|
<< Click to Display Table of Contents >> Navigation: Migration > Kirchhoff PreSTM - gather in/out - migration |
The Kirchhoff PreSTM - gather in/out module performs 2D or 3D Kirchhoff pre-stack time migration (PreSTM) on CMP gathers, writing both the migrated gathers and the migrated stack to the output. The migration is based on Kirchhoff integral theory: each sample in the input data is treated as the apex of a diffraction hyperbola described by the RMS velocity model, and the algorithm stacks contributions from the appropriate input locations to collapse diffractions and move dipping reflectors to their correct positions. All data is read from and written to RAM during processing.
This module is suitable for final or iterative PreSTM workflows where both migrated gathers (for subsequent velocity analysis or AVO work) and the migrated stack are required simultaneously. Use the Kirchhoff PreSTM file in/out variant when the data volume is too large to hold in RAM.
Link to the input seismic gather data (CMP gathers) stored in RAM. The input data must be pre-processed (NMO-corrected statics applied, noise attenuated) before migration. For 3D surveys the data must be regularized.
Link to the RMS velocity model. This model defines the diffraction traveltimes used to construct the Kirchhoff operator. A smooth, spatially interpolated velocity field produces the best migration results.
Optional link to virtual trace headers providing the output CMP geometry for the migrated result. If not connected, the module derives the output geometry automatically from the sorted headers of the input data.
Selects whether the migration is run in 2D or 3D mode. For 2D surveys only in-line traveltimes are computed. For 3D surveys the full 3D traveltime cone is used, requiring regularized input data.
Maximum migration aperture (m). Defines the lateral radius of the diffraction operator. Larger values improve imaging of steeply dipping events but increase computation time. Set this to at least the expected maximum lateral displacement of reflectors at the target depth.
Controls whether the aperture is constant at all times or varies with two-way time. Constant uses the single Aperture value for all times. Time variant scales the aperture as a function of time, which is physically more accurate for shallow events that require a smaller aperture than deep events.
Taper width (m) along the edges of the survey. Within this distance from the survey boundary, the migration aperture is linearly ramped from zero to its full value. This reduces migration operator artefacts (smiles) caused by the truncation of the diffraction operator at the survey edge.
When enabled, a mute is applied to the migration operator based on the ratio of output to input time (stretching factor). This suppresses high-amplitude far-offset contributions where the Kirchhoff operator is strongly stretched, reducing migration noise at shallow times and wide offsets.
Percentage stretch at which the migration operator is muted. Contributions where the traveltime ratio exceeds this value are zeroed. Active only when Use stretching factor is enabled. Typical values range from 30% to 60%.
Near-surface replacement velocity (m/s) used to account for topographic elevation differences between shots and receivers. This is a required parameter; set it to the average velocity used during datum correction or to the shallow weathering velocity.
Scalar multiplier applied to the entire velocity model before migration. A value of 1.0 uses the input model as-is. Values greater than 1.0 increase effective velocities (useful for testing over-migration), while values less than 1.0 decrease them.
Strength of the anti-aliasing filter applied to the migration operator. Higher values produce a stronger low-pass filter that prevents spatial aliasing of the migration operator at far offsets. Increasing this coefficient reduces high-frequency noise at the cost of some bandwidth at large offsets. Typical values range from 0.5 to 1.0.
Maximum frequency (Hz) passed to the RHO filter applied as part of the Kirchhoff integral. This acts as an output bandwidth limit; set it to the highest usable frequency in your data to avoid passing migration noise at frequencies above the signal bandwidth.
Maximum propagation angle (degrees) of the migration operator relative to the vertical. Contributions from the diffraction cone beyond this angle are excluded. Reducing this value limits the operator to near-vertical energy and suppresses migration noise from steeply dipping events outside the angle of interest.
Output bin size (m) along the inline direction. Set this to the nominal CMP spacing of the survey. This value is used to define the output image grid for the migrated result.
Output bin size (m) along the crossline direction. For 2D data this is not applicable. For 3D data set this to the nominal CMP spacing in the crossline direction.
Performance boost factor that reduces computation time by selectively skipping less-significant contributions to the migration sum. Higher values produce faster results with a small reduction in image accuracy. A value of 1 gives full-quality migration. Use higher values for QC runs and set to 1 for final production migration.
Selects whether the module runs on the CPU or on an available GPU. GPU execution can significantly reduce migration run time for large 3D surveys.
Number of CPU threads used for parallel migration. Set to the number of physical cores available for best performance.