
✔ Improved migration results
Better velocities = better images.
✔ Supports exploration and development
FWI reduces uncertainty in drilling decisions.
3. How FWI Works
A. Initial Model
FWI requires a reasonable starting velocity model.
B. Forward Modeling
Synthetic seismic data is generated using wave‑equation modeling.
C. Residual Calculation
Differences between synthetic and recorded data are computed.
D. Gradient Calculation
The model is updated using adjoint‑state methods.
E. Iteration
Steps repeat until convergence.
4. Types of FWI
Time‑Domain FWI
High accuracy, computationally heavy.
Frequency‑Domain FWI
Efficient for low frequencies.
Reflection FWI
Uses reflections when diving waves are absent.
Elastic FWI
Inverts for multiple elastic parameters.
5. Challenges
Cycle skipping
Noise sensitivity
Limited low frequencies
High computational cost
Conclusion
FWI is revolutionizing seismic imaging. Its ability to produce high‑resolution velocity models has made it a cornerstone of modern processing workflows.
