|
<< Click to Display Table of Contents >> Navigation: Migration > 2D PostSTM - Finite-Difference |
Performs 2D Post-Stack Time Migration (PostSTM) using the Finite-Difference (FD) method based on the Claerbout one-way wave equation. The migration collapses diffraction hyperbolas and repositions dipping reflectors to their true subsurface positions in the time domain, using an interval velocity model to propagate the wavefield downward step by step through the earth.
Unlike the Kirchhoff method, the FD approach handles lateral velocity variations and steeply dipping events more accurately. It is particularly effective for data acquired along a straight 2D line where structural dips are moderate to steep and the subsurface velocity field varies laterally. Input data must be a post-stack 2D seismic section together with an interval velocity field, both referenced to a common datum.
The input post-stack 2D seismic section referenced to the processing datum. The data must be fully stacked (zero-offset equivalent) and corrected to a flat datum before migration is applied. Ensure that NMO correction and any necessary datum static corrections have already been applied upstream in the processing flow.
The interval velocity field used to drive the wavefield extrapolation. Must be an interval velocity model (not RMS velocities) defined at the same datum as the input stack and sampled along the 2D line. Interval velocities can be derived from RMS velocities using the Dix conversion formula, or supplied directly from a velocity analysis workflow.
Selects the finite-difference approximation order. The 15-degree approximation is the simplest and most stable, suitable for gently dipping geology. The 45-degree approximation is the most commonly used option, providing a good balance between accuracy and computational cost for reflectors up to ~45°. The 65-degree approximation migrates steeply dipping events accurately and is recommended for thrust belts, salt flanks, or other geologically complex areas.
The spatial sampling interval between consecutive CMP traces along the 2D line, in distance units (metres or feet). Must match the true CMP spacing of the input stacked data. An incorrect value will cause the migration aperture and dip accuracy to be wrong, producing under-migrated or over-migrated results.
The vertical step size (ms) used during downward wavefield extrapolation. A smaller step produces a more accurate migration result but increases computation time. A larger step is faster but may introduce numerical dispersion, particularly in zones of rapid velocity change. The step should not exceed the dominant period of the seismic data. Typical values: 2–8 ms.
Number of zero-amplitude traces padded at each end of the 2D line before migration to prevent edge-reflection artifacts during wavefield extrapolation. A minimum of 20–50 blank traces is recommended; increase this value when imaging steep dips or deep targets. Blank traces are trimmed from the output after migration.
A dimensionless scaling factor applied to the interval velocity model before migration. Values greater than 1.0 over-migrate; values less than 1.0 under-migrate. Useful for rapid migration velocity sensitivity tests without modifying the velocity model itself. For production migration, set to 1.0 unless a systematic velocity bias has been confirmed through analysis.