|
<< Click to Display Table of Contents >> Navigation: Velocity > Create Layers From Digit Velocity |
Note: This module is deprecated and is retained for legacy workflow compatibility only. It is not recommended for use in new processing projects.
Create Layers From Digit Velocity converts a digitized depth-velocity model into a set of discrete geological layer boundaries. The module scans the input model for significant velocity contrasts between adjacent depth samples, identifies locations where the velocity changes abruptly, and groups those change-points laterally into continuous reflector segments (layers).
Once raw layer boundaries are detected, the module applies a moving-average depth smoothing pass to reduce digitizing noise along each boundary. A derivative-based calculation then characterizes the dip angle and curvature of each smoothed horizon, producing auxiliary diagnostic point clouds that can be displayed in the section viewer. Only layers that span a minimum number of traces are retained, preventing short isolated features from being promoted to full layer objects.
This module is typically applied after digitizing an interval-velocity or RMS-velocity model, when the interpreter wishes to extract explicit horizon objects for use in depth-domain modeling workflows.
The digitized depth-velocity model from which layer boundaries will be extracted. This is a 2D data gather whose axes represent trace position (horizontal) and depth (vertical), with sample values holding velocity at each grid node. The model is typically produced by digitizing a velocity field or by importing an ASCII velocity grid. Both axes must be populated with valid depth coordinates before running this module.
The minimum velocity difference (in the native units of the input model) between two vertically adjacent samples that is treated as a genuine layer boundary. Sample pairs whose velocity difference is smaller than this threshold are ignored; only pairs exceeding the threshold produce a boundary spike that seeds a new layer. The default value is 8.
Reduce this value to detect weaker, subtler velocity contrasts and extract more layer boundaries. Increase it to suppress noise-driven boundaries and retain only the strongest, most significant velocity interfaces. Setting the value too low will generate a large number of spurious thin layers; setting it too high will merge distinct geological boundaries into a single layer.
The number of points used in the moving-average window that smooths the depth coordinate of each detected layer boundary. After the raw boundary is traced laterally through the model, this smoothing pass averages the depth values over the specified window length to remove high-frequency digitizing jitter. The default is 10 points; the minimum accepted value is 1 (no smoothing).
Increase this value to produce smoother, more geologically plausible horizon shapes. Use a smaller window when the velocity model itself is already smooth, or when you need to preserve sharp lateral depth variations such as fault steps.
The minimum number of trace positions that a detected boundary segment must span in order to be retained as a valid layer. Boundary segments shorter than this count are discarded after the lateral tracing step. The default is 50 points; the minimum accepted value is 1.
This parameter acts as a lateral continuity filter. Increase it to suppress small discontinuous patches and retain only regionally continuous reflectors. Decrease it when working on a small model area or when you need to capture shorter, locally significant boundaries such as those associated with channel features or small faults.
The window length (in points) used to calculate first- and second-order derivatives along each smoothed layer boundary. These derivatives are used to compute the local dip angle and radius of curvature at every point on the boundary, which are written to the Angle Points and Radius Points diagnostic outputs. The default is 10 points; the minimum accepted value is 1.
A larger window produces smoother derivative curves and more stable curvature estimates, but may blur abrupt dip changes. Use a smaller window when you need to resolve rapid lateral variations in reflector dip or curvature.
A collection of unsmoothed layer boundary objects extracted directly from the velocity contrast detection step. Each layer object contains the list of (X, Y, Z) positions of the boundary points and is identified by a sequential layer number. This output preserves the original detected boundary shapes before any depth smoothing is applied.
A collection of smoothed and derivative-annotated layer boundary objects. Each boundary has been subjected to moving-average depth smoothing (controlled by the Window for layer smooth parameter) and is enriched with dip angle and curvature values computed at every point. This is the primary layer output for use in downstream depth modeling.
A depth-domain gather in which non-zero values mark the locations of detected velocity contrasts before lateral grouping. The amplitude at each non-zero sample is proportional to the normalized velocity difference (reflection coefficient) across the boundary. This output is useful for quality-controlling the sensitivity threshold: a display of this gather shows exactly which depth-trace positions triggered boundary detection.
A depth-domain gather in which each detected raw layer is painted with a unique amplitude value, allowing all layers to be visualized simultaneously as a color-coded section. Points belonging to the same layer share the same amplitude, making it straightforward to assess layer connectivity and coverage across the model.
The same color-coded layer section as the Layers Gather, but built from the smoothed boundary positions. Comparing this gather against the Layers Gather provides a visual check on how much the smoothing step has altered the raw boundary geometry.